Fuels and Combustion Marion L. Smith, Karl W. Stinson.

»*OPERTT Of
ARTE '.VERITAS
FUELS AND COMBUSTION
Marion L. Smith
Assistant Professor of Mechanical Engineering
The Ohio State University
Karl W. Stinson
Professor of Mechanical Engineering
The Ohio State University
FIRST EDITION
New York Toronto London

McGRAW-HILL BOOK COMPANY, INC.
1952
139
Copyright, 1952, by the McGraw-Hill Book Company, Inc. Printed in the
United States of America. All rights reserved. This book, or parts thereof,
may not be reproduced in any form without permission of the publishers.
Library of Congress Catalog Card Number: 51-12949

East Engto-
Library
TP
3»3
THE MAPLE PRESS COMPANY, YORK, PA.

PREFACE
It has been the observation of the authors and their colleagues that the
general treatment of fuels and combustion is many times slighted in engi
neering curricula, even though the combustion process is a vital link in
power cycles. Frequently the topics of fuels and combustion are pre
sented, in part, through courses in steam power, internal-combustion
engines, heating and ventilating, and laboratory, with little coordination
or continuity of the material presented in the various courses. The
result many times is an incomplete coverage of either fuel technology or
combustion, and usually a meager concept is obtained of the actual
chemical processes of burning.
This textbook has been written to present to undergraduate engineer
ing students and to engineers fundamental and factual information con
cerning solid, liquid, and gaseous fuels and the problems associated with
their combustion. The general topics of fuel technology and the rela
tionships of air, fuel, combustion products, and the heat released, are
presented, the better to understand the problems involved in the applica
tion of combustion to furnaces, heaters, oil and gas burners, reciprocating
internal-combustion engines, gas turbines, and rockets. Fundamental
concepts of modern theories of oxidation are covered in such a manner
that they are within the grasp of students who are not chemistry majors.
While the book was developed for use in a fuels and combustion course
which would follow fundamental thermodynamics and precede the appli
cation courses of steam power, internal-combustion engines, and heating
and ventilating, it is so written that it can be used as a text at other levels
by omitting certain sections.
Since many aspects of this subject have been studied intensively for
many years, the authors have made liberal use of and reference to the
widely diversified literature in the field in their effort to present a modern,
concise, and yet fairly comprehensive treatment of fuels and combus
tion. The authors wish to express their sincere thanks to the many
writers and manufacturers referred to in the text for their contributions.
Special recognition is extended to Dr. C. E. Boord, Dr. K. W. Greenlee,
John Corsiglia, Prof. F. W. Marquis, Prof. Paul Bucher, and John Apple-
gate for their timely advice and criticism during the preparation of the
manuscript. The cooperation and suggestions offered by Prof. A. I.
Brown and other members of the Department of Mechanical Engineer
ing of The Ohio State University are gratefully acknowledged.
Columbus, Ohio Marion L. Smith
May, 1952 Karl W. Stinson
CONTENTS
Preface v
1. Fuels 1
2. Stoichiometric Analysis 38
3. Thermochemical Analysis 69
4. The Process of Combustion 114
5. Physical Properties of Fuels 166
6. Gas and Oil Burners 197
7. Coal-burning Equipment 236
8. Combustion in Engines 288
Index 331
vii
CHAPTER 1
FUELS
As an introduction to a study of fuels and their combustion, it is only

proper that consideration be given to the amounts and types of fuels
which are consumed in this country each year. It is interesting also to
have some conception of the reserves of those natural resources known as
fuels, so that a reasonable estimate may be made of the supply available
for future generations.
FUEL RESERVES AND USE OF ENERGY IN THE UNITED STATES
The fuel situation is a topic which has received a great deal of study
from the U.S. Bureau of Mines, the various oil companies, private
research organizations, and other interested parties. The extent of the
national fuel reserves, because of the very nature of the problem, cannot
be estimated with very great accuracy. Yet, it is desirable to have some
knowledge of the available fuels and their relative costs.
Figure 1-1 shows the quantities of energy consumed from the various
mineral fuels and water power in the United States from 1889 through
1949. This includes energy used for all purposes such as the generation
of electric power, and power for industry, the heating of homes, driving
automobiles, trains, and airplanes. It can be observed that the use of
energy has increased about fivefold since 1900 and has approximately
doubled in the past 30 years.
The estimated mineral fuel reserves of the United States are given in
Table 1-1. This shows the actual reserve and rate of consumption as well
as the equivalent reserve and rate of consumption in terms of 13,000 Btu
per pound bituminous coal.
The last column shows an estimate of the probable number of years the
supply of the mineral fuels now in use will be available, assuming that the
rate of annual consumption remains constant. In summary, the table
shows that the supply of coal will last approximately 1800 years, oil from
petroleum about 12 years, and natural gas about 30 years. A graphical
comparison of the yearly fuel consumption with the known reserves as of
Jan. 1, 1948, is shown in Fig. 1-2.
The reserves of coal have been established rather accurately for a num
ber of years and are probably fairly stable and reliable. On the other
1
40,000
35,000 A
A A
30,000
n ital en ergy
/
w
fine luc ting wc ter po 1

at pi ailing fuel ec, uivaler
2 25,000
rA
N
m
J
J / J>> J
CO
\
•
V
20,000
V u
/ n r\
15,000 A. t» 'tuminc us coc if
>M
N
/
and li jnife
10,000
5,000
/ r
Pe troleun u total c
/
— —
ry .J
An fhratii -yWa) fir 0Jtural gas^
poi>
1899 1904 1909

-m-r — Wit TTT
1913 'BÔ 1925 1930
tiit 0TT7TTTT
1935 1940
I I
1945
1111
1950
Fig. 1-1. Annual consumption of energy in the United States, 1899 to 1949. Energy from
water power computed at prevailing central-station equivalent Btu per kilowatthour.
(" Bituminous Coal and Lignite in 1949," U.S. Bureau of Mines, October, 1950.)
Petroleum 0.4% Oil shale 1.6%
0Anthracite coal 4% Anthracite coal 0.5%^ Natural gas 0.5%
SOURCES OF 1949 PROPORTIONS OF ESTIMATED

FUEL CONSUMPTION FUEL RESERVES, 1948
Fig. 1-2. Comparison of annual fuel consumption of various fuels with known fuel reserves
in United States.
FUELS 3
hand, the figures on the reserves of pe'troleum and natural gas include
only those oil fields which have been proved. New oil fields have been
brought in more rapidly than petroleum has been used in the past. Con
sequently, while the use of petroleum has more than tripled during the
past 30 years, the proved crude-oil reserves are greater now than in 1920.
Table 1-1. Estimated Mineral-fuel Reserves of the United States, Jan. 1,

1948*
Equivalent billions of net

tons bituminous coal, Reserve
Assume d 13,000 Btu/lbf divided
Reserve,
annual by
Description Jan. 1,
produc Fraction Assumed annual
1948 Reserve,
tion of total annual produc
Jan. 1,
fuel re produc tion
1948
serves , % tion
Coal, billions of net tons:

Anthracite 7.5 0.060 7.3 0.6 0.06 162
Bituminous 700 0.600 728 55.6 0.62 895
Subbituminous 409 0.014 299 22.8 0.01
Lignite 470 0.004 242 18.5 0.002
Total 1586. 5J 0.678 1276 97.5 0.69 1850
Petroleum, billions of bbl. . 24.7 2.0 5.7 0.4 0.46 12.7
Oil from oil shale, billions
of bbl 92 % 2.0 21.2 1.6 (0.46) 46.1
Natural gas, trillions of
cu ft 165.9 5.6 6.4 0.5 0.22 29.1
Total, all fuels 1309 1.37 955
* Carl E. Miller, The General Fuel Situation, Combustion, October, 1948.

t Heating values used :
Anthracite coal 12,900 Btu/lb
Bituminous coal 13,500 Btu/lb
Subbituminous coal 9 , 500Btu/lb
Petroleum 6,000,000 Btu/bbl
Lignite 6,700 Btu/lb
Natural gas 1 ,000 Btu/cu ft
X Data from Dr. A. C. Fieldner, Chief, Fuels and Explosives Branch, U.S. Bureau of Mines, corrected
to Jan. 1, 1948.
Figure depicts the increase in proved reserves of petroleum as com

1-3
pared to the increased annual consumption. From this comparison, it is
considered quite likely that the supply of oil from petroleum will be ade
quate for the next 20 to 25 years at least, and perhaps for a much longer
period. Likewise, new gas fields are discovered each year, and it is con
ceivable that the natural-gas supply will be adequate for the next 50 years.
Apparently it is reasonable to assume that a substantial supply of the
solid, liquid, and gaseous fuels used today will exist for a good many
4 FUELS AND COMBUSTION
years, especially if synthetic liquid fuels from coal and oil shale are used
to supplement the supply of petroleum.
The use of foreign oil might also aid in conserving the petroleum
reserves of this country. However, it has been estimated that the United
States has about 35 per cent of the proved world reserve of crude oil, thus
no substantial help from foreign fields is completely dependable, espe
cially in a time of world crisis when the drain on petroleum reserves is
Production
2
0 i
2
10
12
14
16
no ie
20
Proved reserves
22
24
26
1918 '20 '24 '28 '32 '36 '40 '44 •48
Year
Fig. 1-3. United States production
of crude petroleum compared to proved reserves.
("Petroleum Data Book, 1947," and Mineral Industry Surveys, U.S. Bureau of Mines.)
greatest. It is interesting to observe that the United States also enjoys

the enviable position of possessing about 40 to 50 per cent of the world's
coal supply.
Figure 1-4 shows the power output by different fuels from all private
and public electric utilities from 1920 through 1950. It can be seen that
water power is used to generate about 30 per cent of the power produced
by electric utilities. However, the energy produced by hydroelectric
stations represents less than 5 per cent of the total amount of energy
used yearly for all purposes in the United States, as is shown in Fig. 1-1.
The principal users of bituminous coal and the relative amounts they
consumed in 1950 appear in Table 1-2. In contrast to the diversified
market for bituminous coal, anthracite is used for space heating almost
entirely. Table 1-3 shows the percentage yields of the major petroleum
FUELS 5
Table 1-2. Bituminous Coal Consumption by Users, 1950*
Millions of Per cent of

Consumer
net tons total
Railroads 65 0 14 3
Coke, gas, and chemical manufacturers 103 3 22 8
7 7 1 7
Cement mills 7 9 1 7
94 2 20 8
86 6 19 1
Electric-power utilities 88 3 19 4
Others 0 8 0 2
Total 453 8 100.0
* Bituminous Coal Institute.
Table 1-3. Percentage Yields from January, 1951, Petroleum Production*

Product Per Cent
Gasolines 41 .0
Kerosene 6.2
Distillate fuel oil, for diesel fuels and oil burners 21 .7
Residual fuel oil, heavy fuel oils for diesels and oil burners 20 . 8
Lubricating oil 2.5
Miscellaneous 7.8
* Mineral Industry Surveys, U.S. Bur. Mines, Monthly Petroleum Statement 336.
Table 1-4. Natural-gas Consumption by Users in 1949*

Consumer Per Cent
Residential 19.1
Commercial 6.7
Refineries 8.1
Industrial 45.7
Field 20.4
• Mineral Industry Surveys, U.S. Bur. Mines, Mineral Market Rept. MMS 1930.
products from the January, 1951, refinery production in the United

States. By far the greatest fraction of all petroleum production is made
into gasoline.
The 1949 consumption of natural gas by users is shown in Table 1-4.
One-fifth of the gas coming from the wells was used in the fields to provide
the heat and power needed for different operations there.
From Fig. 1-2, it is obvious that this country is using its better grades
of coal almost exclusively. In addition, the fractions of oil and gas con
sumed far exceed their relative supply. Thus, ultimately, it will be
necessary to design equipment to handle the poorer grades of fuel, and it
may become necessary to develop methods for producing new supplies of
liquid and gaseous fuels synthetically, if fuels are to be used in these con
venient forms. The abundance of coal and oil shale as compared to the
proved reserve of petroleum explains why so much interest has been
shown recently in the development of synthetic liquid and gaseous fuels.

With the coming of atomic energy, it is possible that within a few years
much energy will be supplied from fissionable materials. Present devel
opments are reported to be along the lines of applying atomic power to
1920 22 24 26 28 1930 32 34 36 38 1940 42 44 46 48 1950

Fig. 1-4. Consumption of fuel for the production of electric energy. (Courtesy of Federal
Power Commission.)
the generation of electricity in large central stations and of providing

motive power for marine vessels and guided missiles. Experts in the
field are predicting that atomic piles will not be producing any appreciable
quantity of central-station power soon.
Thus, barring an unforeseen revolutionary development in the atomic
field, it seems that the present fuels and types of combustion equipment
FUELS 7
will for a good many years to supply the bulk of the heat and
be used
power needed in this modern civilization.
SOLID FUELS
The solid fuels used in this country primarily are bituminous and
anthracite coal, coke, wood, and charcoal. Of these, the consumption of
coal far exceeds the combined demand for the others, and most of the
discussion here will be devoted to coal.
Origin of Coal. Geological studies prove definitely that coal is of plant
origin. During the initial period of coal formation, heavy vegetation in
swamps and bogs built up a thick layer of partially decayed vegetal
residue which was probably soon covered with water. Over a period of
many centuries the decayed matter accumulated to a depth of several
feet before it was buried by the violent earth upheavals, floods, and
glaciers of the formative period of this planet.
Then began the secondary transformation period wherein the decayed
vegetation was subjected to extreme temperatures and crushing pressures.
It is believed this second stage must have lasted several hundred thousand
years to transform the soggy swamp decay into the solid mineral known
today as coal.
In the course of the coalification, or coal-forming, process the moisture
and gaseous inclusions in the original decayed vegetation were slowly
excluded from the coal bed as a result of the pressures and high tempera
tures exerted by the earth. The extent to which coalification has pro
gressed determines the rank of a coal. The major ranks of coal from
lowest to highest, are
1. Lignite
2. Subbituminous
3. Bituminous
4. Semianthracite
5. Anthracite
There are no clear-cut dividing lines between coals of the various ranks,
and different physical properties are sometimes listed to help classify fuels
which are borderline cases between lignite and bituminous, or bituminous
and anthracite. The method recommended by the American Society for
Testing Materials (ASTM) and the American Society of Mechanical
Engineers (ASME) to classify coals according to rank appears in Chap. 5.
The accumulation of decayed vegetation first came to the stage where
it was known as peat. Peat is not a coal but represents the first stage in
the formation of coal. In the early stages of transformation, varying
amounts of water, methane, and carbon dioxide were gradually elimi-
8
nated. Further increases of pressure and temperature were then respon

sible for driving off much oxygen, volatile matter, and moisture as the
peat was changed to lignite, and finally to bituminous coal. A few coal
fields, principally in Pennsylvania, were evidently subjected to the coal-
forming processes to a greater extent than most other areas, and in the
coal seams there the volatile matter of the fuel was reduced still lower to
form the anthracite coals. The composition changes which occurred
during the coalification process are shown graphically in Fig. 1-5.
100 1 1 1 1 1 1 1 1
Wood Peat Lignite Subbi- Semi Anthracite

Bituminous
tuminous anthracite
Fig. 1-5. Progressive changes during the coalification process. Percentages given are on
the moisture, ash, and sulfur-free basis. (Adapted from " The Making, Shaping, and
Treating of Steel," Carnegie Steel Co., 1920.)
Production of Coal. In the preceding sections, the extent of the coal

reserves of the United States was discussed, but no mention was made of
the areas in which these reserves were located. Figure 1-6 shows the
states having the greatest coal reserves and also reveals the ranks of these
fuels. For comparison with the location of reserves, Table 1-5 gives the
production of coal by states in 1949. It should be noted the greatest
coal-producing states are Pennsylvania, West Virginia, Kentucky, and
Illinois, with Ohio and Indiana closely following, while the greatest
reserves are in Wyoming, North Dakota, Montana, and Colorado. This
lack of development of the western coal fields is due principally to the
facts that (1) the demand for coal is greatest in the highly industrial
East and (2) the rank of the eastern coals is considerably higher and the
quality is more adaptable to present combustion equipment.
Although most of the coal produced comes from underground mines, an
increasing proportion is being mined yearly by strip mining. Table 1-5
also presents the extent of this type of mining in the various states in 1949.
Coal veins in the United States vary from several inches to 40 ft in thick
ness, but the average coal seam is only about 4 ft thick. The thickness of
the vein and its depth below ground are two of the major economic factors
FUELS 9
Wyoming
N.Dakota
Colorado
Montana
Illinois
W Virginia
Kentucky
Pennsylvania
Ohio
Utah
Missouri
Alabama
Washington
Oklahoma
Indiana
Texas
Towa
Kansas
Tennessee
Virginia
Maryland
New Mexico
Others
150 200 250 300
Billions of tons
Fig. 1-6. Coal reserves by states, 1948.
Table 1-5. Coal Production by States, 1949*
Millions of net tons

Rank of Per cent of U.S.
State
coal production
All mining Strip mining
Pennsylvania Anthr. 42 7
Bitum. 89 2
131 9 22 2 27.4
West Virginia Bitum. 122 6 13 8 25.5
Kentucky Bitum. 62 6 10 4 13.0
Illinois Bitum. 47 2 13 9 9.8
Ohio Bitum. 31 0 18 3 6.5
Bitum. 16 6 9 7 3.5
Virginia Bitum. 14 6 1 1 3.0
Bitum. 12 9 1 8 2.7
Utah Bitum. 6 2 0 0 1.3
Wyoming Bitum. 6 0 1 1 1.2
Colorado Bitum. 4 6 0 3 1.0
Bitum. 4 2 0 5 0.9
Others Bitum. 20 2 13 0 4.2
480 6 106 ~0~ 100.0
* Mineral Industry Surveys, Bituminous Coal and Lignite in 1949, U.S. Bur. Mines, October, 1950.
which determine what type of mining will be most advantageous in a

particular location.
All mining has become more mechanized each year. As of 1949, 67
per cent of all coal mined underground was loaded mechanically, and 91
per cent was cut mechanically from the vein. Modern mining machines
such as the one shown in Fig. 1-7 are used extensively to cut the coal from
the vein and load it automatically on rail cars in the mine. Such mining
methods helped to increase the production per man-day, so that in 1949
the national average coal production was 6.43 tons per miner per day, the
Fig. 1-7. The Colmol at work. This modern machine mines and loads coal in one oper
ation without the use of explosives. {Courtesy of Jeffrey Mfg. Co.)
highest rate of production for any country in the world. This figure is
still low as compared to the 15.33 tons per man-day produced by strip
mining, which gives a comparison as to the relative ease of mining by
stripping and other means.
An increasing percentage of coal is being cleaned and sized at the mine.
In the most modern mining operations, the coal is conveyed to a coal
tipple at the mine entrance. Here the large lumps are crushed, and the
entire production of coal is screened to yield products of a uniform size.
Much of the coal is washed and cleaned to remove as much of the rock,
sulfur, and other mineral content as possible. It is then loaded on rail
cars or trucks for the consumer.
Types of Coal. Coals may be classified as banded or nonbanded,
according to the type of their petrographic, or rock, structure. Banded
FUELS 11
coal is not a homogeneous substance, but consists of alternate layers or

bands of bright-black, dull-black, and gray metamorphosed vegetal
material. Banded coals exist in all ranks of coal from lignite to anthra
cite. The appearance of the bands is attributed to different kinds of
wood and plant substances in various stages of decay.
The U.S. Bureau of Mines further classifies the banded coals as (1)
bright, (2) semisplint, and (3) splint coals, depending upon the pre
dominant plant residues of which they are composed. Figure 1-8 shows
a block of bright, banded, bituminous coal. The extent and appearance
of the bands are evident. The nonbanded coals are the cannel or bog-
Fig. 1-8. A bright, banded, bituminous coal. (Courtesy of U.S. Bureau of Mines, Depart
ment of the Interior.)
head types. Figure 1-9 is a picture of a lump of cannel coal. It is uni

form and compact in structure with little or no evidence of bands. Both
cannel and boghead coals break with sharp clean fractures as shown and
are greasy black in appearance. They are high in volatile matter and
produce high yields of tar and light oils upon distillation.
Composition of Solid Fuels. Solid fuels are composed of carbon and
hydrogen, with various amounts of oxygen, nitrogen, and mineral matter
included. The chemical constituents of coals, coke, peat, etc., are
determined by the proximate analysis and the ultimate analysis, and typical
analyses and heating values of various solid fuels are listed in Table 1-6.
These analyses are reported on the basis of the fuels as received at the
laboratory and the heating values given are for both moist and dry fuels.
Ultimate Analysis. The ultimate analysis of a coal is a precise chemical

determination in which the six basic components of the fuel — carbon,
hydrogen, oxygen, nitrogen, sulfur, and ash — are listed in per cent by
weight. The percentages of hydrogen and carbon are found by burning a
0.2-g sample in a strongly oxidizing atmosphere and weighing the quan
tities of carbon dioxide and water vapor emitted as the products of com
bustion of the sample. The carbon dioxide is absorbed in a solution of
Fig. 1-9. A cannel coal. (Courtesy of U.S. Bureau of Mines, Department of the Interior.)
potassium hydroxide for weighing, while the water vapor is collected in a

calcium chloride tube.
The percentage of ash in the coal is found by heating a 1-g sample to
approximately 1340 F. In the presence of oxygen, this high temperature
assures the complete burning of all combustible in the fuel, and the
residue is ash.
The percentages of nitrogen and sulfur are found by complex chemical
analyses on different small samples of the fuel.The percentage of oxygen
is computed as the difference between the sum of the percentages of the
other five items and 100. Thus, the percentage of oxygen, as listed in
the ultimate analysis, is subject to the accumulative errors of the other
determinations.
Proximate Analysis. The proximate analysis is a simpler test which
shows a few important components of a coal sample without resorting to
FUELS 13
oooo ooooooooooooooo
HCOCOO 00H^00NCNotOll0 00CNC4MO00
00 CS 00 H 0*-*-*CO-*COmNCOCONNNHn
o o
O rH
ooooooooooo
HOCOlOOSCOOOCO'tiOlO
o o o
02 o o
CO CN CO CN CO
< I
o NONoOotOoOlNeiONOO
doe*
•0>CO CN
tOOONHliJt-WôoS03NollOCO
0* r- 00OoMOCNCOCOiOMO 0 o CO
© ©' nHHoOH^dddd -i-io
o a rH CO N U3 U3 "3 o CO 00 rH oo co 03
I a o o rH i-t rH i-l rH o O
1>J LI
rH rH rH rH © 1—
1
~5 to
0
C3
CO i-H © co CO X CN rH o "* © o> t~ o o tN CN
I
to CO a: CO U3 U3 »o c CO 03 CN oo O
o4
5x
CN
o o CO N OiCOoOONCOOoNCOiOOOOWCO
tjîio-jlioiOiOtOCNCOCOCOON«
Sf
rr? CO t>- 03 CO
N tO H NCOrHOJM'ÔOOOt^lNiONOOOo
OS O o
lO
CN CO CN
lO CN
OJOC0O100O00CO
NOONNCOtOIOM
o
00
O CO 113CO 0*
0* 00 CO 00
0 © CN CNcONooOCOOCtiMO © ©
0 co tOOONHCONW-^COCSOJ O
cd
00 CO
6S
• 0 CN oo o 00 O! 03 N rH 03 03 t~ • rH oo 03
carb on
r*> , rH
o rH t- CN 00 00 OS iO 00 tO 0 t- to rH
g
a "3
N
Fix.
0 rH co CO CO K3 •* co oo tO 0 oo CO
, • © CN 10 CN 00 00 0 rH CN U3 rH co 0 N rH
.
•
. CO oo
0 CN CN © tO rH CO 113CO O O CO 00
rH
0 55
CN rH CN CN CO CO -T CO 06
0 00 © O
1
rH 00 rH cn rH CO co rH
•
00 00 CO
0
CO QJ
•
0 © CN CN
cd
cd
cd
. CO CO (N 00 CN 03 CN CO CO
!
, 0 us co rH rH rH
*<3
o
oj
f '
03
*1
>
Koa
Ph Ph
r?3
o
eq"
"iS
^
!> <
<!
Ph
CO
b
b
>
>
>
rg
i
cd
g
g
1
1
0°
s
44
* SrSssHHHtS
CO u
-9
5
"8
I5
&-a
5
O O «
8g
£S
Ph «
<?
,9
g
CO
» co P5 U
rS
»3
Ph PQ
the accurate chemical analysis required for the ultimate. It lists the four
items: moisture, volatile matter, fixed carbon, and ash in per cent by
weight. The percentage of moisture is found by heating a 1-g sample of
the fuel to about 225 F to evaporate the water and by measuring the loss
in weight of the original sample. By heating another 1-g sample in an
airtight crucible for 7 min at about 1750 F, the volatile matter and the
moisture are expelled from the sample without igniting the fixed carbon.
The percentage of volatile matter is the loss in weight upon heating the
sample minus the percentage of moisture already found. The percentage
of ash is measured in the same manner as was explained for the ultimate
analysis. Obviously, the same percentage of ash should be reported in
the proximate as in the ultimate analysis. The remaining item, fixed
carbon, is computed as 100 minus the percentages of moisture, volatile
matter, and ash.
The heating value of the fuel and the percentage of sulfur are often
given when reporting a proximate analysis of a coal in addition to the
above four items which total 100 per cent. The heating value of the fuel
is determined by burning a 1-g sample in a bomb calorimeter with a large
excess of oxygen to ensure complete combustion, as discussed in Chap. 3.
From the heat emitted following the combustion process, the heating
value, or calorific value, of the fuel can be computed. Under the condi
tions existing in the bomb calorimeter, the sulfur in the fuel burns and
forms sulfuric acid upon uniting with the water vapor. The amount of
sulfur in the original fuel sample can be determined accurately by measur
ing the quantity of sulfuric acid washed out of the bomb calorimeter
following the combustion process. If the percentage of sulfur were to be
included in the proximate, it would be accounted for twice, since it already
has been reported, part being included in the volatile matter and part in
the ash.
The moisture as determined for the proximate analysis consists only of
the free moisture, or surface moisture, present with the solid fuel. Addi
tional hydrogen and oxygen may be contained in the fuel even after the
free moisture is driven off, but they probably are not united to form
water. The volatile matter consists of all gaseous products which can
be expelled by heating the fuel. Fixed carbon is that carbon in the solid
state which remains once the volatile matter and moisture have been
removed. It does not represent all of the carbon in the coal, as listed in
the ultimate analysis, because volatile matter contains considerable
carbon in the form of complex hydrocarbon compounds.
PETROLEUM FUELS
Origin of Petroleum. Of the several theories advanced concerning the
origin of petroleum, the theory of vegetal origin is the most widely
FUELS 15
accepted today. It
holds that enormous quantities of gelatinous vegeta
tion accumulated in clays and sands along seacoasts or lakes. Shifting
currents then buried the organic material in layers of sediment which pro
tected it from the normal processes of decay. Subsequent changes in the
earth structure eventually subjected the entrapped vegetal matter to
enormously high pressures and temperatures which brought about a type
of semidestructive distillation. This along with other metamorphic proc
esses resulted in the formation of petroleum.
Geologic studies indicate that petroleum probably was not formed in
the pools where it is found today. As mentioned above, the deposits of
crude petroleum were undoubtedly formed near seashores, but the action
Fig. 1-10. Typical oil-pool formations. In the center, pool water has forced the petroleum
and gas up into a dome, or anticline, in the porous oil rock layer. Hard cap rock prevents
the escape of the oil. A fault, at the left, or a stratigraphic trap, as shown at the right,
similarly traps oil pools between water under high pressure and hard rock strata.
of the surrounding water gradually shifted the location of the oil pool.
After of many centuries, the oil was forced through layers of
a period
porous rock strata until it finally became entrapped under a dome capped
by hard rock. This made it impossible for the gas or oil to escape, and
the water kept the pool of oil under pressure. A sketch of typical oil
rock formations is shown in Fig. 1-10.
Production of Crude Oil. In drilling for oil, the oil-retaining cap rock
above the deposit is penetrated, and the existing pressure forces the oil,
gas, and some water to the surface. An attempt is made to locate the
bottom of the well at the richest part of the oil-bearing rock. This
procedure makes it possible to keep out much of the water which would
otherwise come up with the oil and gas. As the petroleum is removed
from the underground deposit, the water moves in to take its place.
Eventually the natural pressure decreases to the point where it will no

longer force the oil and gas to the surface. When this occurs, gas or
water may be pumped back into the well to increase the pressure and
force the crude oil from the pool.
Under the natural pressure of the well, which may run to 1000 psi or
higher, great quantities of hydrocarbon vapors are absorbed in the liquid
petroleum and much liquid is held in suspension in the gases. As the
crude oil and gas are pumped from the well, they are passed through a
separator, or trap, where the liquid and the gases are separated. After a
preliminary purification process in a field treating plant, the crude oil is
pumped through pipe lines to the refinery. This treating plant reduces
the dirt and water in the crude to less than 3 per cent so as to minimize
pipe-line troubles.
Fuel Chemistry. Any petroleum product is composed of a large num
ber of different hydrocarbon compounds, each of which has individual
properties and characteristics.
Fortunately, almost all the many hydrocarbons present in fuels may be
classified into one of five chemical groups: the paraffins, isoparaffins,
olefins, naphthenes, and aromatics. The chemical, physical, and com
bustion characteristics of a compound tend to be similar to those for the
other compounds within the same group. Members of many other
organic chemistry groups may be present in fuels in small amounts while
others appear as intermediate products of combustion.
Before we discuss the properties of any of the groups, a few terms com
mon to the chemistry of fuels should be mentioned. Carbon atoms have
a valence of 4, while hydrogen atoms have a valence of 1. There is a
tendency for the carbon to pick up other atoms until all four of its valence
bonds are used, in which case the resulting compound is said to be
saturated. Sometimes a carbon atom does not have all the other atoms it
could hold, but two or more valence bonds are shared with another carbon
atom. In this case the compound is said to be unsaturated because it is
possible to add more atoms to the molecule.
Hydrocarbons are known as chain compounds or ring compounds
depending upon their molecular structure. Compounds of the same
chemical group that have the same number of carbon and hydrogen atoms
but different molecular structures are called isomers.
Normal Paraffins. Normal, or straight-chain, paraffins are stable,
saturated compounds having the general chemical formula CnH2n+2.
The molecular structures of three of the members of the normal paraffin
family are shown in Fig. 1-11.
Each compound in this group has a name ending in -one. The names
of some of the compounds in the series with their formulas and a few
FUELS 17
physical properties appear in Table 1-7. Methane has one carbon atom
in the molecule, formula CH4, and each member in the family has one
more carbon atom in the molecule than the compound preceding it. A
large portion of any hydrocarbon fuel is composed of normal paraffins.
H HHHHH HHHHHHHH I
I I I I I I I I I I I I I
H— C— H H— C— C—C— C— C— H H—C— C—C— C— C—C— C— C—H
H
I
Methane
HHHHH
I I I
Pentane
I I
HHHHHHHH
I I I I 1
Octane
I I I
Fig. 1-11. Molecular structures of three normal paraffins.
Isoparaffins. Any compound with four or more carbon atoms may

possess the same number of carbon and hydrogen atoms as a normal
paraffin, but have them arranged in a different manner from that shown
above. Thus octane, CsHi8, may occur as a straight chain or in the more
compact molecular structure shown in Fig. 1-12. The isoparaffins may
carry the same names as the related normal paraffins with the prefix iso-
being added, or they may be named by the number of carbon atoms
occurring in the straight chain and the number, location, and name of
attached groups. Thus, in Fig. 1-12, the 2, 2, 4 denotes that two methyl,
CH3, groups are attached to the second atom in the pentane straight
chain, and one methyl group is attached to the fourth. There are still 8
H— C— H
II H H H
00 C C C C C— H
ll H H
H— C—H H— C—H
I 1
H H
Fig. 1-12. Molecular structure of isooctane (2,2,4-trimethylpentane).
carbon and 18 hydrogen atoms present in the molecule, hence this is an

isooctane, an isomer of normal octane. It should be apparent there are
many ways in which the molecular structure could occur for the larger,
heavier molecules. Each isomer may differ only slightly from the others
in physical and chemical characteristics, such as boiling point, specific
gravity, or conductivity, but the combustion characteristics of the isomers
may vary widely. For example, 2,2,4-trimethylpentane (isooctane) is a
very smooth-burning fuel in a gasoline engine and has been chosen as the
standard for 100 octane gasoline. In contrast, normal octane is a rather
poor engine fuel and has a low octane rating.
CMOCO'f^COCONWHHHHHOOOfllOlOO SNNKNSS Tj)©lOi-t

Tf>
Tf COCOCO
I o o§ OO©OOOO'J"fCBCSiOiOiOMMO'*N'il © O © © OOQ
©©©©©5 © -
©©©
© ~
© u
©—c
mil
OONWNb.SCOCOOO000000©©iO>HOS iO © CO'O |
Ol COCOCO
©O000000CO«t*t*OOCMCMCMTjtTjtiOO)OcO l> b. N N N r»
iO if; if; itSto 'f; >O
COGOCOGO00COCO
O
© OSa 'O 1
lOCOCMCN
b-OSCDO -*CO
COOStOCO© t» Ot-OS cO—I cm oScmoo-**
>ocm co aSo iOiO CO if; cm CMb-© b-COOS
r'--'CM'-'tMCMCMCOCMCOîO© rH-HCM 0HCMCM — — CMCM
1 I 1+ I I +
tO00MOcOiO00«0i0'-i0S<Nb.C0>O00 © © CO
©b-tOb-iOi-îOOSb-^CNO'O-6 >-"l*b- CMCMiO © CMiH © iOCOCMOS CMOSCo
OSOSO.-iU30iOcO,<fCO'-it-CO©©WCM'*©b. N© cmcoîraco
NNCONNclNi-iHH hh i-> CMCOCMCMCMrH
I I I I I I I 1 I 11 M I I I I + I I I I I 1+ l + 1 + i i
......
...
iSeOCM"* O co —i
3OSCMCMCO© b- 00
0U3iO>O©©©©©©t>.b-©I>b-b-b-t>tN.
>OCM«3 CM>QOSiO
Mb- CO O300OS© I
000000
0©©©©©©©©©©0©0©©©©0 0oooooo
©0000b-OS©OSCM^tD'-f00>OOSC0©>OU) COCOiOCM©CM
MS*
3
www w,ww
oooooo dwo666o6c5c5666(5 66ooo66 666 666 o666
s s I
aj to d
ill
C ci c x"c
a>- .
aj5 !
111!
£ o £
s!l
o>u cj
o O-T
'2 H e.2 i« c« s4n sn" s e e e OO&

o i
S
FUELS 19
Olefins.Olefins are unsaturated paraffin compounds. The maximum

amount of hydrogen is not present, which causes some carbon atoms to
share two or more valence bonds. The members of this family have
names similar to the paraffins except the suffix -ene is used. The struc
ture of pentene is shown in Fig. 1-13. The general formula for the mem-
H H H
H— C— C—C =C— C—H
I I I I I
H H H H H
Fig. 1-13. Molecular structure of the olefin, pentene.
bers of this group is CnH2n. Olefins are formed principally in the refining
process called cracking. As an example, when the heavy paraffin Ci6H34
is subjected to cracking, a reaction of the following nature may occur:
CieH34 — ► CsHi8 + CsHi6

Paraffin Paraffin Olefin
The and the still more unsaturated diolefins CnH2n-t, are

olefins,
unstable compounds and are thought to be the cause of much gum forma
tion in gasoline. However, since they possess very good combustion
characteristics, substantial quantities appear in modern high-octane
gasolines.
H
1
C
H— G/ \)—H
H-A
\ C
✓
A-H
Fig. 1-14. Molecular structure of the naph- Fig. 1-15. Molecular structure of the
thene, cyclopentane. aromatic, benzene.
Naphthenes. Naphthenes are saturated, stable compounds with a ring

structure. Figure 1-14 shows the cyclopentane molecular structure. As
can be seen from the figure, the naphthenes are saturated compounds
even though the general chemical formula is C„H2n, the same as for the
unsaturated olefins. This is because the ends of the chain are tied
together. This reduces the potential of the carbon for taking on hydro
gen. The names of the members of this family are the same as those of
the corresponding paraffin except the prefix cyclo- is added to denote the
ring structure of the naphthenes. The physical properties of this group
are similar to those of the normal paraffins, but the combustion properties
are more like those of the isoparaffins.
Aromatics. The aromatics are unsaturated but stable ring compounds.
Benzene, C6H6, is the most characteristic member of the group, and all
aromatics consist of some variation of the benzene ring shown in Fig. 1-15.
The general chemical formula for this group is C„H2n-6. An inspection
of the figure shows benzene to be unsaturated. Unlike the olefins, there
is no tendency for oxygen atoms to be picked up to take the place of the
double bond. This group has very desirable combustion characteristics
for use in spark-ignition engines, and members of the group are often
added to gasoline to produce high-octane fuels.
Composition of Crude Oils. Crude oil is a liquid consisting chiefly of
hydrocarbons with traces of sulfur, nitrogen, oxygen, and a few impurities
such as water and sediment. The chemical analyses of representative
petroleums from various oil fields are shown in Table 1-8. This shows
that the percentages of carbon and hydrogen present in the crudes from
all these sources are essentially the same. However, an analysis which
shows merely the percentage of each element in the crude oil is somewhat
misleading. When the crudes are analyzed on the basis of the hydro
carbon families present and the amounts of the different compounds
in those families, it is found that oils from the different sources vary
considerably.
Table 1-8. Typical Analyses of Representative Crude Oils
Per cent, by weight
Source
Impurities, including
Carbon Hydrogen
sulfur, oxygen, nitrogen
Pennsylvania 85 14 1
West Virginia 84 13 3
Ohio 84 13 3
California 82 10 8
85 11 4
Mexico 84 10 6
Russia, Baku 87 12 1
Petroleum from the Pennsylvania fields is composed principally of

paraffin hydrocarbons and yields what are known as "paraffin-base"
products. Crudes from the west coast area contain a larger percentage
of naphthenes than the Pennsylvania oils and hence are termed "naph-
thene-base," or sometimes "asphalt-base," crude oils. Even in west
coast crudes the largest percentage of compounds may belong in the
FUELS 21
paraffin family, but the number of naphthenes present is much greater

than in the Pennsylvania petroleums. Oils from the Texas-Oklahoma
area, commonly called mid-continent oils, are known as "intermediate-
base" crudes because their composition is between that of the paraffin-
base and the naphthene-base crudes. Crude oils from other fields
throughout the world are usually classified as belonging to one of these
three types.
Fig. 1-16. Flow diagram of a petroleum refinery. (After API.)
Refining of Petroleum. The refining of petroleum is one of the most

advanced of modern industrial techniques. In the fractional distillation
process, crude oil is separated into many parts, these fractions are treated
and further processed, and the final products of gasoline, fuel oil, etc.,
are obtained by blending proper amounts of the refined fractions.
Since petroleum consists of hydrogen and carbon compounds, refining
processes involve the separation of the light compounds from the heavy,
and the conversion of one hydrocarbon compound to another. The
refining technique has been so perfected that it is now possible to form
almost any product from any crude oil in the desired proportions,
although in many cases it would not be economical.
A few of the fundamental processes employed in modern refineries are
described in the paragraphs which follow. A simplified flow diagram
showing where each process fits into the over-all production is given in
Fig. 1-16.
Fractional Distillation. The first step in the refining of crude oil is the
fractional distillation process. Primary fractions, obtained from a frac
tionating tower, and the order in which they occur are shown diagram-
matically in Fig. 1-17. In this simplified representation of the process,
Vapor f~
-Light gases to
,
vapor recovery
Icondenser
|_ -Straight -run
gasoline
V. -Naphtha
-Kerosene
Crude oil
I -Light gas oil

Tube-type
evaporating
A
furnace -Medium gas oil
-Heavy gas oil
-Residue
Fig. 1-17. Straight-run fractions obtained from the fractional distillation process.
liquid petroleum is heated to a temperature of about 700 F in a tube

heater, which evaporates most of the crude oil. The vapors are passed
into the fractionating column where the oil is separated into several
fractions. The heavier compounds have a high boiling point and, as the
vapors rise in the tower and are cooled, the heavy fractions precipitate
first and are removed from the tower. The light hydrocarbons rise to
the top of the column before condensing, while compounds with inter
mediate boiling points are removed at the intervening stages. The
liquids obtained from the various stages of the distillation process are
known as straight-run, or primary, fractions. In the refinery, fractional
distillation is carried out in two or more towers, either at atmospheric
pressure or in a high vacuum.
Sometimes the straight-run fractions can be used as they are, but usu
ally they are improved by further treating. These fractions provide the
charging stock for the other refining processes where the nature of each
FUELS 23
fraction may be completely changed to increase the yield of a specific

product.
Process. A hydrocarbon fraction such as heavy gas oil can
Cracking
be broken down into a number of lighter fractions by a process known as
cracking. This process makes it possible to increase the yield of gasoline
or light fuel oil per barrel of crude. The thermal cracking process is
accomplished by subjecting the gas oil to temperatures of 1000 F or over
and pressures of 1000 to 1500 psi. Under these conditions, the heavy
hydrocarbon compounds are broken to form new compounds, some
lighter and some heavier than the originals. Many of the lighter com
pounds fall into the range for gasolines and light fuel oils. A few are too
light to condense at atmospheric temperature, and they form refinery
gases which must be further processed. The heavy residues are used to
form coke or some other by-product. Cracking may be achieved at
lower temperatures and pressures if a suitable catalyst is present. There
are a number of commercial catalytic cracking processes developed, using
various catalysts, which give a higher yield of premium gasoline from the
heavy fractions than the thermal method.
Polymerization. Light hydrocarbon gases resulting from the frac
tional distillation or cracking processes may be built up into heavier
compounds by a process of polymerization. Thus, many gaseous prod
ucts of the refinery are utilized which might otherwise be wasted. As an
example of the type of reaction followed here, many light compounds,
such as CH4, C2H4, C3H6, and C3H8, may be polymerized, or joined
together, to form heavier molecules such as C6Hi2, C6Hi4, and C7H14.
The initial and the final compounds may be members of the normal-
paraffin, isoparaffin, olefin, naphthene, or aromatic families, depending
upon the reactions followed. The products from the polymerization
reaction are controlled by regulating the temperature and pressure, and
by the catalyst used in the process.
Absorption Process. The gases leaving the oil well and also many
from the refinery processes contain heavier hydrocarbons in the vapor
state. A portion of these heavier hydrocarbons fall into the gasoline
range. The absorption process is used to recover these vapors. In this
process the gases come into contact with a kerosene or light oil which
absorbs the heavy hydrocarbon vapors but not the lighter hydrocarbons
making up the gas. The vapors are then removed from the absorbing
oil by heating the latter in a steam stripper and driving off the absorbed
vapors. In some cases, part of the heavy vapor is precipitated by flow
ing the natural gas past refrigerating coils before sending it through the
absorption process. If the natural gas is compressed at the well before
entering the pipe lines, much of the vapor may be precipitated at that
point. The liquid obtained from natural gas by precipitation, or by

absorption, is known as natural gasoline.
Propane and butane are removed from natural gas by an absorption
process and are bottled for commercial sale or blended with other petro
leum fractions to increase the volatility of those products.
Hydrogenation. The process hydrogenation may be described as a
cracking process which occurs in an atmosphere of hydrogen. Here cer-
Fig. 1-18. Refinery at Baton Rouge, La., showing three Kellogg catalytic crackers on the
left, an alkylation unit in the center, and the light ends plant on the right. (Courtesy of
Standard Oil Co., N. J.; photo by Libsohn.)
tain unsaturated compounds pick up more hydrogen to become saturated,

and the yield and quality of gasoline are again increased. Temperatures
in this reaction are controlled at approximately 750 to 1000 F while the
hydrogen pressure is held between 300 and 3000 psi. This process gives
a high-octane product with a high boiling point and high flash point,
especially suited for aviation gasoline.
Alkylation. Light undesirable hydrocarbons of one chemical family
may be reformed to heavier compounds of another family by a process
known as alkylation. Thus, light olefins and isoparaffins, such as butyl-
FUELS 25
ene, propylene, and isobutane, are fed into a reactor which converts them
into isoparaffins having a boiling point which falls into the gasoline range.
Here again is a process which produces a high-octane fuel having a high
boiling point and low vapor pressure.
Finished Blended Products. The products from the various refinery
processes are blended together to form liquid fuels with the proper
Fraction number
I 1 I I l
400 450 500 550 600
Evaporation temperature, F
Fig. 1-19. Distribution of hydrocarbon compounds in a straight-run mid-continent gas
oil by boiling point. (Ward and Smith, "Why Diesel Fuels Behave As They Do," paper
presented before the Western Petroleum Refiners' Association, San Antonio, April, 1951.)
physical characteristics to give high-quality gasoline, fuel oil, kerosene,

diesel fuel, etc. Also these liquid products may be put through various
finishing processes to reduce the sulfur and wax contents, inhibit later
gum formation during storage, etc. Each of these fuels contains hun
dreds of different compounds belonging to each of the five principal hydro
carbon families. Although there are many tests for determining the
physical properties of liquid fuels which are discussed more thoroughly
in Chap. 5, the greatest change in the physical characteristics of the
different liquid fuels is their volatility. Gasoline is composed of a large
number of hydrocarbons containing molecules with 5, 6, 7, and 8 carbon

atoms to the molecule which tend to vaporize readily. Kerosene is some
what less volatile and contains a greater number of heavier hydrocarbons.
The heavy fuel oils and diesel fuels are composed principally of hydro
carbons with 12 to 16 or more carbon atoms to the molecule, and as a
result they have much higher evaporation temperatures than gasoline.
Figure 1-19 shows the composition of the various hydrocarbon fractions
which have been determined in a straight-run gas oil used as diesel fuel.
The exact and complete analysis of such a fuel is impossible, but many
hydrocarbons can be detected positively, and the presence of many more
can be established quite definitely.
For the purpose of analyzing the oxygen requirements for combustion,
the analysis of a liquid fuel may be given as:
1. Percentage of hydrogen and carbon by weight

2. Ratio of hydrogen to carbon by weight
3. Ratio of the number of hydrogen atoms to the number of carbon
atoms in the fuel.
Note that none of these methods attempts to show all the constituent
hydrocarbons, but only the hydrogen and carbon contents. Since these
fuels are composed of so many compounds, any formula given to repre
sent the liquid would be merely an average of all the formulas of the com
pounds making up the fuel.
All gasolines are approximately 85.7 per cent carbon and 14.3 per cent
hydrogen by weight and may be represented quite accurately by the
average molecular formula CsHi6. Most kerosene similarly is 85.7 per
cent hydrogen and 14.3 per cent carbon by weight and the formula
Cn.6H23.2 represents all kerosenes. Light and heavy diesel oils and fuel
oils will vary from about 14 to 15 per cent hydrogen. The average
formula for such heavy fuels may vary widely.
GASEOUS FUELS
Gaseous fuels enjoy a number of advantages over solid and liquid fuels
for many applications. They may be easily piped into a furnace so that
no physical handling of the fuel is necessary. Generally, gases are free
from ash or other foreign matter, and they burn completely. Also, the
control of gas flames is relatively easy, and complete combustion with no
smoke can be obtained with a low percentage of excess air. In addition
to these advantages, in many parts of the country today, gaseous fuels
compare favorably in cost per Btu to either coal or oil ; in many instances,
gas is the cheapest.
One of the disadvantages in the use of gas is the difficulty in storing
FUELS 27
any appreciable supply. Large gas holders which store several million
cubic feet of gas under high pressure are used, but even this is not com
parable to the ease of storing large supplies of solid and liquid fuels.
Brief descriptions of the common gaseous fuels used commercially in
this country appear in the paragraphs which follow. Table 1-9 gives
typical analyses of the components, specific gravities, and heating values
of these gases. The specific gravity of a gas is the ratio of the density of
the gas to the density of air at the same temperature and pressure.
Heating values of gases are reported at 60 F and 30 in. Hg, the standard
adopted by the American Gas Association (AGA) for that purpose.
Natural Gas. Of the various gases listed, natural gas is used far more
widely than any of the others. It is compressed to pressures of several
hundred pounds per square inch and sent through pipe lines from the oil
fields of Texas, Louisiana, and Oklahoma to areas in the north central
and northeast portions of the United States. Much of the western United
States is supplied with natural gas from wells in that area, while oil fields
in Pennsylvania and West Virginia have supplied local areas with gas for
many years.
Natural gas is a nearly odorless and colorless gas that accumulates in
the upper parts of oil or gas wells. It consists chiefly of methane, CH4,
with varying quantities of heavier hydrocarbons, and has a heating value
between 1000 and 1100 Btu/cu ft. The gas may contain small amounts
of noncombustible constituents such as carbon dioxide, nitrogen, and
water vapor.
Like petroleum, natural gases from any given region of the country have
certain characteristic analyses. Gases in the east, or Appalachian region,
are usually high in paraffin content. Western gases contain relatively
higher percentages of carbon dioxide. Nitrogen and some helium charac
terize the natural gases from the Kansas, Oklahoma, and Texas regions.
In this country, no natural gas contains carbon monoxide, hydrogen, or
olefin hydrocarbons in any appreciable amounts. Some fields produce
gases containing hydrogen sulfide, H2S. Because of their pungent odor
such gases are called "sour" natural gases. They may be " sweetened "
by removing the hydrogen sulfide.
Much of the gas leaving the liquid and gas separator at the oil well still
contains appreciable portions of liquid hydrocarbons in the vapor phase.
"Dry" natural gas contains less than 0.1 gal of gasoline per 1000 cu ft of
gas. The higher hydrocarbons in "wet" gas are mostly propane, butane,
and pentane, although a small percentage of still heavier compounds may
be present at atmospheric pressure and normal temperatures. The

natural gasoline obtained at the well, by compressing and cooling the gas
to condense the higher hydrocarbons, is often termed casing-head gasoline.
Table 1-9. Typical Analyses of Fuel Gases* 4
Illumi-
ft
II
Btu/cu
nants Sp
Number and type Source 5; N2 CO H2 CH4 C2H6 C3H8 C4H00
and gr
Gross Net
others
Natural Birmingham, Ala. 5.0 90.0 5.0 0.0 1002 "904

Natural Pittsburgh, Pa. 0.8 83.4 15.8 0.5 1129 1021
Natural Los Angeles, Calif. 6.5 77.5 16.0 0.70 1073 51
Natural Kansas City, Mo. 0.8 8.4 84.1 6.7 0.63 54 259
Natural Casing-head type 36.7 14.5 23.5 14.9 10. 4f 1.29 2133 1925
Propane Natural gas 2.2 5.3 0.5 1.90 2908 235
Propane Refinery gas 2.0 72.9 0.8 24. 3i 1.77 2504 2256
Butane Natural gas 6.0 94.0 2.04 3210 295
1. 2. 3. 4. 5. 6. 7. 8. 9.
Butane Refinery gas 5.0 4.7 0. 3§ 2.00 3904 2935
10. Refinery oil Liq. phase crack. 0.2 1.2 6.1 4.4 72.5 1.00 1650 1524
11. Refinery oil Vap. phase crack. 0.2 1.2 13.1 23.3 21.9 39.6 0.89 1475 1351
12. 5il gas Portland 0.5 7.7 54.2 90.1 3.9 0.37 570 510
13. Coal gas Horizontal retort 0.8 7.4 48.0 27.1 3.0 0.47 542 486
14. Coal gas Inclined retort 0.8 7.3 49.5 29.2 3.4 0.47 259 540
15. Coal gas Vertical retort 0.4 13.5 51.9 24.3 3.4 0.42 520 44
0 1 2 1 2 2
16. Coke oven By-product 0.8 4 8 6.3 5.5 32.1 4.0 0.44 525 509
17. Producer Anthracite 8.0 0.1 50.0 23.2 17.7 1.0 0.86 143 133
90. Producer Bituminous 4.5 0.6 50.9 27.0 14.0 3.0 0.86 163 14
19. Blast furnace 11.5 0.0 27.5 1.0 1.02 92 92
.
20. Blue gas (water gas) Coke 5.4 0.7 8.3 37.0 47.3 1.3 0.57 07 52
21. Blue gas (water gas) Bituminous 5.5 0.9 27.6 0.2 32.5 4.6 0.7 0.70 50 239
.
.
22. Carbureted water. . Low gravity 3.6 0.4 5.0 21.9 49.6 10.6 2.5 6.1 0.54 46 45
.
23. Carbureted water.. . Heavy oil 6.0 0.9 12.4 5.8 32.2 13.5 8.2 0.4 40 451
24. Carbureted water High Btu 0.7 0.3 5.8 11.7 0.0 36.1 17.4 0.63 840 770
25. Sewage. „ Decatur, 111. 22.0 6.0 2.0 68.0 0.79 250 41
►
X
F
from
II
Reproduced
f
by permission "Gaseous Fuels," American Gas Association, 09008. CsHu. CiHe. «. CiHs.
CiHa. At 60 and 30 in. Hg.
FUELS 29
Liquefied Petroleum Gas. In the processing of petroleum, considerable

quantities of propane and butane are produced. These gases can be
liquefied at normal temperatures by subjecting them to a moderate pres
sure. They are liquefied and either "bottled" in steel cylinders at the
refinery or shipped in large bulk-type pressure tanks for later redistribu
tion. The recent increase in the use of these gases is due largely to their
flexibility in handling and storage. Liquefied petroleum gases are used
as stand-by supplies for users of natural or manufactured gas, also to aug
ment the heating values of some manufactured gases, and as fuel for stoves,
trucks, busses, and tractors in many parts of the country.
Special precautions must be taken in using propane and butane because
of two physical characteristics not found in the other gaseous fuels: (1)
Because of their high heating value (approximately 2550 Btu/cu ft for
propane and 3200 Btu/cu ft for butane), burners must be specially
adjusted to use them. (2) Since propane and butane are heavier than
air, escaping gas will tend to settle and collect in pockets thus creating an
explosion hazard.
Refinery Oil Gas. During the many operations involved in the refining
of crude oil, light gases are produced which cannot economically be con
verted into gasoline or some other liquid product. These gases are
termed refinery oil gases and consist of various light hydrocarbons with
small percentages of hydrogen and carbon monoxide present. It is not
common to pipe these gases any great distance for distribution, but they
are generally used within the plant to help supply fuel for the many
furnaces and heaters required in a refinery operation.
Coal Gas. The oldest method of manufacturing gas is the high-tem
perature coal-carbonization process. This is still used to produce a large
proportion of the manufactured gas distributed in cities and towns for
domestic and industrial use today. Coal gas is produced in horizontal,
inclined, or vertical retorts in which the primary product is gas; and coke,
which is the secondary product, is below average grade and not satis
factory for metallurgical use. The retorts are fired either intermittently
or continuously, depending on the design, with high-volatile coals which
will give a high gas yield.
Coal is heated to temperatures up to 2700 F in the presence of very
little air. The retorts are heated generally by burning producer gas in
grill work or chambers surrounding the charge of coal, although some of
the coal gas itself may be' fed back into the unit to help maintain the
necessary high temperatures. The complex organic compounds of the
coal decompose at the high temperature and form simpler, volatile prod
ucts and coke. The composition of the coal-gas product is controlled
by the size of charge, the temperatures employed during the carboniza
tion, and the type of equipment. In the vertical retorts, steam is often
added to the base of the coal charge. The steam reacts with the incan
descent carbon according to the equation, H2O + C — » CO + H2. The
addition of steam to the charge decreases the Btu per cubic foot of the
coal-gas product formed, but the volume of gas produced per ton of coal
is increased substantially.
Coke-oven Gas. Coke-oven gas is produced in a manner similar to
that for coal gas. The coke ovens are controlled more closely and at
somewhat lower temperatures, between 1200 and 2000 F, to form a high-
quality coke product. A large share of the coke produced today comes
from by-product coke ovens. These ovens are designed to maintain
fairly constant temperatures throughout the charge of volatile coal and
to remove the coke-oven gas and other by-products quickly. They are
heated with producer gas or blast-furnace gas. The composition of coke-
oven gas varies with the size of the charge, the length of the coking period,
and other similar factors. Some of the valuable by-products also derived
from the coking process are ammonia and ammonium salts, light oils
(including benzene, toluene, and light paraffins and olefins), and coal tar.
Each of the manufactured gases contains some compounds called
illuminants, as shown in Table 1-9. Any hydrocarbon compound which
produces a bright yellow flame upon combustion could be called an
illuminant, but the chief ones are ethylene, propylene, butylene, benzene,
toluene, xylene, and acetylene.
Producer Gas. Producer gas is formed in gas generators having a
general construction similar in principle to that shown in Fig. 1-20.
Either anthracite or low grades of bituminous coal are fed to the generator
where the coal is partially burned with a supply of air entering the base
of the combustion zone. The technique and equipment required to
produce gas from bituminous coal differ from those for anthracite; like
wise the composition of the finished producer gas varies widely.
The oxygen in the air entering the combustion zone first unites with
the hot carbon according to the equation, C + O2 — » CO2. The heat
released from this reaction brings the entire charge of coal to a high
temperature, driving off the volatile matter from the coal near the top of
the producer, and leaving incandescent carbon in the middle and bottom
of the unit. As the gases pass on up through the hot coal bed, the carbon
dioxide is reduced to carbon monoxide by the reaction, CO2 + C — » 2C0,
since the oxygen is nearly depleted from the gases by the time they reach
the upper part of the combustion zone.
Because of the large percentage of nitrogen appearing in producer gas,
resulting from the air fed to the producer, the heating value of the gas is
low, being about 130 to 150 Btu/cu ft. The Btu value of the finished
FUELS 31
gas may be increased by spraying steam into the coke charge of the gas
producer occasionally. The steam reacts with the coke to form carbon
monoxide and hydrogen. The addition of steam to the process must be
employed only intermittently, since this reaction absorbs a considerable
quantity of heat and may cool the coke bed to the point where all reac
tions will cease.
The low heating value of producer gas does not make distribution of
this gas in long pipe lines feasible. It is used extensively for firing open
Coal
in,
Producer gas out, intermittently
continuously
Water-cooled
poker
Rotating
shell
Space for Water, to guench

continuous ash and seal shell
ash removal
Air and steam

continuously
in,
Fig. 1-20. Diagram of a gas producer.
hearth furnaces in the steel industry. It may be used hot just as comes
it
from the gas generator, or the tar and other impurities may be "scrubbed
out." Recent research indicates that the use of small, efficient, high-
output gas producers may be a feasible way of utilizing coal, while still
maintaining many of the advantages of burning gaseous fuel in com
bustion equipment.
Blast -furnace Gas. Blast-furnace gas one of the by-products of the
is
blast furnace operation. It contains large per cent of nitrogen and has
a
the lowest heating value of any of the commercial gases. The Btu per
cubic foot of blast-furnace gas frequently runs below 100. For this
reason used near the area in which produced to operate gas
it
it
is
is
engines, to heat by-product coke ovens, and for various heating processes.
Carbureted Water Gas. The manufactured gas produced most widely

for commercial distribution is carbureted water gas. Figure 1-21 shows
a diagrammatic view of a generator to produce such a product. The
charge of coal or coke is heated to incandescence by burning the coal
vigorously during the "blow" run when air is blown through the fuel bed.
After the coal charge has burned and has been heated sufficiently, the air
supply is shut off. Steam is then forced either up or down through the
bed of hot coke during the "make" run, and water gas, or blue gas, is
generated by the reaction of the water with the hot carbon according to
Waste gas from
blast run
,
t,
Fig. 1-21. Diagram of a carbureted water-gas generator.
the equation: H20 + C — > CO + H2. This reaction cools the charge
within 3 or 4 min. Then the "make" run is stopped, and more air is
forced through the fuel bed to burn and thus heat the coke during
another "blow" run.
Thus, very little nitrogen is included in the finished water gas generated
during the "make" run, because air is not admitted to the fuel bed during
that period. Part of the volatile matter of the coal is included in the
product gas, however. Because of the high carbon monoxide content of
the gas, which burns with a short, blue flame, water gas is often called
blue gas.
The heating value of blue gas is approximately 270 Btu/cu ft, as it
FUELS 33
comes from the gas generator. For commercial distribution, it is desir

able to have a product with a higher heating value. The increase in Btu
value is achieved by enriching the water gas with gas oil (or fuel oil) by
spraying the liquid oil onto a checkerwork of heated firebrick, called a
carburetor. Sometimes the oil is sprayed into the fuel bed of the gas
generator itself. The intense heat cracks the heavier hydrocarbons of
the gas oil into lighter hydrocarbons which remain in the vapor phase in
the carbureted water gas. The heating value of this product could be
maintained at almost any value by regulating the amount of gas oil added
in the carburetor, but values of 500 to 600 Btu/cu ft are common.
Sewage Gas. The decomposition of sewage sludges produces a gas
that is approximately two-thirds methane and one-third carbon dioxide.
Heating values of the gas vary from 600 to 750 Btu/cu ft. The amount
of this gas produced in normal sewage-disposal plant operations varies
from 0.3 to 1.0 cu ft per capita of population per day. Sewage gas is
seldom distributed commercially but is used to supply heat and power to
operate city sewage-disposal plants.
SYNTHETIC FUELS
The obvious unbalance between the use of liquid and gaseous fuels and
their reserves, as compared to the use and reserves of coal, has caused
much concern as to what other fuel sources might provide suitable sup
plies of liquid and gaseous fuels to meet future demands. This specula
tion seemed especially urgent during times of temporary petroleum
shortages such as were experienced in 1918-1919 and 1945-1947. Follow
ing each of these periods, new reserves of crude oil were discovered, and
new production and refining methods have enabled the supply of liquid
fuels from petroleum to more than meet the demand.
However, research departments of various private corporations and the
United States government have continued investigations of possible sub
stitutes for petroleum products. Some of the developments receiving
the widest attention and showing more or less promise are discussed here.
Oil Shale. The reserves of oil shale are large in many states of the
union, especially in Colorado. It is estimated that at least 100 billion
barrels of oil can be obtained from known United States deposits of oil
shale. The U.S. Bureau of Mines has established a pilot plant at Rifle,
Colo, to investigate and develop a method for recovering this oil. Oil
shale is a rock bearing anywhere from 7 to 50 gal of oil per ton of shale.
The quantity of liquid is so low that it cannot be forced from the rock by
pumping it from the ground as is done with oil-bearing sands. Neither
will a high pressure exerted on the shale by water or mud displace the oil.
The production of liquid fuels from oil shale may be divided into two
basic operations: (1) mining of the shale and recovery of the crude shale
oil and (2) refining the crude oil obtained. With modern mechanical
mining and quarrying techniques, it has been found that the shale may
be mined, crushed, and conveyed to the recovery plant at a relatively low
cost per ton. A simplified flow diagram showing the processes used to
recover the oil from the shale and refine the crude shale oil appears in
Fig. 1-22. The oil is removed from the shale by passing heated air over
Air
Spent shale
Processing
for
Sized oil shale -Hot air, gases, and oil-
pipe-line
Retort transportation
OIL-RECOVERY PHASE
Hydrocarbon Fuel gas
> I Catalytic
}*- Gasoline
cracking
-Naphtha
Diesel fuel
Heavy gas oil
Ammonium sulfate
Heavy residues converted to coke

Coke
PRODUCT-REFINING PHASE
Fiq. 1-22. Recovery of liquid fuels from oil shale.
the raw shale in a retort. The oil evaporates into the air and is later
condensed as crude shale oil, which is rich in olefinic and ring compounds
and somewhat troublesome to refine.
By the method shown in Fig. 1-22, the crude shale oil is hydrogenated
to form a product which can be further refined by conventional refinery
processes into the usual commercial qualities of gasoline, kerosene, fuel
oil, etc. There are various processes by which the crude shale oil might
be refined.
Synthesis of Coal. A number of processes have been developed for
converting coal into liquid fuels. Figure 1-23 shows a simplified flow
FUELS 35
diagram of one of these. This portrays the Fischer-Tropsch method

which has been used extensively in other countries during the past decade.
Basically, the process involves the forming of a synthesis gas in a gas
producer by partially burning coal with insufficient air. This yields a
product high in CO and H2 content. The synthesis gas is then passed
through a synthesis reactor where, in the presence of a catalyst and the
proper temperature and pressure, the CO and H2 are reformed to hydro-
Synthesis Gas-purification
process
r
Steam
Oxygen 2100 F
500 psi
Steam
Dried pulverized Fluid-fuel-bed

coal combustion chamber
SYNTHESIS -GAS GENERATION AND PURIFICATION
Synthesis gos
Hydrocarbons^ -Fuel gas
■Light polymer
Polymeri -1 , ■Butane
600 to 700 F zavion - Heavy
Product polymer
recovery
Synthesis - Gasoline
reactor c_5_a Isomeri-
Steam -Heater oil
Heavier zofion - Heovy gas oil
SYNTHESIS PRODUCT REFINING
Flo. 1-23. Liquid fuels from synthesis of coal.
carbons of the paraffin, olefin, and alcohol families. These products are
further refined by conventional processes to form suitable liquid fuels.
The Bergius hydrogenation process is another means for converting coal
into liquid fuels. It employs a hydrogenation process on a powdered coal
paste. The U.S. Bureau of Mines has established a pilot plant for
investigation of all phases of coal synthesis at Louisiana, Mo.
Synthesis of Natural Gas. A synthetic gasoline can be produced from
natural gas using much the same process as that described for the syn
thesis of oil from coal. In both of these processes, a low-percentage Btu
conversion is obtained since part of the original coal or gas must be burned
to form the synthesis gas. The production of liquid fuels from natural
gas is reported to be able to compete economically with the production of
petroleum fuels. Unfortunately, the reserves of natural gas are small as
compared to those of oil shale and coal, so that the latter seem to offer
the best possibility for supplying unlimited quantities of liquid fuels for
many years.
Alcohol and Vegetable Oils. The U.S. Department of Agriculture has
established a pilot plant at Peoria, 111., to investigate the possibility of
producing liquid fuels from agricultural products. These fuels might
take the form of natural oils, such as corn oil, cottonseed, or peanut oil, or
alcohols formed from organic compounds. While these compounds have
a few select uses at the present, no means seem in sight for processing a
satisfactory fuel in any form from agricultural products which can begin
to compete economically with present fuels.
Underground Gasification of Coal. Attempts are being made to gasify
coal underground and bring up much of the heat content without mining
the coal. Tests have been made to study the feasibility of this plan at
Gorgas, Ala., under the sponsorship of the Alabama Power and Light Co.
and in cooperation with the U.S. Bureau of Mines. Similar tests have
been and are being made in other countries.
The general plan is to force air or oxygen into an enclosed vein of burn
ing coal, let the ground act as a gas producer, and take the product gas
out of the mine a few hundred feet down the vein from the point of the
admission of air. So far it has been proved that the coal will burn, and
rather completely, under such circumstances. The results obtained by
the end of the second Gorgas test in 1950, however, were not too encour
aging in that a gas low in Btu content was produced. At any rate, here
is a possible new development which would throw many new factors into
the economics of the use of fuels and power generation.
Problems
1. Distinguish between proved reserves and estimates of fuels.
2. Why is sulfur listed separately when a proximate analysis is reported?
3. Correlate the percentages of volatile matter and fixed carbon for the various
ranks of coal.
4. In general, what conclusion may be drawn in regard to the effect of the composi
tion on the heating value on a dry basis for solid fuels?
5. Why is there less difference between the fixed carbon in the proximate analysis
and the carbon in the ultimate analysis for a high-rank than for a low-rank coal?
6. Discuss the economics of the hypothetical problem: What would happen if
Pennsylvania, West Virginia, Kentucky, and Ohio should suddenly be depleted of
their coal resources?
7. Compare the origin of coal to the origin of petroleum.
FUELS 37
8. Show structural formulas for: nonane; re-heptane; 2,3-dimethylpentane;

3-methylheptane ;2,2,3-trimethylbutane.
9. Construct molecular structures for cycloheptane, heptylene, toluene, methyl
alcohol, butyl alcohol, acetylene.
10. Correlate the number of carbon atoms in the molecule to the boiling point for
the hydrocarbons listed in Table 1-7; the freezing point.
11. What general relationship exists between the API gravity of a compound and
the number of atoms in the molecule?
12. From a study of Table 1-7, should a heavy fuel oil require an appreciably
different quantity of air for combustion than a light naphtha?
13. What general observation may be made concerning the latent heat of vaporiza
tion of hydrocarbons?
14. List the significant fuel properties of the hydrocarbon families present in
petroleum fuels.
16. Discuss what is meant by a paraffin base oil.
16. Distinguish between natural gasoline and casing-head gasoline.
17. Would natural gasoline tend to form gum? Explain.
18. Discuss the difference between a straight-run and a cracked fuel oil as far as the
types of hydrocarbons present are concerned.
19. Why is a process of polymerization necessary in oil refining if a cracking process
is also used?
20. What is accomplished in the alkylation process?
21. Discuss the need for finishing processes in oil refining.
22. From Table 1-9, compare the principal components of manufactured and
natural gases.
23. What is the boiling point of propane at atmospheric pressure? of butane? of
re-pentane?
24. How could the heating value of a butane gas be safely reduced to 500 Btu/cu ft?
25. What are liquefied petroleum gases?
26. Why is water gas more desirable for distribution than producer gas?
27. Discuss the difference between water-gas generation and the manufacture of
producer gas.
28. Comparing the data from Tables 1-6 and 1-9 and Fig. A-l in the Appendix,
correlate the costs of coal per ton, gas per 1000 cu ft, and liquid fuel per gallon, which
yield the same cost per Btu of fuel.
References
1. "Gaseous Fuels," American Gas Association, 1948.

2. Griswold, J., "Fuels, Combustion, and Furnaces," McGraw-Hill Book Company,
Inc., New York, 1946.
3. Johnson, A. J., and G. H. Auth, "Fuels and Combustion Handbook," McGraw-
Hill Book Company, Inc., New York, 1951.
4. Obert, E. F., "Internal Combustion Engines," 2d ed., International Textbook
Company, Scranton, Pa., 1950.
5. "Mineral Resources of the U.S.," U.S. Bureau of Mines, 1948.
6. Wilson, P. J., and J. H. Wells, "Coal, Coke, and Coal Chemicals," McGraw-Hill
Book Company, Inc., New York, 1950.
CHAPTER 2
STOICHIOMETRIC ANALYSIS
Stoichiometric calculations deal with the weights of materials and quan

tities of energy entering and leaving chemical reactions. Fundamental
stoichiometric calculations dealing with combustion are idealized chemical
reactions wherein fuel and air combine chemically to form the products of
combustion. Actually, the burning process is extremely complicated;
in fact, no theory exists at the present time which completely explains all
the phenomena of combustion. When fuel is burned very rapidly under
high pressures, as in the cylinder of an internal-combustion engine, the
products of the reaction vary with the physical conditions of temperature
and pressure in the cylinder. In other types of combustion equipment,
the degree of mixing of the fuel and air is a controlling factor in the reac
tions which occur, once the fuel and air mixture is ignited. What is
more, it is thought that all burning is the result of a series of very com
plicated and rapid chemical reactions.
In this chapter, however, it is assumed that the fuel is completely mixed
with air, or oxygen, and that combustion proceeds directly to the simple
end products of the complete chemical reaction, namely, carbon dioxide
and water. Later, the effects of incomplete mixing of fuel and air, high
temperatures and pressures of the combustion gases, and rapid cooling of
those gases will be discussed. The simplified reactions used in the
stoichiometric calculations which follow will be found to yield results
which are sufficiently accurate for most types of design or efficiency
computations.
FUNDAMENTAL PHYSICAL LAWS
The calculations used in analyzing combustion processes are compara
tively simple, once the physical principles involved are thoroughly under
stood. The need for a thorough understanding of the seven laws enumer
ated below cannot be emphasized too strongly. These laws are the
premises upon which all calculations that follow are based. In their
simplest form, the seven fundamental physical laws of immediate concern
are:
1. Conservation of mass. Matter can neither be destroyed nor created.
2. Conservation of energy. Energy can neither be destroyed nor
created.
38
STOICHIOMETRIC ANALYSIS 39
3. Law of combining weights. All substances combine in accordance

with simple, definite weight relationships.
4. Ideal-gas law. The volume of an ideal gas is directly proportional
to its absolute temperature and inversely proportional to its absolute
pressure.
5. Avogadro's law. Equal volumes of perfect gases under identical con
ditions of temperature and pressure have the same number of molecules.
6. Dalton's law. The total pressure of a mixture of gases is the sum of
the partial pressures which would be exerted by each of the constituents if
each gas were to occupy alone the same volume as the mixture.
7. Amagat's law. The total volume occupied by a mixture of gases is
equal to the sum of the volumes which would be occupied by each con
stituent when at the same temperature and pressure as the mixture.
Conservation of Mass. The total mass of material that enters into a
process remains unchanged, even though there may be some rearrange
ment of atoms or molecules. In the normal thermodynamic processes
associated with the heat power field, this concept is easily understood
because the molecules remain unchanged. In the combustion process,
the atoms rearrange themselves to form new molecules and liberate chem
ical energy in doing so. However, every atom in the material that enters
the process must be accounted for in the material that leaves, and there
is no change of weight during the process.
This concept of the conservation of mass may be illustrated by an
internal-combustion engine operating on an air/fuel ratio of 15 to 1 by
weight. In burning 1 lb of fuel, 15 lb of air are used and 16 lb of exhaust
products are formed. The chemical energy in the 1 lb of fuel is released
as heat energy which is then changed to mechanical work or dissipated as
heat losses from the cycle.
Hydrogen burns to form water according to the chemical equation
2H2 + O2-» 2H2O

Thus two molecules of hydrogen plus one molecule of oxygen react to give
the end product — two molecules of water. A "material balance" of the
equation indicates that four atoms of hydrogen and two atoms of oxygen
enter into the reaction, and these same atoms again appear in the
products of the reaction, although their grouping, or combination, is dif
ferent. Likewise the sum of the weights of hydrogen and oxygen entering
the combustion process must equal the weight of water formed. As long
as there are no changes in the atomic structures of the elements involved,
the material balance must exist.
Conservation of Energy. The summation of all energy entering a proc
ess is exactly equal to the summation of energy leaving. The total of the
potential, kinetic, heat, chemical, and electrical energy in the material

entering must appear in the material leaving, although the proportionate
amounts of each form of energy may change. The combustion process is
characterized by the change of chemical energy into heat energy.
Law of Combining Weights. It has been found experimentally that
elements combine in simple and constant proportions to form definite
compounds. These proportions by weight are always exact ratios of the
molecular weights of the constituents entering into the reaction. Because
of this method of combination, it is convenient to state chemical equations
in terms of the number of molecular weights, or moles, of the elements or
compounds in the reaction.
The molecular weights are based on an atomic weight of oxygen of
16.0000, so chosen because the atomic weight of the lightest element,
hydrogen, is then approximately 1. The atomic weight of carbon is 12,
because the carbon atom is three-fourths as heavy as the oxygen atom.
The atomic weights used in combustion calculations are listed in Table
2-1. As can be seen from the values listed in the table, the approximate
atomic weights are so near to the exact values that they can be used in
most cases with sufficient accuracy.
Table 2-1. Atomic Weights of Common Elements
Element Exact atomic weight Approximate atomic weight
Hydrogen 1.0080 1
Carbon 12.01 12
Nitrogen 14.008 14
Oxygen 16.0000 16
Sulfur 32.06 32
A term called the pound molecular weight, or pound mole, is applied to

that mass of a substance which has a weight in pounds equal to its
molecular weight. Since hydrogen has a molecular weight of 2, a pound
mole of hydrogen weighs 2 lb. The pound mole is a very convenient term
to use in combustion calculations because the elements combine in simple
and constant proportions. A better understanding of its advantages and
use will be obtained in the calculations which follow later in the chapter.
The term pound mole is often abbreviated to mole when there is no ques
tion as to whether the term refers to a pound mole as compared to a gram
mole. The gram mole is the quantity of material required to have a
weight in grams equal to its molecular weight. In this book, all weights
will be given in pounds, and where the term mole is used it will always
refer to the pound mole, or pound molecular weight. A term pound atom
could likewise be applied to that quantity of a material which has a weight

in pounds equal to its atomic weight.
Ideal-gas Law. The ideal-gas law, in equation form, is
PV = NRT
where P absolute pressure, lb/sq ft
—
V = volume, cu ft
N = number of moles of gas
R = universal gas constant, 1545 ft-lb/(mole) (R)
T = absolute temperature, R
This law is the statement of the physical phenomenon first observed by
Boyle and Charles. For 1 mole of gas,
~ijT
= R
The universal gas constant R has been determined very accurately for
many gases and is approximately 1545 ft-lb/(mole) (R). It should be
noted that the specific gas constant, whose units are ft-lb/ (lb) (R), is equal
to the universal gas constant divided by the molecular weight of the gas.
Thus, for oxygen which has a molecular weight of 32.0 lb per mole, the
specific gas constant R' is
R' - molecular wt "

3^
" fMb/(lb)(R)
While the universal gas constant does not vary appreciably, the specific
gas constant for each gas is different since the gases have different
molecular weights.
Oxygen has a molecular weight of 32. Therefore 1 mole of oxygen
weighs 32 lb, and the volume of 1 mole of oxygen at the scientific "stand
ard" temperature and pressure (32 F and 14.7 psia) would be
NRT X ...
-T
1 X 1545 492
" 359
V
T7 = = CU ft
14.7 X 144
A pound mole of any ideal gas must occupy the same volume at stand
ard conditions since all terms in the ideal-gas equation are constant at the
same temperature and pressure. Experimental evidence substantiates
this pressure-volume-mass relationship for all gases that approximate the
ideal gas.
Appreciable deviations from the ideal-gas law will occur with real gases
is,
under conditions of high pressure or low temperature, that in the

vaporous region near saturation conditions for the gas. For highly super
heated gases at normal pressures, the deviations are usually insignificant,
and the ideal-gas law can be applied with sufficient accuracy for combus
tion computations.
Avogadro's Law. From the preceding article it is seen that the volume
occupied by 1 mole of any ideal gas is the same under like conditions of
temperature and pressure. Also, the weight of a mole of gas is directly
dependent upon the weight of the molecules making up the gas. It there
fore follows that a mole of any gas always has a fixed number of molecules.
Avogadro stated this relationship. It has been further proved that a
mole of gas contains 2.7 X 1026 molecules. The full importance of
Avogadro's law is not readily recognized. However, in the calculations
which follow, the volume and mole relationship should become evident as
the law is applied.
Dalton's Law. In a mixture of gases, each gas fills the entire volume.
In Fig. 2-1 three gases are considered — first alone, and then as a mixture.
Gas A + Gas B + Gas C = Mixture
P = 14.7 psia P = 14.7 psia P = 14.7 psia P =44.1 psia

V = 359 cu ft V = 359 cu ft V = 359 cu ft V - 359 cu ft
N = 1 mole N = 1 mole N = 1 mole N = 3 moles
5P = 32 F 71 = 32 F r = 32 F T = 32 F
Fig. 2-1. Partial pressures of a gaseous mixture.
If gases A, B, and C are 1 mole each of carbon dioxide,

nitrogen, and
oxygen, respectively, their weights will be 44, 28, and 32 lb in the order
given. The 3 moles of gases, when combined, will have a total weight of
44 + 28 + 32 = 104 lb. If each gas exerts a pressure of 14.7 psi alone,
it will exert the same pressure on the vessel containing the mixture if the
final volume and temperature are the same. Thus the total pressure
(14.7 + 14.7 + 14.7 = 44.1 psia) of the mixture is equal to the sum of
the pressures created by each constituent.
The above is an example of Dalton's law of partial pressures which
states that the total pressure of a gaseous mixture is the sum of the partial
pressures which would be exerted by each of the constituents if each gas
were to occupy alone the same volume as the mixture. Or, in equation
form,
PT = PA + Pb + Pc
where PT is the total pressure, and PA, Pb, and Pc are the partial pressures
exerted by gases A, B, and C.
If the ideal-gas law is applied to gas A,
PAV = NART
(1)
For the mixture, the same law states
(2) PTV = NTRT

And dividing Eq. (1) by Eq. (2),
PAV = NART
PTV NTRT
which, when simplified, becomes
=
(3) pi wT
Solving for the partial pressure of A and substituting the numerical values
from the example of Fig. 2-1,
PA = X PT = X = 14-7
WT \ 44-1 Psk
The fraction N'A/NT is often called the fraction. The partial pres
mole
sure of a gas present in a gaseous mixture is equal to the mole fraction of
that gas times the total pressure of the mixture. Although Dalton's law
of partial pressures can be applied in principle to real gases, the expression
in Eq. (3) is precise only for ideal gases because it was derived from the
ideal-gas law. However, Eq. (3) may be used for all gases within the
limits of accuracy required for combustion calculations.
Example 1. A tank having a volume of 1000 cu ft contains 44 lb of

carbon dioxide, 120 lb of nitrogen, and 25 lb of oxygen. What pressure
will be indicated by a gage attached to the tank if the gas temperature is
130 F?
Solution. For the CO2 :
Ar = wtofCO2 =
44 ,
N -. -n — \ mole
molecular wt 44
NRT
V
p 1 X 1545(460 + 130) . .
Fc(H 6-33 psm
1000 X 144
For the N2:

N = I2^g = 4.29 moles
p 4.29 X 1545(460 + 130) 0_ .

Pt" = 27 13 pSia
1000 144 X
For the 02:
N = 2^2 = 0.78 mole
D =
0.78 X 1545(460 + 130) ,
= 4 nr
.
Po' 95 pSia
1000 X 144
Therefore, the total pressure is
PT = 6.33 + 27.13 + 4.95 = 38.41 psia
and the gage would indicate (38.4 — 14.7), or 23.7 psig.

A simpler method would be to sum up the number of moles present and
solve directly.
NT = Ncoi + A^n, + No,

= = 6.07 moles
4^4 + 120As + 25A2
PTV = NTRT
=
6.07 X 1545 X 590
= 38.4 psia
PT
1000 X 144
Amagat's Law. Amagat's law is similar to Dalton's in that it concerns

gaseous mixtures. However, instead of considering the additive effect of
partial pressures, it states a similar principle of additive partial volumes.
Amagat's law states: "The total volume occupied by a mixture of gases
is equal to the sum of the volumes which would be occupied by each con
stituent when at the same temperature and pressure as the mixture."
The physical significance of this law may be visualized by referring to
the gaseous mixture of Fig. 2-2. If the gases making up the mixture were
separated into individual containers so that each gas was at the same
temperature and pressure as the mixture, the resulting volumes would be
as shown.
Mixture Gas A
Gas B
+ Gas C
P = 44.1 psia P 44.1 psia P = 44.1 psia P = 44.1 psia

V = 359 cu ft V 119% cu ft V = 119% cu ft V = 119% cu ft
N = 3 N 1 A' = 1 .V = 1
T = 32 F T = 32 F T = 32 F T = 32 F
Fig. 2-2. Partial volumes of a gaseous mixture.
From Fig. 2-2 it is evident that the total volume VT is equal to the sum
of the partial volumes VA, VB, and Vc- Following a derivation similar to
that for Eq. (3),
(4) VA = ~ VT = I X 359 = 119

| cu ft
Thus, the partial volume of a gas present in a gaseous mixture is equal to

the mole fraction of that gas times the total volume of the mixture. As
with Eq. (3), Eq. (4) applies to ideal gases but yields sufficient accuracy
for combustion calculations with any gas.
Example 2. What is the volumetric composition of the gaseous mix

ture of Example 1? ("Volumetric composition" is merely the partial
volumes expressed as percentages of the total volume.)
Solution:
1.0
Fcoj X 1000 = 164.7 cu ft = 16.47%
6.07
4.29
= X 1000 = 706.8 cu ft = 70.68%
6.07
0.78
X 1000 = 128.5 cu ft = 12.85%
6.07
1000.0 cu ft 100.00%
FUNDAMENTAL LAWS APPLIED
The application of the above physical laws to the solution of problems

dealing with the stoichiometric relationships of fuels, air, and combustion
products is presented in the following sections.
Combustion with Oxygen. Since the combustible matter in fuels is
composed mainly of carbon and hydrogen, combustion calculations deal
mostly with the different relationships between carbon, hydrogen, and
oxygen and the combustion products which appear as a result of the
reactions.
The reaction between hydrogen and oxygen may be stated in equation
form as:
2H2 + 02 -> 2H20
This is the same as saying that two molecules of hydrogen plus one mole
cule of oxygen will unite to form two molecules of water. While it must
be remembered that a chemical reaction between two substances is actu
ally a combination of the individual molecules of those substances, it is
usually more convenient to work with molal units. The equation also
states that 2 lb moles of hydrogen plus 1 lb mole of oxygen form 2 lb
moles of water. Also from Avogadro's law it can be concluded that 2
vol of hydrogen plus 1 vol of oxygen will form 2 vol of water vapor. Since
each mole of hydrogen weighs 2 lb, a mole of oxygen weighs 32 lb, and
each mole of water weighs 18 lb, it follows that 4 lb of hydrogen plus 32 lb

of oxygen gives 36 lb of water.
In summary, the combustion equation for hydrogen says:
2H2 + 02 -> 2H20

2 molecules H2 + 1 molecule 02 — > 2 molecules H20
— > 2 moles
2 moles H2 +1 mole 02 H20
2 cu ft H2 +1 cu ft 02 -» 2 cu ft H20 (vapor)
4 lb H2 + 32 lb 02 -> 36 lb H20
These relationships must be thoroughly understood as they are the basis

for all combustion calculations. Other fundamental combustion equa
tions are:
C + 02 -» C02
CO + ko2 -> C02
c + H02 — CO
s + o2-> so2
The same reactions of the elements with oxygen will occur even if the
elements are combined in a chemical compound; thus the combustion
equation for the hydrocarbons methane and octane can be written,
CH4 + 202 -> C02 + 2H20

C8H18 + 12.502 -» 8C02 + 9H20
Note that it is not always necessary to balance the equation in such a way
that all the coefficients are whole numbers. It is just as true that 1 mole
of C0 unites with % mole of 02 as if it were written 2 moles of CO unite
with 1 mole of 02. This simplifies balancing the equations. Also it
should be observed that the combustion of a hydrocarbon fuel can be
expressed as a simple combination with oxygen to form carbon dioxide
and water as the combustion products.
Combustion with Air. In most applications of combustion, fuel is
burned in the presence of air. This means that much nitrogen from the
air must accompany the oxygen which goes into the chemical reaction.
However, the nitrogen, being inert, does not enter into the reaction.
Table 2-2 shows the average composition of dry air by volume. This
Table 2-2. Composition of Dry Air*
Per cent by volume Per cent by volume
Nitrogen 78.03 Hydrogen 0.01

Oxygen 20.99 Neon .... 0.00123
Argon 0.94 Helium . . 0.0004
Carbon dioxide 0.03 Krypton . 0.00005
Xenon . . . 0.000006
* International Critical Tables.

analysis of the atmosphere remains rather constant throughout the world

at the surface of the earth and up to elevations yet attainable by man.
The small percentages of the rare gases present in the atmosphere are
also inert. They act very much like nitrogen and do not enter the com
bustion reaction. Therefore, the composition of air by volume can be
taken as 79 per cent nitrogen and 21 per cent oxygen with sufficient
accuracy for most engineering calculations. The ratio of nitrogen to
oxygen in air is then 7
%i = 3.76, which means 3.76 moles of nitrogen
accompany each mole of oxygen when a fuel is burned in air. The com
bustion equation for methane in air could then be written :
CH4 + 2O2 + (2 X 3.76)N2 -» CO2 + 2H2O + 7.52N,

The molecular weight for a mixture of gases can be computed as the
sum of the mole fractions of the gases times their respective molecular
weights. In equation form,
0 0 ,
+ +
m--wTMA wTMb wTMc+
where M„ = average
molecular weight of mixture
MA, MB, Mc molecular weights of gases A, B, and C, respectively
=
NA, NB, Nc = number of moles of gases A, B, and C, respectively

NT = total number of moles of gas
Since the temperature and pressure of each constituent gas in the mix
ture are the same, the mole fraction is also the per cent by volume of each
constituent.
The molecular weight of a mixture of gases such as air, for examp1e, can
be computed as
A^
100 100
= 0.21 X 32 + 0.79 X 28
= 28.9
Because the argon and the small amount of carbon dioxide in the air
are somewhat heavier than nitrogen, the actual average molecular weight
of air is closer to 29.0 than the 28.9 obtained above.
In addition to the various components of dry air listed in Table 2-2,
there is always a certain amount of moisture in the atmosphere. The
weight of water vapor present per pound of dry air under various atmos
pheric conditions may be determined from a psychrometric chart (Fig.
A-8 in the Appendix) . From the chart, the weight of water vapor in 1 lb
of dry air may be found by reading vertically from the dry-bulb air
temperature to the appropriate relative humidity line and then reading
horizontally to the right. Relative humidity is the ratio of the actual to
-18 FUELS AND COMBUSTION
the saturation quantity of water vapor per unit volume, at the same
temperature. Thus, as read from the psychrometric chart, the maximum
amount of water vapor the air could hold at 70 F would be 0.01(5 lb per
pound of dry air. Since the air generally has a relative humidity con
siderably less than 100 per cent, air used for combustion frequently has
much less than 1 per cent moisture by weight. The error introduced by
neglecting the moisture in the air in combustion calculations is so insig
nificant for most practical problems that it has been omitted from the
examples in this book, although for certain conditions it may be desirable
to include the moisture.
The examples which follow show methods for calculating the various
relationships among fuel, air, and combustion products which must be
known in many types of engineering work. In these problems the
molecular weight of air is taken to be 29.0 lb/mole, and the moisture in the
air used for combustion is neglected.
Combustion of Liquid Fuels. If a liquid fuel can be represented by a
chemical formula, one combustion equation for the chemical reaction can
be written which shows the quantity of oxygen required to burn the fuel,
the combustion products formed, and the nitrogen from the air which
must accompany the required oxygen. For the combustion of the hydro
carbon octane CsHi8 with air, the equation would be
isHi8 + 12.502 + 12.5 X 3.76N2 -> 8C02 + 9H20 + 47N2

The chemical equation is balanced by noting that eight oxygen mole
cules (802) would be needed to burn the eight carbon atoms in the octane
molecule to eight carbon dioxide molecules (8C02). In equation form,
8C + 802 -> 8C02
Likewise, nine oxygen atoms, or four and one-half oxygen molecules

(4.502), would be required to burn the 18 hydrogen atoms in the original
fuel molecule to nine water molecules.
18H + 4.502 -> 9H20
Thus, (8 + 4.5) or 12.5 oxygen molecules (12.502) must be taken from

the air for complete combustion of octane. This oxygen would be
accompanied by 3.76 X 12.5 nitrogen molecules, and the nitrogen would
appear unchanged in the combustion products.
Theoretical air is defined as the exact quantity of air required for the
chemical reaction. Sometimes combustion occurs with insufficient air
to burn the fuel completely to carbon dioxide and water vapor. Usually
more air is present than is actually needed to enter into the chemical
reaction. The quantity of air present beyond that actually required is
called excess air. A mixture of theoretical air and fuel is termed a

chemically correct, A mixture of 1 mole of
or a stoichiometric, mixture.
octane, 12.5 moles of oxygen, and
nitrogen, as stated in the
47 moles of
combustion equation for octane above, would be a chemically correct
mixture.
The ratio of air to fuel in any mixture may be stated as the volumes (or
moles) of air present per volume (or mole) of fuel, or as the weight of air
per weight of fuel. For the volume (or mole) relationship, the theoretical
air/fuel ratio for octane, is:
A _ 12.5 moles 02 + 47 moles N2 _ 59.5 moles air _ cu ft air
"
F 1 mole C8Hi8 1 mole fuel cu ft fuel vapor
It important to note that the volume relationships hold only if the
is
fuel is in the vapor state, as a mole of liquid fuel would occupy considerably
less than 359 cu ft under standard conditions.
On a weight basis, the air/fuel ratio is:

A _~ 59.5 moles air X 29.0 lb/mole air
~ _ lb air
1 1 mole fuel X (8 X 12 + 18 X 1 j lb mole fuel lb fuel
As discussed in Chap. 1, the liquid fuels used most commonly today,
namely, gasoline, fuel oil, kerosene, etc., are not made up of a simple com
pound such as octane. Instead, these fuels are composed of hundreds of
different hydrocarbon compounds, each having a different molecular
formula, molecular weight, and physical and combustion characteristics.
For the purposes of these calculations, a liquid hydrocarbon fuel may be
represented by one simple formula which gives the relationship between
the number of hydrogen atoms and the number of carbon atoms.
Thus, while a gasoline may be composed of several hundred different
compounds, a formula CxHy can be used to write a combustion equation
for the fuel and determine the relationships of air, fuel, and combustion
products.
The composition of a liquid fuel may be given in several ways; two
common methods are (1) to give the ratio of hydrogen to carbon by
weight and (2) to list the percentage of hydrogen and the percentage of
carbon by weight in the fuel. Refined petroleum fuels essentially are
entirely composed of hydrogen and carbon compounds, so either of the
above mentioned methods gives an accurate statement of what is avail
able for combustion in the liquid. The ratio of hydrogen to carbon by
weight may be changed to the ratio of hydrogen atoms to carbon atoms
by dividing each element by its atomic weight. For example, if a liquid
fuel is composed of 8.72 lb of hydrogen per 48 lb of carbon,
8.72 lb hydrogen/ 1
_ 8.72 lb hydrogen _ 8.72 atoms hydrogen
48 lb carbon/12 48 lb carbon 4 atoms carbon
A representative chemical formula for

the fuel can then be written.
Such a formula cannot be used to find the average molecular weight of
the liquid fuel unless the average number of carbon atoms in the molecule
has been determined chemically, but it is useful in predicting the air-
fuel-combustion products relationships.
Many times the ratio of hydrogen atoms to carbon atoms is such that
the representative formula cannot be expressed easily with integer sub
scripts. This should cause little trouble, however, since it is no more
difficult to balance combustion equations for C4H8.72, or CH2.i8, than for
C8Hi8, once it is realized that such a thing can be done.
"<- \ \ /
Example 3. Determine the products of combustion and the air/fuel

ratio by weight when a liquid fuel of 16 per cent hydrogen and 84 per cent
is,
carbon by weight is burned with 20 per cent excess air, that 20 per cent
more air than theoretically required.
is
Solution. To solve this problem advantageous first to find
is
it
a
representative formula for the fuel burned. In this case, the ratio of
H/C by weight 16 lb H/84 lb C. By dividing the weight of each ele
is
ment by its respective atomic weight, an atomic ratio of H/C can be
found. Thus, or shows the fuel could be represented

'
&toms
(}
by the chemical formula The theoretical combustion equation
is
C7H16.
+ 3.76 X UN,-* 7CO2 + 8H2O + 41.36N2

+
C7H16 11O2
excess air the ratio of the excess air supplied to the air
is
Percentage
of
theoretically required times 100. The equation balanced for the theo
is
retical air, but since the problem states 20 per cent excess air supplied,
is
the coefficients of the O2 and N2 on the left side of the equation should be
multiplied by 1.20. Again the N2 unchanged during the reaction, but
is
since more O2 supplied than needed for the chemical reaction, some
is
is
O2 (20 per cent of theoretical O2) appears in the products of combustion.

The combustion equation with 20 per cent excess air
is
+ X X 3.76 1.20)N,
X
+
C7H16 (11 1.20)O2 (11

-» 7CO2 + 8H2O + 2.2O2 + 49.6N2
+ 49.6)29.0 lb air/mole
F A
. (13.2
weiShtJ
y
X lb fuel/mole
{
12 + 16
X
1)
1(7
= 18.2 lb air/lb fuel
Dew Point of Combustion Products. It sometimes desirable to know

is
the dew-point temperature, or the temperature at which the water vapor

in the products of combustion begins to condense, since detrimental
is
it
to duct work, breechings, and mufflers to have water deposited on them.

The dew point is reached when water vapor begins to condense and form
droplets of liquid. Theoretically, the dew point is the temperature at
which the saturation pressure for water, as shown in thermodynamic
tables, equals the partial pressure of water vapor in the gaseous mixture.
The partial pressure exerted by the water vapor in the mixture of com
bustion products of CH4 burned with theoretical air, if the gases are at
atmospheric pressure, would be:
CH4 + 202 + 2 X 3.76N2 -> C02 + 2H20 + 7.52N2

2 moles H20
. -PHiO X 14.7 = 2.8 psia
(2 + 1 + 7.52) moles total
In Table 2-3, the temperature corresponding to a saturation pressure of
2.8 psia for water vapor is found to be 139 F. If the temperature were
reduced below this value, the table shows that less than 2.8 psia water-
vapor pressure could be maintained, hence some of the vapor would con
dense to liquid. Thus, the dew-point temperature for the conditions
stated is 139 F.
Sulfur compounds in the flue gases tend to cause condensation of water
vapor sooner than would be expected according to the principle of partial
pressures. Thus, the actual dew point for a gas containing sulfur dioxide
or sulfur trioxide would be somewhat higher than that computed above.
Table 2-3. Saturation Temperatures and Pressures for Water*
Temperature, Pressure, Temperature, Pressure,

F psia F psia
32 0.08854 100 0.9492

35 0.09995 105 1.1016
40 0.12170 110 1.2748
45 0.14752 115 1.4709
50 0.17811 120 1.6924
55 0.2141 125 1.9420

60 0.2563 130 2.2225
65 0.3056 135 2.5370
70 0.3631 140 2.8886
75 0.4298 145 3.281
80 0.5069 150 3.718

85 0.5959 155 4.203
90 0.6982 160 4.741
95 0.8153 165 5.335
•Abstracted by permission from J. H. Keenan and F. G. Keyes, "Thermodynamic Properties of

Steam,"John Wiley & Sons, Inc., New York. >.
Flue-gas Analysis. While there are many instruments for analyzing

the products of combustion, the Orsat apparatus is by far the most com
monly used. A sketch showing the essential features of such an appa
ratus appears in Fig. 2-3. It consists of three absorption pipettes and a
measuring burette housed in a wooden frame. A leveling bottle is used
to create a hydrostatic head to force the flue-gas sample from the measur
ing burette into the absorption pipettes and back again. Each of the
absorption pipettes contains a chemical reagent which absorbs one of the
constituents of the flue gas. The decrease in volume of the flue-gas
Fig. 2-3. Orsat apparatus.
sample, after one of the component gases has been absorbed in a pipette,
gives the percentage by volume of that component.
To analyze a sample of flue gas, more than 100 volumes of the gas are
drawn into the measuring burette by lowering the leveling bottle.
Enough of the sample is then expelled to the atmosphere through the
three-way petcock to give exactly a 100-vol sample at atmospheric
pressure in the measuring burette. The petcock leading to the carbon
dioxide absorption pipette is next opened, and the entire sample of flue
gas is forced into the pipette by raising the leveling bottle. The gas
sample is allowed to stand in this pipette, which contains a 20 per cent
solution of potassium hydroxide, about 30 sec to allow most of the carbon

dioxide to be absorbed. The leveling bottle is then lowered until the
reagent in the pipette rises to the original level. The number of volumes
of the flue-gas sample absorbed in the first pipette can then be read on
the scale of the measuring burette. The water level in the leveling bottle
is held equal to that in the measuring burette to ensure that the gas in the
burette is at atmospheric pressure when a reading is taken. The flue-gas
sample is forced back into the carbon dioxide absorption pipette again to
make certain all of the carbon dioxide has been absorbed. This process
is repeated until two identical successive readings are obtained on the
measuring burette.
The remainder of the flue-gas sample is then forced into the second
pipette. Here an alkaline solution of pyrogallic acid absorbs the oxygen.
Just as before, the gas sample is forced into the oxygen absorption pipette
several times to ensure complete removal of the oxygen. The reading
obtained on the measuring burette is now the sum of the percentage of
carbon dioxide and the percentage of oxygen in the original flue gas. The
percentage of oxygen is found by subtraction.
Finally, the sample is forced into the third pipette where an acid solu
tion of cuprous chloride absorbs the carbon monoxide from the gas sample.
Again the volume of gas absorbed is measured in the burette.
When the percentages of carbon dioxide, oxygen, and carbon monoxide
have been determined, the remainder of the gas sample is assumed to be
nitrogen.
As shown in Fig. 2-3, liquid reagents in the carbon monoxide and
oxygen absorption pipettes are protected by an extra pipette which acts
as a water seal. Both of these solutions absorb oxygen and, if they were
exposed to the atmosphere, would soon be saturated with oxygen.
The Orsat apparatus gives an analysis of the flue gas in per cent by
volume on a dry basis. Actually the gas sample is saturated with water
vapor at all times, but the water jacket surrounding the measuring
burette keeps the sample at room temperature throughout the analysis.
This results in a constant partial pressure of water vapor. Thus, while
there is always water vapor present, when the volume of gas sample
decreases, the quantity of water vapor also is decreased to maintain the
same partial pressure of water vapor. Therefore, the water vapor has no
effect on the analysis, and the results appear exactly as they would if
dry gas had been analyzed.
If a more complete analysis of the combustion products is desired, other
absorption pipettes may be added to the Orsat to enable accurate meas
urement of additional constituents, such as hydrogen or methane, which
may be present in small amounts under certain conditions.
Automatic instruments are on the market which give continuous read

ings of the percentage of carbon dioxide, carbon monoxide, oxygen, or
combustible. Many of these are recording instruments. While they
operate on several different principles, practically all of them give the
analysis in per cent by volume on the dry basis.
Combustion of Gaseous Fuels. Gaseous fuels are usually mixtures of
several combustible components. The composition of these fuels is
always given in per cent by volume of each constituent in the mixture.
Table 1-9 lists typical compositions of a number of gaseous fuels. The
principal constituents of gaseous fuels are the lighter hydrocarbons,
hydrogen, and carbon monoxide. The latter two constituents are present
in varying amounts in all manufactured gases, but not in petroleum gases.
Since the compositions of gases are given in per cent by volume, the
number of moles of each constituent in 100 moles of gaseous fuel is the
same as the per cent by volume. It is convenient to choose 100 moles of
fuel as the basis for the computation, and find the amount of air required
to burn 100 moles of the gas and the number of moles of combustion
products formed.
Example 4. A gaseous fuel of the following composition by volume

burns with 10 per cent excess air.
CO2 = 5.5% CO = 38.3% CH4 = 0.4%

O2 = 0.1% H2 = 52.8% N2 = 2.9%
Flue gases leave the heater at 750 F and 14.7 psia. Determine the com
position of the flue gases, the air/fuel ratio by volume, and the volume of
flue gases formed per cubic foot of dry fuel gas supplied at 60 F and
14.7 psia.
Solution. When working a problem of this type, we usually find it
advantageous to arrange the calculations in tabular form. From the
basic combustion equations, determine the moles of oxygen theoretically
required to burn each constituent and the moles of combustion products
which result. Thus, from the basic equation, CO + Ô2 —* CO2, it is
seen that 19.15 moles of oxygen are theoretically required to burn 38.3
moles of carbon monoxide to 38.3 moles of carbon dioxide. The 0.1
mole of oxygen present in the fuel will enter into the reaction once the
fuel is heated to combustion temperature. The 0.1 mole of oxygen is
therefore entered in the
" Moles of oxygen required" column as a negative
value, indicating that it is not required from the atmosphere. The
carbon dioxide and nitrogen are inert constituents and are carried over to
the combustion products unchanged.
Basis: 100 moles fuel gas
Moles of combustion products

Moles
Moles O2 req'd.
fuel
C02 N2 H20
CO2 5.5 5.5

02 0.1 -0.1
CO 38.3 19.15 38.3
H, 52.8 26.4 52.8
CH4 0.4 0.8 0.4 0.8
N2 2.9 2.9
100.0 46 . 25 theor. 44.2 2.9 53.6
+4.63 (10% excess) 4.63* 191. 3f
50 . 88 supplied 194.2
♦Excess oxygen resulting from the 10 per cent excess air.

fNi with the air = 50.88 X 3.76 = 191.3 moles.
Orsat analysis,
Total flue g ases, moles Vol % (wet) Dry flue g ases, moles
vol % (dry)
CO2 = 44.2 14.9 C02 = 44.2 18.2

O2 = 4.63 1.6 02 = 4.63 1.9
N2 = 194.2 65.4 N2 = 194.2 79.9
H2O = 53.6 18.1
296.63 100.0 243.03 100.0
As shown in the above example, the dry flue gases are determined by
omitting the water vapor from the combustion products.
—b vol — * 100/21 moles air _ 2 ' 42

moles air
F * 100 moles gas moles fuel
cu ft air
= 2.42
cu ft fuel
750 + 460
296.6 X X 359
Vol flue gases 32 + 460 = 6.9
cu ft flue gas, 750 F
Cu ft dry fuel 60 + 460 cu ft fuel gas, 60 F
100 X X 359
32 + 460
Combustion of Coal. There are several methods of reporting the com

position and burning characteristics of coal as discussed in Chap. 1.
However, to make stoichiometric calculations, the ultimate analysis is all
that need be known. The ultimate analysis is an accurate chemical
analysis which lists the per cents by weight of the five principal elements —
carbon, hydrogen, oxygen, nitrogen, and sulfur — and the ash in the fuel.
Since this analysis is given on the weight basis, it is convenient to choose
100 lb of coal as the basis for the combustion calculations. However, in

order readily to determine the oxygen required for combustion and the
moles of combustion products, it is necessary to know the number of
moles of combustible elements in the 100 lb of coal. Therefore, the
weights of the constituents are divided by their respective molecular
weights to find the number of moles of the elements present. From this
point, the problem is handled similarly to the combustion of a gaseous
fuel.
Example 5. Determine the flue-gas analysis, the air/fuel ratio by

weight, and the volume of combustion products at 500 F when coal of the
following composition burns with 50 per cent excess air. Assume that
all fuel is burned and that dry air is used for combustion.
Carbon = Nitrogen = Ash =

72% 1.4% 11.0%
Hydrogen = Oxygen = Sulfur =
6% 8% 1.6%
Solution:
Basis: 100 lb of coal
Moles of products/
Lb/100 Moles/ 100 Moles 02 theor. 100 lb coal
lb coal Hi coal req'd. /1001b coal
CO* H20 so2 N2 02
Carbon .... 72.0 + 12 6.0 6.0 6.0

Hydrogen . . 6.0 + 2 3.0 1.5 3.0
Oxygen .... 8.0 + 32 0.25 -0.25
Nitrogen . . . 1.4 28 0.05 0.05
Sulfur 1 . 6 h- 32 0.05 0.05 0.05
Ash 11.0
100.0 7.30 6.0 3.0 0.05 0.05 0.0
41.17* 3.65f
41.22 3.65
(7.30 X 1.30)3.76 = 41.17 moles Nj with air.

*
t Excess O2 = 0.50 X 7.30 = 3.6S moles.
Analysis of wet flue gas,

Vol % (wet) Orsat analysis Vol % (dry)
moles (wet)
C02 = 6.0 + 53.92 11.13 C02 + S02 = 6.05 + 50.92 11.88

H20 = 3.0 -h 53.92 5.56 02 = 3.65 -h 50.92 7.17
S02 = 0.05 53.92 .09 CO = 0.0 -h 50.92 0.0
N2 = 41.22 -h 53.92 76.45 N2 = 41.22 50.92 80.95
02 = 3.65 -h 53.92 6.77 50.92 100.0
53.92 100.00
Sulfur dioxide is absorbed in the first pipette with the carbon dioxide in
the common Orsat apparatus. For that reason the percentage of carbon
dioxide appearing in the results of the Orsat analysis is actually the sum
of the percentages of carbon dioxide and sulfur dioxide present. The
moles of these two gases are added in the above example and listed as
carbon dioxide in the gas analysis.
A _
~ (7.30
X 1.50 X 100/21 moles air) 29.0 lb air /mole
F 100 lb coal
= 15.12 lb air/lb coal
Vol combustion products = 53.92 moles X 359 X

492
= 37,800
cu ft of flue gases
100 lb coal
Combustion with Insufficient Air. In certain applications of combus

tion, and particularly in spark-ignition engines, it is sometimes desirable
to burn slightly rich mixtures. For example, the air/fuel mixture which
gives maximum power in gasoline engines provides only about 85 to 90
per cent of theoretical air. Thus, with insufficient air, the fuel is not
oxidized completely, and the products of combustion passing out the
exhaust manifold always contain carbon monoxide. Example 6 shows
an approximate method for computing the combustion products from rich
mixtures.
When making this type of calculation, we may assume, first, that all of
the hydrogen will burn to water, next that the carbon will all burn to
carbon monoxide, and finally that any oxygen remaining will unite with
some of the carbon monoxide to form carbon dioxide. This is a logical
assumption since the relative affinities of hydrogen, carbon, and carbon
monoxide for oxygen are in that respective order. As shown in the
example below, it is helpful to find the number of oxygen atoms actually
available for the reactions, and then to run an oxygen balance to determine
the number of oxygen atoms used for each step of the reaction.
Example 6. Compute the composition of the exhaust gases resulting

from the combustion of C8H18 with 85 per cent theoretical air.
Solution. The first step is to find the amount of air theoretically
required for combustion. Writing the theoretical equation for the com
bustion of CsHis,
C8H18 + 12.5O2 + 12.5 X 3.76N2 -» 8CO2 + 9H2O + 47N2

it is found that 12.5 oxygen molecules (or 25 oxygen atoms) are required
for complete combustion of one molecule of the fuel. Here only 85 per
cent of that number, or 0.85 X 25 = 21.25, is

available and, since this is
not enough to complete the reaction, it is necessary to find how far the
combustion will be carried. An oxygen balance is established to deter
mine where each atom goes and how many are left for further combustion.
Oxygen balance
-
atoms available
21 . 25
9.00 to burn 18H to 9H2O

12.25 left over
- 8.00 to burn 8C to 8CO
-
4 . 25 left over
4 .25 to burn 4.25CO to CO2
0.00 left
Assuming the hydrogen gets its share of oxygen atoms first, the number
needed to burn the hydrogen to water vapor is subtracted from the total
number of oxygen atoms available. The number needed to burn the
carbon to carbon monoxide is then deducted. This leaves only 4.25
oxygen atoms to complete the combustion of the 8CO formed. There
fore, all of the CO cannot be burned to CO2, but that reaction will proceed
as long as oxygen is available. As shown, the 4.25 oxygen atoms left
will burn 4.25CO to 4.25CO2. The rest of the 8CO formed in the second
step of the problem remains as CO in the exhaust.
-
The molecules of CO2
in the exhaust would be 4.25 and the CO, 8 4.25, or 3.75. The final
combustion equation for C8Hi8 with 85 per cent theoretical air could then
be written
C8H18 + 0.85 X 12.5O2 + 39.9N2

-» 4.25CO2 + 3.75CO + 9H20 + 39.9N2
The dry analysis of the exhaust gases would be,
Components Moles % by vol, dry
CO2 4.25 8.9

CO 3.75 7.8
N2 39.9 83.3
47.9 100.0
Dissociation of the combustion products at the extremely high tem

peratures encountered in the spark-ignition engine causes an even greater
percentage of CO in the exhaust than is shown by the above example.
This phenomenon is discussed more fully in Chap. 3.
Combustion Analysis from the Orsat. In the previous problems, the

quantity of air supplied for combustion has been specified. Actually the
measurement of the quantities of air entering a furnace or combustion
chamber for most types of combustion equipment is extremely difficult,
and it is seldom that such measurements are made in practice. There
fore, the amount of air entering the combustion process is not usually
known. In contrast, it is possible to analyze the products of combustion
rather easily. From the composition of the combustion products, the
air/fuel ratio used for combustion can be computed. Also the composi
tion of the fuel itself may be analyzed rather closely. From this stand
point, the problems in Examples 7, 8, and 9 are more typical of the type of
stoichiometric analysis required in engineering practice than the pre
ceding examples.
Essentially, the method of attacking a problem of this type involves the
setting up of a balance, or equation, between some quantity leaving the
combustion process and the same quantity entering the reaction. For
example, the amount of carbon appearing in the combustion products
should be equal to the amount of carbon in the original fuel, providing
none is lost during the process. The amount of hydrogen in the original
fuel could similarly be used as a balance except that the Orsat analysis
does not determine the percentage of water vapor, and hydrogen com
pounds do not, therefore, appear in any of the data from the flue-gas
analysis.
The following two examples show the procedure for making a balance
between the carbon in the combustion products, as determined by the
Orsat, and the carbon in the original fuel burned. It will be found that
the approach in each case is similar.
Example 7. A water gas at 70 F and 30 in. Hg is burned with dry air

supplied at 70 F The fuel-gas composition is CO2 = 6.0%,
and 30 in. Hg.
N2 = 5.5%, H2 = 48.0%, O2 = 0.5%, CH4 = 2.0%, CO = 38.0%.
The Orsat analysis of the flue gases showed CO2 = 15.50%, O2 = 4.76%,
and CO = 0.20%. Find the percentage of excess air supplied for com
bustion and the volume of flue gases at 450 F and 14.7 psia formed per
cubic foot of fuel.
Solution. The carbon balance is used to equate the moles of carbon in
the combustion products with the moles of carbon in the original gaseous
fuel. The Orsat analysis shows 15.50 per cent CO2 and 0.20 per cent CO,
or 15.50 + 0.20 = 15.70 per cent of the dry flue gases contain carbon.
The number of moles of carbon in 100 moles of fuel gas is 6 in the CO2,
2 in the CH4, and 38 in the CO, giving a total of 46.
Since the carbon in the dry flue gases must balance the carbon in the
fuel gas,
46 0.1570 X (moles of dry flue gas/100 moles fuel)

Moles dry flue gas _ 46
= 293
100 moles fuel 0.1570
The Orsat analysis shows that 4.76 per cent of the dry flue gases are
oxygen, hence
Moles 02 in flue gases

= 0.0476 X 293 = 13.95
100 moles fuel
All of this oxygen is not excess, because 0.0020 X 293, or 0.586, mole of
C0 is also in the flue gases. This C0 would have required 0.586/2, or
0.293 mole of 02 had it burned completely to C02. The moles of excess
02 are then 13.95 — 0.293 = 13.66. Only that oxygen which would have
appeared in the combustion products had the fuel burned completely to
C02 and H20 is considered as excess. The number of moles of 02 theo
retically. required for combustion can now be computed.
Basis: 100 moles fuel gas
Moles Moles 02 required Moles H20 formed in combustion products
co2 6.0
02 0.5 -0.5
N2 5.5
CH4 2.0 4.0 4.0
H2 48.0 24.0 48.0
CO 38.0 19.0
100.0 46.5 52.0
The per cent of excess air is then

13.66 moles 02 excess
= 29.3%
46.5 moles 02 theoretical
The number of moles of water formed by burning 100 moles of fuel gas is
tabulated above. Hence, total moles of wet flue gases = 293 moles dry
flue gas + 52 moles H20 = 345 moles.
Cu ft flue gases formed

Cu ft fuel
345 moles wet flue gases X 359 X 910/492
5.94
100 moles fuel X 359 X 530/492 X 29.92/30
Example 8. A coal having the following composition was burned in a

furnace.
Carbon = 71.89% Oxygen = 9.01% Sulfur = 1.91%

Hydrogen = 7.66% Nitrogen = 2.68% Ash = 6.85%
The dry refuse taken from the ashpit tested 36.0% carbon and 64.0% ash.
The Orsat analysis gave CO2 = 11.50%, O2 = 5.39%, and CO = 1.32%.
Assuming that dry combustion air entered the furnace at 68 F and
29.26 in. Hg abs, find: (a) percentage of excess air, (b) cubic foot of air
supplied per 100 lb of coal fired.
Solution. In this case a part of the carbon was lost in the ashpit before
it had a chance to burn. The amount lost must be determined. Equat
ing the ash in the coal fired to the ash in the refuse from the ashpit,
Lb ash = 0.64 X (wt of refuse) = 6.85 lb ash/100 lb coal fired

Wt of refuse 6.85
= 10.70 lb
lb coal fired
100 0.64
Wt carbon lost in ashpit _
= 10.70 X 0.36 = 3.85 lb
100 lb coal fired
The carbon lost in the ashpit must be subtracted from the carbon in the
coal fired to find the amount burned.
Basis: 100 lb coal fired
Lb Moles fuel burned Moles O2 theoretically required
Carbon 71.89
-3.85
68.04 h- 12 5.67 5.67
Hydrogen .... 7.66 + 2 3.83 1.91
Oxygen 9.01 32 0.28 -0.28
Nitrogen 2.68 + 28 0.10
Sulfur 1.91 -s- 32 0.06 0.06
Ash 6.85
100.00 7.36
Balancing the carbon in the flue gases with the carbon burned in the
coal, 1.32 + 11.50 = 12.82 per cent of the dry flue gases must contain
the 5.67 moles of carbon burned.
Moles C = 0.1282 X (moles dry flue gas) = 5.67
Moles dry flue gas _ 5.67 _

~ ~
100 lb coal fired 071282
This figure could be determined slightly more accurately by including the

moles of sulfur with the carbon, since the SO2 formed will appear as CO2
in the Orsat analysis. Then
Moles dry flue gas _ 5.67 + 0.06
= 44.7
100 lb coal fired 0.1282
The moles of O2 in the flue gases are
0.0539 X 44.7 = 2.41 moles
Moles of CO equals 0.0132 X 44.7 = 0.59 mole, and the moles of 0S

excess equals 2.41 — = 2.12 moles.
0.59/2
2 12
Excess air = =
28.8%
The volume of air supplied per 100 lb of coal fired equals
7.36 X ^X 1.288 X 359 X X = 17,800 cu ft
Fuel Analysis from Exhaust Composition. When combustion occurs

in compact volumes, during very short times and with little excess or even
insufficient air, combustion products other than those which can be
measured with the Orsat will be present. Such is the case with rockets,
jet engines, spark-ignition and compression-ignition engines, etc. One
technique for obtaining a complete and accurate exhaust analysis utilizes
a number of chemical absorption pipettes in addition to the Orsat-type
apparatus. Other techniques based on varied physical properties of the
exhaust gases may be employed to obtain an accurate and complete
analysis.
Usually small amounts of hydrogen, methane, and other hydrocarbon
compounds will be present in the exhaust from a spark-ignition engine.
Methane is formed as an intermediate product of combustion, and
hydrogen appears because of dissociation of water vapor at the high
temperatures encountered in the cylinder. Because of the very short
time allowed for the mixing of the fuel and air, and for combustion, these
components are present in the exhaust even though they would burn if
enough time were available for the reaction and if there were a homo
geneous mixture of the fuel and air. Figure 8-7 shows the percentages of
the various components present in the exhaust from a spark-ignition
engine with different air/fuel ratios. Note that a small quantity of
oxygen is present even when the mixture is rich, principally because of
the heterogeneous nature of the charge in the cylinder, and dissociation.
Some carbon monoxide is present when excess air is available, for the same
reason. Exhaust analyses from other types of engines are comparable.
The example which follows shows a method for computing the mixture
composition used in an engine, provided the exhaust analysis is known.
This procedure is called the carbon-hydrogen balance method and
accounts for all of the carbon and hydrogen appearing in the exhaust
products. It is especially useful when the composition of the fuel burned

is not known.
Example 9. Determine the air/fuel ratio by weight used in an engine if

the products of combustion show CO2 = 6.0%, O2 = 0.2%, CO = 13.4%,
H2 = 8.0%, and CH4 = 0.5% by volume on a dry basis. Since the con
stituents listed in the exhaust-gas analysis do not total 100 per cent, the
remainder of the gas may be assumed to be N2.
Solution:
Basis: 100 moles exhaust gas
Moles
CO2 = 6.0
O2 = 0.2
CO = 13.4
H2 = 8.0
CH4 = JX5
28.1
N2 = 100 - 28.1 = 71.9 moles
The following combustion equation is written on the basis that 71.9

molesof N2 were used and 71.9/3.76 moles of O2 were involved in the
combustion reaction:
CxH, + O2 + 71.9N, -» 6.0CO2 + 0.2O2 + 13.4CO + 8.0H2

+ 0.5CH4 + 71.9N2 + WH2O
Since thenumbers of carbon and hydrogen atoms in the fuel are unknown,
they are shown as unknowns x and y. A carbon balance is established
to determine the number of C atoms appearing in the exhaust products
(which must also appear as x, the number of C atoms, in the fuel) . Like
wise a hydrogen balance is used to find the original quantity of hydrogen
y from the hydrogen shown in the exhaust products, keeping in mind
that much hydrogen is present in H2O in the exhaust gas which does not
appear in the dry exhaust-gas analysis.
Carbon balance:
x = 6.0 + 13.4 + 0.5 = 19.9 carbon atoms

in in in
C02 CO CH4
Hydrogen balance: It is necessary first to determine the water vapor

formed by finding where all the 71.9/3.76 = 19.12 moles O2 have gone.
19.12 moles 02 = 6.0 + 0.2 + ™

13 4
+
W
y (H20)
W moles H20 formed in combustion products
— 12.44
y = (2 X 12.44) + (2 X 8.0) + (4 X 0.5) = 42.88 hydrogen atoms

in in in
H20 H2 CH,
The ratio of hydrogen atoms to carbon atoms in the fuel is
42.88/19.9 = 2.16
Air ~ (19.12 + 71.9)29.0 jb air

Fuel 19.9 X 12 + 42.88 X 1 lb fuel
Problems
1. A tank contains 40 lb of air, 20 lb of C02 and 30 lb of S02. Compute the volume

of the tank if the total pressure of the mixture is 50 psia at 68 F. What is the partial
pressure of each constituent and the average molecular weight of the mixture?
Ans. 261 cu ft.
2. If 50 lb of air per minute flows through a duct with temperature and pressure
conditions of 200 F and 30 in. Hg abs, respectively, what is the rate of flow in cubic
feet per minute? Ans. 829 cfm.
3. A tank that has a volume of 500 cu ft contains 120 lb of a mixture of C02 and N2.
If a gage attached to the tank indicates a pressure of 20 psig for the mixture and if the
temperature is 80 F, what is the partial pressure of each gas?
Ans. 26.2 and 8.5 psia, respectively.
4. A gasoline can be represented by the formula C7H16. Write the theoretical
combustion equation for this fuel and, assuming that the fuel and air are at the same
temperature and pressure, determine the volume of air required per cubic feet of fuel
vapor and the pounds of air per pound of fuel burned.
5. A kerosene, with a representative formula of CgHi9, is burned with 15 per cent
excess air. Calculate the cubic feet of combustion products at 300 F and 14.5 psia
resulting from the burning of 1 lb of kerosene.
6. Find the air/fuel ratio by weight when benzene, C6H6, burns with theoretical
air. Also calculate the dew point at atmospheric pressure of the combustion products
if an air /fuel ratio of 40 to 1 by weight is used. Ans. 13.3 air /fuel.
7. Ethyl alcohol, C2H60, has a specific gravity of 0.785. How many cubic feet of
air per gallon of liquid fuel are used when the ethyl alcohol burns with 75 per cent
excess air, if the air for combustion is at 70 F and 14.2 psia?
8. Liquid cetane, Ci6H34, has a specific gravity of 0.744. How many cubic feet of
air per gallon of liquid fuel are present when the cetane is burned in an engine with
200 per cent excess air? The air used for combustion is at a temperature of 70 F and
14.5 psia. Also find the average molecular weights of the initial air /fuel mixture and
the final combustion products.
9. Compute the theoretical air/fuel ratio by weight for the combustion of C7H8.
Write the theoretical combustion equation and determine also the weight of H20
formed in the combustion products per pound of fuel burned.
10. It is found that a liquid fuel of composition carbon 87 per cent, hydrogen
13 per cent by weight, can be burned in a spark-ignition engine if air/fuel ratios
between the limits of 12 to 1 and 21 to 1 by weight are admitted to the cylinders.

Calculate:
a. The percentage of theoretical air supplied at the combustion limit on the rich
side.
6. The composition of the exhaust gases in per cent by volume when the richest
mixture is burned.
c. The average molecular weight of the combustion products from both the lean
and the rich mixtures.
11. A gasoline having a composition 15.2 per cent hydrogen and 84.8 per cent
carbon by weight is burned in an engine. Write the combustion equation and deter
mine the volumetric analysis of the combustion products if the fuel burns with 15 per
cent excess air. Ans. C02 = 11.18%.
12. Calculate the following items for the burning of octane, CsHu, with 10 per cent
excess air:
a. Air /fuel ratio by weight and volume.
b. Pounds of dry exhaust gas formed per pound of fuel burned.
c. Moles of O2 in exhaust gas per pound of fuel burned.
d. Pounds of H20 in exhaust gas per pound of fuel burned.
e. Volume of exhaust gas in cubic feet at 500 F and 14.7 psia per gallon of liquid
fuel burned (sp gr of fuel = 0.7). Ans. (6) 16.18 lb; (e) 2510 cu ft/gal.
13. A liquid fuel tests H/C = 0.169 by weight. If this fuel is burned in an engine
with an air/fuel ratio of 17 by weight, determine the wet and dry volumetric analysis
of the exhaust gases and the percentage of theoretical air used in the combustion.
Find the dew point of the products of combustion and a representative molecular
formula for the fuel.
14. A liquid petroleum fuel having a ratio of H/C by weight of 0.174 is burned in a
heater with 20 per cent excess air. Determine:
o. Air /fuel ratio by weight.
b. Weight of water formed by combustion per pound of fuel burned.
c. Volume of flue gases at 500 F and 28 in. Hg per pound of fuel burned.
16. A liquid fuel having an H/C ratio of 0.171 by weight is burned with an air/fuel
ratio of 19 to 1 by weight. Compute:
a. Pounds of H20 formed per pound of fuel.
b. Cubic feet of oxygen in the combustion products per pound of fuel burned, if
the combustion gases leave at 450 F and a pressure of 16 psia.
c. Give the average molecular formula for the fuel.
16. If pentane, CsHn, is burned with a 20 to 1 air/fuel ratio by weight, calculate:
a. Per cent excess air used.
b. Volumetric analysis of the flue gases on a wet basis.
c. Orsat analysis of the combustion products.
d. Per cent by weight of each constituent in the wet flue gas.
e. Dew point of the flue gases.
/. Volume of air at 80 F and 14.7 psia supplied per gallon of liquid fuel, if the
specific gravity of the fuel is 0.626.
17. Natural gas having the following composition by volume:
CO, = 0.4% C2H6 = 4.1% CH4=92.1% N2 = 3.4%
burns with 30 per cent excess air. Determine the volumetric analysis of the flue
gases on a wet and a dry basis; calculate the volume of dry air at 70 F and 27 in. Hg
used to burn 1000 cu ft of gas at 68 F and 29.92 in. Hg; and find the dew point and
the average molecular weight of the combustion products.
Ans. 13,620 cu ft air per 1000 cu ft gas.
18. A gas furnace requires a heat input of 500,000 Btu/hr. Gas is supplied at
70 F and 31 in. Hg abs. Twenty per cent excess air is available for combustion at
70 F and 14.7 psia. The gas has a composition in per cent by volume as follows:
CH4 = 90.0% C2H6 = 5.0% N, = 5.0%
The net heating value of the fuel gas is 1002 Btu/cu ft of gas at 60 F and 30.0 in. Hg
abs pressure. Compute:
a. Cubic feet of gas per hour under the supply conditions.
6. Cubic feet per minute of air required.
c.Orsat analysis of the flue gases.
d. Dew point of the combustion products.
19. A gas of the following composition in per cent by volume is burned in a heater
with 25 per cent excess air:
CO2 = 6.5% CH4 = 77.5% C2H6 = 16.0%
The air for combustion is supplied at a pressure of 14.7 psia and 90 F while the fuel
gas enters at 1 psig and 60 F. Calculate:
a. Volume of air supplied for combustion in cubic feet per cubic foot of gas supplied.
b. The cubic feet of fuel gas per minute required to supply 200,000 Btu/hr heat
release if the heating value of the fuel is given as 1073 Btu /cu ft of gas at 70 F and
30.0 in. Hg.
20. gas having the following composition
A coke-oven in per cent by volume is
burned in a furnace with 15 per cent excess air:
CO2 = 2.2% CO = 6.3% C2H6 = 2.0%

02 = 0.8% H2 = 46.5% C3H8 = 2.0%
N2 = 8.1% CH4 = 32.1%
Calculate:
a. The Btu per cubic foot of combustible air /gas mixture at 80 F and 14.2 psia, if
the fuel gas has a heating value of 509 Btu/cu ft at 60 F and 30.0 in Hg.
b. The weight of combustion products formed per Btu of heat released.
21. A mixture of 4.7 per cent propane, C3H8, by volume in air flows through the
tube of a bunsen burner. Calculate the percentage of theoretical air in the mixture.
22. Natural gas at 80 F and 14.7 psia, relative humidity of 80 per cent, is compressed
to 300 psia and the gas cooled back to 80 F. How many pounds of water would be
condensed per 1,000,000 cu ft of gas in the original state?
23. Compute the weight of air theoretically required to burn 1 lb of coal having the
following composition:
Carbon = 66.10% Hydrogen = 6.03% Oxygen = 11.91%

Nitrogen = 1.04% Sulfur = 2.40% Ash = 12.52%
What would be the Orsat analysis of the dry flue gases produced?
24. If 20 per cent excess
air were supplied for combustion of the coal in Prob. 23
and the coal were burned at a rate of 3 tons/hr, compute the cfm capacity required
of a fan used to supply the combustion air. Assume the air to be dry and at a pressure
of 14.7 psia and temperature of 70 F.
26. Ten pounds of coal per minute having an ultimate analysis as follows is com
pletely burned in a small power boiler:

Twenty-five per cent excess air is supplied at 70 F and 28 in. Hg. The flue gas
leaving the unit is at 550 F and 28 in. Hg. Compute:
a. Air /fuel ratio by weight.
b. Pounds of C02 formed per minute.
c. Volume of flue gas leaving unit, in cfm.

d. Per cent oxygen in the dry flue gas.
e. Dew point of the combustion products at atmospheric pressure.
26. A coal of the following ultimate analysis is burned in a furnace:

a. Compute the cubic feet of air entering the furnace per 100 lb of coal fired if the
air is at 30 in. Hg abs pressure and 70 F, and 50 per cent excess air is supplied for
combustion.
b. Compute the Orsat analysis of the flue gases.
27. Two hundred pounds of coal per minute having the following ultimate is burned
in a boiler:
a. What would be the rating in cubic feet per minute for a forced-draft fan designed
air at 90 F and 14.7 psia to this boiler?
to supply 35 per cent excess
6. How many pounds per minute of dry flue gases pass out the stack if the Orsat
analysis reads: CO2 = 11.50%, 02 = 5.39%, and CO = 1.32%? Assume that all
of the carbon is burned.
28. A gas of the following composition in per cent by volume is burned in a furnace:
CO = 21% H2 = 31% N, = 48%
The Orsat analysis of the combustion products reads: C02 = 8.8%, O2 = 6.3%, and
CO = 0.0%. Compute the percentage of excess air used for combustion.
29. Natural gas with an analysis as follows: CO2 = 0.4%, CH4 = 97.3%, N2 = 2.3%,
is burned in a furnace. The Orsat analysis of the flue gases shows: C02 = 8.5%,
02 = 5.6%, CO = 0.1%, and N2 = 85.8%. Find the percentage of excess air
supplied to the furnace.
30. An underfeed stoker feeds coal to a boiler at the rate of 250 lb/hr. The ultimate
analysis of the coal is:
Carbon = 75.2% Oxygen = 7.4% Sulfur =

2.1%
The dry refuse removed from the ashpit shows a combustible content of 20 per cent.
An Orsat analysis of the flue gases gives the following composition:
C02 = 11.5% 02 = 7.9% CO = 1.0%

With the above data, calculate:

a. Percentage of heat lost due to unburned combustible in the refuse if the heating
value of the coal is 12,580 Btu/lb.
6. Cubic feet of air supplied per minute if the pressure is 14.65 psia at 70 F.
c. Total volume of flue gases flowing through the stack per minute if the tempera
ture is 450 F and pressure is 14.7 psia.
31. Write the combustion equation for C8H18 if 90 per cent theoretical air is sup
plied. Calculate the dry analysis of the exhaust products.
32. What percentage of CO is present in the exhaust of an engine if benzene, C6Hi,
is burned with an air /fuel ratio by weight of 9 to 1? Also find the dew point of the
exhaust gases for this situation.
33. A liquid fuel tests H/C = 0.169 by weight. If this fuel is burned in an engine
with an air/fuel ratio of 13 by weight, determine the wet and dry volumetric analysis
of the exhaust gases and the percentage of theoretical air used in the combustion.
Find the dew point of the products of combustion and the average molecular formula
for the fuel. Ans. 87.8% theoretical air, CH2.o3 (or any multiple).
34. Compute the air /fuel ratio by weight used in an engine if the exhaust-gas
analysis in per cent by volume dry shows: C02 = 12.4%, 02 = 3.2%, CO = 0.1%,
H2 = 0.2%, and CH4 = 0.0%. Find also the atomic ratio of H/C of the fuel used.
Ans. Air /fuel = 17.4, H/C = 2.18.
35. Same as above except the exhaust analysis is CO2 = 8.8%, CO = 9.8%,
O2 = 0.3%, H2 = 4.3%, CH4 = 0.3%.
References
Faires, V. M., "Applied Thermodynamics," 2d ed., The Macmillan Company,

New York, 1947.
Griswold, J., "Fuels, Combustion, and Furnaces," McGraw-Hill Book Company,
Inc., New York, 1946.
Kowalke, O. L., "Fundamentals of Chemical Process Calculations," The Macmillan
Company, New York, 1947.
Lichty, L. C, "Thermodynamics," 2d ed., McGraw-Hill Book Company, Inc., New
York, 1948.
Obert, E. F., "Thermodynamics," McGraw-Hill Book Company, Inc., New York,
1948.
Obert, E. F., "Internal Combustion Engines," 2d ed., International Textbook Com
pany, Scranton, Pa., 1950.
CHAPTER 3
THERMOCHEMICAL ANALYSIS
Many of the important calculations concerning combustion processes

involve the heat released from the reactions and the temperatures attained
by the products of combustion. Before these determinations can be
made, certain fundamental principles of thermodynamics must be under
stood. A review of some of these concepts follows.
The principal thermodynamic functions are pressure, P; temperature,
T; volume, V; internal energy, E; entropy, S; and enthalpy, H. The
total enthalpy may be defined as*
A change of total enthalpy is a function of the heat capacity and the

temperature change only. Total temperatures are used in calculating
total enthalpy changes. The total temperature is the static temperature
plus the theoretical temperature rise which would be realized by converting
isentropically the kinetic energy of a moving fluid into a higher pressure
and higher temperature fluid at zero velocity. A thermometer, thermo
couple, or any other temperature-measuring device reads very close to
the total temperature when placed in a moving fluid. For example, the
actual observed temperature for a gas stream having a total temperature
of 2000 F moving at 500 ft/sec would be between 1997 and 1999 F.
Simple correction factors may be applied to obtain the actual total tem
perature from the observed temperature. In nonflow processes, there is
no kinetic energy, and the total temperature and the static temper
ature are equal. Thus, with this concept of enthalpy, the kinetic
energy effects are always included in the calculations with no extra
complications because total temperatures, which are very near the actual
observed temperatures, are used in the computations.
Combustion calculations frequently necessitate the use of the mole
concept, yet engineers generally prefer to have the final results in terms
of the weight of a substance. To provide a thorough training in making
computations on both the weight and the molal bases, examples and
problems are worked using both. To convert from one basis to the other,
only the molecular-weight factor is needed.
*
Symbols are defined at the end of this chapter.
69
FUNDAMENTAL RELATIONSHIPS
From the nonflow energy equation,
Q = AE + W
for the constant-volume process, where W = jP dV/J = 0,
Q = AE
Thus, the heat transferred Q in a constant-volume process is equal to the
j
change of internal energy during the process. Likewise in the nonflow
constant-pressure process, where W —

—j—
= ^*^2
j— — ,
_(,1+^)_(,, + P£)
Substituting the total-enthalpy expression into the preceding equation,
it follows that for the nonflow constant-pressure process, where the kinetic
energy is zero,
Q = H, - Hx = AH
Hence, in the constant-pressure process, the transferred heat is equal to

the total enthalpy change during the process. This latter derivation can
be made also from the general energy equation for any steady-flow proc
ess. Thus,
Ifthe change of potential energy is small compared to the other energy

quantities and if no work is done during the process, the equation can be
written
Again substituting total enthalpy, H = E + (PV/J) + (wuz/2gj),
Q = H2
- H, = AH
In summary, for a constant-volume process,
(1) Q = AE
and for a constant-pressure process or a steady-flow process,
(2) Q = AH
THERMOCHEMICAL ANALYSIS 71
Note that these derivations are made from the general energy equation and
the nonflow energy equation so that the relationships will apply equally
well to real fluids, to the so-called perfect gas, and to irreversible as well as
reversible processes.
Specific Heat. The specific heat, also often called heat capacity, of a
substance is defined to be the amount of heat required to raise a unit
quantity of the substance one unit of temperature. In equation form,
n
C =
dQ dQ
TT-TTF or
N dT w dT
where C = specific heat or heat capacity

w = weight of substance, lb
N = number of moles of substance
From above, it can be seen the units for specific heat may be Btu/
(mole)(R), or Btu/ (lb) (R). The specific heat per pound is equal to the
specific heat per mole divided by the molecular weight of the medium.
Since Q is not a thermodynamic state property but varies with the type
of process, the value of C will depend on the process as well as on the state
of the substance being heated. There are an infinite number of different
specific heats which could be determined, but only those values for
constant-volume and constant-pressure processes are important.
For a nonflow constant-pressure or a steady-flow process
Q = NfCp dT = AH
or on a per mole basis,
Q = fCp dT = Ah
Therefore,
dh = Cp dT
Cp = or to be precise Cp =
since dh is not an exact derivative of temperature for real gases but varies
slightly with pressure changes. Likewise, for a constant-volume process,
on a per mole basis,
Q = fCv dT = Ae
\dTjv
These equations show the importance of specific heat for determining the
thermodynamic properties of internal energy and enthalpy. The latter
two properties are likewise useful in computing the energy changes
associated with thermodynamic processes.
Methods of Determining Specific Heat. The specific heat for a perfect

gas with monatomic molecules is constant at all temperatures and can be
calculated from equations based on the kinetic theory of gases. How
ever, for real gases the heat capacities do not remain constant and must
be determined by experimental means. As this subject is especially
fundamental to thermodynamics, much research has been performed, and
reliable data are available, concerning the specific heat values for different
substances. Although there are varying techniques which are followed
to perform the actual test work, the methods of experiment may be
classified under two headings : the thermal methods and the spectroscopic
method.
In the thermal methods, a given quantity of the medium being tested
is subjected to a known source of heat, such as an electric resistor, and
the temperature rise of the medium is noted.* From these data an
instantaneous value of specific heat can be determined which is
r = dQ
NdT
or the actual heat required to raise 1 mole of the substance 1 deg at the
temperature of the test. The values for instantaneous specific heat are not
constant at all temperatures for real gases. A plot of the experimental
instantaneous values of Cp versus temperature is made and the best curve
drawn through the test points. A mathematical equation for the line is
determined. This equation is usually in the form of Cp = a + bT + eT2
or Cp = a + (b/T) + (e/T2), where the constants a, b, and e are chosen
to make the plot of the equation fall as close as possible to the best curve
through the test points. Instantaneous values of specific heat for several
gases are plotted in Fig. 3-1.
The spectroscopic method for determining specific heats has come into
use more during the past few years because of the greater accuracy which
can be attained, especially in the higher temperature ranges. Relation
ships derived from quantum mechanics are used to calculate the instan
taneous specific heats from the spectroscopic data observed. Again a
mathematical equation is chosen which follows the test values as closely
as possible. Equations for specific heats as reported by R. L. Sweigert
and M. W. Beardsley of the Georgia School of Technology are given in
Table 3-1.
Reasons for Variation of Specific Heat with Temperature. As stated
previously, C„ is a function of the internal energy, and Cp is a function of
*
For a more complete explanation of experimental procedure to be followed in the
thermal methods, see M. W. Zemansky, "Heat and Thermodynamics," 2d ed.,
McGraw-Hill Book Company, Inc., New York, 1943.
enthalpy which in turn makes it a function of the internal energy. The

internal energy of a gas is dependent upon the state of molecular activity.
The individual molecules of a gas may possess kinetic energy due to
translatory or straight-line motion of the entire molecule, to rotational
motion of the atoms within the molecule about the center of gravity of
16 1 ■
—i
1 1 1 1 1 —i
3000 4000 5000 6000 7000

Temperature, R
Fig. 3-1. Variation of instantaneous molal specific heats with temperature.
the molecule, and to vibrational motion of those atoms with respect to

each other. The greater the sum of these energies, the greater the
internal energy of the gas.
The translatory motion of the molecules is the only one of the three
forms of molecular kinetic energy which is proportional to the tempera
ture. When rotational and vibrational motion of the atoms within the
molecule are absent, such as in a monatomic gas, a given quantity of heat
energy added to the gas increases the translatory motion of the molecules
in direct proportion to the temperature rise. As the translatory motion

of the molecules is proportional to the temperature rise at all tempera
tures, the values of specific heat for a monatomic gas are essentially
constant.
Table 3-1. Instantaneous Values of Specific Heat*
(Based on Spectroscopic data at zero pressure)
Max
Gas or Range, R
Equation, Cp in Btu/(lb mole)(R) error,
vapor (
= 460 + F)
%
540-5000 1.1
02
. n.„5 - ™ + J- S + 5000-9000 0.3
3.47 X 10' 1.16 X 106

N2 540-9000 1.7
CO CP - 9.46 - 3 29 * 103
+
1 07 * 106
540-9000 1.1
Cv - 5.76 +
0.578
7 +
20
540-4000 0.8
100Q ^_
H2
Wor+^-SS^-4^
= 5-76 + 4000-9000 1.4
H20 C, = 19.86 540-5400 1.8

--^+Zf>
C02 Cr = 16.2
- 653 * 103
+
106
540-6300 0.8
CH4 Cp = 4.52 + 0.0073771 540-1500 1.2
C2H4 Cp = 4.23 + 0.01 17771 350-1100 1.5
C2H6 Cp = 4.01 + 0.016367' 400-1100 1.5
CijHi8 Cp = 7.92 + 0.0601 T 400-1100 Est. 4
C12H26 CP = 8.68 + 0.08897/ 400-1100 Est. 4
* K. L. Sweigert and M. W. Beardsley, Georgia School Technol., Bull. 2, 1938.
The rotational motion of the atoms within the molecule starts at very
low temperatures and increases slightly throughout any temperature
rise. In contrast, vibrations between atoms do not exist appreciably in
most gases until temperatures well above room temperature have been
THEHMOCHEMICAL ANALYSIS 75
reached. Then the vibrations increase at rates dependent upon the

temperature and the particular molecular structure involved. At very
high temperatures there may be less increase in rotational and vibra
tional motion than at lower temperatures for a corresponding tempera
ture rise. Since additional heat energy is required to activate the rota
tions and vibrations, the instantaneous specific heat of such gases must
vary with the temperature.
Effect of Pressure on Specific Heat. The internal energy of a gas is
affected by the potential energy of the molecules, or the attractive and
0.31
1000 2000 3000 4000

Temperature, F
Fig. 3-2. The' effect of temperature on Cv, C„, and k for dry air at various pressures.
(Ellenwood, Kulik, and Gay, The Specific Heats of Certain Gases over Wide Ranges of Pres
sures and Temperatures, Cornell Univ. Bull. 30, October, 1942.)
repulsive forces which the individual molecules exert on each other. In

the so-called perfect gas, it is assumed there are no intermolecular attrac
tions or repulsions, hence the pressure has no effect upon the specific heat,
internal energy, or enthalpy functions. In all real gases there are such
forces acting between molecules, and the pressure affects the values of
specific heat to the extent that a change of pressure changes the relative
distance between molecules which in turn varies the magnitude of the
intermolecular force fields.
Figure 3-2 shows the variation of the specific heat of air with tempera
ture at different pressures. The effect of pressure on other common gases
is similar. As the pressure approaches zero, the molecules of the gas are
farther apart and their effect on each other is less. Therefore, at low
pressures most real gases follow very closely the perfect-gas law at tem
peratures well above the critical. The equations for specific heat gen
erally used are based upon experimental values calculated either at zero
pressure or at 1 atm of pressure.
As shown in Fig. 3-2, the error involved in neglecting the effect of
pressure on specific heat at high pressures and high temperatures is small,
about 2 per cent. At temperatures less than 1000 F and at very high
pressures, this error is appreciable.
However, in the usual situations
encountered, the pressure is not
high at low temperatures. There
fore, little error is involved in
neglecting the effect of pressure on
specific heat for real gases in most
engineering applications.
Computations Using Variable
Temperature, R Specific Heat. Once the relation
Fig. 3-3. Relationship between instan
taneous and mean specific heats.
ship between Cp and temperature
has been established, the heat trans
ferred in a constant-pressure process can be determined by integrating the
expression,
" "
(3) CpdT
In Fig. 3-3 the area 1-2-b-a represents the transferred heat Q in a constant-
pressure process.
This integration tends to become very laborious when many calcula
tions are involved, as the equations in Table 3-1 will verify. It is simpler
to establish a mean value of specific heat between the temperature limits
concerned so that a simple equation
Q = NCpm(T* - Ti)
can be used to determine Q accurately. The area e-f-b-a in Fig. 3-3 would
represent Q in this case, and Cpm must be of such a value that the two
areas are equal. Hence,
Q = N [T'cpdT =
NCîT* - 7\)
or,
(4)
I 2
1
—
f'
i 1 jTi dT
Tables or graphs of Cpm may be calculated, such as Fig. 3-4, and the values
may be used to shorten calculations considerably.
Thermodynamic Tables. Tables relating the thermodynamic proper

ties of substances have been constructed based on the fundamental
thermodynamic relationships.4* When such tables are complete, accu
rate calculations may be made with a minimum of tedious work by choos
ing the appropriate value of internal energy, enthalpy, or other thermo
dynamic property directly from them. Table 3-2 lists enthalpy values of
different gases in Btu per pound mole of gas. Absolute zero has been
chosen as the base temperature for Table 3-2, and a value of enthalpy
listed at a given temperature represents the heat required to heat at con
stant pressure a gas from its standard state at absolute zero to the
temperature indicated. Values of enthalpy may be computed using any
other temperature as a base by subtracting the enthalpy value at the
desired base temperature from the higher value. Thus, the enthalpy at
3000 R above a temperature of 0 R may be converted into enthalpy above
a base of 400 R by the relationship
#3000-400 = #3000-0 #400-0
where enthalpy value at 3000 R above a base of 400 R, etc.

#3000-400 =
Enthalpy values (based on static temperatures) may be used to deter

mine values of internal energy with the relationship
h = e +
T
Since most real gases approximate ideal gases, especially at high tempera
tures and low pressures,
, RT
e =
h--r
8 Example 1. Determine the heat added to 1 mole of C02 at constant pressure to
raise the temperature from 140 to 3140 F by use of the equation for instantaneous Cp,
-™
the graph of Cpm, and the enthalpy table.
Solution. Using the equation for Cp in terms of temperature from Table 3-1,
« - ' -
CO" +-#ii>
- -
16.271 6.53 X In T
1.41 X lO'peooR
J 600 R
103
T
16.2(3600) - 6.53 X 103 In 3600 - 1 41
:
X
36qQ
10*1
J
- 16.2(600) - 6.53 X 10s In 600 - 1.41 X 10*1
[ 600
- 38,860 Btu/mole
With the graph of Cpm, Fig. 3-4,
Q = AH = NCpm(T2 - Ti)
*
Superior numbers refer to references listed at the end of the chapter.
Table 3-2. Enthalpy Values of Gases*
(Btu/lb mole of gas above zero It)
Temp, Air o, CO CO,

R N2 Hs H20 CH4t CsHiet CiaHnf
300 2,074 S 2,073 G 2,082 0 2,063 S 2,081 9 2,108 2 2,367 6

400 2,767 S 2,769 1 2,777 0 2,710 2 2,776 .9 2,874 7 3, 163 8
500 3,461 3 3,466 2 3,472 2 3,386 1 3,472 1 3,706 2 3,962 0 4,006 13,550 27,079
537 3,718 0 3,725 1 3,729 S 3,640 3 3,729 5 4,030 2 4,258 3 4,312 15,160 30,432
560 3,877 9 3,886 6 3,889 5 3,798 8 3,889 5 4,235 8 4,442 8 4,509 16,200 32 , 624
600 4,156 3 4,168 3 4,167 9 4,075 6 4,168 0 4,600 9 4,764 7 4,851 18,070 36,428
700 4,854 2 4,879 3 4,864 9 4,770 2 4,866 0 5,552 0 5,575 4 5,772 23,180 47, 132
800 5,556 7 5,602 0 5,564 4 5,467 1 5,568 2 6,552 9 6,396 9 6,768 28,900 59 , 008
6,265 6,337 6,268 6,165 6,276 7,597 7,230 7,838 35,170 72,028
1 ('.
900 1 9 1 .3 4
9 9
1000 6,981 7,087 6,977 6,864 6,992 8,682 8,078 8,990 41,940 86, 135
2
9
5
5
2
1100 7,705 7,850 7,695 7,564 7,716 9,802 8,942 10,222 49,210 101,280
9 9 9 0 0
1 8 5 4 0
9 8 8 0 7
1 1 6 8 8
0 7 9 3 6
1 4 9 8 4
5 8 7 8 6
1200 8,439 8,625 8,420 8,265 8,450 10,955 9,820 11.528 56,900 117,310
1300 9,181 9,412 9,153 8,968 9,194 12,136 10,714 12,907 64 930 134, 160
, ,
1400 9,933 10,210 9,896 9,673 9,948 13,344 11,624 14,356 73 370 151,600
1500 10,694 11,017 10,648 10,381 10,711 14,576 12,551 15,872 81,990 169,820
1600 11,464 832 11,409 11,092 11,483 15,829 13 494 17,450 90,870 188,620
1
1
0 5 0 4 9
0 1 8 6 5
4 6 :i 9 7
1 9 8 4 5
2 8 2 3 4
4 3 1 4 5
7 8 5 4 0
,
,
1700 12,242 12,655 12,178 807 12,264 17, 101 14,455 19 088 100,200 208,050
1
1
,
, , ,
1800 13,028 13,485 12,956 12,526 13,053 18,391 15,433 20 782 109,800 228,020
1900 13,821 14,322 13,741 13,250 13,849 19,697 16,427 22 539 119,600 248 400
,
2000 14,621 15,164 14 534 13 980 14 653 21,018 17,439 24,304 129,600 269,450
,
,
,
2100 15,428 16,010 15,334 14,714 15,463 22,352 18,466 26 148 139,700 290,970
5 9 2 8 0
8 4 0 4 3
4 7 5 8 9
4 2 8 6 9
2 (1 3 0 7
2 9 1 3 0
3 6 8 4 5
,
2200 16,240 16,862 16,139 15,454 16,279 23,699 19,510 28,041 150,000 312,480
2300 17,058 17,718 16,951 16,199 17,101 25,056 20,570 29,982 160.700 333,300
2400 17,880 18,579 17 767 16,950 17,927 26,424 21,645
,
2500 18,708 19,443 18,589 17,707 18,758 27,801 22,735
2600 19,540 20,311 19,415 18,469 19 594 29 187 23 839

0 2 6 0 0
6 5 8 2 1
7 1 8 0 4
4 8 2 0 3
3 2 0 2 5
5 5 1 4 8
9 5 8 8 7
,
2700 20,376 21,182 20,246 19,237 20,434 30,581 24,957
2800 21,215 22,057 21,081 20,011 21,277 31,982 26,088
2900 22,059 22,936 21,919 20,791 22 123 33,391 27,231
,
3000 22,906 23,817 22,761 21,576 22,973 34 806 28,386

,
3100 23 756 24,702 23,606 22,367 23 826 36,227 29 552

4 9 2 4 3
8 3 0 2 0
2 6 7 7 9
5 0 2 2 8
3 9 0 5 5
9 9 5 1 7
9 7 0 0 8
,
, ,
3200 24 609 25,590 24,455 23 164 24,681 37,654 30,730

,
3300 25,465 26,481 25,306 23,965 25,539 39,086 31,918

3400 26,323 27,375 26,159 24,771 26,399 40,523 33,116
3500 27,185 28,273 27,015 25,582 27,261 41,965 34 323
,
3600 28,049 29 173 27 874 26,398 28,126 43,411 35,540

1 2 9 4 1
9 3 8 2 0
5 1 8 5 5
2 9 3 5 6
1 6 1 5 9
4 8 9 1 4
4 0 0 6 0
,
,
3700 28,915 30,077 28,735 27,218 28,993 44,860 36 765

,
3800 29,783 30,984 29 597 28,042 29 862 46,314 37,998

,
,
3900 30,654 31,893 30,462 28,871 30,732 47,771 39,240

4000 31,526 32,806 31,329 29 703 31,605 49,231 40,489
,
4100 32,401 33,721 32 198 30,539 32,479 50,695 41,745

6 2 2 4 1
9 !) 0 4 4
6 >)5 8 S
7 9 8 4 8
8 0 1 9 5
81 9 1 0
n 1 1 0 1
,
4200 33 278 34,639 33,068 31 ,379 33,354 52 162 43,008

, , ,
,
4300 34,156 35,561 33,939 32,223 34,231 53 632 44,278

4400 35,036 36,485 34,813 33,070 35,109 55 105 45,553
4500 35,918 37,411 35,687 33,921 35,988 56,581 46,835
4600 36 802 38,341 36 563 34,775 36,869 58,059 48,123

7 8 9 (1 3
(1 3 9 l 7
4 0 5 9 6
4 1 8 7 3
8 1 5 1 8
3 9 4 0 7
3 1 6 6 4
,
4700 37,687 39,273 37,441 35,633 37,751 59,541 49,416

4800 38,573 40,208 38,319 36.493 38,633 61,024 50,715
4900 39,462 41,146 39,199 37,356 39,517 62,511 52,019
5000 40,351 42,086 40,079 38,223 40,402 64,000 53,327
5100 41,242 43,029 40,961 39,092 41,288 65 490 54,640

8 2 1 8
2 2 5 5
(1 1 0 9
0 2 3 1
0 3 4 6
0 7 4 3
6 0 3 3
5200 42,134 43,974 41,844 39,965 42,175 66,984 55,957

5300 43,028 44,922 42,728 40,840 43,063 68,479 57,278
5400 43,922 45,900 43,641 41,734 43,981 70, 110 58,644
Abstracted by permission from J. H. Keenan and J. Kaye, "Gas Tables," John Wiley Sons, Inc.,
*
&
New York, 1948.

Data from API Project 44.
t
0 1000 2000 3000 4000 5000
Temperature, F
Kig. 3-4. Mean molal specific heat at constant pressure between 60 F and abscissa temper
ature. (Computed from data from API Project 44.)
The value of Q could be found directly if the graph were plotted for values of Cpm
above a base of 140 F. However, only values between 60 F and the abscissa tem
peratures are plotted. Therefore, it is necessary to find AH between 140 and 60 F
and subtract it from the value of AH between 3140 and 60 F.
Q = AH = Aff3M0-6O
- - JVC2(140 -
—
= JVCi(3140 60) 60)
where Ci = Cpm between 3140 and 60 F

C2 = Cpm between 140 and 60 F
Choosing Cpm values from Fig. 3-4,
Q = 12.84(3140
- - -
= 39,547 - 60) 9.00(140
720 = 38,827
60)
Finally with values of enthalpy from Table 3-2,

— ff6oo R
Q = AH =
= 43,411.0 -
36oo k
4600.9
= 38,810.1 Btu/mole
The slight difference in numerical answers shown by the above three methods is due
in part to the fact that the computations have been carried to more significant figures
than the original fundamental data justify, and in part to the different methods of
interpolation used in developing Fig. 3-4 and Table 3-2.
Mixtures of Gases. During the combustion process in burners, fur

naces, internal-combustion engines, rockets, etc., the thermodynamic
working fluid within the equipment changes from a mixture of fuel and
air to a mixture of combustion products. The heat added to a mixture
of gases is the sum of the quantities added to the several components.
If many calculations are to be made with the same mixture of gases, it
may be convenient to find average thermodynamic properties for the
mixture. The entire mixture may then be handled as a single fluid.
Methods were discussed in Chap. 2 for finding the average molecular
weight and average specific gas constant R' for mixtures. The average
mean specific heat for a mixture may be computed as:
AVlp"
AvC - Q
ntAT
where Q = total heat transferred during a given constant-pressure proc
ess with a temperature change of AT
NT number of moles of gas in mixture
=
The average mean specific heat for mixtures may also be computed as
x x
(Jt
(J£-
Qr.
Av = +
+
cJ)jB +
• •
x
(5) e,m^
where each term represents the mole fraction of particular gas times its
a
mean specific heat for the temperature range.

Example 2. Find the average Cpm for the combustion products of methane with
theoretical air between the temperatures 60 and 3000 F.
Solution. The combustion products from methane are
CH4 + 202 + 2 X 3.76N2 -> C02 + 2H20 + 7.52X2
Choosing values of Cpm from Fig. 3-4 for C02, H20, and N2, between the temperatures
60 and 3000 F,
1 mole CQ2 „ _„ 2H20 7.52

Av n
.
Cpm =
(T^py'+T^ôieTtota-! X 12.76 +
ia52
X 1013 + X 7M
UX52
= 8.75 Btu/(mole)(R)
The average molecular weight of the combustion products is
1 X 44 + 2 X 18 + 7.52 X 28 290.6 lb
= 27 62
"10.52 lb/mole
10.52 moles
Hence,
AV^ =
8^2^
The heat added to the products of combustion of 1 mole of CH4 with stoichiometric
F, could then be found as
air, to heat them from 60 to 3000
Q = NCpm AT = 10.52 X 8.75(3000 - 60)

■ 270'600KU
= wCpm AT = 290.6 X 0.317(3000 ' ni ' - ' aim

*
270,600 Btu
HEATING VALUES OF FUELS3 .. ^ isr.

,. i,*.
. , fHV
'
.. y,*
"y^1
The heating value of a fuel is the energy released per unit quantity of fr
fuel when the combustible is burned and the products of combustion are /
'
cooled back to the initial temperature of the combustible mixture. Heat- y*"*
ing values are given in Btu per mole or Btu per pound for liquid and solid
^fuelgj_ while the values*7or gaseous fuels are given in 'Btu per 'mole or B~tu
per cubic foot. Other names used for the heating value of a fuel are the
heat of combustion, calorific value, and heatof_reaction. Actually the heat
of reaction is a broader term which is applicable to chemical reactions
other than the oxidation of a fuel, but when the term is used in connec
tion with the combustion process, it is synonymous with the heating value
of the fuel.
By burning a small quantity of fuel in a calorimeter, where the heat
release can be measured precisely, heating values and related information
may be checked. There are two types of calorimeters in general use
today, and the specifications for their use are given in the Standards of
the American Society for Testing Materials.
Constant-volume (Bomb) Calorimeter. To determine the heat released
during a constant-volume combustion process, the bomb calorimeter is
used. A diagrammatic sketch of such a device is shown in Fig. 3-5. A

known quantity of fuel is placed in the fuel pan, and the bomb is filled
with a large excess of oxygen at a pressure of 20 to 40 atm to ensure com
plete oxidation of the fuel. The bomb is surrounded by a water jacket
which absorbs the heat liberated when the fuel is ignited by the electrical
ignition wire. The entire calorimeter is insulated so that very little heat
is lost to the atmosphere, and the temperature rise of the known weight of
. Thermometer
Stirring
device
with motor 8
Fig. 3-5. Diagrammatic sketch of a constant-volume calorimeter.
water and metal parts surrounding the bomb makes it possible to deter
mine the heat liberated during the chemical reaction. While it is impos
sible to cool the products of combustion completely back to the initial
temperature, only a small temperature rise occurs in the water jacket,
and the heat remaining in the products which is not transferred to the
cooling water is negligible.
Constant-pressure Calorimeter. For the steady-flow constant-pres
sure combustion process, heating values are measured in the constant-
pressure calorimeter shown diagrammatically in Fig. 3-6. Here gas and
air at room temperature enter the calorimeter in a steady-flow process,
the volume of the gas being measured by a gas meter. Complete com-
bustion takes place within the calorimeter, and the products are cooled to
the initial temperature of the gas/air mixture by means of the water
jacket surrounding the combustion chamber. The weight of water flow
ing through the calorimeter for a given length of time is measured, and the
temperature rise of that water is noted on the thermometers. From
these data the heat removed from the combustion products can be calcu
lated. This is equal to the heating value of the fuel.
\Thermomefer
Wet- lest meter

Fig.3-6. Diagrammatic sketch of a constant-pressure calorimeter.
Heating Values at Constant Volume and Constant Pressure. A

thermodynamic analysis of the combustion process in the calorimeter
shows that a number of qualifications must be made to establish the exact
significance of the heating value of a fuel.
Applying the nonflow energy equation to the constant-volume calorim
eter process:
Q. = AE + W
and, since W = 0 for the constant-volume process,
Qv = AE = E, - Em
where AE = total change in internal energy

Ep = total internal energy of products
Em = total internal energy of mixture
The only need for introducing the term total internal energy is to dis
tinguish between the sensible internal energy of the initial mixture and
the total internal energy of that mixture, which is the sum of the sensible
internal energy and the latent energy which can be released when the
chemical reaction occurs. The latter is called the chemical energy of the
fuel. Thus
Em = Em + 4>
where ^
= chemical energy
Esm = sensibleinternal energy of mixture, which is a function of
temperature only
The nonflow energy equation for the constant-volume calorimeter proc
ess can then be rewritten
Qv = Ep — Em = Esp —
(Esm + ^)
= E,p — E,m — ^
(6)
In the above equations note that the sensible internal energy of the
products is the same as the total internal energy of those products, since
the chemical energy has already been released in the combustion process.
It should also be observed that the heating value of the fuel Q„ will be a
negative number when it is defined by such an expression. This is
logical by thermodynamic notation because the heating value of the fuel
is removed from the products of combustion and absorbed by the jacket
water surrounding the bomb calorimeter. In thermodynamic equations,
when heat is removed from a substance during a process, the algebraic
sign attached to this transferred heat is negative.
The chemical energy differs from the heating value of a fuel by an
amount equal to the difference between the sensible internal energy of
the products and the sensible internal energy of the initial mixture, when
they are at the same temperature. From Eq. (6) it is evident that the
heating value and the chemical energy of a fuel can be computed, one
from the other, if values of internal energy are known. Ordinarily the
values of chemical energy are computed from the experimentally deter
mined heating values. Chemical energy values are not used frequently
in calculations, and the term is introduced here only to bring out its rela
tionship with the heating value.
For the constant-pressure calorimeter process, the general energy
equation reduces to
El +
^ +
T0 + Q = E2 + Pp + 2gJ
since the change in potential energy of the gases passing through the
calorimeter is negligible and no external work is done. Here again the
term E is used to denote total internal energy. Since
Ep = E,p and Em = Em + ^
the above equation can be written
*.„+("!
or
Hm + * + Qp
and
(7) Qp = H,p - Hm- +
The notation for sensibje enthalpy is introduced here to show that it is

composed entirely of sensible heat and does not include any stored
chemical energy.
For the constant-pressure combustion process, it is evident that the
heating value of the fuel differs from the chemical energy by an amount
equal to the difference in enthalpy of the products and the initial mixture.
Therefore the difference between the heating value at constant pressure
is,
and the heating value at constant volume
Qp-QV = - - - (E.p - -
*)
(H.T H.m E.m

ifi)
- - E,p + E.„
(e.
(e.
=
PJ
-J
+ +
+
|^)m
Since the velocities of gases entering and leaving the constant-pressure
calorimeter are small and nearly equal,
-(7).
which the flow work required to change the volume of the gases follow
is
ing the reaction. Substituting PV = NRT in the above,
or
(8) Q„
- Qv = (Np - Nm) ~
In computing the difference between heating values at constant pres
sure and at constant volume, the moles of mixture or products which
appear in the liquid or solid state at the temperature of the test can be
neglected because the volume of the liquid or solid is almost negligible

compared with the volume of the gases.
Example 3. Compute the higher heating value of methane at constant pressure

if the higher heating value at constant volume is —380,598 Btu/mole at 77 F.
Solution:
CH4 + 202 -» C02 + 2H2O
At of 77 F the water vapor in the products of combustion will con

a temperature
dense to the liquid state, so it is not counted in the total number of moles of gaseous
products.
Qp - Q„ = (Np
- Nm) ^
(1.0 - 3.0)
= X ^j-
. _2 x
587 R = _m ^
- to
7
Qp = Qv 2132
380,598 - 2132
382,730 Btu/mole, or = -23>861 Btu/lb fuel

16 04
rigorous thermodynamic analysis of the calorimeter process reveals

A
that the heating values are negative numbers, denoting that the heat
is
leaving the combustion products. In engineering practice the heating
values are often expressed as positive numbers.
From the above example, can be seen that the difference between the
it
heating values at constant volume and at constant pressure very small

and negligible in most cases. is
is
Higher and Lower Heating Values. When the products of combustion

F,
in a calorimeter are cooled to 77 practically all the water resulting from

the combustion process will be condensed, and the latent heat of evapora
tion of the water vapor will be given off to the water jacket of the calorim
eter. Usually few drops of water are added to saturate the original
a
mixture with water vapor so all of the water vapor formed by combustion
will condense. The heating value so obtained known as the higher
is
heating value or gross heating value. The lower (or net) heating value
is
a
fictitious quantity which would be obtained the water vapor in the

if
products of combustion could be cooled to 77 and still remain com

F
pletely in the vapor state. In other words, the lower heating value
is
equal to the higher heating value minus the latent heat of evaporation of
the water vapor formed. Keenan and Keyes in "Thermodynamic
"
Properties of Steam show that the latent heat of evaporation at constant
pressure (h/g) 1050.4 Btu/lb at 77 and the latent heat at constant
is
volume (e/0) 991.3 Btu/lb. This the equivalent of hfa per mole of
is
is
18,924 and of efg per mole of 17,860 Btu.

The state of the fuel (liquid or gas) is another factor which affects heat
ing values similarly. Thus, the heating value for 1 lb of octane in the
vapor state will be greater than the heating value of the same fuel in the
liquid state by an amount equal to the latent heat of evaporation of the
fuel itself. For precise specifications of the conditions under which a
heating value was obtained, it is necessary to include the initial state of
the fuel and the state of the products of combustion.
Example 4, Calculate the lower heating value at constant pressure in Btu per
pound for methane CH4, if the constant-pressure higher heating value (liquid H20
in products and gaseous fuel in the original mixture) is —23,861 Btu/lb of fuel.
Solution. From the reaction equation,
CH4(g) + 2O2 CO2 + 2H20(1) (normal state at 77 F)
the weight of water formed per mole of fuel burned is obtained.
2 moles H20 2 X 18.016 lb H20 =

2.2464 lb H2O/lb fuel
1 mole fuel 1 X 16.04 lb fuel
h/, = 1050.4 Btu/lb H2O
and the latent heat at constant pressure of the water vapor formed per pound of fuel is
1050.4 X 2.2464 = 2359 Btu/lb fuel

Lower heating value 23,861 + 2359 = -21,502 Btu/lb fuel
A comparison of both higher and lower heating values at constant

volume and constant pressure for two common hydrocarbons is tabulated.
Heating Value at 77 F, Btu per Pound of Fuel
-Higher HV — Lower HV
Constant Constant Constant Constant

pressure volume pressure volume
Methane, CH4 23,861 23,728 21,502 21,502

re-Octane, C8Hla 20,591 20,549 19,100 19,142
Effect of Temperature Level. Ordinarily in the calorimeter process

the combustion products are cooled to very nearly the san*e temperature
as that of the fuel/air mixture before combustion. It is necessary to
know this temperature to report heating values precisely, since the heat
ing value depends on the difference of enthalpy, or internal energy, of the
exhaust products and the initial mixture. In other words (E,p — E,m)
may be different when both the products and mixture are at 1500 F than
when they are both at 70 F.
By convention, chemists determine most heating values at a tempera
ture of 77 F (25 C) ; mechanical engineers sometimes use 68 F as the refer
ence temperature; and in the gas industry 60 F is most commonly used,

although determinations occasionally are reported at other temperatures.
The difference between constant-volume heating values at two tem
peratures is derived as follows:
Q.r, = (E,p - E,m -

and
QvT, = (E,p — E,m — \p)T,
QvT, — QnT, = (E,pT, — EspTl) — (E,mT, — E,mTl) — tyr, — ^r,)
The chemical energy depends only on the fuel and the heat energy which
will be released during the reaction. Hence it is not affected by tempera
ture change and cancels out of the above equation.
Then
(9) QvT, — QvTt = AB,p(T,_r,) — A£,m(rr-r,)
For the heating value at constant pressure, a similar derivation yields,
QpT, — QpT, = Aff^Ts-T,) —

AHsm(Tl-Tl)
Example 5. The lower heating value (gaseous fuel and vapor H20 in the products)
of CH4 at constant pressure is 21,502 Btu/lb at 77 F. What will be the constant-
pressure lower heating value of CH4 at 1500 R?
Solution. The stoichiometric combustion equation of the fuel
CH4 + 202 + 2 X 3.76N2 -» C02 + 2H2O + 7.52N2
shows the number of moles of mixture and products. Since the same number of moles
of nitrogen appear in the products and the mixture, the numerical values of enthalpy
change for the nitrogen will cancel, and it may be omitted from the calculations.
Using Table 3-2 to find the enthalpy changes of the mixture and products per mole
of fuel:
AH,p = (Ah,)co, + 2(AA,)h:o
where Aft, = sensible enthalpy change per mole of gas between 537 and 1500 R.
AH,p = (14,576.0 - 4030.2) + 2(12,551.4 - 4258.3)

= 27,132.0 Btu
AH,m = (Ah,)cH,
- + 2(11,017.1 - 3725.1)
+ 2(Ah,)ni
= (15,872 4312.0)
= 26,144 Btu
AH,p - AH,m = 27,132.0 - 26,144 = 988
Btu/mole fuel
QQQ
=
I^T
16.04
= °2-0 Btu/lb fuel
QpT, — QpTi = AH,p — AH,m

QpT, = -21,502 + 62.0 = -21,440 Btu/lb fuel
Example 5 shows that the heating value of a fuel varies with the tem
perature at which it is determined, but the change is extremely small and
need be included only in the most exacting calculations. In the example,
the change in heating value was only about 0.2 per cent for a temperature
change from 77 to 1040 F. It
should be apparent that the difference
between the heating values of a fuel tested at 60 F, 68 F, 77 F, etc., must
be exceedingly small. In many cases the limits of accuracy of the experi
mental data do not warrant such differentiation.
In summing up the discussion of heating values of fuels, we might con
clude that for most types of engineering work either the constant-volume
or constant-pressure heating value at 77 F may be used with sufficient
accuracy for calculations, regardless of the type of process or the tempera
ture at which the combustion occurs. However, a differentiation
between the higher and the lower heating value must be made.
Table 3-3 gives the heats of reaction for many commonly used fuels.
The values listed in the table and used in the examples are given to more
significant figures than the experimental data warrant in order to point
out the relationships between the different heats of reaction.
Heat of Formation. The heat released or absorbed when a compound
is formed from its elements is called the heat of formation. In the cases of
C + O2 -» CO2 + (-169,182) Btu/moleat constant pressure and 77 F
and
H2 + HO2-»H2O +( - 122,890) Btu/mole at constant pressure and 77 F

the heats of formation of CO2 and H2O simply give the heating values
(heats of combustion) of the elements carbon and hydrogen. For the
formation of fuels such as CH
C + 2H2 -» CH4 + (-32,179) Btu/mole CH4
the heats of formation can be useful in predicting the heating value of a

compound since the same heat is absorbed in breaking down a compound
as was released in forming it (or vice versa). An exothermic reaction is
one which gives off heat to the surroundings. When this occurs, the heat
carries a negative sign by thermodynamic notation. An endothermic
reaction is one which absorbs heat as it progresses, and the heat transfer
is designated by a positive sign. The heating value of a given compound
of carbon and hydrogen can be computed by adding the heats of forma
tion of CO2 and H2O from the original carbon and hydrogen in the fuel
and subtracting algebraically the heat of formation of the original hydro
carbon. Heats of formation of some hydrocarbon compounds are given
in Table 3-3.
Example 6. Calculate the heat of reaction of CH, with air, at 77 F and constant
pressure of 1 atm, from the heats of formation of Table 3-3.
Table 3-3. Heats op Reaction*
— Ah combustion, ]
or heating value,
Mole Ah] Ah
Name Formula cular evapora Btu/lb
formation!
tion,
weight Btu/lb mole
Btu/lb
Higher Lower
Methane CH4 16 04 23,861 21,502 -32,179

Ethane C2H6 30 07 22,304 20,416 -36,401
Propane C3H8 44 09 147.1 21,646 19,929 -44,647
n-Butane C4H10 58 12 155.8 21,293 19,665 -53,626
Isobutane C4H10 58 12 141.4 21,242 19,614 -56,576
n-Pentane CsHij 72 15 157.5 20,914 19,340 -74,400
n-Hexane C6H14 86 17 157.4 20,771 19,233 -85,480
n- Heptane C7H16 100 20 156.8 20,668 19,157 -96,470
n-Octane CsHis 114 22 156.1 20,591 19,100 -107,500
2, 2, 4-Trimethylpen- CsHis 114 22 132.2 20,556 19,005 -111,500
tane
n-Nonane C9H20 128 25 155.7 20,531 19,056
n-Decane C10H22 142 28 155 20,483 19,020
n-Tetradecane C14H30 198 38 154 20,358 18,927
n-Hexadecane C16H34 226 43 154 20,318 18,898
re-Pentatriacontane C35H72 492 93 20,350 19,032
Ethylene C.H4 28 05 21,625 20,276 22,478
Propylene C3H, 42 08 21,032 19,683 8,776
Isobutene C4H8 56 10 157.7 20,716 19,367 -6,017
Octene C sHie 112 21 158.0 20,349 19,000
Cyclopentane C5H10 70 13 174.7 20,175 18,825
Cyclohexane C6Hi2 84 16 168.8 20,026 18,676
Benzene C6H6 78 11 186.3 17,986 17,259 21,079
Toluene C7H8 92 13 177.3 18,245 17,424 5,157
o-Xylene CsHio 106 16 175.9 18,438 17,547 10,510
Methyl alcohol CH30H 32 0 502 9,770 8,644
Ethyl alcohol C2HsOH 46 0 396 12,780 11,604
Propyl alcohol C3H-OH 60 0 295 14,500 13,300
Butyl alcohol C4H9OH 74 1 254 15,500 14,284
Acetylene C2H2 26 04 21,460 20,734 97,485
Hydrogen H2 2 016 60,958 51,571 0
Carbon (solid, C 12 01 To CO2 14,087 0
graphite)
Carbon (coke) C 12 01 To CO2 14,540 0
Carbon (coke) C 12 01 To CO 4,440 0
Water, liquid H20 18 016 -122,890
Water, vapor ...... H20 18 016 -103,968
Carbon monoxide. . CO 28 01 To C02 4343 . 6
Carbon dioxide CO2 44 01 -169,182
* Most of these data obtained from Selected Values of Properties of Hydrocarbons, Nat. Bur. Stand
ards, Circ. C461, API Research Project 44.
t Latent heat of evaporation of fuel at constant pressure, 77 F
t — Aft combustion, or —heating value at constant pressure, 77 F, each compound in its normal state
at 77 F. (E.g., CHi is a gas it 77 F, so heating value is for gaseous fuel. CsHis is ordinarily a liquid
at 77 F, so the heating value for it is reported for liquid fuel.)
§ Ah formation, or heat of formation of compound from its elements at constant pressure, 77 F.
THERM0CHEM1CAL ANALYSIS 91
Solution:
CH4 + 202 + 2 X 3.76N2 -> 2H20 + C02 + 7.52N2
The nitrogen does not affect the heating value and can be neglected. It may be
reasoned that the CH< must be broken down into C + 2H2 so the latter are free to
react with oxygen and form C02 and 2H20. The same quantity of heat must be
absorbed in the endothermic reaction of breaking CH4 down as was released in the
exothermic reaction of formation. Hence, the heat of formation of CH4 is subtracted
algebraicallyfrom the sum of the heats of formation of C02 and 2H20.
Heat of reaction of CH4 with 02 = 2(-122,890) + (-169,182) - (-32,179)

At 77 F and constant pressure H* to H2O C to COa from CH4
= -382,783 Btu/mole fuel
382,783
= -23,861 Btu/lb CH4
16.0425
Computing Heating Values. The heating value of a given fuel can

always be found by burning a small quantity of it in a calorimeter, but it
is not always practicable or desirable to do this. Methods for calculating
an accurate heating value of the fuel then become useful. There are
various methods used for such calculations, some based on fundamental
derivations and others empirical, but only one method each for solid,
liquid, and gaseous fuels will be given here.
To find the heating value of a gas if the constituents of the gas are
known, use the heats of reactions from Table 3-3 for each of the com
bustible components, and find the heat which would be released from the
burning of that part of the fuel. The sum of the heat released from all
such reactions is the heating value for the fuel.
Example 7. Compute the heating value in Btu per mole and in Btu per cubic foot
.ât 14.7 psia and 60 F for a manufactured gas of composition as follows:
C02 = 4.6%, N2 = 7.0%, H2 = 49.7%, C2H6 = 2.5%, 02 = 0.4%, CO = 24.9%,

CH4 = 10.9%
Solution:
1
Basis: 1 mole fuel gas
Component Moles present Heat of reaction, Btu/mole Heat released, Btu
CO 0.249 4,343.6 X 28.01 30,294

H2 0.497 60,958 X 2.016 61,076
CH4 0.109 23,861 X 16.04 41,718
C2H6 0.025 22,304 X 30.07 1 16,767
Total. $6f74 149,855
(h&t
149,855
Heating value per cu ft fuel gas = = 395 Btu/cu ft
359 X 520/492
i-/ CO 0
3 a o 0 0
/ / jj V O O
For coal, Dulong's formula is often used to compute the heating value
when the ultimate analysis of the coal is known.
(10) Heating value, Btu/lb = 14,540C + 60,958 ( H - ^

I + 4050S
where C = fraction of carbon in fuel, by weight

H fraction of hydrogen in fuel, by weight
=
S fraction of sulfur in fuel, by weight

=
0 = fraction of oxygen in fuel, by weight

This equation is based on the heats of combustion of carbon, hydrogen,
and sulfur. It is assumed that any oxygen present in the coal is already
united with hydrogen in the form of moisture, and no heating effect is
obtained from the hydrogen in this form. Sometimes the coefficients in
Dulong's formula are varied slightly from those given above to help
account for the incomplete heat liberation from the fuel which usually
exists to some degree.
The heating values of liquid hydrocarbons are frequently calculated
from formulas based upon the specific gravity or API gravity of the fuel.
These formulas are for the most part empirical in origin but depend on the
fact that the greater the percentage of hydrogen in a fuel, the lighter the
liquid because the hydrogen is much lighter than the carbon. By the
same token, a fuel high in hydrogen has a high heating value because
hydrogen has a heat of reaction per pound some four times that of carbon,
as seen in Table 3-3. The heating values for different API gravity fuels
are shown graphically in Fig. A-l in the Appendix.
Some of the formulas used are:
(11) HHV* = 22,820 - 3780rf2 Btu/lb

where d = specific gravity at 60/60 F.
HHV] = 18,650 -
(12)
HHV = 18,440
+ 40 (API
(API
gr.
- 10) Btu/lb for fuel oil
Btu/lb for kerosene
(13)
(14) HHV =
+ 40
18,320 + 40 (API
gr.
gr. - 10)
10) Btu/lb for gasoline
where API gr. is the API
gravity of the fuel at 60/60 F.
Since the above equations give only close approximations to the HHV
of the fuels, it is impossible to designate the temperature at which the
value is exact or to stipulate the condition of constant volume or constant
pressure.
An interesting relationship between the heat released and the air
required for combustion of various fuels is often overlooked. Investiga-
*
C. S. Cragoe, Not. Bur. Standards, Misc. Pub. 97, 1929.
t Am. Chem. Soc, vol. 30, 1908.
tion of this relationship shows that approximately the same quantity of

heat is released per pound of air required for combustion, regardless of
the fuel. Also, the Btu per cubic foot of air/fuel vapor mixture is
approximately constant for most fuels. It is true there is a variation in
these two quantities for fuels which differ widely in their ratios of hydro
gen to carbon, but there is more consistency between the above-mentioned
relationships than is generally realized.
THEORETICAL FLAME TEMPERATURE

The theoretical flame temperature which should be reached following
the combustion of a certain mixture of air and fuel can be computed by
assuming that the reaction goes to completion in an adiabatic process.
In other words, all of the heating value of the fuel is used to heat the
products of combustion.
For the nonflow, constant volume, adiabatic process
Q = AE + W
but Q = 0 for the adiabatic and W = 0 for the constant-volume stipula
tions. Then
(15) AE = 0 and EpT, = EmTl
This formula states that the total energy of the products of combustion
at the high temperature is equal to the total energy of the initial mixture.
Substituting Ep - Esp and Em = E.m + \p gives
This equation would be satisfactory except that values of chemical energy

for fuels are not so widely used as heating values. Recall
Qv = (E.p - Em)Tl - $
(E,p - E,m)Tl -
or
) = Qv
Substituting this expression for chemical energy into the equation above
EspT, = —
EgmT^ "h EspTl EamTl Qv
and rearranging
= '
E8pTt E,pT1 Qv
This equation shows that all of the heating value of the fuel goes to heat
the products of combustion from the initial to the final temperature.
Actually the combustion reaction is progressive, and the products are
heated gradually as the heat is released, as represented in Fig. 3-7. How
ever, when theoretical flame temperatures are computed, it may be
assumed that the entire chemical reaction goes to completion at the

temperature of the fuel/air mixture. The energy thus released then
heats the products of combustion from the low temperature to the final
flame temperature. Thus, to find the theoretical combustion tempera
ture, it is necessary to find the temperature where
(16)
Here again the heating value of the fuel carries the algebraic negative
sign. This makes the internal energy of the products at the theoretical
flame temperature equal to the sum of the heating value of the fuel and
the internal energy of the combustion products at the initial temperature
of the air/fuel mixture. It should be observed that the lower heating
Theoretical flame temperature
;W2) Products,
c
° w
s
/ ^sr'*^ Actual burning

and heating path
\%
E o
8*
Effective path
Fig.
Mixture
\ @
Initial temperature
A Products,
3-7. Effect of the heat released during combustion in heating the products of com
bustion to the final theoretical flame temperature.
value must be used in this type of problem since the water vapor in the
products is very definitely in the vapor state at the flame temperature.
Since the products of combustion are a mixture of several gases, the
determination of the theoretical flame temperature can be found by a
trial-and-error method.
A derivation similar to the above for constant-pressure or steady-flow
adiabatic combustion would yield
(17)
Example 8. Compare the theoretical flame temperatures which should be obtained

from the complete combustion of octene C8Hi6, with theoretical air at constant pres
sure if the air temperature before ignition is" (a) 77 F and (fe) 1600 R, or 1140 F.
Solution. The combustion equation of octene with theoretical air is
C8H16 + 1202 + 12 X 3.76N2-^ 8CO2 + 8H20 + 45.12N2
From Table 3-3, the lower heating value of octene at constant pressure is 19,000
Btu/lb, which corresponds to 19,000 X 112 lb/mole, or 2,128,000 Btu/mole of fuel.
The computations will be based on the numerical values per mole of fuel. If the
initial temperature is 77 F,
H.pT, = ff„P77 - (-2,128,000)

To find the enthalpy of the products at 77 F above a base of 60 F (the base tempera
ture of the Cpm graph), use Fig. 3-4 and the relationship
Between 77 and 60 F,
'
Av = (V, —jTj Cpm\ + ( —jTf + (\ ~Xr
CPm
Nt /cO2 \ Nt Cpwt^
/Hso Nt Cpro^
/n
8
61.12
,8.80+^7.94+^6.95
1
61.12
'
61.12
= 0.1309 X 8.80 + 0.1309 X 7.94 + 0.7382 X 6.95
= 7.32
A#77-6o = 61.12 X 7.32(77 - 60) = 7600 Btu/mole fuel

Hence AH of the products above 60 F is
H,pT, = 7600 + 2,128,000 = 2,135,600 Btu
A certain amount of trial and error is involved in finding a temperature at which the
mixture of combustion products has the enthalpy value stated. To be able to com
pute a reasonable average Cpm of the products following combustion, assume that the
flame temperature is 3800 F. Then between 3800 and 60 F
Av Cpm = 0.1309 X 13.17 + 0.1309 X 10.63 + 0.7382 X 8.02

= 9.04 Btu/mole of products
H.pTl = NTCpm AT = 2,135,600 Btu
M - 61.12
2,135,600
= 3865
X 9.04
Hence with this trial, 772 = 3865 + 60 = 3925 F. As the average C,,,„ was computed
for 3800 F, another more accurate value of Cpm may be computed. This time assume
T, is 3920 F. Then,
Av Cpm = 0.1309 X 13.21 + 0.1309 X 10.70 + 0.7382 X 8.04

= 9.07 Btu/ (mole) (R)
AT -_ 2,135,600
61.12 X 9.07
-_ 3852
or
r2 = 3852 + 60 = 3912 F
If the initial temperature of the air and fuel were 1600 R, or 1140 F, before ignition,
then between 1140 and 60 F
Av Cpm = 0.1309 X 11.04 + 0.1309 X 8.69 + 0.7382 X 7.22

H,pTl = 61.12 X 7.91(1140 - 60 F) - (-2,128,000)
= 2,650,000 Btu
Assume that the flame temperature is 4750 F. Then between 4750 and 60 F
Av Cpm = 0.1309 X 13.50 + 0.1309 X 11.11 + 0.7382 X 8.18
2,650,000 Btu
= =
61.12 X 9.26
or
T2 = 4682 + 60 = 4742 F
The theoretical flame temperatures computed in the above example

show that the amount of preheat of the air before ignition influences
greatly the final flame temperature. It might also be observed that com
bustion temperatures at constant volume are somewhat higher than for
burning at constant pressure since nearly the same heating value of fuel
is released but no energy is absorbed in the flow work of expanding the
products of combustion.
CHEMICAL EQUILIBRIUM AND DISSOCIATION

The theoretical combustion temperatures determined in Example 8 can
not be reached under actual conditions because the combustion process
does not go to completion at these high temperatures. Even while the
combustion reactions are proceeding, the resultant products dissociate
to a certain degree into other components. Thus, the equilibrium
equation,
CO + KO2 ^ CO2
shows that the reaction may proceed both to the right and to the left at
the same time. At agiven condition of temperature and pressure, an
equilibrium among the gases is reached so that the amounts of CO, O2,
and CO2 present will remain constant. The extent of dissociation of the
CO2 will give an indication of the degree of completeness of the combus
tion. Just as heat is released in the exothermic process of forming the
CO2, much of the heat is absorbed when the CO2 is broken down into CO
and O2. In other words, dissociation reactions of combustion products
are endothermic. Therefore, the theoretical combustion temperature can
never be attained if dissociation exists.
The three equilibrium reactions most commonly considered in com
bustion calculations are,
—
H2 + y2o2 H2O
CO + ko2 ^ co2
CO2 + H2 ?± H2O + CO
The last equation is known as the water-gas reaction and must be taken
into account since both CO2 and H2O are present in practically all com
bustion products.
Other dissociation reactions which occur to a minor extent in high tem
perature combustion are
H2^±H + H
O2^±O + O
H2O ^ H + OH
N2^N + N
H2O + HN2 ^ H2 + NO
The percentage of dissociation present at a given temperature and

pressure might be determined by experimental means. However, from
fundamental thermodynamic data, an equilibrium constant can be calcu
lated which provides a means for predicting the degree of dissociation for
a given condition.
To establish a concept of the relationship between the equilibrium con

stant and the components of a reaction, consider the hypothetical chem
ical equation,
9A + rB ^ sC + tD
which says q moles of A unite with r moles of B to form s moles of C and
t moles of D. Then
(18) K ~
(wi?
where K
is the equilibrium constant and (PA) 9 is the partial pressure in
atmospheres of gas A, raised to the q power, etc. There are other equa
tions used for applying the equilibrium constant, but the above method is
convenient for combustion calculations and will be followed here.
The numerical value of the equilibrium constant for a given reaction
will remain the same at any one temperature and will show the propor
tional amounts of constituents present. For instance, if the amounts of
A and B remain constant and the quantity of D increases, then the
amount of C present must decrease in order to keep the value of K a
constant.
Equilibrium constants vary with temperature, and values of K for some
of the more important combustion reactions are given in Table 3-4.
While the theory involving the method of calculation of the equilibrium
constant is beyond the scope of this book, it may be interesting to know
that equilibrium is related to the entropy change during the reaction.
All self-propelled irreversible processes proceed to the point of maximum
increase of entropy. Hence, the combustion reaction proceeds until the
change of entropy is greatest. At this point an equilibrium among the
constituents is established.
The following examples will serve to demonstrate the use of the equi
librium constant.
Example 9. Experimental evidence shows that carbon dioxide gas becomes 10
per cent dissociated into carbon monoxide and oxygen at a temperature of 4300 R
when the pressure of the mixture is 1 atm. Determine the equilibrium constant for
the equation, CO + HO2 ^ CO2 at this temperature.
Solution. On the basis of 1 mole of CO2 gas to start with, when 10 per cent of the
CO2 dissociates by the equation
CO + J^02 ^ CO2
Table 3-4. Equilibrium Constants for Some Combustion Equations*
Reactions
Temp, R
CO + HO, ^ co2 H2 + K02 ^ H2O C()2 + H2 ^ CO + H20
3000 22,000 72,500 3.32

3200 6,200 23,000 3.72
3400 2,000 8,200 4.14
3600 720 3,300 4.55
3800 300 1,500 4.94
4000 130 710 5.29

4200 65 370 5.60
4400 35 200 5.90
4600 20 120 6.13
4800 11 .5 73 6.36
5000 7.1 46 6.55

5200 4.6 31 6.75
5400 3.0 20 6.93
* Computed from data from B. Lewis and G. von Elbe, J. Am. Chem. Soc, vol. 57, p. 612, 1935.
only 0.90 mole of C02 remains, During this reaction 0.10 mole of CO is formed along
with 0.10/2 or 0.05 mole of 02. Thus the total number of moles of gases present in
the mixture would be
0.90 + 0.10 + 0.05 = 1.05 moles

C02 + CO + O, = total
The partial pressure of each of the constituents would then be
0.90
Pco, X 1 atm = 0.857 atm
1.05
O05
P01 = X 1 atm = 0.0476 atm
1.05
0.10
Pco X 1 atm = 0.0952 atm
1.05
Since
K =
Pco(Po*)«
for the equilibrium equation as it is written,
0.857
K = = 41.3
(0.0952) (0.0476) V>
For a temperature of 4300 R the amounts of C02, CO, and O2 present in a mixture of
gases will always be such that the equilibrium constant will have a value of 41.3.
At this point it should be mentioned that the form in which the equilibrium equation
is written must be known before the value of an equilibrium constant can be used.
If the dissociation of CO2 had been written
2CO + O, ;=± 2C02

then the equilibrium constant would be
K _ _(Pco2)\
(Pco)2Po
and the value of K at this temperature would be

2
(0.857) = 1705
2
(0.0952) (0.0476)
Thus the equilibrium equation as well as the value of the equilibrium constant must
be specified if the constant is to be significant.
Example 10. Calculate the percentage of dissociation of 1 mole of CO2 at 5000 R
if the pressure on the CO2 mixture is 1.7 in. Hg abs.
Solution. The equilibrium equation is CO + KO2 ^ C02.
Let X = fraction of C02 dissociated, then on the basis of 1 mole of CO2 initially
1 — X = moles CO2 remaining

X = moles CO formed by dissociation
jr
= moles 02 formed by dissociation
0g
The total number of moles in the mixture is then
(1-Z)+Z+|-1+|
and t*he partial pressures of the constituents in terms of the pressure P are
Pco2 =
\+X/2P
Pco =
1 +X/2P
P
Po>
- X/2 P
1 + X/2P
Setting up the equilibrium equation
Pco,
K =
Pco(Po2)H
\i +x/2/ ;
x pY
( P) ( XV
-X
Simplifying,
1
- X)'
and
(1
K* =
x2
Then
K2 =
(1
- 2X + X2)(2 + X)
X3P
and
K*X*P X* 3X + -
(K'P - -
= 2
1)X> + 3X 2 = 0
This last equation is convenient to use since it satisfies all equilibrium conditions for
the stated reaction CO + ^j02 ^ CO2. If the temperature is known, the value of K
can be found from the table. Then the percentage dissociation may be determined
for any pressure.
At 5000 R, K = 7.1 from Table 3-4. The pressure in atmospheres at 1.7 in. Hg
abs is 1.7/0.492 X 14.7 = 0.05 atm. Substituting in the above equation
X - 1]X3 + 3X
- 2 = 0
+ SX -
[(7.1)' 0.05
1.52X" 2 = 0
This cubic equation may be solved by trial and error.

First assume X = 0.60.
Then
1.52(0.60)' + 3(0.60) - 2 = 0.33 + 1.8 -2
= 0.13 ^ 0
Next try X = 0.57.
1.52(0.57)' + 3(0.57) - 2 = 0.282 + 1.71 - 2

0.008
This is fairly close to the correct value of X, and the dissociation of CO2 under these
conditions is 57 per cent. .
The above example shows that the effect of dissociation is very pro
nounced at low pressures. This is somewhat the case when CO2 and H2O
appear in the normal products of combustion of fuel with air. The nitro
gen thoroughly dilutes the combustion products, and the partial pressures
of the CO 2 and H2O in the products are rather small. Since little dissocia
tion occurs at temperatures less than 3000 R, even at low pressures, the
dissociation phenomena come into importance only in internal-combus
tion engines, rockets, welding equipment, etc., where the temperatures
attained are quite high. In most types of atmospheric-pressure heating
or furnace units, the temperatures are too low to make the effects of dis
sociation significant.
When fuel is bu-ned with excess air, the excess oxygen present tends to
drive the equilibrium reaction to completion, even at elevated tempera
tures. The following example shows that dissociation of CO2 at 5000 R
is reduced appreciably if excess oxygen is supplied for combustion.
Example 11. Determine the degree of dissociation of CO2 at 5000 R and atmos
pheric pressure, if 100 per cent excess oxygen is supplied for the reaction.
Solution:
CO + 02 ^ C02 + HOt
The equilibrium equation for this reaction is
Pco,(Po,)H Pco,
K = =
PcoPo, PcoWo,)*
The latter expression is the same as the equilibrium constant for the equation
CO + y2Q2 ^ C02
Thus, the problem is the same as for the stoichiometric equation except that the partial
pressure of oxygen is now greater.
Let X = fraction dissociated from 1 mole of CO2 = moles CO, then
1 -X = moles C02
X/2 = moles O2 from dissociation
J£ = moles 02 excess
Hence
(1
—
X) + X +
XI
-_- +
^
= 1-5 +
X
~2
= total moles products
Pco' = P
1.6 + X/2
Pco = P
1.5+ X/2
X/2 + 1/2 X + 1
°" = =
1.5 + X/2 3 + X
-X
and
l-X
1
„
A = 1.5+ X/2
\?
1.5 + X/2
- 2X
Squaring,
„, (1 + X')(3 + X) = X3 + X2 5X + - 3
X*(X + 1)P + X')P
K*X'P + K*X'P -
X' X* + 5X 3 = 0 - -
(K*P - 1)X3 + (K*P 1)Z2 + 5X 3 = 0 - -
At 5000 R, K = 7.1 from Table 3-4. For a pressure of 1 atm,
- l]X' - - 3 = 0
[(7.1)' + [(7.1)2 1}X* + 5X
49.4Z' + 49.4X* + 5X - 3 = 0
Solving by trial and error, if X = 0.188,
49.4(0.188)s + 49.4(0.188)* + 5(0.188)

- 3 = 0.02
Hence, under these conditions, the CO2 is 18.8 per cent dissociated. In
Table 3-5, at 1 atm pressure and 5000 R, the dissociation of CO2 is 28.4
per cent. Thus, the addition of excess oxygen reduces the dissociation
materially, even at high temperatures.
Table 3-5 summarizes the effects of temperature and pressure on the
dissociation of carbon dioxide and water. Dissociation below 3000 R is
negligible and, as the pressure is increased, the reaction is driven near
completion even at high temperatures.
At lower temperatures, the combustion reactions tend to progress
slowly. If a mixture of gases at high-temperature equilibrium is cooled
very rapidly, it is possible that the combustion process might not be com
pleted before the temperature drops to the point where no more burning
would take place. The combustion products would then show a certain
amount of dissociation even at low temperatures. Such a condition is
known as "frozen equilibrium." This situation exists in the gasoline
engine where a rather large percentage of C0 is present in the exhaust
even when excess air is used for combustion.
Table 3-5. Dissociation of C02 and H20 at Various Pressures, Per Cent*
Pressure, atm
Temp, R
0.1 0.2 0.3 1.0 10 30
Carbon dioxide to CO and 02
3000 0.35 0.30 0.21 0.16 0.075 0.052

4000 10.0 8.0 7.1 4.8 2.2 1.57
5000 50.0 42.8 39.0 28.4 14.9 10.4
Water vapor to H2 and 02
3000 0.16 0.12 0.11 0.07 0.03 0.02

4000 3.3 2.7 2.3 1.57 0.73 0.51
5000 18.9 15.4 13.6 9.3 4.4 3.1
* Calculated from the equilibrium data of Table 3-4.
Figure 3-8 shows the percentages of different component gases com

puted to be present in the combustion products in the cylinder of a spark-
ignition engine under various conditions. In making these analyses, the
dissociation of H2 to H, 02 to O, H20 to OH and H, and N2 to N were
included as well as the more common dissociation reactions.
Maximum Adiabatic Flame Temperature. The fundamental energy
equations for computing flame temperatures hold even if dissociation does
That
is,
exist.
AE = or AH =
0
and
Ept, = or HpTl = HmTl
EmTl
For actual conditions at high temperatures, HpT, must include the sensi
ble enthalpy and the unreleased heating value of the constituents in the
equilibrium mixture. The mathematical determination of an accurate
maximum flame temperature, including all effects of dissociation,
is
beyond the scope of this book, but a few of the factors to be considered in
such an analysis will be discussed. From Table 3-5 apparent that
it
is
the degree of dissociation increases markedly as the temperature rises.

At the same time, if carbon and hydrogen are held in equilibrium at some
stage of incomplete combustion, less heat will be released from the fuel,
and the resulting temperature of the products of combustion will be
lower.
The problem of calculating the flame temperature of a fuel then
involves the simultaneous solution of the expressions relating dissociation
phenomena and the equations showing the relationship between thermal
3,000 4,000 5,000 3,000 4,000 5,000 3,000 4,000 5,000 90 100 110
Temp.,R Temp.,R Temp^R % of theaair
Fig. 3-8. Composition at equilibrium of the gases in the cylinder of a spark-ignition engine
under various conditions. (Hershey, Eberhardt, and Hotlel, SAE Trans., vol. 31, 1936.)
and chemical energies of the fuel and products. In the burning of com
mon fuels with air, C02, H2O, 02, N2, GO, and H2 will all be present in
the combustion products. At extremely high temperatures of combus
tion, additional dissociation reactions produce small quantities of NO,
OH, H and O, as evidenced in Fig. 3-8. Thus, several equilibrium equa
tions must be considered simultaneously when the dissociation of the
combustion products of hydrocarbon fuels is analyzed. Since nitrogen
comprises by far the greatest percentage of the products for combustion
with air, the partial pressures of some of the constituents such as C02,
H2O, CO, and H2 will be very low, especially if the combustion reaction
occurs at atmospheric pressure. These low partial pressures tend to
increase the degree of dissociation, as shown in Table 3-5.
Table 3-6 shows the actual maximum flame temperatures which could
be expected from the combustion of some specific compounds in air.
Acetylene is shown to have a higher flame temperature than any of the
other hydrocarbon gases. This helps account for its widespread use in
welding. If these fuels were burned in oxygen instead of air, the maxi
mum temperatures would go much higher since nearly the same heat would
be liberated from the reaction and no large volume of nitrogen would be
heated. When acetylene is burned with a theoretical oxygen/fuel ratio,
the peak flame temperature goes to 5630 F. However, at the higher
temperatures the effect of dissociation becomes much more pronounced.
Table 3-6. Maximum Adiabatic Flame Temperatures at Atmospheric Pressure*
for Selected Gases
Gas Symbol Max flame temp, Ft
Carbon monoxide CO 3960

Hydrogen H2 3960
CH4 3416
Ethane C2H6 3442
CH, 3497
C4H10 3447
C<Hio 3442
Acetylene C2H, 4250
*J. Am. Chem. Soc., vol. 53, p. 876, March, 1931.

t For air/fuel ratio giving maximum measured flame temperature. >
Combustion Charts. Computations of high-temperature combustion

phenomena are exceedingly tedious and time-consuming if done by
classical thermodynamic methods, since several equilibrium reactions are
progressing simultaneously and account must be taken of variable specific
heats for the varying gas composition. The complexity of such problems
has prompted the development of numerous charts and tables to facilitate
accurate numerical analyses. References to several such methods which
include the effects of dissociation and variable specific heats are pre
sented.4-8 Two of these procedures are discussed here.
If the temperature of combustion is under 3000 R, dissociation effects
need not be considered. Dissociation may also be neglected if fuel is
burned at higher temperatures and the products of combustion are then
slowly cooled by dilution with cooler air until the final temperature is
under 3000 R. If the chilling of the high-temperature products is too
rapid, "frozen equilibrium" may account for some dissociation at the
lower temperatures. Figure 3-9 has been prepared to permit rapid calcu
lations of temperatures of combustion at constant pressure for given
fuel/air ratios.6 The peak temperatures of this chart push into the dis
3400
Chart fuel-air ratio, f

Fia. 3-9. Fuel /air ratio, /, for ideal constant-pressure combustion as a function of the
initial temperature./ = KmKiJCvJ.. (Bogart, Okrent, and Turner, NACA TN 1655,
1948.)
1
sociation range slightly, but the equilibrium effects are still small at
3000 F. A large-scale chart of Fig. 3-9 is included inside the back cover.
It is based upon a fuel having a lower heating value of 18,700 Btu/lb and
a ratio of hydrogen to carbon by weight of 0.176. The temperature rise
created by burning a given fuel/air ratio/' of this assumed fuel with dry
air may be read from the ordinate. The temperature rise for any fuel
may be found by using the appropriate factors, plotted on the chart,
which make corrections for the heating value of the new fuel Kh, and the
hydrogen/carbon ratio in the new fuel Km. A correction factor Kw is also
included to adjust for the effect of water vapor which might be present in
the combustion air. The chart fuel/air ratio /', based on the assumed
chart fuel, is related to the fuel/air ratio for any hydrocarbon fuel /, by
the relationship
(19) / =
KuKhKmf
The following examples illustrate the use of the chart.
Example 12. Calculate the fuel /air ratio required to give a temperature of 2000 R
when liquid octene CsHi6 is burned with dry air at constant pressure. Initial air
temperature is 600 R.
Solution. From Table 3-3, the lower heating value of liquid octene at constant
pressure is 19,000 Btu/lb. The ratio of hydrogen/carbon by weight is 16 lb hydrogen
per 8 X 12 lb carbon, or 0.167.
Using Fig. 3-9, for a temperature rise from 600 to 2000 R, or 1400 R, the chart shows
that a fuel/air ratio /' of 0.0204 would be required. The chart fuel/air ratio /' is
based on the chart fuel of H/C = 0.176 and heating value = 18,700 Btu/lb. For
octene, the correction factor for the heating value of the fuel Kk is 0.983. The cor
rection factor for the hydrogen /carbon ratio of the fuel Km is 0.999. Since dry air is
used for burning, the water correction factor K„ is 1.00.
The true fuel/air ratio needed to reach a temperature of 2000 R is then
/ == K,„KkKwf
0.999 X 0.983 X 1.00 X 0.0204 = 0.02003
This corresponds to an air/fuel ratio of 50 to 1 which, when compared with the

theoretical air /fuel ratio (Table 1-7) of 14.7, shows that the octene must be burned
with 240 per cent excess air to keep the final combustion temperature at 2000 R.
Example 13. Determine the adiabatic flame temperature when benzene C6H6
is burned with 80 per cent excess air at constant pressure. The air for combustion
enters the combustion chamber at 130 F and contains 0.04 lb water vapor per pound
of dry air. ^
Solution. The theoretical air/fuel ratio for benzene is 13.3 to 1 (Table 1-7), hence,
for 80 per cent excess air, the air/fuel ratio is 1.80 X 13.3, or 23.9. The fuel/air
ratio actually used is 1 /23.9, or 0.0418. The actual fucl/air ratio must be corrected
to the chart fuel /air ratio /'.
To find the correction factors, the flame temperature must be assumed. In this
case, assume a combustion temperature of 3000 R. For benzene, lower heating
value is 17,259 Btu/lb (Table 3-3) and H/C is 0.083 by weight. The weight of
moisture is given as 0.04 lb moisture per pound of dry air. From Fig. 3-9, the correc
tion factors are Kk = 1.094, Km = 0.971, and K„ = 1.04.
Chart fuel/air ratio/' =

L094 ^\ x m = 0.0378
Figure 3-9 shows that the temperature rise for these conditions is 2380 R. Hence,
the combustion temperature is
590 R + 2380 R = 2970 R, or 2510 F
When combustion temperatures rise well above 3000 R, dissociation

becomes appreciable, and the type of chart required to plot the various
thermodynamic relationships becomes more complex and somewhat less
Fig. 3-10. Sketch of a chart relating the thermodynamic properties of a mixture of com
bustion products at high temperatures.
flexible in its applications. Hottel and his coworkers have prepared

charts relating the thermodynamic properties of combustion products7,8
at high temperatures. Figure 3-10 is a sketch showing the variables
plotted on these charts, and a set of five such charts is included inside the
back cover. Each chart is constructed for the products of combustion of
1 lb of air with a given percentage of theoretical fuel. The fuel used,
octene C8Hie, is considered to have properties representative of most
gasolines. While all five charts are constructed for fuels having the
general formula C„H2„, it has been established that computations involv
ing fuels with other hydrogen/carbon ratios can be handled on the charts
with good accuracy.9
One pound of air is used for the basic working fluid for these charts
since the number of moles of gases varies with dissociation. The charts
are constructed for 80, 90, 100, 110, and 120 per cent of theoretical fuel;
is,
that the chart for 90 per cent theoretical fuel for 1/0.90, or 111 per
is
cent of theoretical air. In the high-temperature combustion range, the
combustion products behave thermodynamically like perfect gases, and
if
any two of the properties — pressure, temperature, volume, internal energy,
enthalpy, or entropy — are known, the other four can be determined.
In the charts, Es of the products C02, H2O, 02, and N2 set equal to
is
zero at 560R. H, of the products at 560 R then PV/J, or NRT/J,
is
where N equals the number of moles of gases per pound of air. Thus,
E, = and H, averages about 41 Btu per pound of unit mixture at 560 R.
0
unit mixture that mass of mixture which contained lb of air before

A
is
1
combustion. The total internal energy at any temperature equal
is
to the sensible internal energy Es released in cooling the products at con
stant volume to 560 R, plus the heating value at constant volume Qv of
8
the unburned portion of the products. Thus,
= E. Q,
E
As a result of assigning E, = at 560 R, the internal energy of combustion

0
Qv at 560 R numerically equal to the chemical energy. Consequently,

is
since
PV
H. = E.+
f
and
PV PV
t-j~
H = = E. +
^
+ +
E
Qv
H = H.
+
Qv
From previous discussions can be shown that the total enthalpy of

it
burned products in equilibrium may be computed from either of the

following
:
Hp = H,mT, + Qv
Hp = H,pTl
+
Qp
where Hp = total enthalpy of burned mixture at equilibrium temperature

H,mTl = sensible enthalpy of unburned mixture atinitial temperature
HspTl = sensible
enthalpy of products of complete combustion at
precombustion temperature
In each of the above equations the heating value of the fuel expressed
is
as positive number to be consistent with the charts. While Qv repre

a
sents the heating value at 560 R, heating values at 77 may be used

F
without impairing the accuracy.

The following example will serve to illustrate the use of this type of
chart for computing adiabatic flame temperatures.These charts can be

used in many types of high-temperature combustion calculations.
Example 14. Calculate the adiabatic flame temperature for octene C8Hi8, burning
with theoretical air at atmospheric pressure, initial mixture temperature, 1600 R.
Solution. The flame temperature for octene burning with theoretical air is in the
region where dissociation must be considered. Hence, charts which include this
effect will be used in the calculations. The combustion equation for octene is
C8H16 + 1202 + 45.12N2-> 8C02 + 8H20 + 45.12N2
For 1 mole of fuel, the change of sensible enthalpy of the fuel/air mixture between 560
R is
and 1600
Fuel: - =
- 3886.6)=
(90,870 16,200) 74,670
12(11,832.5
02:
- 3889.5)
95,351
N2: 45.12(11,409.7 = 339,311
XH = 509,332
Weight of air required to burn 1 mole of fuel is
(12 + 45.12) moles air X 29.0 = 1656.5 lb
Hence, AH of the mixture between 560 and 1600 R is
509,332
,
' Btu -
„„_ . D1 ,,,
- ., — = 307.5 Btu /lb unit mixture
.. . .
1656.5 lb air
The moles of initial mixture per pound of air equal
(1 + + 45-12) = 0.0351
?;
1656.5
Since E, = 0 at 560 R on the charts, H, at 560 R equals
= 0.0351 X 1.986 X 560 - 39.0

Therefore, the chart value of H,m at 1600 R equals
307.5 + 39.0 = 346.5 Btu/lb unit mixture

The theoretical fuel/air ratio, by weight, is
112. lb fuel
0.0677
1656.5 lb air
The charts are based on computations involving the lower heating value of fuel at
constant volume. From Table 3-3, the lower heating value of liquid fuel at constant
pressure is 19,000 Btu/lb. If the fuel is assumed to be in the vapor state at 1600 R,
its heating value would be increased by the amount of the latent heat of evaporation
of the octene, to give 19,158. A further correction to constant- volume conditions
may be made by applying Eq. (8).
Op - Q, = (iV, - Nm)
~
= (16-13)L986 X 560
= 30 Btu
Q, = -19,158 - 30 - -19,188 Btu/lb

Hence the lower heating value at constant volume, vapor fuel, is 19,188 Btu/lb.
Note that little error would be introduced by neglecting to correct the heating value
listed in Table 3-3 to constant volume. The heating value of fuel available per pound
of air in the theoretical mixture is
0.0677 X 19,188 = 1299 Btu/lb air
Since H = H. + Q„, ., , Q
H = 346.5 + 1299 = 1645.5 Btu/lb unit mixture
From the chart for theoretical air /fuel mixtures, F = 1.0, at a pressure of 14.7 psia,
H = 1646, T = 4550 R, or 4090 F.
Note that this flame temperature, calculated with the aid of a chart which takes
proper account of dissociation, is about 650 F lower than the theoretical flame tem
perature determined in Example 8, where no effects of dissociation were considered.
Problems
1. Using the equation from Table 3-1, calculate the instantaneous specific heat
at constant pressure of CO2 at a temperature of 3600 R. Ans. 14.50 Btu/(mole) (R)
2. What is the average of the instantaneous specific heats at constant pressure for
C02 between the temperatures 3600 and 600 R? Ans. 11.87 Btu/(mole)(R)
3. Calculate the mean specific heat at constant pressure for C02 between the
temperatures 600 and 3600 R. How does this compare with the average of the
instantaneous values found in Prob. 2? Ans. 12.95 Btu/(mole)(R)
4. Check the value of mean molal specific heat of O2 at 4000 F given in Fig. 3-4,
using the equation for instantaneous specific heat given in Table 3-1.
6. Check the value of Cpm of nitrogen at 3000 F given in Fig. 3-4, using the instan
taneous specific heat data of Table 3-1.
6. How much heat must be removed from 88 lb of CO2 to cool the gas at constant
pressure from 3400 to 500 R? Use the values of Cpm plotted in Fig. 3-4 to work this
problem and then check by using the enthalpy data in Table 3-2.
7. Show the justification for using a mean value of Cp of 0.24 Btu/(lb)(R) to find
the heat loss per pound of dry flue gas from a boiler when the flue gases test. 10% CO2,
8% O2, 0.5% CO. The flue-gas temperature at the breeching is 550 F, and the air
entering the furnace is 80 F.
8. Calculate the percentage of error in using a constant value of C„ = 0.1715
Btu/(lb)(R) to heat the products of combustion of octane with theoretical air from
1000 to 3000 R at constant volume instead of taking into account the variation of
specific heat with temperature.
9. A gaseous mixture consisting of 82 moles of N2, 16 moles of CO2, and 2 moles of
O2 is cooled at constant pressure from 2000 to 400 F. Compute the heat removed,
the average Cpm in Btu/(lb) (R) and Btu/(mole) (R), and the average molecular weight
for the mixture.
10. Find the enthalpy change of the theoretical products of combustion of CH4
in cooling from 2500 to 500 F. Express the answer in Btu per mole of CH4 and in
Btu per cubic foot of CH4 at 14.7 psia and 60 F. Also compute the mean specific
heat per pound and the average specific gas constant for the mixture of combustion
products.
11. Find the higher heating value at constant volume for C2H6 if the higher heating
value at constant pressure is 22,304 Btu/lb at 77 F.
12. Compute the difference in the higher heating value at constant volume and at
constant pressure for octane at 77 F.
13. Using the data from Tables 3-2 and 3-3, calculate the lower heating value at
constant pressure of CsH18 at 1500 R.
14. Check the lower heating value of C2H5OH at constant pressure in Table 3-3,
using the value given for the higher heating value.
16. Check the same as above for C6Hu.
16. Compute the higher heating value of octane for gaseous fuel at constant pres
sure and a temperature of 77 F from the data of Table 3-3.
17. Determine the heating value per cubic foot of gas fired and the heating value
per cubic foot of theoretical air/gas mixture supplied to the furnace for the following
water gas, assuming that the gas and air enter the furnace at 60 F and 14.7 psia.
Gas composition, by volume: CO2 = 5.5%; 02 = 0.1%; CO = 38.3%; H2 = 52.8%;
CH4 = 0.4%; N2 = 2.9%. Ans. 298 Btu/cu ft gas and 93.1 Btu/cu ft mixture.
18. Determine the theoretical flame temperature (complete combustion assumed)
for CH4 with theoretical air used for combustion in a constant-pressure process where
the air enters at 77 F.
Same as above except that 50 per cent excess air is used.
19.
20. Same as Prob. 18 except that the burning is at constant volume.
21. Same as Prob. 18 for C10H22.

22. Compare the heating value per cubic foot of a chemically correct mixture of
fuel vapor and air at 77 F and 14.7 psia for the following fuels: CH4, C8H18, C2H2,
C6H6, and Ci6H31.
How many pounds of air per 10,000 Btu of heating value must be supplied
23.
theoretically to burn each of the following fuels: a typical low-volatile bituminous

coal, a high-volatile bituminous coal, a gasoline, a medium fuel oil, a heavy fuel oil,
natural gas?
24. A mixture of oxygen, hydrogen, and water vapor is in equilibrium, and
experimental data show that the H2O is 4 per cent dissociated at a pressure of 14.7
psia. Determine the equilibrium constant for the reaction
H2 + HO, ^ HsO
and find the temperature of the mixture.

Find the degree of dissociation of H20 at 4800 R and a pressure of 147 psia.
25.
26. What percentage of dissociation of H20 will exist if equilibrium is established
at 3000 R, 14.7 psia, with 200 per cent excess oxygen present?
27. Calculate the percentage of dissociation of CO2 at 4600 R at pressures of 0.05,
0.10, 0.20, and 0.30 atm.
What percentage of theoretical air would be needed to maintain a temperature
28.
in the combustion products of 2600 F for a gasoline having an average molecular
formula of C9Hi9 and a lower heating value of 18,900 Btu/lb? Combustion is at
constant pressure and the initial air temperature is 240 F.
29. Plot
a graph of maximum adiabatic flame temperatures at constant pressure
versus the percentage of excess air used for combustion of heptene, CjHh, for an
initial mixture temperature of 100 F. Use the appropriate charts included in the
back of the book.
30. Compare the maximum flame temperatures at constant volume and at constant
pressure for ethyl alcohol if 110 per cent of theoretical fuel is used for combustion
and the initial mixture temperature is 400 F.
31. Compare the maximum flame temperatures at constant volume for propane
if 120 and 90 per cent of theoretical fuel are used for burning mixtures initially at
960 F.
References
1. Keenan, J. H., "Thermodynamics," John Wiley & Sons, Inc., New York, 1941.
2. Obert, E. F., "Thermodynamics," McGraw-Hill Book Company, Inc., New York,
1948.
3. Zemansky, M. W., "Heat and Thermodynamics," 2d ed., McGraw-Hill Book
Company, Inc., New York, 1943.
4. Keenan, J. H., and J. Kaye, "Gas Tables," John Wiley & Sons, Inc., New York,
1948.
5. Hottel, H. C, and C. N. Satterfield,
Generalized Thermodynamics of High
Temperature Combustion, ASME
Trans., vol. 70, p. 667, 1948.
6. Bogart, D., D. Okrent, and L. R. Turner, Thermodynamic Charts for the Com
putation of Fuel Quantity Required for Constant-pressure Combustion with
Diluents, NACA TN 1655, July, 1948.
7. Hottel, H. C, G. C. Williams, and C. N. Satterfield, "Thermodynamic
Charts for Combustion Processes," John Wiley & Sons, Inc., New York, 1949.
8. McCann, W. J., L. R. Turner, and E. A. Bauer, Thermodynamic Charts for
Internal-combustion-engine Fluids, NACA TN 1883, 1949.
9. Hottel, H. C, and H. C. Tsien, Effect of Hydrogen-carbon Ratio of Fuel on the
Validity of Mollier Diagrams for Internal Combustion Engines, J. Inst. Aeronauti
cal Sci., vol. 5, no. 5, 1938.
Symbols
c. - specific heat at constant volume, Btu/(mole)(R) or Btu/(lb)(R)

= specific heat at constant pressure, Btu/(mole)(R), or Btu/(lb)(R)
Cpm = mean specific heat at constant pressure, Btu/(mole)(R), or Btu/(lb)(R)
E = total internal energy, Btu
E, sensible "internal energy, Btu
e = total internal energy per mole, or per lb
e. sensible internal energy per mole, or per lb
e/«
= internal energy of evaporation, Btu /mole, or Btu /lb
H total enthalpy, Btu
H, = sensible enthalpy, Btu
h total enthalpy per mole, or per lb
h. = sensible enthalpy per mole, or per lb
enthalpy of evaporation, Btu /mole, or Btu /lb
J Joule's constant, 778 ft-lb/Btu
K = equilibrium constant
m = mixture (used as a subscript)
N = number of moles of substance
P = pressure, abs, lb/sq ft, or atm
P. = partial pressure of gas, (a), atm
Vproducts (used as a subscript)
Q = transferred heat, Btu
Q. = heating value at constant volume, Btu /mole or Btu /lb
Q, = heating value at constant
pressure, Btu /mole or Btu /lb
Qr, -
heating value at temperature 7\
B = universal gas constant, 1545 Btu/ (mole) (R)
R degrees Rankine
T temperature, R or F
u = velocity, ft/sec
V = volume, cu ft
v = volume per mole or per pound, cu ft
W = work, Btu
w = weight of substance, lb
z = elevation, ft
ii = chemical internal energy, Btu
CHAPTER 4
THE PROCESS OF COMBUSTION
For many centuries man has studied the nature of fire in an attempt to
control it and use it more efficiently. At one time, it was thought there
were only four elements which composed all nature: fire, water, air, and
earth. In fact, fire has been regarded with awe by men throughout
history, for the useful effects it could perform as well as for the terrible
destruction it might cause.
Experiments with fire were conducted by Robert Boyle and his students
as early as 1630, more than a century before the discovery of oxygen by
Lavoisier. Later in the seventeenth century the phlogiston theory of
burning became popular. According to this principle, everything con
tained an unknown ponderable substance called phlogiston which caused
a flame to appear when it escaped rapidly from whatever was burning.
It was not until the latter part of the eighteenth century that Lavoisier
was able to supplant that theory with the oxygen theory of combustion.
Since the time of Boyle, investigators have spent hundreds of man-
years studying the effects of the numerous variables on burning. Never
theless, many aspects of combustion are still only partly understood. In
the chapter which follows, some modern theories on the subject are
presented along with experimental results showing the actual effects of
different physical factors on the burning process.
Combustion. Combustion is the rapid, high-temperature oxidation of
fuels. Since most fuels used at present consist almost entirely of carbon
and hydrogen, burning involves the rapid oxidation of carbon to carbon
dioxide, or carbon monoxide, and of hydrogen to water vapor. The
combustion reaction takes place in the gaseous phase, except for the
burning of the fixed carbon in solid fuels. Even in the latter case, the
oxygen and the combustion products exist as gases, and only the fixed
carbon itself is present as a solid.
Flame may be defined as gas rendered luminous by the liberation of
chemical energy. The flame front is the surface or area between the
luminous region and the dark region of unburned gas which exists in all
combustion reactions in the gaseous phase. Since the gases may not
become luminous instantly, it is expedient to visualize the burning zone
as consisting of a luminous zone and a reaction zone.1 Ignition and most
114
THE PROCESS OF COMBUSTION 115
of the oxidation occur in the latter zone, while completion of burning and
emission of light take place in the luminous zone. Generally, the locally
available supply of oxygen is consumed in the reaction zone. It is
difficult to make a clear distinction between these regions because the
total thickness of the burning zone may vary from a few thousandths of
an inch to an indefinite thickness, depending upon the turbulence and
the homogeneity of the gases.
The above definitions provide for the persistence of flame until lumi
nosity and radiation cease, even after the chemical reaction has proceeded
to equilibrium. The continuance of luminosity, called afterburning, is
evident in various types of combustion, especially in spark-ignition
engines. A fast-burning mixture has a very thin reaction zone, and ig
nition, combustion, and luminescence occur almost simultaneously. Fig
ure 4-8 shows a diagrammatic sketch of the various zones which exist at
the area supporting the combustion process.
TYPES OF APPLIED COMBUSTION PROCESSES

Fuels burn under diverse conditions in apparatus and machines utiliz
ing the combustion process. At the present time it is not possible to
resolve the process of burning into several well-defined laws which can
always be applied to the combustion process in its many different forms.
Therefore, it is convenient to list the various types of flames and study
them individually, even though there is not a clear-cut distinction between
some of them.
Stationary Flames. A stationary flame is one in which the flame front
is more or less stationary in space; the unburned gases flow toward the
reaction zone at the proper speed to maintain the position of the flame.
This type of flame may be further classified as combustion (1) in which
the fuel is premixed with air or (2) in which the fuel and air enter the
combustion area separately. The latter is called a diffusion flame as it
becomes necessary for the oxygen to be diffused into the reaction zone
and mixed with the fuel before burning can occur.
Stationary flames are utilized at atmospheric pressure or at other
pressures, higher or lower. Gas burners, pulverized-coal burners, and
some oil burners employ this type of flame. The flow of mixture to these
flames may be either laminar or turbulent. If there is a great deal of
turbulence, the reaction zone and resulting flame front may be irregular
and rather unsteady. Such turbulence may create what appears to be a
solid cone of flame in the vicinity of the combustion.
Some oil and gas burners operate with a diffusion flame. In these cases
considerable turbulence of the air and fuel in the combustion chamber is
employed to ensure fairly rapid mixing. A true diffusion flame creates a
much longer zone of reaction, and there are comparatively few applica
tions where diffusion of the gases alone is relied upon to provide mixing of
the fuel and oxygen.
Explosion Flames. Explosive burning occurs in homogeneous mix
tures of fuel and air, and the flame front progresses rapidly through the
mixture. This type of flame may again be further classified as a (1) con
stant-pressure or (2) a constant-volume reaction.
The former is the type of combustion which exists during a mine explo
sion, in which a flash fire races through the mine from the point of igni
tion. While there is a localized pressure disturbance at the reaction zone,
no appreciable rise in pressure is created in the mine shaft. In constant-
pressure explosions the chemical reaction is generally completed at the
flame front and little or no burning takes place after the passage of the
flame. Characteristics of such flames have been studied rather carefully
in glass tubes.
The explosive flame in the cylinder of a gasoline engine is essentially a
combustion process at constant volume. In this application, the mixture
of air and fuel is ignited with a spark, and the flame front spreads very
rapidly across the cylinder. The high degree of turbulence in the cylin
der aids in the rapid propagation of the flame, resulting in a very irregular
flame front and reaction zone.
Because of the widespread interest in this application of combustion,
many fundamental flame studies have been made on explosion flames in
constant-volume cylinders or bombs. Factors which apparently have no
effect on stationary flames at constant pressure seem to have considerable
importance in explosion flames where the pressure is rising rapidly.
Detonation is a special type of explosion characterized by a tremen
dously increased reaction rate accompanied by an ultra-high velocity per
cussion wave within the cylinder and an abnormal rate of pressure rise.
Detonation is considered more thoroughly in Chap. 8, since it is a problem
encountered in internal-combustion engines.
MECHANISM OF COMBUSTION
The term mechanism of combustion refers to the reactions by which fuel

is transformed chemically to combustion products. In Chap. 2, calcula
tions are made to determine the relationships between air, fuel, and com
bustion products. For the combustion of octene, C8Hi6, the theoretical
combustion equation reads,
C8H16 + 1202 -> 8C02 + 8H20
According to present theories of chemical reaction kinetics, it is necessary

for molecules of one substance to collide with molecules of another with
an energy equal to or greater than a certain critical minimum before any

chemical reaction can take place. The theoretical combustion equation
for octene written above implies that if the reaction were to occur in one
step, it would be necessary for 1 octene molecule to collide simultaneously
with 12 oxygen molecules before the reaction could take place. Even
though a molecule of a gas may collide with other molecules several
billion times a second, it is still highly improbable that 13 molecules in the
right proportions (1 of octene and 12 of oxygen) would collide simul
taneously. In fact, a study of the problem from the standpoint of the
mathematics of probability shows that the "probability" of more than
three molecules colliding simultaneously is negligible. Hence, any com
plex reaction involving many molecules must occur as a series (or chain)
of different reactions starting with the initial substances and proceeding
to the final products.
Chain Reaction Theory. A self-sustaining chemical process which con
sists of a series of different reactions in which intermediate products are
formed in one step and destroyed in a succeeding step is known as a chain
reaction. The intermediate products formed are known as chain carriers
since they help carry the reaction to completion. Chain carriers may be
free atoms of diatomic gases (such as H and 0), free radicals (like OH,
CHO, CH, etc.), or some organic compound (such as formaldehyde,
HCHO). A free radical is a group of atoms which carries one unpaired
electron. In other words, a free radical has a free valence bond. The
hydroxyl free radical, OH, for example, may unite with a free hydrogen
atom to form a water molecule, H2O, or it may enter into many other
reactions. A chain carrier may exist only a minute fraction of a second.
Billions of chain carriers are formed and instantly destroyed during the
course of a chain reaction.
Any chain reaction consists of an initiation phase, a propagation phase,
and a termination phase. In the first phase the chain carriers are formed
which promote the propagation phase. Combustion may be terminated!
by a chain-breaking reaction in which some of the chain carriers are taken
out of play by another substance which reacts with or adsorbs the chain
carriers. A cold combustion chamber wall in an oil burner, for instance,
apparently adsorbs enough of the chain carriers to stop the combustion of
fuel oil near the surface. As a result there is a strong tendency to deposit
partially burned fuel, or soot, on combustion chamber walls when they^
are cold.
Some intermediate reactions occur in such a manner that several chainj
carriers are formed with each step. Such a reaction is known as a chain-
branching reaction. Each of these new chain carriers may then branchy
out and start a new series of reactions of its own. If two or three new
chemical reactions are started each time one is completed, it is easy to vis
ualize how chain branching can speed up a reaction to explosive violence.
During the past 50 years considerable research work has been per
formed in an attempt to establish the mechanism, or series of individual
intermediate reactions, by which fuels burn. Unfortunately, this is an
extremely complex problem affected by a large number of physical factors.
Even today the mechanism involved in the burning of any but the
simplest of fuels is not definitely known. Spectroscopic analysis of the
reaction zone and flame front shows the presence of certain intermediate
products. The various reactions which occur can be predicted by apply
ing theories of reaction kinetics to qualitative and quantitative data con
cerning the products of the reactions, reaction rates, etc.
In the sections which follow, recently proposed mechanisms are pre
sented along with some experimental evidence to help support these
theories. Only those reactions are given here which are reasonable in
energy release and which do not involve too complicated an atomic
rearrangement. By the very nature of this problem, it becomes obvious
that such experimental work would be performed in the physical chem
istry laboratories.
Hydrogen. The mechanism of the combustion of hydrogen with
oxygen is presented first, not only because hydrogen is one of the simpler
fuels, but because the chain reactions involved in its burning are the most
thoroughly understood.
During the initiation phase of the combustion reaction, active chain
carriers are formed by such reactions as
H2^ 2H
02-> 20
H20 -> OH + H
The preceding reactions may result from intermolecular collisions, thermal

dissociation, or excitation from an electric spark. These active particles
also may be diffused into the unburned gaseous mixture from an adjacent
flame.
B. Lewis and G. von Elbe2 in their studies found the following propaga
tion reactions to be predominant in the combustion of hydrogen :
(1) OH + H2 -> H20 + H

(2) H + 02^ OH + 0
(3) 0 + H2^ OH + H
The OH free radical must be present to promote the reaction in Eq. (1).
The products of Eq. (1) are H20 and a free hydrogen atom which reacts
in step (2) with an oxygen molecule to form another OH radical and a free
oxygen atom. Both the second and third steps are branching reactions
as in each equation two chain carriers appear in the products while only
one enters into the reaction. It is quite possible that a free hydrogen
atom also can initiate the reaction chain. Following is a scheme showing
how chain branching builds up with the above three reactions:
1st Link 2d Link

H + 02 ^ OH + 0
I ±5! > OH + H
±^2 > H20 + H
The firstlink in the chain [Eq. (2)] destroys one chain carrier — a free
hydrogen atom. By the end of the second link of the chain, composed of
Eqs. (1) and (3), two new chain carriers have been formed, each of which
may start its own new chain. Thus, combustion proceeding by the
above reactions quickly accelerates to explosive velocity. When a mix
ture of hydrogen and oxygen is ignited at temperatures above about
750 F, flame envelops the mixture almost instantaneously. The rate of
the reaction may be slowed, or even stopped, however, by introducing
chain-breaking reactions rapidly enough to destroy some of the chain
carriers as they are formed.
It has been mentioned that cold wall surfaces apparently act to adsorb
chain carriers and thus slow or stop chain reactions. The effect of this
phenomenon can be observed by mixing stoichiometric portions of hydro
gen and oxygen at room temperature and pressure. If such a mixture
is not ignited, the number of chain carriers adsorbed at the walls of the
vessel is large as compared with the number of "successful" collisions of

the chain carriers with hydrogen and oxygen molecules, and the mixture
does not react.
The term "successful" is used advisedly, as not all collisions of mole
cules and chain carriers result in a chemical reaction. The molecules in
any gas are moving at various speeds, some very rapidly, a larger number
at a speed close to the average of the group, and a few very slowly, as
shown in Fig. 4-1. When the temperature rises, the percentage of

rapidly moving molecules increases.
In order that a chemical reaction may occur, it is necessary for the
molecule or molecules involved to be sufficiently activated, i.e., to attain
the necessary energy level. One means by which this level may be
reached is by intermolecular collisions in which the combined kinetic
energies of the colliding molecules equal or exceed the requirement.
Thus, at higher temperatures the number of successful collisions per
second is much greater than at low temperatures because more molecules

are traveling at the higher velocities and hence possess the requisite
kinetic energy.
If there is no ignition of a hydrogen/oxygen mixture at ordinary
temperature and pressure, the number of successful collisions of the mole
cules in the mixture is not great enough to offset the destruction or
adsorption of chain carriers. Under such conditions, oxidation of the
hydrogen proceeds, but at such a slow rate that it may take years for
the formation of a detectable amount of water. As the temperature is
600 1200 1800 2400

Molecular velocity U, ft/sec
Fig. Relative number of molecules
4-1. having a given molecular velocity, U, at different
temperatures.
increased to around 750 F, the rates of the reactions increase because of

an increased number of successful collisions and also because the chain-
branching reactions become predominant. An explosion results. Thus,
two factors seem to be present to create explosive reactions: (1) a thermal
effect and (2) a chain-branching effect. Chemical kinetic calculations
show that as little as 1 per cent branching in the hydrogen-oxygen chain
reactions would result in speeding the combustion to explosive velocity.
If hydrogen and oxygen are admitted at a very low pressure to a vessel
and heated to a temperature even above 750 F, the relatively small num
ber of molecules present in the rarefied gases does not yield sufficiently
frequent successful collisions of chain carriers and reactant molecules to
offset the adsorption of chain carriers at the wall surfaces. Hence, rapid
burning is impossible. As the pressure is increased, more and more mole
cules collide per second. At some critical pressure, known as the first
explosion limit, the effect of chain branching becomes greater than the
effect of chain breaking, and an explosion occurs.
chain-breaking reactions at the walls of the container are an
Since
important factor controlling the rate of the combustion, the size of vessel,
type of surface, and the condition of the surface are all variables affecting
the first explosion limit. Because of the poisoning of the wall surface,
or saturation of the wall surface with molecules which do not permit the
wall to adsorb chain carriers in its normal manner, even similar wall
materials do not always act to give the same pressure at the first explosion
ZONE TK
Pressure, psia (not plotted to scale)

Fig. 4-2. Zones of slow combustion and explosion for hydrogen for constant mixture
compositions at temperatures above the ignition temperature.
limit. Thus, the first explosion limit is difficult to determine and is very
much dependent on test apparatus and procedure.
An excellent example of the complicated nature of the combustion
process is the fact that there is a pressure, somewhat higher than the
pressure at the first explosion limit, above which the chemical reaction of
hydrogen and oxygen no longer proceeds with explosive violence. This
critical pressure is called the second explosion limit; above this limiting
pressure the oxidation of the hydrogen proceeds slowly, as shown in Fig.
4-2. Oddly enough, vessel dimensions and wall surface conditions have
little influence on the second explosion limit. Instead, this phenomenon
seems to be created by a chain-breaking reaction of the form
(4) H + 02 + M -» H02 + M
where M represents any kind of molecule, such as H2, 02, H2O, N2, C02,
etc. Ordinarily H02 is a very unstable combination of hydrogen and
oxygen because more energy is released upon its formation than is
required to bring about its decomposition. However, if any other mole

cule is present to absorb the excess energy released from the formation of
HO 2, the latter may exist as an intermediate product in the chain reac
tion. In Eq. (4), M represents any "third body" which might act as an
energy acceptor and stabilizer to expedite the reaction. Equation (4)
serves to break the hydrogen-oxygen reaction chain by removing free
hydrogen atoms from the mixture. The H02 radicals are adsorbed on
the wall surfaces and, as the surface becomes saturated, the HO2 radicals
are destroyed by the reaction
(5) 2H02 H202 + 02

The H202 and 02 then leave the wall and reenter the gaseous mixture.
Thus, the net effect is the removal of H atoms from the gaseous mixture,
which results in breaking, or at least considerably slowing, the reaction
chain.
The two reaction paths open to the hydrogen atoms are then:
— ±h. „ OH + H
> H20 + H
+°, (Chain-branching path)
I
-> 0 + OH
I
' ±^ > H20 + H
H
+°2 M
I 8'
> H02 + M (Chain-breaking path)
Neglecting the effect of other reactions, at pressures between the first

and second explosion limits more than one-third of the hydrogen atoms
react according to the chain-branching reaction. Whereas, at pressures
above the second explosion limit the possibility of a hydrogen atom
colliding with an oxygen molecule near a third molecule, which absorbs
the bond energy of the creation of H02, is great enough so that more than
two-thirds of the hydrogen atoms follow the lower path. Thus, at the
second explosion limit, reaction between oxygen and hydrogen proceeds
in such a way that Eq. (4) becomes more dominant than Eq. (2).
Besides Eq. (5), two other reactions may occur with the H02 which
alter the second explosion limit. If the surface is not very efficient in
adsorbing H02, Eq. (5) may not proceed and the H02 radicals may
escape. In the gaseous phase the following reaction is possible.
(6) H02 + H2 -» H202 + H

The above reaction helps nullify the chain-breaking ability of Eq. (4),
and the second explosion limit is related to the degree to which Eq. (6)
occurs. This effect is usually not very great, however. Clean quartz
and certain types of clean glass surfaces apparently have lower adsorptive
qualities and allow Eq. (6) to proceed to a greater degree than do vessels
coated with certain salt solutions.
In addition to the reaction path denoted by Eq. (6), it is possible occa
sionally for the H02 to react according to Eq. (7), even in vessels with
excellent adsorptive efficiency.
(7) H02 + H202 -» H20 + 02 + OH
The above reaction nullifies the chain-breaking powers of Eq. (4), since
the OH radical will form another H atom by the path of Eq. (1).
The reaction of Eq. (5) is the most common of the three possible paths
for the destruction of H02. When it occurs, two-thirds of the free hydro
gen atoms must react according to Eq. (4) to keep the burning from
progressing to explosion, since it removes the free hydrogen atoms from
the gaseous mixture. The reaction of Eq. (6) may develop in the gaseous
phase when the walls fail to adsorb the H02 radicals adequately. As
Eq. (6) shows, another H atom is released by this reaction and the effect
of the chain breaking of Eq. (4) is impaired. Under these conditions it
becomes necessary for more than two-thirds of the free hydrogen atoms
to follow the path of Eq. (4) to offset their formation again by Eq. (6).
Thus, the wall surface has some effect on the second explosion limit,
although this effect is not so marked as it is for the first explosion limit.
It is significant that the primary reaction product, water, itself poisons
the surface of certain vessels. The poisoned surface apparently hinders
the adsorption of H02. The reaction rate progressively increases as the
wall prevents the effectiveness of chain-breaking Eq. (4) . If water vapor
is sprayed over the walls at the beginning of a test, the reaction rate is
rapid from the start.
When the path of Eq. (7) is followed by an H02 radical, the chain-
breaking tendency of Eq. (4) is again impaired. In this case three-
fourths instead of two-thirds of the H atoms must follow Eq. (4) to com
pensate for the later formation of the OH radical.
The intermediate product H202 formed by either Eq. (5) or (6) has
four reaction paths by which it may be destroyed:
(7) H202 + H02 -» H20 + 02 + OH

(8) H202 (by dissociation) -» 20H
(9) 2H202 (by decomposition) -> 2H20 + 02
(10) H202 + H + 02 -» H20 + 02 + OH
The manner of destruction of H202 also affects the over-all mechanism.

If H202 is destroyed by the reaction of Eq. (9), where no OH chain carrier
is formed, the chain-breaking power of Eq. (4) is maintained. By any of
the other three possible paths, another OH radical is formed. Under the
latter conditions, a larger number of H atoms must follow the chain-
breaking path Eq. (4) to prevent an explosive reaction. Equation (8)
may occur if an H202 molecule collides with some other molecule with
sufficient energy to split it into two OH free radicals. When the proper
catalytic actions are in force, H202 vapor may decompose by the path of
Eq. (9). The nature of the molecular collisions controls the number of
H202 molecules following Eqs. (7) and (10).
A third explosion limit leading to a very violent reaction exists at pres
sures considerably above the second explosion limit when the diffusion of
H02 to the wall, or its destruction at the wall, is impeded by any of the
following changes:
1. Increase of pressure — by adding H2, 02, or even an inert gas to the

system
2. Increase of vessel diameter
3. Poisoning of the surface
It is theorized that any combination of the above conditions leads to an

increased tendency for the H02 radicals to follow Eq. (6).
H02 + H2 -» H202 + H
This, of course, destroys the chain-breaking effect of the formation of

H02, chain branching becomes predominant, and the reaction accelerates
to explosive velocity.
The individual reactions of Eqs. (1) through (10) make up a chain of
reactions, called a reaction mechanism, for the combustion of hydrogen.
Unfortunately, no experimental means has been devised for studying the
various possible reaction paths independently of each other. The reac
tions denoted by the above equations are the most probable chain of
reactions leading to the formation of water. Many other reactions are
possible but they occur infrequently.
Figure 4-2 shows the significance of the various explosion limits. The
principal reactions which seem to govern the rate of burning in the
various zones are summarized in the following equations:
Zone I. Wall surfaces adsorb OH, H, O chain carriers rapidly enough

to prevent sustained chain reaction.
Zone II. OH + H2 -» H20 + H
+
0,-» OH + 0
+ + (chain-branching
H2 H2 mechanism)
U OH + H
-> H20 + H
Zone III.Reactions of Zone II

and H + 02 + M -» HO2 + M
2H02 (adsorbed at the wall) — » O2 + H2O2 (chain-breaking mechanism)
2H202 (decomposed at the wall) — » 2H2O + O2
Zone IV. Reactions of Zone II and H + O2 + M -» HO2 + M
H02 is destroyed by some combination of Eqs. (5), (6), (7), and (10).
Net effect is that chain-breaking mechanism is no longer effective.

Hence, chain branching predominates.
01 I I I 1 1 I I I
1000 1020 1040 1060 1080 1100 1120 1140 1160
Temperature, F
Fig. 4-3. Second and third explosion limits for mixtures of 2H2 + O2 in salt-coated vessels.
Calculated curves for vessels of diameter d are shown in solid lines. Experimental data:
T. °i X for d = 3.94-, 2. 91-, and 1.54-in. diameter, respectively, (von Elbe and Lewis,
J. Chem. Phys., 1942.)
Figures 4-3 through 4-6 are presented to show the effects of several
variables on the second and third explosion limits. No curves are given
for the first explosion limit since the nature of the adsorption of chain car
riers at the wall surfaces makes a systematic investigation of this limit
extremely difficult.
Figure 4-3 shows the second and third explosion limits for vessels of dif
ferent diameters at various temperatures. The area to the right of any
curve is the explosive region, whereas to the left of the curve is an area of
126 FUELS AXD COMBUSTION
slow or no combustion. The second explosion limit is represented by the

lower line leading up and to the right. For the section of the curve lead
ing up and to the left, the pressure at any given temperature is the third
explosion limit for the conditions stated.
The curves of Fig. 4-4 show the effect of various mixtures on the second
explosion limit. Figures 4-5 and 4-6 depict the reaction rates for different
temperatures and different H2 and O2 mixtures. The reaction rates were
measured in millimeters of Hg pressure drop per minute. The pressure
Inert gas I
added to ll
3 n
/M // -/
Mixtures of /Helium
1
H2 and Og
20 / ^{Nitrogen
H2O added to
\ 2H2+O2
IX)
Carlxxii^-^^
dioxide
.4 i i i i ! 1
—
1 1 1 1 1
0 02 0.4 Q6 Q8 0 0.04 0.08 0.12 0 02 0.4 0.6 03
fraction
Mole Mole fraction Mole fraction
of 02 of H20 of inert gas
Fig. 4-4. Second explosion limit pressures as functions of mixture composition. Tem
perature. 986 F. spherical vessel, 2.9 in. diameter, salt-coated pyrex surface. Calculated
curves are solid lines. Experimental data: X, °, +, •. (von Elbe and Lewis, Chem. J.
Phys., wji. 10, p. 366, 1942.)
drop in the reaction vessel occurs because li2 vol of hydrogen and oxygen
are required to form 1 vol of water vapor. Thus, the faster the reaction,
the greater the rate of pressure drop. Since only a comparative reaction
rate is significant, a millimeter of Hg pressure drop is plotted on the graph
as a "unit" of reaction rate.
The solid line curves in Fig. 4-3 represent values calculated from theo
retical equations, assuming that the combustion mechanism consists of
some combination of Eqs. (1) through (10). As can be seen, the calcu
lated values compare very closely to experimental data in most cases. No
other combination of possible reactions yields such excellent agreement.
100
;
40
-
o
;
4
X
*
- X
in
c
r x X
.4
-
X 8
x 9^
o ^ X
S .i
a:
:
.04
s x
- ^ 1/amete r
in.
:
.006
8 10 12 14 16
Pressure, psia
Fig. 4-5. Reaction rates for mixtures of 2H2 + O2. Calculated curves, solid lines. Experi
mental data: X, pyrex vessel coated with potassium chloride; A, pyrex with barium chloride.
(von Elbe and Lewis, J.
Chem. Phys., 1942.)
10 IT—
Second explosion limits
-20% H2 in
'
mixture
-40% H?
6 8 10 12 14 16
Pressure, psia
Fig. 4-6. Reaction rates for various ratios of hydrogen and oxygen. Calculated curves,
solid lines. Experimental data: •, 80% H2 in H2 and 02 mixture; X, 66.7% H2; +, 40%
Hi; * 20% H2, 2. 91-in. -diameter pyrex vessel, potassium chloride coated, at 986 F. (von
Elbe and Lewis, J.
Chem. Phys., 1942.)
The slow combustion reactions, such as are obtained in Zones and I III
of Fig. 4-2, are the subject of much fundamental research at the present
time. Explosive reactions are extremely difficult to handle experimen
tally, and it is hoped that, if the mechanism of slow combustion can be
discovered, the factors which control more violent reactions may be better
understood. However, the mechanisms which prevail under slow com
bustion conditions are quite different from those which lead to explosions.
Carbon Monoxide. The mechanism whereby oxygen unites with car
bon monoxide to form the ultimate combustion product, carbon dioxide,
has not been investigated so thoroughly as that of hydrogen/oxygen mix
tures. Lewis and von Elbe3 29 have suggested the following reactions
as possible in the combustion of CO and 02 mixtures free of water and
other hydrogen compounds:
(11) CO + O + M-» C02 + M

(12) 02 + 0 + M->03-|-M
(13) CO + 03 -> C02 + 20
(14) CO + 03 + M -> C02 + 02 + M
Equation (11) denotes the simple uniting of a free oxygen atom with a CO
molecule to form C02. The third body M must be present at the time
of collision to dissipate the heat released in the reaction. It is possible
that C02 is also formed by the reactions of Eq. (13) or (14). In these an
ozone molecule, 03, formed as in Eq. (12), acts as a chain carrier to pro
mote the formation of C02. Note that Eq. (13) is chain branching, as
each of the resulting oxygen atoms is free to start its own chains via Eq.
(12), while Eq. (14) is chain breaking.
The combustion of carbon monoxide is extremely sensitive to the pres
ence of water vapor in the mixture. Jost4 has pointed out that several
times as much electrical energy is required to ignite a completely dry mix
ture of CO and 02 as is necessary if a very small amount of water vapor is
present. In addition the resulting flame velocities and tendencies toward
violent explosion, or detonation, are greater if the CO and 02, or C0 and
air mixture is saturated with water vapor. Hence, it seems reasonable
that a different combustion mechanism may exist for the oxidation of CO
in the presence of hydrogen, or hydrogen compounds. The following
have been suggested3,4 as some of the possible reactions occurring in this
mechanism:
(15) CO + 02 + H -» C02 + OH
(16) C0 + OH -» C02 + H
(17) H + 02 + M -> H02 + M
(18) C0 + H02 C02 + OH
It can be noted that the above mechanism is of a type similar to that
existing in the combustion of hydrogen.
Elementary Carbon. The combustion mechanism for elementary car

bon in the solid state, asit appears in the fixed carbon in coal, coke, or
charcoal, is not thoroughly understood, and various investigators have
arrived at different conclusions as to the predominant reactions. The
basic mechanism involves the diffusion of gaseous oxygen to the surface
of the solid carbon where the oxygen molecules react to form a primary
product, which may be either carbon monoxide or carbon dioxide. This
gaseous product must then diffuse from the surface to allow more oxygen
to contact the surface molecules of carbon.
It is generally agreed by authorities that two distinctly different types
of reactions are involved, the one mechanism prevailing at temperatures
over 1800 F and the other at lower temperatures. In either case, the
speed of the actual chemical reaction is so great, when compared to the
rate of diffusion of gases to and from the carbon surface, that diffusion
controls the rate of burning almost entirely.
Since the temperatures of carbonaceous fuels used in furnaces are
almost always above 1800 F at the time of burning, it appears that the
high temperature mechanism is the one of most interest to engineers.
Strickland-Constable6 has reported that at temperatures between about
1650 and 2000 F the reaction rate increases with temperature, and above
2000 F it appears to remain fairly constant. Some C02 may be detected
as a primary product at the surface, but the concentration of the C02 is
low as compared with the C0 at all times. Evidently the oxidation reac
tion at the surface proceeds only to the formation of CO, and the gaseous
CO is oxidized at some point beyond the carbon surface. These investi
gations were carried on at pressures of 0.009 psia or lower to reduce the
burning rate considerably in an attempt to learn more of the mechanism
of the reaction. It is believed the same mechanism would prevail at simi
lar temperatures at atmospheric or higher pressures.
Methane. The combustion of methane in air has been studied inten
sively by a number of investigators, and the possible mechanisms are
fairly well established.
Etienne Audibert6, Inspector general of French Mines, has concluded
that there are four different mechanisms whereby methane and air can
unite, but that only a branched-chain mechanism produces active com
bustion, as follows:
(19) -»
02 + CH3 CHsOO
(20) CH4 + CH300 -» CH3 + CH3OOH
(21) CH3OOH -» CO + 2H2 + O
(22) CH4 + O -» CH3 + OH
The net effect of the above series of reactions may be represented by

the following equation, which sums up the action of the primary chain:
(23) 2CH4 + 02->2C0 + 4H2
The primary products C0 and H2 are then oxidized to C02 and H20 by
the secondary mechanisms:
(24) CO + OH H + C02
(25) 02 + CO + H -» OH + C02
(26) H2 + OH -» H + H20
(27) 02 + H2 + H-> OH + H20
The CH3OOH (methylhydroperoxide) is termed a propagating center,

which serves as a carrier to promote the chain. It normally breaks up
[Eq. (21)] into two primary products and an oxygen atom, thus branching
the chain. To ignite a mixture of methane and air, it seems necessary to
produce a few CH3OOH molecules to act as propagating centers. This
may be accomplished by heating the mixture to the ignition temperature
or by some auxiliary means. However, it appears quite certain that the
critical factor in ignition is the production jaf propagating centers by
inducing reactions [Eqs. (19) and (20)] and not the temperature of the
mixture. This was partially proved by A. Van Tiggelen when he ignited
a methane and air mixture by adding a small portion of acetone and
decomposing it with radiation of a suitable wavelength. The heating
effect of the fuel and air mixture was negligible, and yet the flame was
produced vigorously once the necessary flame-propagating centers were
formed. The action of an electric spark probably decomposes the CH4 to
hydrocarbon radicals which induce the chain-branching mechanism.
To prevent ignition, it is only necessary to prevent the formation of the
propagating centers. This can be accomplished by adding a few parts,
per million of ethylene dibromide to the mixture, whether the attempted
ignition is by spark or flame. It is believed that miners' safety lamps,
and other such flame-arresting devices, destroy chain-propagating centers
by a secondary catalytic reaction occurring at the surface of the wire mesh.
Lewis and von Elbe7 have proposed a somewhat different mechanism
for the oxidation of methane. In Eq. (28) formaldehyde acts as the
initiating reaction agent to form active free radicals which propagate the
reaction chain.
(28)' HCHO + 02 -> free radicals -> OH

(29) OH + CH4 -» CH3 + H20
(30) CH3 + 02 -> OH + HCHO -> CHO -» H20
(31) CHO + 02 -> C0 + H02
(32) CHO + 02 + M -> CH03 + M
(33) CH03 + HCHO -> 2C0 + H20 + OH
A highly branching reaction which is possible is
(34) HO2 + HCHO + O2 -* CO2 + 3OH
At pressures of 150 to 2250 psi, some methyl alcohol, CH3OH, appears in

the combustion products during the burning of methane. Thus, it is
highly probable that additional mechanisms exist at the higher pressures.
Jost4 has suggested that reactions involving the formation and decompo
sition of CH3OO, the methyl peroxide radical, make up the combustion
mechanism at high pressures.
Higher Hydrocarbons. In the preceding section the complexity of the
combustion mechanisms was evident, even for a simple hydrocarbon like
methane. With the more complex hydrocarbons, the number of reac
tions which may occur during oxidation becomes much greater. In fact,
the more complex the fuel, the less certainly a combustion mechanism can
be established. Hence, it can be easily understood why the exact process
is not known by which a complex hydrocarbon, such as 2,2,4-trimethyl-
pentane (isooctane), is oxidized into carbon dioxide and water vapor.
Nevertheless, several investigators have proposed chain mechanisms
which likely occur during the burning of hydrocarbons. One of the
3,1,4
schemes proposed follows.
A paraffin hydrocarbon, which may be represented by an alkyl radical,
R, with a hydrogen atom attached,* may be converted to an alkyl radical
and a hydrogen atom by high-temperature decomposition. Or, the
*
An alkyl . radicalis any paraffin hydrocarbon minus one hydrogen atom. For
example, the alkyl radical of the normal paraffin pentane,CsHu, would be CsHu,
and could be represented graphically as
H H H H H
H— •
(free)
Since the alkyl radicals of the many compounds react similarly under many
circumstances, alkyl radicals are often simply represented by R. Thus, a paraffin
hydrocarbon could be represented as RH, showing that no hydrogen atom is missing.
Some organic compounds involving alkyl radicals which appear as intermediate
or final products of combustion are:
1. Alcohols R— OH
2. Aldehydes R— CHO
3. Acids R— COOH
4. Peracids RCO(OOH)
5. Ketones R— CO— R
6. Hydroperoxides R— OOH
7. Dialkyl peroxides R— OO— R
hydrocarbon may react with an OH radical to form an alkyl radical and

water according to the equation
(35) RH + OH -» R + H20
The alkyl radical may then be oxidized to an aldehyde
(36) R + 02 — OH + R'CHO
in which the symbol R' represents the alkyl radical with one less CH2
group than the original alkyl radical, R. The aldehyde thus formed may
then react with OH to yield a carbonyl radical and water
(37) R'CHO + OH -» R'CO + H20
Further oxidation of the carbonyl radical yields the peracid radical
(38) R'C0 + 02 -» R'C0(OO)
which reacts directly with a new hydrocarbon molecule to form peracid

and a new alkyl radical.
(39) R'CO(OO) + RH -> R'C0(OOH) + R
The peracid formed is very unstable at high temperatures and decom

poses according to the equation
(40) R'C0(OOH) -» R"CHO + C0 + H20

The R"CHO is the aldehyde of the next lower alkyl radical than R'. In
other words, the aldehyde R'CHO has been oxidized to the aldehyde
R"CHO through the peracid path.
The net effect of the peracid oxidation chain is progressively to shorten
the length of the alkyl radical in the aldehyde. This can be shown by the
following summarizing equations:
(C7H15)CHO + 02-+ (C6H13)CHO + C0 + H20

(C6H13)CHO + 02^ (C«Hn)CHO + CO + H20
,
(C5Hn)CHO + 02 -> (C4H9)CHO + C0 + H20
The above has been termed the aldehyde degradation mechanism.

The oxidation occurs on the aldehyde derivative of the alkyl group; suc
cessive CH2 groups are picked off the molecule with each step in the chain
reaction, forming one C0 and one H20 each time the aldehyde is degraded.
o
o
T
M
o
t
w
X
o
w
o
o
o
w
g
w
w
o
o
a
o
a
o
«
«5
Just as in the combustion of hydrogen and methane, many other reactions

probably occur simultaneously with the chain listed above. In fact, it is
thought that oxidation paths may progress through the aldehydes, the
hydrocarbons, the peracids, or the peroxides. Certain of these reaction
paths lead to chain branching and a much increased reaction rate, while
others may tend to inhibit the oxidation.
The preceding page shows a tentative scheme for the oxidation of
propane.8,1 This mechanism was designed to describe the results of a
number of investigations in combustion of propane specifically, and to
point out some facts which seem evident in connection with the com
bustion of higher hydrocarbons in general.
^
It will be noted that one of the recurring reactions is the formation of
an aldehyde from the basic propane molecule. The aldehydes are con
tinually being formed and destroyed throughout the series of reactions.
In the side reaction listed at the bottom, methane appears as one of the
products of the combustion of propane. If the methane itself were not
completely oxidized during a subsequent reaction, a trace of it would
appear in the combustion products. Indeed, in the exhaust of spark-
ignition engines, where complex hydrocarbons are burned under high
pressures in very short periods of time, a fraction of a per cent of methane
is always present in the exhaust products.
The branching path mechanism shows various peracids and peroxides
being formed as intermediate products. These are destroyed in reac
tions which lead to chain branching. The speed of the oxidation reaction
will be controlled somewhat by the relative tendency to follow chain-
branching paths, or other paths continuing with a slower reaction rate.
In the case of combustion of normal paraffins, the oxidation apparently
follows quite closely the aldehyde degradation mechanism, with the
formation of extremely unstable peracids in the intermediate reaction
steps. Likewise, the straight-chain portions of isoparamns seem to oxi
dize by the aldehyde degradation chain, but the more compact part of
the molecule may be oxidized by reaction paths which lead to the forma
tion of ketones and alcohols which result in reduced oxidation rates.
C. E. Boord, et al., have discussed the possible steering mechanism which
directs combustion into certain reaction paths.8 Thus, the isoparaffins
apparently do not follow the same reaction paths and are somewhat more
resistant to rapid oxidation than the normal paraffins. This tendency
in general seems to hold with all compact fuel molecules, including the
isoparamns, naphthenes, and aromatics. Their greater resistance to
extremely rapid oxidation is evident in the reduced tendency of these
compact fuel molecules to detonate in the cylinders of spark-ignition
engines.
A peculiar phenomenon often displayed by the heavier normal paraffin

compounds is shown in Fig. 4-7. The reaction rate does not increase uni
formly with mixture temperature but increases, drops off considerably,
and then rises again with increasing temperature. The pressures at
8
II
i
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
I
75[
i
|
|
70
1
65
60
500 550 600 650 700 750 800 850 900 950 1000 1050
Temperature,
F
Fig. 4-7. Ignition region of 3.1 per cent hexane in air mixture. Ignition lag in seconds
indicated in numbers along curve. (Townend, Cohen, Mandle-Kar, Proc. Roy. Soc. (Lon
don), A 146, 113, 1934.)
which the reactionwill become explosive for the various temperatures are
quite unpredictable, and in this case give rise to peninsula-shaped
a
explosion region in the vicinity of 700 F. Such irregularities occur fre

quently in experimental combustion data and are probably caused by
different chain paths coming into effect with new conditions of tempera
ture and pressure.
The numbers on the curve represent the time in seconds which was
required to heat the fuel and air mixture before combustion started.
This delay period is known as the ignition lag and is discussed later.
A region of cool flames just outside the explosion area is shown on the
graph of Fig. 4-7. Cool flames are gentle, incomplete combustion reac
tions which emit little light and heat. Under special conditions, a gaseous
mixture of fuel and air may react at rather low temperatures in such a
manner that no flame front or reaction zone is visible. This type of
burning is called homogeneous combustion because the oxidation of the
carbon and hydrogen is apparently occurring at many different points in
the mixture simultaneously but at a very slow rate. This type of reaction
is also called slow combustion. The temperatures thus attained are only
600 to 1000 F, and much less light is emitted than by the usual rapid
oxidation of fuel.
FLAME PROPAGATION
The speed with which the combustion process occurs influences mark
edly the efficiency with which the heat released by the chemical reaction
can be used. With greater rates of heat release, higher peak temperatures
can be obtained. The rate of heat transfer varies with the temperature
difference between the heat source and the body receiving the heat. If
higher temperatures can be achieved through a more rapid combustion
process, it is often possible to transfer a greater portion of the heat
released to some other medium to perform a useful purpose. Likewise,
the burning rate will dictate the volume of combustion chamber required
to burn a given quantity of fuel. Following the present trend toward
the development of more compact and efficient machines, the rate of heat
release is being pushed to the maximum. This requires a high flame
speed.
Burning Rates. The spatial velocity of a flame is the velocity with
which the flame moves through space. While this velocity may be
measured, it is not a fundamental fuel property since the spatial velocity
will depend to a large extent on the type of vessel in which the flame exists.
The spatial velocity can be resolved into two important components:
the transformation velocity and the gas velocity. The transformation
velocity is the speed, relative to the unburned gases, with which the flame
front moves from the burned to the unburned gases. It is measured in a
direction normal to the surface of the flame front. In other words, if
the mass flow of the gases could be prevented, then the transformation
velocity would be the actual rate of advance of the flame into the unburned
gas.
i There is always a certain amount of expansion of the heated gases at
the vicinity of the flame, which gives rise to a movement of gases in thatl
region and displaces the flame front. The velocity at which the un burned
gases approach the burning zone is called the gas velocity. The spatial
velocity is the vector sum of the gas velocity and the transformation
velocity. In the case of a stationary flame, the spatial velocity is zero. I
Theory of Flame Propagation. Cool flames are the topic of much

research by those studying combustion from the viewpoint of the chem
ical kinetics involved.9 While many fundamental data may thus be
obtained, this type of burning reaction does not hold much interest for
the engineer for two reasons: (1) this is not the type of reaction found in
Burning zone
I
Fresh gas Heating Reaction Luminous- Burned gas
mixture zone zone flame zone
Most of Combustion
chain completed (
reaction here \
occurs
here
(
V
Fig. 4-8. Diagram of a quiescent flame and the temperature variation within.
the applications of combustion to engineering apparatus, and (2) the

fundamental data obtained from studies of homogeneous combustion
have not as yet yielded much information which can be used to apply the
normal combustion process more successfully. Hence, flame propaga
tion in rapid combustion only will be treated here.
The flame front and reaction zone for normal quiescent rapid combus
tion of a fuel and air mixture may be represented by a diagram such as
Fig. 4-8. From the diagram it appears that unburned gas must be
heated from the initial temperature To to some elevated temperature Tt,
the ignition temperature, before the reaction can start at the reaction
138
8
zone. Within the very narrow reaction zone, the chain reaction proceeds
to chemical equilibrium. In the region directly behind the reaction zone,
called the luminous flame zone, the radiations which appear as the visible
portion of the flame are emitted. The reaction zone and the luminous
flame zone are almost superimposed, the total width of both regions being
no more than a few ten-thousandths of an inch for rapid, quiescent com
bustion of premixed fuel and air.
Present developments in gas turbines, furnaces, heaters, internal-
combustion engines, etc., need the maximum possible rate of heat release
during the combustion process. To accomplish this, many analytical
studies have been made, and considerable research work is in progress, to
establish specifically which factors control the rate of burning, or the
flame speed, in a given application.
An analysis of the problem shows that at least four factors influence
the rate of flame propagation in varying degrees:
1.Mechanism of the reaction

2. Kinetics of the individual reactions in the mechanism
3. Diffusion of chain carriers, or propagating centers, from the reaction
zone into the unburned gases
4. Rate of heat transfer from the reaction zone to the adjacent heating
zone of the unburned gases
No present theory of combustion is sufficiently complete to establish

definitely which of the above factors are most important in controlling
the flame speed. The mechanisms of the chemical reactions followed in
rapid burning are still the subject of considerable research. It is an
exceedingly complex problem, not only because of the speed of the
burning process, but also because so many physical variables, such as
temperature, pressure, humidity, air/fuel ratio, and vessel size, affect the
mechanism so markedly. Therefore, definite knowledge concerning this
process is quite meager, although it seems that the path by which the fuel
is oxidized should determine to a degree the rate of the burning.
The subject of chemical kinetics deals with the manner and speed with
which a given chemical reaction will progress. If the mechanism of the
combustion process were known, or could be assumed, it would then seem
logical that the burning rate could proceed no more rapidly than the
slowest reaction in the chain mechanism. Studies of kinetics establish
the rates of burning by estimating the speed of each individual reaction
in the chain mechanism. The general form of the equation used in this
type of analysis is the classical Arrhenius equation:
(41) W .= Ke-*'RT
where W = reaction rate

K constant, depending upon concentration of reacting sub
=
stances, and collision frequency of molecules

E = activation energy, or energy required to initiate reaction
R = universal gas constant
T = absolute temperature
Probable reaction rates have been used in helping to establish the
mechanisms for combustion of the various fuels given in the preceding
paragraphs. The study of combustion from the viewpoints of the
mechanism involved and the kinetics of those various reactions is an open
field for research. Much work has been done in this area, and extensive
studies of kinetics and mechanisms of reactions are under way. Unfortu
nately, the science of combustion kinetics is still not developed to the
point where it can be applied to predict combustion reaction rates,
in advance. However, a potential for future improvement seems
to lie here, and the engineer should be aware of these fundamental
studies.
Regardless of the mechanism of the chain reaction or the kinetics of I
the individual steps in the chain, it has been fairly definitely established
that in most types of combustion the chief factors which control the burn
ing speed are the rate of diffusion of propagating centers into the unburned
charge and the rate of heat transfer to the unburned gases. Early)
theories of flame propagation, as proposed by Mallard and Le Chatelier in
1883 and expanded upon later by L. Crussard, E. Jouguet, W. Z. Nusselt,
P. J. Daniell, and others, were based upon the assumption that thiel
unburned gas must first be brought to its ignition temperature by direct
conduction of heat from the reaction zone to the gas being heated in the
adjacent layer. These may be classified as thermal theories of flame_\
propagation.10
Mathematical expressions for the transformation velocity, developed
by the above-named researchers and others, based on the thermal theories
of flame propagation, show that
where VT = transformation velocity

K = thermal conductivity
C = mean specific heat
= maximum flame temperature
T/
Ti = ignition temperature
To = initial temperature
C Numerous investigators have proposed to include in the equation the

effect of diffusion of active centers (free atoms and radicals) into the
' unburned gases. Jost and Muffling11 have stated that the concept of an
ignition temperature in the thermal theories has little meaning, as ignition
of a given gas may occur at a number of temperatures, depending upon
other physical factors. On this ground Jost4 has written that a flame
propagation theory involving only heat transfer by conduction is surely
inadequate. Jost and Muffling estimated that more than ten times as
many active centers get to the cold edge of the flame front by diffusion as
would be produced there by thermal effect alone. It is known that, if the
f~mixture of unburned gases next to the flame front is heated to a high
temperature, significant reactions will occur which lead to the formation
of active chain carriers (OH, O, H, etc.). On the other hand it is quite
possible that such chain carriers (or active centers of flame propagation)
might be projected by diffusion into the unburned gas zone. These
could initiate the reaction in that layer of gas without the addition of an
Appreciable amount of heat from the adjacent flame. A close correlation
between experimentally measured and calculated flame speeds for hydro
carbons has been made recently, based on the diffusion theory.12 How
ever, J. W. Linnett13 has summed up current thinking on this point by
stating that conceivably both diffusion and heat transfer influence the
propagation of the reaction, and their relative importance may vary from
one case to another.
To illustrate how it is quite possible to arrive at varying conclusions in
attempting to prove the above theories, the following example is offered.
Tanford and Pease14 calculated equilibrium concentrations of free OH
radicals and free H and O atoms in CO flames when varying amounts of
N2, H2O, and H2 were added. It was observed that at increased burning
velocities of the CO flame there was a corresponding increase in Pa, the
partial pressure (i.e., concentration) of the hydrogen atoms. They con
cluded that the propagation of the flame must depend upon the diffusion
of light, fast-moving hydrogen atoms ahead of the flame. From their
calculated data, which neglected the effects of chemical kinetics entirely
and considered only gas flow and diffusion, it was noted that the partial
pressure of hydrogen atoms was proportional to the actual flame speed of
the mixture. Thus, it could be theorized that the increased number of
hydrogen atoms diffused more rapidly into the unburned layer of gases
adjacent to the flame, which resulted in propagating the combustion
reaction more rapidly. Their calculated flame speeds correlated rather
closely with those experimentally observed.
On the other hand, M. F. Hoare and J. W. Linnett15 have pointed out
that in the combustion of the same mixtures of CO, the burning rate was
greater for increased partial pressures of CO, O2, and H2. An increase in
the amounts of any of these three constituents would tend to make the
flame hotter. Thus, they concluded, the increase in burning velocity
could also be explained by an increased rate of heat transfer to the
unburned gases. Likewise, their calculations agreed quite well with the
experimental data.
Evidently both factors aid in controlling the rate of flame propagation,
and the relative importance of each as yet can only be speculated upon.
METHODS OF MEASURING FLAME SPEEDS

Since the fundamental importance of the flame speed has long been
recognized as one of the basic parameters affecting rate of heat release,
much experimental work has been conducted over a period of the past 60
or 70 years in an attempt to discover the effects of physical variables on
the transformation velocity. Many experimental techniques have been
employed in these studies, but nearly all of them can be classified under
one of the following headings.
Bunsen-burner Method. This method has been employed by many
investigators since Robert Wilhelm von Bunsen developed the type of
burner in 1855 which has continued to bear his name. Bunsen made the
first recorded measurements of flame velocity, although he did not employ
what is now called the bunsen-burner method. Gouy, in 1879, developed
a form of the experimental technique discussed here. Other researchers
including Michelson, Mache, Smith, Pickering, and many others, have
made various improvements on the technique through the years.
A few principles of combustion in the bunsen burner and the basic
theory involved in computing transformation velocity are depicted in
Fig. 4-9. Gaseous fuel and some air are mixed in the mixing tube. The
air which is premixed with the gas before combustion is called primary air.
In the bunsen-type burner, the primary air inducted through the air
shutter is normally less than the amount of air stoichiometrically required.
At the inner cone of flame, the combustion proceeds until the oxygen in
the primary air is exhausted. Some CO, H2, and any other unburned
products rise into the outer envelope of the flame where secondary air,
diffused into the envelope from the surrounding air, completes the com
bustion. Active free atoms of H, O, N and the free radicals OH, C2, CH,
CHO, etc., are thought to be the factors which lend color and lumines
cence to the flame. The thin surface of flame at the inner cone is termed
the flame front, and the transformation velocity is the rate at which the
fresh gas mixture entering the flame front is consumed. Since the flame
front is stationary, the spatial velocity is zero. Hence, the transforma
tion velocity is vectorially equal and opposite to the velocity of the fresh
mixture issuing from the burner. If an image of the flame is projected

on a screen, photographed, or viewed through a telescope, the exact shape
of the inner cone can be established and the relationship between trans
formation and flow velocities can be computed.
Different investigators have used somewhat different approaches to
the problem of calculating the transformation velocity.1617 One simpli-
Fig. 4-9. Elements of a bunsen burner and the method of measuring transformation
velocity.
fled method which yields quite uniform results is based on the equation:
VT = Vav sin a
(43)
where VT = transformation velocity

Vav = average velocity of premixed gases leaving burner tube
a = angle inner cone makes with the vertical
Assuming the flow of unburned gases through the tube to be laminar, the
velocity distribution in the tube approximates a parabola, and the
average velocity of the mixture is equal to the local velocity at 0.7r, where
r is the radius of the tube. After calculating the average flow velocity
of the unburned gases in the mixing tube from the continuity equation,
q = A V, and measuring the angle of the inner cone, we can compute the
transformation velocity directly from Eq. (43). Transformation veloci
ties of several gases measured by the bunsen-burner method appear in
Fig. 4-10. Results reported recently by W. C. Johnston,18 using a
technique similar to that just described, are given in Figs. 4-17 and 4-19.
£D 4
Ac2 H2 /
I
E
3
C3H8
___C0
0 15% 30% 45% 60% 75%

Per cent gas in mixture
Fig. 4-10. Transformation velocities for several gases in air measured by the bunsen-
burner method. Temperature of mixtures, 86 F and 122 F. (Smith and Pickering, J.
Research, Nat. Bur. Standards, 1936.)
Soap-bubble Method. About 30 years ago F. W. Stevens developed a

relatively simple method for measuring the transformation velocity in an
explosive mixture at constant pres
sure. No simpler or more funda Glass filling
mental technique has been developed tube
Wire -ring
to date, although other investigators
support
have made minor improvements in
his procedure. In it a soap bubble is
filled with a homogeneous mixture of
Soap
fuel and air, and the explosive mix
bubble
ture is ignited at the center of the
bubble by an electric spark. Figure
1-11 shows a soap bubble filled with a
combustible mixture just before

Electrode
ignition. The homogeneous mixture
is admitted through the glass tube.
Fig. 4-11. Soap bubble containing ex
plosive mixture ready to be fired. (Fiock
The gold wire ring near the top helps and Marvin, Chem. Rev., vol. 21, 1937.)
support the bubble, while the igni
tion is provided through the spark gap in the electrodes at the center.
Bubbles are blown to about 4 in. diameter before ignition.
The flame spreads as a sphere from the point of ignition. Since the
soap bubble expands as the temperature of the gases within rises, the
combustion is essentially at the constant pressure of the surroundings.
35
The chief advantage of this method
is its simplicity. No rapid rate of
30
pressure rise is involved, and the
25
shape of the container (the soap
y Nr
I 20
Spatia 1 veloc bubble) does not interfere with the
normal path of the flame. This
technique does have the disadvan
.£> 15
tage that the percentage of water
10 vapor in the mixture cannot be
— Trarsformo tion ve locity— controlled, because the amount pres
ent is influenced by the surrounding
soap film.
30 40 50 60 70 80
Volume per cent CO in mixture Spatial velocities are measured
Fig. 4-12. Variation of spatial velocity from high-speed photographs taken
and transformation velocity with compo of the explosion. Transformation
sition of CO and O2 mixtures containing
2.7 per cent H»0 vapor, as measured in a velocities are computed from the
soap-bubble explosion. (Fiock and Mar measured spatial velocities, taking
vin, Chem. Rev., vol. 21, 1937.)
into account the expansion of the
burned and unburned gases upon heating. Figure 4-12 shows typical
results obtained from the soap-bubble method for measuring spatial
velocities by E. F. Fiock and C. F. Marvin.19
Glass-tube Method. Much of the early flame research was carried on
in glass tubes whose lengths were
much greater than their diameters. /-Transformation velocity
If a tube is filled with an explosive

mixture and ignited at the open
end, the speed of flame travel can
be observed through the sides of
Unburned
the tube and can be photographed gas
to preserve a record of the flame

characteristics. Mallard and Le
Chatelier reported results of flame
studies with this type of appara Fig. 4-13. Shape, of a flame in a tube, showing
the relationship transformation
tus as early as 1883. A sketch velocity and spatial between
velocity.
showing the usual shape of the
flame front and the relationship between spatial velocity and transforma
tion velocity is given in Fig. 4-13.
While this method is quite simple to carry out, there are a number of
disadvantages involved. vDifferent values for transformation velocity,
are obtained if the flame is progressing upward instead of downward/

horizontally, or at an inclined angle. Transformation velocities for open-
end tubes differ from those for closed-end tubes. In addition, there is an
appreciable effect on the shape of the flame front by wall friction and heat
transfer for small tubes. These effects become less for the larger tubes,
but then the effect of convection currents in the hot gases becomes
appreciable. Likewise, there is always a certain tendency for the gases
12 16 20 24 28 32 36 40
Tube diameter, in.
Fig. 4-14. Spatial velocities for various methane/air mixtures in glass tubes of different
diameters. (Coward and Hartwell, J. Chem. Soc, 1932.)
in the tube to pulsate, much as in an organ tube. This further compli-j

cates the data.
Thus, in many of the earlier investigations the values of transformation
velocities were found to be 50 to 100 per cent greater in tubes than as
measured by other methods. The effects of tube diameter and mixture
composition on combustion velocities as measured in long glass tubes
are shown in Fig. 4-14. However, by using a revised method recently,
transformation velocities in glass tubes have been found comparable to
those measured by other techniques.20
Spherical-bomb Method. Any method for measuring flame speed

inside a constant-volume vessel is complicated by the rapid pressure rise
created within the container. Nevertheless, this type of explosion has
been the subject of considerable study because these are essentially the
conditions under which combustion takes place in the cylinder of an
internal-combustion engine. It has been found that the shape of a
flame front always tends to approach the shape of the vessel which con
tains it. Hence, a spherical bomb affects the travel of the flame less than
any other shape of container.
Studies of explosions in such a vessel were conducted at the National
Bureau of Standards, under the auspices of the National Advisory Com-
Electronic pickup
for time and
pressure signals
Fig. 4-15. Diagrammatic layout for spherical-bomb explosion studies. (After Fiock,
Marvin, Caldwell, and Roeder, NACA Tech. Rept. 682, 1939.)
mittee for Aeronautics.21 A simplified sketch of the apparatus used in

this study appears in Fig. 4-15. Through the window of the bomb, the
flame movement was recorded on a film moving at a constant known
speed in the camera. Pressures within the cylinder were measured by
means of the indicator at the bottom of the bomb. These pressures were
likewise recorded photographically so that the film carried a time-pressure
and a time-flame displacement record of the explosion. From these
records, values of spatial velocity, gas velocity, and transformation
velocity were computed, taking into account the pressure rise, expansion,
and compression of gases in the bomb. Figure 4-16 shows the results
obtained from the explosion of a mixture of carbon monoxide and oxygen
with 2.7 per cent water vapor which was fired at an initial pressure of
5 psia. Note that, for this type of combustion, the spatial velocity is the
sum of the gas velocity and the transformation velocity. The trans
formation velocity is constant for much of the reaction, but it increases

gradually toward the end of the explosion as the pressure and tempera
ture of the. last part of the charge to burn are increased due to the com
pression of the unburned gases. Most of the spatial velocity is created by
the expansion of the high-temperature burned gases. Toward the end of
the burning process, outward expansion of the gases decreases to zero.
The behaviors of the hydrocarbons studied by this method were so
similar that high accuracy of measurement was necessary to distinguish
between the various fuels. The transformation velocities of benzene,
25 1 1 1
—
1 i 1 1 1 1 1 1
0 1 2 3 4 5
Flame radius, in.
Fig. 4-16. Flame speeds in CO and O2 mixtures as measured in a spherical bomb. Initial
mixture pressure, 5 psia. (Fiock, Marvin, Caldwell, and Roeder, NACA Tech. Rept. 682,
1939.)
w-heptane, and isooctane rated in that order, but there was very little
difference in the actual values for the three fuels. , The addition of tetra-
ethyllead fluid to the n-heptane mixture produced no appreciable change

in transformation velocity. In fact, after a thorough examination of all
the characteristics of normal constant-volume burning, none could be
correlated with the relative tendency of the above-named fuels to detonate
when used in a spark-ignition engine.1 Pressure records and photo
graphs of the flame front showed evidence that burning did not reach
chemical equilibrium at the reaction zone, but that some afterburning, or
subsequent chemical reaction, was present. This resulted in lower
pressures than would be predicted on the basis of complete combustion.
EFFECT OF PHYSICAL VARIABLES ON TRANSFORMATION VELOCITY

As the reader has probably gathered from the preceding discussion, it is
extremely difficult, and in some cases impossible, to get consistent values
of the transformation velocity for a given fuel using different methods of
testing. Part of this difficulty lies in the fact that the wall surfaces,
shape, and size of the container play such an important factor in con
trolling the combustion reaction. Also the physical variables of tempera
ture, pressure, air/fuel ratio, humidity, etc., have an effect on combustion,
and it is nearly impossible to reproduce all these factors from one method
of testing to another.
In the sections which follow, an attempt is made to show the general
effect of each of these physical variables on transformation velocity, and
90k
mixture
£
o
5
o A, r5 99/
i
D
E \\
\
o
/ J fI /
3
834 7F
1
v.f
1
N
i
*
A
tmo spht vie

A
pressure
8 22
10
14 18 26
Air- fuel ratio,
by
weight
Fig. 4-17. Transformation velocities for mixtures of natural gas and air at various temper
atures, measured by a bunsen-burner method. (Johnston, SAE J., December, 1947.)
to present some of the experimental evidence which seems to establish

the trend in each case. Because of the diversified manner in which com
bustion studies have been conducted, the experimental data presented
are somewhat inconsistent, because frequently more than one variable
was permitted to change during the tests. The data which follow are
presented only to point out certain trends.
Temperature. Figures 4-17 and 4-18 are typical of much experimental
data which show that as the initial temperature of fuel/air mixture
a
rises, the transformation velocity increases. The data for these curves
were obtained from tests conducted by W. C. Johnston,18 and S. Sherratt
and J. W. in which the bunsen-burner method for measuring

Linnett22
flame speeds was used. A preheating of only 65 F caused an increase of
30 per cent in the transformation
velocity of a 31.7 per cent coal gas

to
in air mixture (Fig. 4-18). This .O
may be another argument in favor
of the thermal theories of flame
propagation. An increase in the

maximum flame temperature also
resulted in an increased transforma 40 80 120 160 200
Temperature, F
tion velocity.
Fig. 4-18. Variation of transformation
Pressure. Gaydon and Wolfhard velocity with temperature for a 31.7 per
found that, for stationary flames of cent coal gas in air mixture, measured by a
bunsen-burner method. (Sherratt and Lin
hydrocarbons premixed with air, nett, Trans. Faraday Soc., 1948.)
the flame speed was independent of
pressure between 1 and 0.01 atm absolute pressure. However, J. W.
Linnett and P. J. Wheatley report that their experience and that of
20
19 I
u
CD
J"
■1.9ps/'t7
.2
13
4A psia
15
8etic
g
1
14
61 ?ps ia-'
13
12
Temperature --6 04 F
18
16
10 12 14
8
by
Air/ fuel ratios, weight

Fig. 4-19. Transformation velocities for acetylene at various pressures, measured by a
bunsen-burner method. (Johnston, SAE J., December, 1947.)
several others indicated that the transformation velocity should vary

approximately inversely as the fourth root of the ratio of absolute pres
sures. Figure 4-19 shows that the transformation velocity of acetylene
increases with decreased pressures, although not strictly according to the
inverse fourth-root relationship suggested. The data of Fig. 4-19 were

obtained by the bunsen-burner method.
The reaction rates for pentane in oxygen (Fig. 4-20) show the reverse
phenomenon; i.e., at higher pressures the reaction rate increases.
Hence, it can probably be concluded that other physical factors affect
the reaction rate of fuels much more than pressure, and the effect of
pressure on transformation velocity cannot be predicted unless the rela
tive effects of the other variables are known.
600 650 700 750 800 850 900 950 1000 1050 1100
Temperature, F
Fig. 4-20. Relative reaction rates at different pressures for an 11.1 per cent pentane in
oxygen mixture. {Neumann and Aivazov, Nature, vol. 135, p. 655, 1935.)
Mixture Composition. An analysis of the graphs of transformation

velocity plotted against fuel/air composition, as in Figs. 4-10, 4-17, and
[4-19, shows that the maximum flame velocity always occurs with a mix-
(
ture which is slightly rich. The term rich mixture implies that there is
not enough air to burn the fuel completely. In general, hydrocarbon
fuels burn most rapidly with about 85 to 95 per cent of theoretical air
available. As more air is supplied and the mixture becomes lean, the
transformation velocity decreases.
Wall Surfaces. Most types of wall surfaces of glass tubes, cylinders,
bunsen-burner
i mixing tubes, etc., have a retarding effect on the burning
n
i
rate. Evidently walls have a combined effect of chilling the adjacent
gases by excessive rates of heat transfer and of adsorbing some of the
chain carriers. Both of these effects would slow or stop the combustion
reaction and have a detrimental effect on good combustion. The larger

the vessel, cylinder, or tube in which the combustion occurs, the less will
be the wall effect. Probably for this reason certain designs of furnaces,
diesel engines, and gas heaters will operate satisfactorily when built in
large sizes whereas a unit of similar design, but smaller, may not function
well.
Certain types of refractory walls have just the opposite effect, however.
Such walls are often used as an inner lining for combustion chambers and
become heated almost white hot under heavy firing conditions. The
irradiant brickwork transfers heat to the unburned gases by convection
and radiation and aids in the ignition of the gaseous fuel. In addition,
there seems to be a certain catalytic reaction occurring at the refractory
surface which promotes rapid, complete combustion. Firebrick refrac
tory is often placed around the mouth of oil and gas burners to provide
positive ignition and to expedite complete, rapid combustion, even when
some other material is used to line the walls of the combustion chamber.
In fact, there is a type of burning, called surface combustion, in which
the gaseous fuel/air mixture is projected over the surface of porous
refractory. The exact mechanism of this type of burning is little under
stood, but the fuel burns completely and swiftly within a few hundredths
of an inch of the brick wall, resulting in unusually high rates of heat
release and heat transfer to the refractory. A number of designs of
industrial processing furnaces operate on this principle.
Turbulence. Of all the variables listed here, turbulence has by far the
greatest effect on the rate of flame propagation. This is true whether
burning is of the stationary, constant-pressure type, as in a bunsen
burner, or whether it is explosive in nature, as in the cylinder of an engine.
G. H. Markstein and M. Polanyi10 have depicted the action of turbulence
in flame propagation by the diagram of Fig. 4-21. Thus, the effect is to
make the reaction zone much deeper and more or less irregular. Not
only is the surface area of the flame front increased, but the turbulence
increases the effective diffusion of the active centers from the reaction
zone into the unburned gases. By the proper use of turbulence, fuels
which show a transformation velocity of only a few feet per second under
quiescent conditions can be made to burn at spatial velocities of a hundred
feet per second or more.
No satisfactory experimental technique has yet been devised to
measure turbulence quantitatively. Its effect can be noted qualitatively
in Fig. 4-22, which shows the variations in length of time and number of
crank angle degrees required for combustion within an engine cylinder for
different intake air velocities and different engine speeds. Increasing
either of these factors results in greater turbulence within the cylinder.
Pockets of un burned gas

in flame zone
Fig.4-21. Combustion zone in a highly turbulent gas. (Markstein and Polanyi, Cornell
Aeronaut. Lab., Bumblebee Series, Rept. 61, 1947.)
8 Best)
(Marvin
NACA at NBS
600 700 800 900 1000 1100 1200 1300

Engine speed, rpm
o
E
E
o
(Rabezzana 8 Kalmar)
AC Spark Plug Co.
600 700 800 900 1000 1100 1200 1300

Engine speed, rpm
(Schnauffer)
Germany
50 75 100 125
Intake velocity, ft /sec
Fig. ';
Effect of turbulence on burning rates in spark-ignition engines.
4-22. (Fiock, Chem-
ical Background for Engine Research,1' Jnterscience Publishers, 1943.)
Limits of Inflammability. Any mixture which will burn without addi

tional air or fuel is termed a combustible mixture. It is known from
observations that certain mixtures will not ignite, even though both fuel
and air are present. In a previous section it was pointed out that a mix
ture slightly richer than stoichiometric would burn with the greatest flame
speed; i.e., no other mixture of the same components would burn so fast.
The lower limit of inflammability is the leanest mixture (least percentage
of fuel in the mixture) which will support combustion. The upper limit
is, of course, the richest mixture which will propagate a flame. Various
other terms such as inflammation limits and explosive limits are often
used interchangeably with the term limits of inflammability.
Like other experimental results from combustion, the limits of inflam
mability are affected by the method of testing. The following factors
appear to affect the limits:24
1. Direction of Flame Propagation. Wider limits are obtained for
upward propagation of flame than for horizontal or downward flame
travel.
2. Design of Test Apparatus. Turbulence of the mixture reduces the
lower limit slightly. Hence, most tests are conducted in long glass tubes
to obtain a more or less quiescent mixture.
3. Diameter of Tube. It is found that the limits of inflammability
become wider with increased diameters up to tube diameters of about 2 in.
Larger diameter tubes apparently affect the experimental explosion limits
very little.
4. Length of Tube. The combustion tube must be at least 3 ft in length
to ensure that the mixture is capable of supporting a flame after the heat
of ignition has been dissipated.
5. Temperature at Time of Ignition. Temperature changes of a few
degrees at the time of ignition have little effect on the limits. At tem
peratures considerably above room temperature, the limits are somewhat
wider than those reported in Table 4-1.
6. Pressure. As with temperature, small pressure changes have a
negligible effect on the explosion limits. At high pressures or very low
pressures, each fuel is affected differently and the limits may be greater,
less, or the same as the values in Table 4-1. No extensive systematic
studies of inflammability limits at pressures differing widely from atmos
pheric have been made.
7. Humidity. A normal amount of water vapor in the mixture prior
to ignition seems to have no effect on the lower limit, but it acts to lower
the upper limit.
Table 4-1 gives the lower and upper limits of inflammability for a
number of fuels for upward propagation in a 2-in. -diameter tube. All
data are for mixtures at room temperature and 1 atm pressure.
While the exact controlling factors on inflammability limits are not
completely understood, the following trends are apparent for hydro
carbons. At the lower limit, the heating value per cubic foot of com
bustible mixture is approximately constant. It follows that the flame
temperature reached by burning such mixtures would be constant, which
seems to indicate that the limiting factor for combustion of lean mixtures
is of a thermal nature. Flame temperatures for these limiting lean mix

tures of hydrocarbons calculate to be about 2550 F for all hydrocarbon
fuels, including methane. There are fewer correlated data for the upper
limit, but apparently it is caused by a lack of oxygen to generate the
necessary chain carriers to support the reaction.
The approximate inflammability limits for a mixture of gases can be
computed from Le Chatelier's rule for mixtures:
, 100
=
, , ,
(Pi/N1) + (P2/N2) + (P3/N3) +
where Pi, Pi, P3, etc., are the percentages of each combustible gas present
in the original mixture, free from air and inerts. N\, Ni, N3, etc., are
the respective limits of inflammability of the gaseous components. The
above relationship has been found to give reliable results from mixtures
of hydrogen, carbon monoxide, and methane, and for mixtures of various
paraffin hydrocarbons. However, for certain mixtures of organic solvents
with hydrogen, etc., Le Chatelier's rule has been found somewhat in
error. It should be applied with caution and used only to obtain an
estimate of inflammability limits.
One of the most useful applications of the limits of inflammability data
is in the control of hazardous industrial processes involving highly inflam
mable gases and vapors. In the use of combustible solvents, etc., the
percentage of the fuel vapor in the air and vapor mixture at any one time
must be kept well below the lower inflammability limit or well above the
upper, if costly accidents are to be avoided. Sometimes inert gases, such
as nitrogen and carbon dioxide, are added to dilute the inflammable mix
tures and reduce the percentage of fuel vapor below the inflammation
limits.
Ignition Temperature. Considerable experimental work has been con
ducted to establish the ignition temperature, which may be defined as the
minimum temperature at which a given fuel can be ignited. Some of the
important experimental techniques employed to find ignition tempera
tures are summarized:
1. A heated metal rod or strip is inserted into a mixture of fuel and air.
2. The explosive mixture flows through a tube of known maximum

temperature.
3. The explosive mixture is introduced into a container of known
temperature.
4. A small quantity of liquid fuel, or a stream of a combustible gas, is
sprayed into a space of known temperature containing air or oxygen.
5. The explosive mixture is compressed adiabatically until ignition
occurs.
Table 4-1. Limits of Inflammability of Selected Fuels in Air*
Limits,
% by volume of fuel
vapor in mixture % fuel vapor in j
Type and name Formula stoichiometric
mixture
Lower Upper
Paraffins:
Methane CHi 5.00 15.00
Ethane. . . . : CsHa 3.22 12.45
Propane C3H8 2.37 9.50
Butane C4Hio 1.86 8.41
Isobutane iso-C4Hio 1.80 8.44
Pentane C6His 1.40 7.80
Isopentane iso-CaHi2 1.32
Hexane CsHh 1.25 6.90
Heptane CvHia 1.00 6.00
Octane C8Ht8 0.95
Nonane CsH20 0.83
Ole5ns:
Ethylene C2H4 2 75 28.60
Propylene C3H6 2 00 11 . 10
Butylene CiH* 1 70 9.00
Amylene C&H10 1 60
Acetylenes:
Acetylene C2H2 2.50 80.00
Aromatics :
Benzene C«H< 1 41 6.75
Toluene C7H8 1 27 6.75
o-Xylene C»Hio 1 00 6.00
Naphthenes:
Cyclopropane C3Hg 2 40 10.40
Cyclohexane CBHi2 1 33 8.35
Methylcyclohexane C7H14 1 15
Terpenes :
Turpentine CioHia 0 80
Alcohols:
Methyl CH40 6 72 36.50
Ethyl 3 28 18.95
Propyl C,H»0 2 55
Iso propyl CiH.O 2 65
Butyl C4H10O 1 70
Isobutyl CiHioO 1 68
Amyl CH12O 1 19
Isoamyl C5H12O 1 20
Allyl CjH,0 2 40
Aldehydes :
Acetaldehyde C2H.0 3 97 57.00
Crotonaldehyde C.HeO 2 12 15.50
Furfural CsH.02 2 10
Paraldehyde C.H12O3 1 30
Miscellaneous:
Hydrogen H2 4 00 74.20
Carbon monoxide. . CO 12 50 74.20
Ammonia NHi 15 50 27.00
* G. W. Jones, Chem. Rev., vol. 22, p. 1, 1938.

H. F. Coward and P. G. Guest26 conducted a series of ignition tempera

ture tests on natural gas and air mixtures by inserting heated metal rods
and strips into a vessel containing the mixture. Figure 4-23 shows the
effect of various metallic surfaces on the ignition temperature as well as
the effect of the gas mixture on ignition. The higher temperature
required of the platinum to produce ignition may indicate that it is such
a good catalyst that the mixture near the surface tends to become
completely oxidized without propagating the flame. Evidently the effect
of active radical producing reactions is as important in ignition as temper
ature. Thus, ignition will be more easily achieved in some types of
containers than in others.
2600
2400
li
ar 2200
|
8. 2000
E
1800
1600
0 2 4 6 8 10 12 14 16
Per cent natural gas in mixture

Fio. 4-23. Ignition temperatures of natural gas and air mixtures in contact with heated
rods of various metals. (Coward and Guest, J. Am. Chem. Soc., vol. 49, 1927.)
They found further that much higher temperatures were required to

obtain ignition with very thin metal strips than with larger strips or rods.
The lowest ignition temperatures were obtained when the entire vessel
surrounding the gas was heated. Other investigators have found this
same trend, namely, the smaller the heat source, the higher must be the
temperature of that source to initiate a combustion reaction.
As might be expected, each of the several methods used in determining
the ignition temperature of a fuel yields a different result. Thus, any
experimental ignition temperature is only a relative value, and it must
be defined in terms of the conditions involved. The ASTM, ASA, and
API have adopted a test procedure for determining the ignition tempera
tures of liquid and gaseous fuels based on the fourth method in the pre
ceding summary. In this procedure, a conical pyrex glass flask is heated
in a bath of molten metal. The temperature of the metal bath and the
glass container is held at a known level, and drops of the fuel being tested
are sprayed into the flask. The minimum temperature at which the fuel
can be made to ignite and continue to burn is reported as the ignition
temperature. Table 4-2 gives ignition temperatures for a number of
Table 4-2. Minimum Ignition Temperatures of Selected Compounds*
Ignition
Name Formula Name Formula
temp., F
Acetone C3H60 1042 Methane CH4 1170

Acetylene C2H2 581 Methyl alcohol CH4O 878
Amyl alcohol C6H120 801 Naphtha 450-531
Benzene C6H„ 1078 Naphthalene CioHg 1038
Butadiene C4H6 804 Nonane C9H20 545
Butane C4H10 826 Octane CsHis 446
Isobutane C4H10 1010 Oil, castor 840
Butyl alcohol C4H10O 653 Oil, cottonseed 660
Isobutyl alcohol. . C4HI00 813 Oil, gas 640
Butylene C4H8 829 Oil, lard 650
Carbon monoxide. CO 1128 Oil, linseed 650
Creosote 637 Oil, lubricating 711
Cyclohexane CoHl2 565 Oil, cylinder 783
Cyclopropane CsH6 928 Oil, turbine lubricat
Decane C10H22 482 ing 700
Dodecane C12H26 993 Oil, peanut 833
Ethane C2H6 882 Oil, rosin 648
Ethyl alcohol C2H60 738 Oil, sperm 586
Ethyl bromide. . . . C2H6Br 952 Oil, tung 855
Ethylene C2H4 914 Oil, whale 878
Ethylene glycol . . C2H6O2 775 Paraffin wax 473
Ethyl ether CJI.oO 379 Pentane 527
Gasoline 73 octane 570 Petroleum ether .... 624
Gasoline 92 octane 734 Propane C3H.S 898
Gasoline 100 octane 804 Propyl alcohol C3H,0 822
Glycerine C3H80 739 Isopropyl alcohol . . . C3HsO 853
Heptane C7H16 451 Propylene C3H6 856
Hexane C6Hi4 478 Styrene CsHs 914
Isohexane CoHn 543 Toluene C7H8 1026
Hydrogen H, 1065. Turpentine CioHie 464
Kerosene 491
» Compiled by G. W. Jones, U.S. Bur. Mines Bull., 1946.
fuels determined by this standard procedure. The values of ignition

temperatures thus obtained are useful for comparing one fuel with
another. However, in applying such information, it must be remembered
that a great many factors, such as type of wall surface, size of combustion
chamber, pressure, or mixture composition, affect the actual temperature

at which ignition will occur.
Spark Ignition. The minimum spark energy required to ignite various
mixtures of methane and oxygen has been investigated by M. V. Blanc
and others.26 Later B. Lewis and G. von Elbe made an investigation to
establish an ignition theory based upon this and other experimental
evidence.27
Various carefully measured mixtures of methane, oxygen, and nitrogen
were admitted to an enclosed test bomb. At the center of the bomb were
two 3^6-in--diameter stainless-steel electrodes to ignite the mixture. The
charge on the plates of a condenser was measured and the discharge
10 1 1 1 1
0.05 0.10 0.15
Distance between electrodes, in.
Fig. 4-24.Minimum spark energy for the ignition of stoichiometric mixtures of natural
gas and air at atmospheric pressure as a function of spark gap. Electrodes, }ê-in. stain
less-steel points. (Lewis and von Elbe, Trans. ASME, 1948.)
voltage observed. If the first spark did not ignite the mixture, the
charge on the condenser was increased. By passing a series of sparks
between the electrodes, the minimum spark energy in millijoules required
to ignite the mixture was determined. This minimum energy represented
the energy imparted to the gas in the form of heat and ionization. Figure
4-24 shows the effect of spark gap on the minimum spark energy necessary
to ignite a stoichiometric mixture of methane and air at atmospheric
pressure. As the curve shows, above about 0.08-in. electrode gap, the
minimum ignition energy is approximately constant. Below this critical
distance, greater ignition energies are required, probably because of a
quenching effect due to the electrodes being in close proximity to the
newly created combustion wave. The existence of a range of constant-
ignition energy suggests that the flame is not started uniformly along the
path of the spark but is probably generated at a point source near one of
the electrodes. If the spark brought the gas to a uniform ignition tem
perature along its entire path or excited the surrounding gases equally,
greater spark energies would be required to ignite a mixture when the

spark gap was longer and the spark was passing through a greater mass of
gas. Thus, it seems a virtual point source of ignition is created and most
of the electrical energy is converted to heat energy at a point near the
cathode, or negative electrode. This small electrical charge probably
forms a number of chain carriers through thermal and ionization effects,
and a small sphere of air and fuel is brought to the ignition temperature.
i i i i i i i i
0 10 20 30 40 50 60 70
Per cent Methane
Fig. 4-25. Minimum spark energies required to ignite mixtures of methane, oxygen, and
nitrogen at atmospheric pressure. (Lewis and von Elbe, Trans. ASME, 1948.)
After ignition it is necessary for the combustion within this sphere to

supply sufficient energy to bring the surrounding unburned gases to the
ignition temperature. When the electrodes are moved closer together
than 0.08 in., they are so close to the point source of ignition that they
have a quenching effect on the flame. Different types of electrodes were
found to have varying quenching effects. The effect of mixture com
position on minimum ignition energy is shown in Fig. 4-25. As the mix
ture approaches the limits of inflammability, the energy required for
ignition is much greater.
Figure 4-25 also shows the variation of the minimum spark energy
required with different oxygen and nitrogen compositions. Inert gases
such as nitrogen have a very pronounced effect on this value. For an
air/methane mixture, about 100 times as much spark energy is required
to set off the most easily ignited mixture as is required to ignite a mixture
of pure oxygen and methane.
Ignition Delay. In the preceding sections, no consideration was given
to the time required for the fuel/air mixture to ignite. Ignition studies
show that an ignition lag, or delay period, always exists between the time a
fuel is subjected to an ignition source and the start of rapid combustion.
It is thought that this delay period is required for the necessary active
Ignition
delay
Time, sec
Fiq. 4-26. Effect of temperature and pressure level on ignition delay.
chain carriers to be generated to propagate the combustion reaction.

As proof that the concentration of active particles has an effect on ignition
lag, a small amount of formaldehyde added to a combustible mixture
reduces the delay period markedly. The action of aldehydes as chain-
propagating centers has been long accepted. The length of the ignition
delay is a function of the fuel, the temperature, and the pressure. Like
wise, the temperature at which a fuel can be ignited depends to a large
extent on the length of ignition delay which one wishes to allow.
Figure 4-26 shows diagrammatically the relationships between ignition
delay and the physical factors which affect it. Thus, a combustible
mixture may be heated along the line AB and allowed to remain at
temperature T2. After time BC, the ignition delay period, the mixture
would ignite, and rapid burning would follow. If the same mixture were
heated to temperature T3, the ignition delay, line DE, would be consider
ably shorter. At the still higher temperature, Tt, the ignition delay
period is further reduced. Pressure has much the same effect on ignition
lag. For a given temperature level, a pressure scale might be sub
stituted for the temperature scale on the ordinate of Fig. 4-26, and the
same relationships would prevail. If ignition of a combustible mixture
occurs by the process of adiabatic compression, both pressure and tem
perature increase simultaneously. Both factors then would be acting to
shorten the ignition delay.
The effect of temperature on the ignition delay of stoichiometric mix
tures of benzene, isooctane, and normal heptane, as measured by adia
batic compression, is shown in Fig. 4-27. The ignition lags experienced
at various temperatures and pres
sures for hexane and air mixtures
are shown on the curve of Fig. 4-7,
which shows the definite inter
dependence of pressure and ignition
lag on ignition temperature. Thus,
at 45 psia and 515 F the mixture will
ignite, but it takes 35 sec to pro
duce ignition. Practically speak
ing, the fuel/air mixture cannot be
ignited at 45 psia and 515 F.
The data of Table 4-2 give what
might be termed minimum ignition 0.001
temperatures for the ASTM test 900 800 700 600

Temperature, F
procedure, because the method of
Fig. 4-27. Ignition lags for stoichiometric
testing allows for up to 2 min of mixtures of benzene, isooctane, and n-hep-
ignition lag. The length of the tane as functions of temperature at the end
of adiabatic compression. (Jost and Teich-
ignition delay is not reported in the mann, Naturwissenschaften, 1939.)
test results. Hence, while the igni
tion of fuels may be achieved at these temperatures, it likely would not
be achieved rapidly enough for a practical application.
Autoignition by Adiabatic Compression. Recent tests conducted by
C. F. Taylor, et al., by rapid adiabatic compression show that tempera
ture, pressure, and rate of compression affect the ignition lag and the
pressure-temperature level necessary to achieve combustion. In this
study, continuous instantaneous records of pressure, cylinder volume, and
time were recorded on a high-speed film, and pictures were taken of the
explosion as it developed. The effect of compression ratios and fuel/air
ratios on ignition lag are shown in Figs. 4-28 and 4-29.
Studies of autoignition temperatures and ignition delay may be cor
related with the performance of fuels in engineering combustion equip
ment, especially in the spark-ignition and compression-ignition engines.
In general, gasolines having high-octane ratings also have high auto-

ignition temperatures and comparatively long ignition delay periods at a
given temperature and pressure. On the other hand, good diesel fuels
with high-cetane ratings exhibit relatively low autoignition temperatures
Isooctane
0.03 r6
T, = 150 F
F/A =0.066
P, = 14.7 psia
9 10 II 12 13 14 15
Compression ratio
Fig. 4-28. Effect of compression ratio on ignition delay. (Taylor, Taylor, Livengood,
Russell, and Leary, SAE Trans., 1950.)
/
0.03
o 0.02
I — n-butane
2— benzene
3— n-heptane
r I50F
I50F
I50F
Com
pression
ratio
10.0
12.3
123
P{
15.5 psia
14.7 psia
14.7 psia
4— isooctane I50F 12.3 14.7 psia
0.01
0.000
0.04 0.06 0.08 aio 0.12
Fuel /air ratio
Fig. 4-29. Effect of fuel/air ratio on ignition delay. (Taylor, Taylor, Livengood, Russdl,
and Leary, SAE Trans., 1950.)
and very short ignition delays. In fact, the cetane rating of a fuel is
determined by the length of the ignition delay.
The effects of these and other phenomena of combustion on the design
and operation of fuel-burning equipment are presented in the chapters
which follow.
Problems
1. Distinguish between cool flames and combustion, as defined in this chapter.

2. List 10 applications of combustion and classify them as to the types of combus
tion process involved.
3. Define the terms: chain reaction, chain carrier, branching reaction, and chain-
breaking reaction.
4. What is the difference between a free radical and an ion?
5. In the light of the chain reaction theory of combustion, discuss the probable
role of an inhibitor such as tetraethyllead in suppressing detonation in automobile
engines.
6. List some of the reactions which occur during the initiation, propagation, and
termination phases of the chain reaction of hydrogen with oxygen.
7. List the physical variables which influence the reaction rate between hydrogen
and oxygen, and discuss their effect upon the combustion process.
8. What is meant by the term "poisoning" of wall surfaces?
9. Distinguish between adsorption and absorption.
10. What is a reaction path, and how does it influence combustion?
11. Discuss the significance of the three explosion limits of hydrogen/oxygen
combustion.
12. Give the molecular structures of the following derivatives from the hydro
carbon re-heptane: alkyl radical, aldehyde, acid, peracid, ketone, hydroperoxide.
13. What organic compounds are formed as intermediate products in the course of
the aldehyde degradation mechanism of a hydrocarbon?
14. Establish the equations which relate transformation, spatial, and gas velocities
in different types of combustion processes.
15. If the rate of a chemical reaction follows the Arrhenius equation accurately,
what would be the relative rates of a given reaction at 100, 110, 150, 200, and 500 F,
if all other factors remained constant?
16. Same as Prob. 15 except T = 100, 1000, and 2000 F.
17. Considering the thermal theory of flame propagation, what would be the effect
on transformation velocity of increasing the air preheat temperature for a given
mixture?
18. Referring to Fig. 4-10, what percentage of stoichiometric air is required for
each of the fuels shown to form a mixture which would burn with the maximum flame
speed?
19. A primary air /fuel mixture of 60 per cent air and 40 per cent hydrogen at 122 F,
14.7 psia, is burned in a bunsen burner having a 0.50-in.-diameter tube. Applying
Eq. (43) and Fig. 4-10, what fuel rate in cubic feet per hour and Btu per hour would be
required to maintain the angle a at 45°. What percentage of stoichiometric air
would be supplied as primary air under these conditions?
20. Same as above except the fuel is carbon monoxide and the mixture is 50 per cent
primary air, 50 per cent CO.
21. List advantages and disadvantages of the four methods for measuring trans
formation velocity discussed in this chapter.
22. Tabulate the physical variables which influence burning rates and discuss why
and how each affects transformation velocities.
23. What percentages of stoichiometric air are present in the upper and lower
inflammability limit mixtures for the following fuels: methane, propane, pentane,
heptane, ethylene, benzene, methyl alcohol, hydrogen, acetylene?
24. What would be the upper and lower inflammability limits for a natural gas
having the composition: CH, = 90%, C2H6 = 4%, C2H4 = 4%, N2 = 2.0%?
25. Discuss the reliability of an equation of the type of Eq. (42) in light of the
experimental data concerning ignition temperatures and ignition lags.
26. Why is it reasonable that both temperature and pressure should affect the igni
tion delay of a fuel/air mixture?
References
1. Burk, R. E., and O. Grummitt, eds., "The Chemical Background for Engine
Research," Chaps, by Fiock, Rossini, Whitmore, von Elbe, Lewis, and Beeck,
Interscience Publishers, Inc., New York, 1943.
2. von Elbe, G., and B. Lewis, Mechanism of the Thermal Reaction between
Hydrogen and Oxygen, J.
Chem. Phys., vol. 10, p. 366, 1942.
3. Lewis, B., and G. von Elbe, "Combustion, Flames, and Explosions of Gases,"
Cambridge University Press, New York, 1938.
4. Jost, W., "Explosion and Combustion Processes in Gases," translated by H. O.
Croft, McGraw-Hill Book Company, Inc., New York, 1946.
5. Strickland-Constable, R. F., Kinetics and Mechanism of the Oxidation of
Carbon, Chemistry & Industry, Dec. 4, 1948, p. 771.
6. Audibert, E., The Combustion of Methane-Air Mixtures, Fuel, vol. 27, p. 145,
November-December, 1948.
7. Lewis, B., and G. von Elbe, "Mechanism of Combustion of Hydrocarbons,"
paper presented before the ACS, Division of Petroleum Chemistry, April, 1951.
8. Boord, C. E., K. W. Greenlee, and J. M. Derfer, "Oxidation Reactions as
Related to Hydrocarbon Structure and Engine Knock," paper presented before
the ACS, Division of Petroleum Chemistry, April, 1951.
9. Kahler, E. J., A. E. Bearse, and G. G. Stoner, "Vapor-phase Oxidation of
Hexanes," Ind. Eng. Chem., vol. 43, p. 2777, 1951.
10. Markstein, G. H., and M. Polanyi, Flame Propagation — A Critical Review of
Existing Theories, Cornell Aeronaut. Lab., Bumblebee Series 61, 1947.
11. Jost, W., and V. Muffling, Z. physik. Chem. A, p. 181, 1935.
12. Simon, D. M., "On the Active Particle Diffusion Theory of Flame Propagation,"
paper presented before the ACS, Division of Petroleum Chemistry, April, 1951.
13. Linnett, J. W., The Propagation of Flame. Studies of Shadow Cones & Bubble
Effects, Chem. Age, vol. 61, p. 345, Sept. 10, 1949.
14. Tanford, C, and R. N. Pease, Theory of Burning Velocity, J. Chem. Phys.,
December, 1947, p. 861.
15. Hoare, M. F., and J. W. Linnett, Mechanism of Flame Propagation, J. Chem.
Phys., August, 1948, p. 747.
16. Caldwell, F. R., H. P. Broida, J. J. Dover, and E. F. Fiock, "Apparatus for
Studying Combustion in Bunsen Flames," paper presented before the ACS,
Division of Petroleum Chemistry, April, 1951.
17. Albright, R. E., D. P. Heath, and R. H. Thena, "Flame Velocities of Liquid
Hydrocarbons," paper presented before the ACS, Division of Petroleum Chemis
try, April, 1951.
18. Johnston, W. C, Flame Propagation Rates at Reduced Pressures, SAE J.,
December, 1947, p. 62.
19. Fiock, E. F., and C. F. Marvin, Jr., The Measurement of Flame Speeds, Chem.
Rev., vol. 21, p. 367, 1937.
20. Gerstein, M., O. Levine, and E. L. Wong, The Determination of Fundamental
Flame Velocities of Hydrocarbons by a Revised Tube Method, J. Am. Chem. Soc,

January, 1951, p. 418.
21. Fiock, E. F., C. F. Marvin, F. R. Caldwell, and C. H. Roeder, NACA Tech.
Rept. 682, 1939.
22. Sherratt, S., and J. W. Linnett, The Determination of Flame Speeds in Gaseous
Mixtures, Trans. Faraday Sac., vol. 44, p. 596, 1948.
23. Linnett, J. W., and P. J. Wheatley, Effect of Pressure on Velocity of Burning,
Nature, vol. 164, p. 403, 1949.
24. Jones, G. W., Inflammation Limits and Their Practical Application in Hazardous
Industrial Operations, Chem. Rev., vol. 22, p. 1, 1938.
25. Coward, H. F., and P. G. Guest, Ignition of Natural Gas-Air Mixtures by
Heated Metal Bars, J. Am. Chem. Soc., vol. 49, p. 2479, 1927.
26. Blanc, M. V., P. G. Guest, G. von Elbe, and B. Lewis, Ignition of Explosive
Gas Mixtures by Electric Sparks, /. Chem. Phys., vol. 15, p. 798, 1947.
27. Lewis, B., and G. von Elbe, Ignition and Flame Stabilization in Gases, Trans.
ASME, May, 1948, p. 307.
28. Taylor, C. F., E. S. Taylor, J. C. Livengood, W. A. Russell, and W. A. Leary,
The Ignition of Fuels by Rapid Compression, SAE Trans., April, 1950, p. 232.
29. Lewis, B., and G. von Elbe, "Combustion, Flames, and Explosions of Gases,"
Academic Press, New York, 1951.
30. "Third Symposium on Combustion, Flame, and Explosion Phenomena," The
Williams & Wilkins Company, Baltimore, 1949.
CHAPTER 5
PHYSICAL PROPERTIES OF FUELS
There are many physical characteristics which enter into the problems
of burning fuels in various types of combustion equipment. Thus one
type of coal may be well adapted for a particular type of stoker, whereas
another type of stoker may encounter considerable difficulty in burning
the same fuel. Similarly kerosene, which contains essentially the same
amounts of hydrogen and carbon as gasoline, does not operate satis
factorily in automobile engines.
Many tests for fuels have been adopted as standard by the ASTM, the
AGA, the API, the U.S. Bureau of Mines, and other interested groups.
Some of these tests which measure the physical properties affecting the
handling, storage, and combustion characteristics of fuels are presented
in this chapter.
COAL AND COKE
The proximate and ultimate analyses and the heating values of solid
fuels, discussed in Chap. 1, may be analyzed and reported on a number of
bases. The most commonly used of these are " as mined," "as received,"
"as fired," "moisture-free," "moisture- and ash-free," "dry, mineral-
free," "moist, mineral-free," and "moisture-, ash-, sulfur-free."
The moisture content of the fuel may vary from the "as mined" to the
"as received" or the "as fired" condition. In addition, coal which has
been stored for a considerable length of time may have weathered, and its
composition changed. When a sample of coal is taken for testing, it
should be selected according to the approved methods of the ASTM, and
placed in a closed vessel to prevent any change of moisture before testing.
The "moisture-free" sample gives the per cent by weight of the various
constituents after the moisture has been driven from the coal by heating
it for several minutes at a temperature of 225 F ; or it is the analysis of the
fuel which would have been determined had the sample been tested dry.
If the moisture and ash could be removed from a coal sample, leaving the
rest of the components just as they were, the per cent by weight of the
remaining constituents would be reported on a "moisture- and ash-free"
basis. Usually coal samples are analyzed "as received," or the sample is
dried and analyzed "moisture-free." The analysis on any other basis
can be computed once the composition of the coal "as received
" is known.
166
PHYSICAL PROPERTIES OF FUELS 167
Computing Fuel Analyses to Other Bases. It becomes necessary, occa

sionally, to convert the analysis of a fuel sample to another basis in order
to compare various samples on the same terms. This requires a funda
mental knowledge of percentage calculations. A number of equations
are given below to aid in computing new analyses. The student should
by no means attempt to memorize all, or even any, of these formulas, but
should analyze them to refresh in his mind the simple reasoning involving
percentages upon which these equations are based.
Symbols Used in
Following Equations
the
C carbon in known ultimate analysis, "as received"
= per cent
H = per cent hydrogen in known ultimate analysis, "as received"

0 = per cent oxygen in known ultimate analysis, "as received"
N = per cent nitrogen in known ultimate analysis, "as received"
S = per cent sulfur in known ultimate analysis, "as received"
M = per cent moisture in known proximate analysis, "as received"
VM = per cent volatile matter in known proximate analysis, "as
received"
FC = per cent fixed carbon in known proximate analysis, "as received"
A = per cent ash in known ultimate or proximate analysis, "as
received"
HV = heating value of coal, Btu/lb, "as received"
A subscript 1 on the above symbols denotes the same components on a

new basis.
1. To compute to the "moisture-free" basis from the "as received":
(j
Ci =
m^M x 100
VM
VM> =
Too^Mxm
etc.
r
for all items (C, N, S, A, VM, FC, HV) except hydrogen and oxygen.
Hl _
- - 8M/9)
~ (H
100
M/9)
M X - 100 0l — (0
100 -M X 100
When the moisture content changes from one basis to another, the
weight of hydrogen and oxygen in the moisture must be accounted for to
determine the correct values of those elements in the ultimate analysis.
Since 2 lb of hydrogen form 18 lb of water, }4 of the moisture is hydrogen,
—
{H M/9) represents the free hydrogen present in the original sample,
or the amount beyond that included in the moisture. The free hydrogen
is the part which will remain as a certain percentage of the new "dry"
sample.
2. To compute to the "as fired" basis from the "as received" when
moisture changes:
^qq
Ci = C X _ etc., for all items except hydrogen and oxygen
- Mi + Mi
100
100 -M
9
To compute to the "moisture- and ash-free" basis from the "as
3.
received
"
:
Ci = X Iqq _- êms except hydrogen and oxygen

C
e*c'' ôr a^
—^4'
-M-
100
A
100
- If -
100
A
100
To compute to the "dry, mineral-matter-free" basis from the "as

4.
received":
These calculations are made in accordance with the Parr formulas,
FCl =
rc-°15* X 100
100 — noncoal
where noncoal = M %S + 0.08 (A — 1%S), or approximately

+
A
+
as used, noncoal = M +
1.08A + 0.55$.
The formula for noncoal, or moisture plus mineral matter, takes into
account the fact that part of the sulfur in the coal combined with iron
is
which forms iron oxide upon combustion of the sulfur, resulting in

a
change of weight and composition of the ash during combustion of the

coal. The last expression in the formula also includes the water in the
earthy mineral matter. The above formula for FCi also takes into
account the fact that part of the sulfur remains with the fixed carbon.
VM = 100 — dry, mineral-matter-free FCi

i
Since heating values of "dry, mineral-matter-free" samples are some

times desired, the formula for this computation given,
is
HV - 50 X
8
HV1 =
100 — noncoal
The quantity 50 X S accounts for the loss in heating value when the
sulfur is removed from the sample. While 5000 Btu/lb is greater than
the heating value of sulfur, it gives a close approximation to the actual
change in heating value of the sample.
5. To compute the heating value on the "moist, mineral-matter-free"
basis from the "as received":
The mineral portion of the coal is given by the same type of expression
as in the above paragraph,
Mineral = 1.084 + 0.55S
HVl =
100 - mineral X 100 =
.100 - (1.08A +0.555) X 100
Calculation of Ultimate Analysis from Proximate. While the proxi

mate analysis is comparatively simple to make on a coal sample, the
ultimate analysis is a lengthy and somewhat expensive determination.
An approximate method for finding the ultimate analysis, by computing
it from the proximate, is often very convenient to use. It has been found
that, if samples of coal are taken from different parts of the same bed,
they may have different percentages of moisture, ash, and sulfur, but the
chemical compositions of the remainder of the coal samples will be prac
tically the same. In other words, the analyses on the moisture-, ash-,
sulfur-free basis would be the same for coals from the same bed.
Proximate and ultimate analyses of a large number of coals from all
parts of the country have been made by the U.S. Bureau of Mines and
are published in their Bull. 22, 85, 123, and 193. A correlation of mines
having coals with similar analyses is available in the U.S. Bur. Mines,
Bull. 446. It is therefore possible to calculate an ultimate analysis from
the proximate, if proximate and ultimate analyses of a coal sample from
the same bed are available. Although not absolutely accurate, this
calculation can be made with the assurance of a satisfactory degree of
accuracy for use in computations for efficiency tests of coal-burning equip
ment. The method for making this calculation follows.
Let C, H, 0, and N be the percentages of carbon, hydrogen, oxygen,
and nitrogen in the sample having a known ultimate analysis, and M, A,
and S be the percentages of moisture, ash, and sulfur in the corresponding
proximate analysis. Let A i, Si, and Mi be the percentages of ash, sulfur,
and moisture obtained from the proximate analysis of the new sample;
and Ci, Hi, 0\, and Ni be the percentages of carbon, hydrogen, oxygen,
and nitrogen for the new sample for which the ultimate analysis is desired.
Also, let K be the ratio of moisture-, ash-, sulfur-free coal from the new
sample to the moisture-, ash-, sulfur-free coal from the old sample for
is,
which the ultimate analysis is known; that let

- (Mi + Ai + &)
- (M + A + S)
100
=
100
Then
= CK
Ci
Hl = - %M)K + %M,
Ol - (0 - %M)K + %MX
(H
and
Ni = NK
The percentages of sulfur and ash in the ultimate analysis are the same as
already reported for the proximate.
Example 1. The proximate analysis of a certain coal "as fired" is M = 7.76%,
VM = 35.45%, FC = 41.49%, A = 15.30%, and the sulfur content is S = 3.13%.
Calculate an ultimate analysis for this coal if the proximate and ultimate analyses of
a coal "as fired" from the same bed are as follows:
Proximate Ultimate
M = 6.79% C = 66.40%
VM = 36.70% H = 5.30%
FC = 45.96% 0 = 14.36%
A = 10.55% N = 1.19%
S = 2.20%
S = 2.20% A = 10.55%
-
Solution:
-
100
100 - (7.76
(6.79 +
15.30 + 3.13)
+
10.55 + 2.20)
=
100
100 - 26.19
19.54
=
73.81
80.46
Then
Ci = 66.40 X 0.918 = 60.91
- «f 0.918 +
?f6
Hl =
)
(5.30
= (5.30 - 0.75)0.918 + 0.86
= 4.55 0.918 + 0.86 = 4.18 0.86 = 5.04
+
X
0l _ - 0.918
+
*J^Z»)
(14.36
= (14.36 - 6.04)0.918 + 6.90
= 8.32 X 0.918 + 6.90 = 7.63 6.90 =
14.53
+
AT, = 1.19 X 0.918 =

1.09
Si = 3.13
Ai = 15.30
= 100.00
Miscellaneous Coal Tests. Following are brief discussions of ASTM

tests designed to aid in establishing accurate specifications for coal and
coke:
The ash-softening temperature test analyzes the softening character
1.
istics, or clinkering tendency, of the ash. sample of the ash pulver

A
is
ized, shaped into a cone or pyramid, and heated in an oxidizing furnace

atmosphere to form hard ash cone. This cone then heated to high
is
a
temperatures in a muffle furnace, and the temperature at which softening

of the cone occurs is observed by using an optical pyrometer or a thermo
couple. Figure 5-1 shows the appearance of cones from several different
ashes after being heated to a given peak temperature in the furnace. The
ash-softening temperature is reached when the cones fuse to a spherical
lump as shown by the third and fourth cones in the figure. The second
cone is just approaching the ash-softening temperature while the first
cone shows no deformation. The deformation temperature is reached when
the apex becomes slightly bent. The fifth cone has reached its fluid
temperature and has melted into a small puddle. Ash-softening tempera
tures of coals in the United States average about 2200 to 2300 F but
range from 1900 to 3100 F for different coals. Ash is composed of about
one-half silicon dioxide, and the other half oxides of aluminum, iron, and
Fig. 5-1. Test cones for determining ash-softening temperatures. (Courtesy of ASTM.)
calcium. The chemical composition may be correlated with the ash-
softening temperature.
2. The weathering or slacking index of a coal is an indication of its size
stability when stored exposed to the weather. To determine this
property, a coal sample screened to a 1- to l}4-m. size is air dried, soaked
in water, and again air dried. The sample is screened following the
weathering, and the size degradation is computed as the percentage of
undersize material formed by the test.
3. The grindability index of a coal may be found by either the ball-mill
or the Hardgrove method, and is useful in estimating the power required
to pulverize a coal for use in pulverized fuel burners. In the ball-mill
method, the sample is ground until 80 per cent of it passes through a
No. 200 sieve. A standard ball mill is used, and the results are reported
as 50,000 divided by the number of mill revolutions required to produce
the specific fineness. For the Hardgrove test the sample of coal is sub
jected to 60 revolutions in a miniature pulverizer, and the amount passing
through a No. 200 sieve is weighed. The Hardgrove grindability index
is then calculated as
Index = 13 + 6.93 IF
where W = weight of sample passing through the No. 200 sieve.

4. Drop-shatter tests are used to determine the "size-stability index,"

and give an indication of the decrease in lump sizes that may be expected
during handling. A sample of the coal or coke is tested by dropping it
twice onto a steel plate from a height of 6 ft. After the shattered sample
has been screened, the percentage by weight of small-sized particles
formed through breakage is computed. A similar test is the tumbler
test for coal and coke. It determines the friability, or shattering tend
ency, of coal lumps which are rolled in a porcelain or steel tumbler. It
also is useful in estimating the likely decrease in size upon repeated
handling.
5. The agglomerating and agglutinating values of coal give a measure of
the caking and coking properties of the fuel, respectively. A coal is said
to be agglomerating if the residue from the volatile matter determination
shows swelling or good cell structure and if that residue has sufficient
strength to support a 500-g weight without pulverizing. Some coals are
so deficient in coking and caking properties that only a powdered residue
remains after the test for volatile matter. Such coals are classed as
nonagglomerating and are said to be "free-burning" fuels. In general,
coals with a high oxygen content are nonagglomerating. Certain types
of coal-burning apparatus are better adapted for handling coals which
form a heavy cake, or agglomeration, in the fuel bed whereas caking is
undesirable in other equipment.
The agglutinating value is determined by mixing a small sample of coal
with silicon carbide and subjecting the mixture to a high temperature
until it forms a small, hard button. The load in kilograms required to
crush this button is reported as the agglutinating value. This property
measures both the coking and the caking properties of coal, but it is
especially useful in predicting the adaptability of a coal for making coke.
Classification of Coal by Rank. To facilitate the classification of
various coals according to rank, the ASME and ASTM have adopted a
specification based on the fixed carbon and heating value of the mineral-
matter-free analysis. Table 5-1 gives this classification. The higher-
rank coals are listed according to the percentage of fixed carbon on the
dry basis, while the lower-rank coals are rated by their heating value on
the moist basis. Borderline samples are further differentiated by use of
the slacking, weathering, and agglomerating values.
The exact designation of the rank of a coal may be made by listing two
numbers: (1) the nearest whole per cent of fixed carbon on the dry basis
and (2) the heating value in hundreds of Btu on the moist basis. Both
numbers are enclosed within parentheses indicating that they are calcu
lated on the mineral-matter-free basis. A coal having a fixed carbon
content of 62.4 per cent and a heating value of 14,580 Btu/lb, both
Table 5-1. Classiiication of Coals by Rank after ASME-ASTM Method

(Coals having more than 69 per cent fixed carbon shall be classified according to
fixed-carbon content regardless of heating value.)
"Mineral-matter-free" basis
"Dry
Additional requisite
Rank
physical properties
"
Moist" min
Min % Max % heating value
fixed carbon volatile
matter
I. Anthracite:
98 2
92 8
86 14 Nonagglomeratinc
II. Bituminous:
78 22
69 31
Less than 69 14,000
4. High volatile B Less than 69 13,000
5. High volatile C Less than 69 11,000 a. Agglomerating,
non weathering
6. Agglomerating,
weathering
c. Nonagglomerating,
non weathering
III. Subbituminous:
Less than 69 11,000 Weathering, nonag
glomerating
Less than 69 9500
Less than 69 8300
IV. Lignitic:
Less than 69 Less than 8300 Consolidated
2. Brown coal Less than 69 Less than 8300 Unconsolidated
on a mineral-matter-free basis, could be designated as (62-146) rank

coal.
Grade of Coal. The ASTM grade of a coal is a measure of five of its
physical qualities: size, heating value, ash content, ash-softening tem
perature, and sulfur content, which govern to a certain extent its usage.
The rank of a coal, being an inherent property of the fuel depending upon
its relative progression in the coalification process, should not be confused
with the grade, which may be controlled to a degree by the methods of
mining and preparation.
Size of coal is found by screening a sample to establish the range of sizes
of the coal pieces. The size range is given by stating the diameters of
the smallest and largest round hole screens which will classify at least 80
'
per cent of the sample. Five per cent of the sample may be larger than
the stated upper limit, and 15 per cent may be smaller than the lower.
The heating value is expressed to the nearest 100 Btu on the " as sampled
"
basis. Symbols standardized by the ASTM to define the ash content,
ash-softening temperature, and sulfur content are listed in Table 5-2.
The grade of a coal is given by listing the symbols designating the above
listed five properties of the sample in order. Thus, a typical coal grade
might be written as 2-4 in., 132-A8-F24-S1.6.
Table 5-2. Symbols for Classification of Coals by Grade*

(Analysis expressed on basis of the coal "as sampled")
Ash Softening temp of ash Sulfur
Symbol %, inclusive Symbol Deg F, inclusive Symbol %, inclusive
A 4 0.0-4.0 F 28 2800 and higher SO. 7 0.0-0.7

A 6 4.1-6.0 F 26 2600-2790 S 1.0 0.8-1.0
A 8 6.1-8.0 F 24 2400-2590 S 1.3 1.1-1.3
A 10 8.1-10.0 F 22 2200-2390 S 1.6 1.4-1.6
A 12 10.1-12.0 F 20 2000-2190 S2.0 1.7-2.0
A 14 12.1-14.0 F 20 minus Less than 2000 S3.0 2.1-3.0
.
A 16 14.1-16.0 S 4.0 3.1-4.0
A 18 16.1-18.0 S 5.0 4.1-5.0
A 20 18.1-20.0 S 5 . 0 plus 5 . 1 and higher
A 20 plus 20. 1 and higher-
* ASTM Standards.
When it is desired to report the exact rank of a coal along with its
grade, the symbols designating the rank are reported first. The rank
and grade listing would be (62-146), 2-4 in., 132-A8-F24-S1.6.
LIQUID FUELS
More tests have been devised to analyze the physical and combustion
properties of petroleum fuels than for any other fuels, mainly because
gasolines and fuel oils are used in equipment wherein the fuel requirements
are very exacting. The ASTM has standardized a series of tests to
measure the many important properties of liquid fuels. Anyone wishing
to conduct such tests should consult the ASTM Standards on Petroleum
Products for complete instructions as to the equipment, procedure, and
method of reporting results for the many tests discussed briefly in the
paragraphs which follow. *
Heating Value. The heating value of a liquid fuel is determined by
burning a carefully measured quantity in the presence of oxygen in a
bomb-type calorimeter, as explained in Chap. 3. The heat released is
*
Complete ASTM Standards may be obtained from the American Society for
Testing Materials, 1916 Race St., Philadelphia 3, Pa.
absorbed by a known weight of water in a jacket surrounding the con

stant-volume bomb, and the temperature rise of the water gives an
accurate determination of the heat released from the combustion of the
fuel. Either the higher or the lower heating values may be reported on
the basis of Btu per pound or Btu per mole of fuel. Since the ultimate
aim of any combustion process is to release heat, heating values of fuels
are necessary for computing the fuel requirements for a given application.
Volatility. Petroleum fuels are made up of a large number of different
hydrocarbons each having a different boiling point. Hence, upon heating
a gasoline or fuel oil, it is found that all of the liquid does not pass to the
vapor phase at one temperature, as is the case when a liquid such as
Fig. 5-2. ASTM distillation test apparatus.
water is heated at constant pressure. Instead, a small percentage of the

fuel will vaporize at a low temperature, and a much higher temperature
must be reached before the fuel completely vaporizes.
The apparatus used for the ASTM distillation test is shown in Fig. 5-2.
In conducting the test, 100 ml of fuel are heated in the flask at the left.
The vapors pass through the condenser tube, which is surrounded by
cracked ice, and the condensate drips into the graduate on the right where
the quantity of condensed fuel is measured. A small percentage of the
fuel is lost in this process, part in the very volatile gases which fail to
condense and part in the heavy residue which fails to evaporate.
A reading on the thermometer is taken at the time the first drop of con
densate appears. This temperature is known as the bubble point of the
fuel, signifying that the first bubbles of vapor have passed from the
liquid. Subsequent temperature readings are taken as each additional
10 per cent of the fuel is evaporated. The final temperature required to
evaporate the liquid completely is called the dew point, since at this tern
176 FUELS AND CO M BUST ION
perature a quantity of the fuel in the vapor state would start to condense.
Typical results of the distillation tests of several fuels are plotted on the
graph of Fig. 5-3. Another point of interest regarding fuel volatility,
aside from the vaporization tendencies characteristic of each type of fuel,
is that a liquid which evaporates readily at low temperatures constitutes
a fire hazard, and extra-safety precautions must be taken during storage
800
ol 1 I 1 1 I
0 20 40 60 80 100
Per cent evaporoted
Fig. 5-3. Typical ASTM distillation curves for petroleum fuels.
and handling of the liquid. Also a large portion of a highly volatile fuel
may be lost through evaporation in storage and use.
The distillation curve for a gas oil in Fig. 5-3 shows a comparatively
wide range of boiling points for this oil, and it might be termed a "wide-
cut" fuel, suggesting that the hydrocarbons have been selected over a
wide range of temperatures in the fractional distillation process. The
present trend is toward wider cuts for diesel fuels and jet fuels, since this
allows the available supply of these fuels to be increased with present
refinery techniques and production schedules.
The ASTM distillation process evaporates liquid fuels in the presence

of fuel vapor. In most types of combustion apparatus, the fuel is
evaporated in the presence of air, and a greater percentage of the fuel
will be evaporated at a given temperature than is indicated by the
standard ASTM results. The equilibrium air distillation test may be
used to obtain more accurate information concerning the actual volatility
characteristics of fuels in use, especially for gasoline in the spark-ignition
engine. In this test, the percentages of liquid fuel evaporated in the
presence of flowing air at a controlled temperature are observed over a
wide range of temperatures and fuel/ air mixtures. Comparisons between
the results obtained by the equilibrium air distillation and the ASTM
distillation tests are made in Fig. 5-10. It is easier and quicker to conduct
the ASTM test; hence, it is more often performed and the equilibrium
air distillation temperatures are computed from the ASTM data with the
aid of suitable charts.
Gravity. The specific gravity of an oil is the ratio of its weight to the
weight of an equal volume of water. It is commonly designated as
"sp. gr. 60/60 F," indicating that both the oil and the water are weighed
and measured at a temperature of 60 F.
The oil industry uses a scale adopted by the API for measuring the
density of fuels, giving readings in degrees API. The relationship
between specific gravity and API gravity is:
Deg
6 API =
141 ^
an,an
sp. gr. 60/60 F
- 131-5
Thus a light fuel, which has a low specific gravity, has a higher API
gravity than heavy fuel.
a
The API gravity, or the specific gravity, of a fuel is obtained by
immersing a hydrometer in the fuel and observing the depth to which it
sinks. For an accurate determination, a correction must be made for
the change in density of the liquid at the temperature of the reading from
the density at 60 F due to thermal expansion. Tables are available for
finding the amount of this correction, or the following formula can be
used, with high accuracy between 30 to 90 deg API, for converting degrees
API at some observed temperature to degrees API at 60 F.
Deg API at 60 F = [0.002(60 - observed temp F) + 1]
X [observed deg API]*
The density of a fuel, being directly dependent upon the hydrogen and
carbon content of the liquid, is also related to the heating value of the
* S. R. Beitler and E. J. Lindahl, "Hydraulic Machinery," The Ronald Press
Company, New York, 1947.
fuel. The relationship between the API gravity of a fuel and the heating
value appears in Chap. 3, and in Fig. A-l in the Appendix.
Viscosity. The viscosity, or resistance to flow, of a liquid fuel is usu
ally measured by a Saybolt viscosimeter. A diagrammatic cross section
of such an instrument is shown in Fig. 5-4.
A quantity of the fuel to be tested is placed in the tube and the sur
rounding oil bath held at a constant temperature until the temperature of
the test fuel is constant. The cork is then removed, and the time in
seconds required for 60 ml of the fuel to flow through the orifice is meas
ured. The viscosity is reported as Saybolt Universal seconds (SUS) at
the temperature of the test. Tests are normally conducted at tempera
Oil for
'viscosity test
,Orifice (either
Universal or Furol)
Receiving flask
Fig. 5-4. Cross section of Saybolt viscosity test apparatus.
tures of 70, 100, 130, and 210 F. The viscosity in Saybolt Universal
seconds may be converted to absolute or kinematic viscosity by tables
available for that purpose. More viscous oils which require longer than
1000 sec to flow through the Universal orifice are tested with a larger
orifice. The Furol orifice (abbreviation for fuel and road oils) permits a
flow of approximately ten times the Universal. The test is conducted in
the same manner at temperatures of 77, 100, 122 and 210 F, but the
results are reported as Saybolt Furol seconds (SFS).
Flash and Fire Points. The flash point may be defined as the lowest
temperature at which a fuel will vaporize sufficiently to form a combusti
ble mixture of fuel vapor and air above the fuel. It is found by heating
a quantity of the fuel in a special container while passing a flame above
the liquid to ignite the vapors. A distinct flash of flame occurs when the
flash-point temperature has been reached.
The fire point is the temperature which must be reached before enough
vapors will rise to produce a continuous flame above the liquid fuel. It is
obtained in much the same manner as the flash point. Both the flash
and the fire points give a relative measure of the safety properties of fuels
since a high flash point denotes that a high temperature must be reached
before dangerous handling conditions are encountered. The minimum
flash point permitted in a fuel is usually written into the specifications.
Pour Point. The pour point of an oil is found by cooling a sample in a
test tube until no movement of the oil will occur for 5 sec after the tube is
tilted to a horizontal position. The pour point is reported 5 deg above
the temperature to which the oil must be cooled
to obtain this condition. This test is an in
dication of the low-temperature fluidity of the
Gage
fuel.
Reid Vapor Pressure. The Reid vapor-pressure
bomb affords another method for measuring the
volatility of a fuel. In conducting the test 100 ml
of fuel are sealed in the bomb and the instrument
is immersed in a 100 F water bath. Figure 5-5
illustrates an assembled Reid vapor-pressure Air
chamber
bomb ready for immersion. The portion of the
gasoline which vaporizes rises into an air chamber
above the fuel. The increase in pressure in the
air chamber is measured by a pressure gage
attached to the instrument. The vapor pressure
Gasoline
in the air chamber is reported as the Reid vapor
chamber
pressure of the fuel.
Conradson Carbon Residue. The Conradson
Fig. 5-5. Reid vapor
carbon test is designed to show how much carbon
pressure bomb.
deposit will remain when a sample of fuel oil is
evaporated under specific conditions. The test is performed by heating
10 gr of a fuel in a covered crucible to a very high temperature. The
residue which remains in the crucible following the evaporation is care
fully weighed and the per cent Conradson carbon determined as the per
centage of carbon remaining of the original weight of sample. Light fuel
oils having carbon residues less than 0.05 per cent may be tested by using
the last 10 per cent residuum from a distillation-type test for the carbon
residue test. In this case the Conradson carbon residue is reported in
per cent by weight of the 10 per cent residuum employed as the test
sample. Such a test is especially useful in comparing the clean burning
property of various fuels, and the tendency for oils to crack and form
hard carbon deposits upon being heated.
Water and Sediment. The water and sediment test is conducted to
show the percentage of such impurities in the fuel. A sample of the
liquid fuel is dissolved in benzol and the sample placed in a centrifuge for
10 min at a speed of 1500 rpm. The centrifugal force is sufficient to
separate the heavier water and sediment from the lighter fuel, and the
percentage of these impurities may be read directly from graduations on
the centrifuge test tubes.
Gum Content. The gum content of a gasoline is found by evaporating
a sample of fuel and weighing the amount of gum remaining. A 50-ml
sample of fuel is heated to a temperature of 320 F, and air is blown
over the liquid to promote low-temperature oxidation of the fuel,
resulting in the formation of gum. The results of the test are then
reported in terms of milligrams of gum per 100 ml of gasoline.
Gum is particularly troublesome in the fuel systems of automobiles and
aircraft. Gasoline which may be stored for several months at a time has
inhibitors added during the refining processes to reduce the tendency for
gum formation during storage.
Corrosion Test. Two procedures are available for testing the cor
rosive action of fuels: the copper-strip method and the copper-dish
method. In either case, a clean brightly polished copper surface is
exposed to a sample of heated fuel. Sulfur or sulfur compounds present
in the fuel react with the copper, leaving a gray or black corrosion. To
protect fuel systems, the corrosive action of a fuel must be low.
Ash. The ash content of a fuel is the solid material which remains
after complete combustion of the fuel. This quantity is measured in a
manner similar to the Conradson carbon test except that the tempera
tures of the test are sufficiently higher to ensure complete burning of the
carbon.
Sulfur. The percentage of sulfur in a fuel is determined very accu
rately by an analytical chemical test. Almost all fuels, liquid, gaseous, or
solid, contain small amounts of this element which are very harmful.
Sulfur burns to sulfur dioxide upon combustion of the fuel, and the sulfur
dioxide reacts with the water vapor formed to give sulfurous acid, and in
some cases sulfuric acid. This condition is especially bad when the com
bustion products are cooled enough to precipitate some of the water
vapor. In this case excessive corrosion of the combustion equipment
and exhaust passages will result.
Sulfur can be removed, or at least the amount present can be greatly
reduced, by expensive refining operations. Depending upon the applica
tion for which the fuel is intended, this may or may not be necessary.
Where costly combustion equipment is involved which will show undue
wear or corrosion when high sulfur fuels are used, it is economical to use
premium fuels having a low percentage of sulfur. In other cases, 1 or 2
per cent of sulfur in the fuel may be tolerated.
Octane Number. The octane number of a gasoline is a measure of its

tendency to knock, or detonate, when burned in a spark-ignition engine.
Detonation is the very rapid and uncontrolled burning of the fuel and air
mixture in a cylinder which results in an abnormally rapid pressure rise.
This sets up vibrations of the gases, the cylinder walls, and other metallic
surfaces giving a distinct knock or noise. Combustion phenomena being
what they are, it might be expected that the tendency for any fuel to
detonate varies with such factors as type of engine, engine speed, air/fuel
ratio, temperature of fuel and air entering the cylinder, compression
ratio, and many others. The effect of each of these variables upon actual
engine operation is discussed in Chap. 8.
The reason why some fuels burn smoothly and evenly while others tend
to detonate violently under the same conditions is not thoroughly under
stood. However, it has been proved that detonation tendencies are
related to the autoignition temperature of a fuel and the chain reaction
mechanism by which the fuel burns. A liquid such as isooctane exhibits
very smooth-burning characteristics. In contrast with this the longer
chain compound, normal heptane, displays a very strong tendency toward
detonation. Other organic compounds display a greater or lesser tend
ency toward detonation when burned in an engine.
The octane test is conducted by actually burning a fuel under carefully
controlled conditions in an engine of the type shown in Fig. 5-6 and com
paring its burning characteristics with those of the reference fuels. This
is a comparative test in which the knock tendency of an unknown fuel is
rated relative to the knock obtained with a fuel of known rating used
under exactly the same test conditions. The intensity of knock is
measured by a knockmeter, which indicates the rate of pressure rise
during combustion. During detonation, the rate of pressure rise is much
more rapid than it is for normal smooth combustion.
An octane number scale has been set up in which isooctane (2,2,4-
trimethylpentane) is rated 100 octane, and normal heptane is rated as 0
octane. A gasoline is rated as a 70 octane fuel if its tendency to detonate
in the test engine is the same as that of a mixture of 70 per cent isooctane
and 30 per cent normal heptane.
There are two common procedures for determining the octane rating
of motor gasoline: the motor method and the research method. They
are very similar except that they are conducted at different engine speeds
using different fuel/air mixture temperatures. The operating conditions
are somewhat more severe for the former method, and the octane ratings
obtained by the motor method are generally four to eight numbers lower
than those from the research method. Aviation gasoline is rated by a
third method using rich and lean supercharged mixtures.
Several other terms have come into common usage which relate to the
knock tendency of a gasoline. Some fuels are more sensitive to changes
of the operating conditions mentioned above than others. Thus, if two
gasolines each have an octane rating by the motor method of 74, one of
these may knock worse in a car operating on the road than the other,
simply because the actual operating conditions are not the same as the
Fig. 5-6. Cooperative fuels research (CFR) gasoline test engine. (Courtesy of Waukesha
'Motor Co.)
test conditions. The knock tendency of a sensitive gasoline varies more

with severity of engine operation than a fuel which is not considered
sensitive.
The critical compression ratio of a fuel is the compression ratio at which
the fuel just begins to produce knock in the CFR (cooperative fuels
research) engine under certain specified conditions. A high-octane fuel
also displays a relatively high critical compression ratio.
For modern aviation gasolines with knock tendencies less than for
isooctane, the knock rating may be given as the number of milliliters of
tetraethyllead which must be added to isooctane to obtain the same rela

tive knock as the unknown fuel, for example, isooctane plus 4 ml tetra
ethyllead. The performance number of a fuel likewise is related to detona
tion tendency. It is the indicated horsepower, or power developed
within the cylinder, which a given supercharged engine can produce with
out detonation relative to the indicated horsepower developed by iso
octane. Thus, a 100 octane fuel has a performance number of 100, and
higher quality fuels have performance numbers in excess of 100. This
scale has the advantage that it is more nearly linear in performance
evaluation of a gasoline than any of the other knock rating methods.
For instance, an increase of octane rating from 97 to 100 gives a much
greater increase in power output without detonation than does an
increase from 70 to 73. The performance number more nearly relates
fuel antiknock quality with engine performance.
The relative knock tendencies of various hydrocarbons, as indicated by
their critical compression ratios, are presented in Figs. 5-7 to 5-9. The
first shows the effect of molecular structure on knock for a number of
paraffins. The hydrogen atoms are omitted from the structures for
clarity, and the grouping of the C's shows the structural arrangement.
Figure 5-8 shows the critical compression ratios of several isomers of
cyclopentane while Fig. 5-9 compares knock tendencies of some olefins
with paraffin compounds of the same molecular structure. The olefins
are those compounds having at least one double bond between carbon
atoms. In general, the more compact the molecular structure, the higher
the critical compression ratio and octane rating. By the same token,
the longer the straight-chain section of the molecule, the lower is the
knock resistance. These relationships can best be explained by the
theory that the more compact molecules are oxidized by reaction mecha
nisms which lead to considerably less chain branching than the combus
tion mechanisms for the normal paraffins.
Cetane Number. The cetane number of a diesel fuel is a measure of its
ignition delay when burned in a compression-ignition engine. As will be
seen in Chap. 8, it is desirable for a fuel to autoignite very quickly when
it is injected into the cylinder of a compression-ignition engine. The

time interval between the beginning of injection and the start of combus
tion is known as the ignition delay period. A normal paraffin compound,
cetane, C16H34, has been chosen as the ideal fuel for this scale and is rated
100 cetane, whereas alphamethylnaphthalene, CnHio, has been rated
0 cetane. Just as in the octane rating test for gasoline, the cetane number
of a fuel oil is obtained by comparison between the combustion character
istics of the unknown fuel and mixtures of the reference fuels in a CFR
engine built for compression-ignition tests.
In direct contrast to high-octane gasoline, a high-cetane fuel oil has

a
a low autoignition temperature and a short ignition delay. Since the
normal paraffins excel in these characteristics, a high-cetane fuel contains
mostly normal paraffins. Thus, while there is no direct correlation
5 6 7 8
Number of Carbon Atoms in Molecule
Fig. 5-7. Effect of molecular structure on the knock tendency of paraffins as indicated by
critical compression ratio. (Lovell, Industrial and Engineering Chemistry, 1948.)
between the octane and the cetane scales, a relatively low-octane fuel
would have a high-cetane number. The aniline point gives a measure of
the amount of paraffins in a fuel oil and thus indirectly is related to the
cetane number.
Desirable Properties of Liquid Fuels. A new and specific set of fuel
requirements is imposed by each fuel application. Hence, an oil to be
used in a heavy oil burner will differ widely from the gasoline required for
an aircraft engine. Even within the same type of fuel, such as aviation
gasoline, it is found the properties that give the best performance in
Alaska are not the same as those which are the most desirable for use in
123456789 NUMBER
Fig. 5-9. Critical compression
OF CARBON ATOMS
10
ratios of some olefins as compared to paraffins of the same

molecular structure. (Boord, Greenlee, and Derfer, Ref. 8.)
the tropics. Therefore, all phases of the intended fuel application must
be considered in determining fuel requirements.
Specifications of some of the more common fuels are given in the pages
which follow. As it would be impractical to include the entire permis
sible range of each property for all fuels, an attempt has been made to
present only typical test results or specifications for some common fuels.
Fuel Oils. Any of the petroleum fractions slightly heavier than the
gasolines could come under the general heading of a fuel oil. A classifica
tion of fuel oils as set forth by the ASTM is given in Table 5-3 along with
the specifications recommended by that organization. As can be seen in
the table, in order to qualify for any one grade of fuel oil, the liquid must
meet the standards of only a few of the many fuel tests. Therefore, a
wide variation in the other fuel properties is to be expected, even in fuels
of the same number classification. This is satisfactory since most oil
burners are capable of burning oils of varying quality. The fuel oil must
have a sufficiently high flash point to be safe to handle, and it should be
fluid enough to work well in the equipment at hand. Other specifications
such as water and sediment, ash, and the Conradson carbon are stipulated
to ensure a reasonable degree of cleanliness.
Gasoline. The requirements for a good grade of gasoline are probably
more stringent than for any other liquid fuel. Because of the com
plexity of the spark-ignition engines utilizing gasolines, there are several
very exacting specifications which must be met. Before a gasoline can
be burned in an engine, it must first be vaporized. Hence, a good
vaporization characteristic is one of the first requirements. Once the
fuel has evaporated and has mixed with air, it is very important that the
air/fuel mixture burn smoothly and evenly within the cylinder of the
engine. In other words, the octane number of the fuel must be high
enough to meet the requirements of the particular engine in which it is
• being used. For the past several years, the octane rating by the motor
method has been considered indicative of the actual performance of the
fuel in an engine on the road. However, in the modern high-compression
engines, the octane rating by the research method seems to be a better
index of actual road performance, and the research-method octane rating
is being used more frequently to denote the antiknock property of a fuel.
While many variables affect the combustion performance of a gasoline,
the octane number of the fuel is one of the most important.
Not only must gasoline vaporize and burn smoothly, but it is highly
undesirable to have large amounts of impurities in the nature of gum,
carbon, sulfur, water, sediment, etc., present. Therefore, care must be
taken to keep the impurities to a minimum. Finally, because of the
complex distribution system whereby gasoline gets from the refinery to
the user, good storage stability is necessary.
The volatility of a gasoline is measured by two tests : ASTM distillation
and Reid vapor pressure. In the ASTM distillation test, the initial
boiling point, or bubble point, serves as an index of the "front end"
volatility of the gasoline. It has some relationship to cold engine start
ing, but because the bubble point is much more difficult to measure
Table 5-3. ASTM Tentative Specifications for Fuel 5ils, 1948
(Technical requirements same as National Bureau of Standards Commerrdal Standards, CS12-48.)
Distillation Saybolt viscosity Kinematic viscosity,

Water Carbon
F
tem29 sec. centistoke
Flash Pour and residue Ash, Sulfur, Grav
%
point, point, sedi on 5% % by by ity,
F
F
ment residue, wt. wt. 500
Grade End Universal Furol
3
%
F
F
vol, 5% 90% At 50 At 022
F
F
point at 50 at 022
Go
Min Max Max Max Max Max Max Max Min Max Min
t-i
Max Max, min Max, min Max Min
A
50
0.
No. distillate oil intended f5 Trace 0. 05 520 625 2.2 max 0.5 25
4%
0
4.
vap5izing pot-type burners and 5 min

to
other burners requiring this legal*

grade of fuel
4
A
50
3.
20 0.5 0.85 675 40 max

ts
No. 2. distillate
oil f5 general- 0.0 26
purpose domestic heating f5 use 5
0
in burners not requiring No. legal
fuel oil
No. 5. An oil f5 burner installa 080 20 0.50 0. 5 025 5 26.4 max No limit
5
8.
tions not equipped with preheat 5 min

legal
&
ing facilities
A
0
5.
No. oil f5 burn 030 0.00 0. 5 050 5 32. min 80 max No limit
SO
residual-type
er installations equipped with 5
preheating facilities legal
6.
No. An oil f5 use in burners 050 2.00 300 5 638 max No limit
equipped with preheaters and 92 min
a
permitting high-viscosity fuel
*
Legal as required by local fire regulations, Fire Underwriters, 5 state laws.
00
accurately than the 10 per cent evaporation point, the latter is generally
given as the measure of cold starting characteristics. In comparing
gasolines, the lower the 10 per cent point, the easier an engine will start
with that fuel in cold weather. The 20 to 70 per cent range is important
as an index of the warm-up characteristics of the gasoline, while the 90
per cent point indicates the amount
400
of any very heavy hydrocarbon com
pounds present which do not evapo
rate quickly. The latter may cause
ASTM Distillation A poor mixture distribution in the in
Summer-^/ take manifold and thereby affect
Winter^
/ / acceleration of the engine. In addi
/ / tion, these portions of the gasolines
/ / which do not evaporate may work
/ / down along the cylinder walls and
/ lead to dilution of the lubricating oil.

The Reid vapor pressure is also a
measure of the evaporation char
acteristics of the fuel. Both it and
the 10 per cent point of the volatility
curve show the tendency for the gas
oline to vapor-lock the fuel system.
While it is necessary to have suffi
rS Equilibrium cient gasoline evaporated at a low
air distillation
temperature to ensure good cold-
16 to I A/F ratio
weather starting, it is most important
0
J 20
I
40
I
60
L
80 O0
that the fuel does not form an undue
amount of vapor, lest the fuel supply
Per cent evaporated
Fig. 5-10. Comparison of volatility char lines become vapor-locked and pre
acteristics of average summer and winter vent the flow of gasoline to the engine.
premium gasolines as determined by the
ASTM distillation test, and the equilib In order to meet the requirements of
rium air distillation test for 16 to 1 air /fuel cold-weather starting and vapor lock,
ratios. (After Barber and Macpheraon,
SAE Trans., January, 1950.) the volatility of commercial gasoline
in this country is varied from summer
to winter and from one temperature zone to another within the country.
Figure 5-10 compares average ASTM distillation curves for summer and
winter motor gasolines and shows the difference between ASTM distilla
tion temperatures and computed equilibrium air distillation tempera
tures. The Reid vapor pressure of summer gasolines marketed in 1948
was about 7.7 psi; and winter-grade gasolines averaged about 2 to 3 psi
higher than the summer fuel.
The octane requirement of a gasoline may be met by blending stocks
high in isoparaffin, olefin, and aromatic hydrocarbons, or by adding some

inhibitor, tetraethyllead being by far the most common. Both methods
are used today to give commercial gasolines varying from about 72 to 92
octane number. Specifications for military gasolines for reciprocating
engines are given in Table 5-4. In Table 5-5, average properties of auto
motive gasolines sold commercially in the United States are tabulated.
Table 5-4. Specifications fob Military Aviation Gasolines*

MIL-F-5572 (June, 1951)
Grade
Property
80 91/96 100/130 115/145
10% evaporated point, max F 167 167 167 167

90 % evaporated point, max F 275 275 275 275
Sum of 10 % + 50 % temp, min F 307 307 307 307
End point, max F 338 338 338 338
Lead ml/U.S. gal, max 0.50 4.60 4.60 4.60
Color Red Blue Green Purple
Dye mg/U.S. gal, min 7.93 5.0 10.0 6.0
Dye mg/U.S. gal, max 9.15 5.67 11.67 7.94
Freezing point, max F -76 -76 -76 -76
Octane No. lean, min 80 91 100
Performance No. lean, min (58.3) (76) 100 115
Octane No. rich, min 96
Performance No. rich, min (87 . 5) 130 145
18,700 18,700 18,700 18,700
0.05 0.05 0.05 0.05
Inhibitors permitted f Yes Yes Yes Yes
Accelerated gum (16 hr), mg/100 ml, max 6.0 6.0 6.0 6.0
None None None None
Reid vapor pressure, psi, max 7.0 7.0 7.0 7.0
Reid vapor pressure, psi, min 5.5 5.5 5.5 5.5
Aromatics permitted Yes Yes Yes Yes
* "Aviation Fuels and Their Effects on Engine Performance," Ethyl Corporation,

t Inhibitors may be added not in excess of 1.0 lb per 5000 U.S. gal of fuel.
The presence of large quantities of gum in gasoline usually causes valve

sticking and heavy intake manifold deposits. Tentative ASTM specifica
tions for motor gasoline have set a maximum limit of 5 mg of gum per 100
ml of fuel. Most gasolines on the market today test considerably under
this figure, however.
Sulfur is a very undesirable impurity in motor gasoline. Not only does
it form acid compounds in the engine and exhaust, but it reduces the
susceptibility of tetraethyllead to increase octane number. A few gaso

lines may have sulfur content as high as 0.30 per cent. The great
majority of them contain much less than this, with the average sulfur
content being around 0.08 per cent. Recent research has shown that it
may be possible to design engines to use a higher percentage of sulfur than
is now considered safe.
Table 5-5. Average Properties of Gasolines Sold in the United States,

Winter 1950-1951*
Property Regular Premium
84.5 91.1
79.6 82.9
ASTM distillation temp, F:
91 92
10% point 121 119
223 216
90% point 335 334
End point 401 398
Distillation loss, % 1.5 1.5
Reid vapor pressure, psi 10.9 10.7
Tetraethyllead, ml /gal 1.86 2.24
Gum, mg/100 ml 2 2
* Ethyl Corporation.
Jet Fuels. Requirements for jet fuels are much less exacting than
those for the spark-ignition engine. At the present time, the require
ments for a jet fuel are still in a state of flux, because desirable fuel
properties change with new developments in the gas turbines themselves.
Kerosene, gasoline, and the lighter gas oils have all been used as fuels for
jets and gas turbines.
Because of the nature of the constant-pressure combustion in the gas
turbine, a high-octane fuel is not required. Neither is a highly volatile
fuel necessary. For use in aircraft, a high-heating-value fuel with a low
specific gravity is ideal, since this permits the greatest heat release for the
least weight of fuel. Storage stability and cleanliness are as desirable in
jet fuels as in gasolines. In addition, the freezing point for any fuel
' which is to be used at high altitudes and cold temperatures should be
rather low.
In comparing a volatile fuel, such as gasoline, with a less volatile one,
such as a light gas oil, for use in gas turbines and jets, the following effects
become apparent. The volatile fuel (1) gives easier starting in cold
temperatures, (2) gives a slightly better combustion efficiency, (3) leaves

less deposit in the combustion chamber and on the turbine blades, (4) pro
vides a greater fire hazard, (5) creates a greater danger of vapor lock of the
fuel system, (6) gives high evaporation losses through the breather of the
fuel tank at high altitude. The last two difficulties are practically non
existent with fuels having a low volatility.
Table 5-6. Specifications for Turbine-engine Fuels*
MIL-F-5616, MIL-F-5624A,
grade grade
Property
JP-1 JP-2 JP-3 JP-4
Flash point, min, F 110.0

Reid vapor pressure, psi, min . 5.0 2.0
Reid vapor pressure, psi, max . 2.0 7.0 3.0
Initial boiling point, min F. . . . 150.0
10% evaporated point, max F. 410.0 250
90% evaporated point, max F. 490.0 400.0
End point, max F 572.0 500.0 600.0 550.0
Color White White White
Saybolt, color, max + 12.0 + 12.0
Freezing point, max F -76.0 -76.0 -76.0 -76.0
Sulfur, % by wt, max 0.20 0.20 0.40
Inhibitors permitted f No No Yes Yes
Accelerated gum (16.0 hr) 8.0 20.0 20.0
Corrosion, copper strip None None None None
Water tolerance permitted None None None None
Specific gravity (60/60), max. . 0.850 0.850 0.802 0.825
Viscosity at —40 F, cs, max. . . 10.0 10.0
Viscosity at 100 F, cs, min 0.95
Aromatics % by vol, max 20.0 20.0 25.0 25.0
Bromine No., max 3.0 3.0 30.0 30.0
Lower heating value, Btu/lb, n 18,400 18.400
* "Aviation Fuels and Their Effects on Engine Performance," Ethyl Corporation,

t Inhibitors may be added not in excess of 1.0 lb per 5000 U.S. gal of fuel.
The type of hydrocarbon which makes up the fuel has an effect on com
bustion efficiency and on the smoke and engine deposits created. In this
connection, the order of preference is for paraffins, naphthenes, olefins,
and finally aromatics. Also the straight-run distillates are easier to burn
completely than cracked fuels, probably because of the higher percentages
of olefins and aromatics formed in the cracking process.
As is evident in the military specifications for jet fuels in Table 5-6, the
present trend is toward the use of the heavier, less volatile liquids. Also
the use of wide-cut fuels is permitted. The less volatile fuels have an
advantage in Btu per gallon of fuel, and they are more available since
they are not competing with gasoline for the lighter hydrocarbon com
pounds in the crude.
Diesel Fuels. The term diesel fuel includes almost as wide a variety of
oils as does the classification of fuel oils. In fact they are fuel oils which
have a few characteristics which make them more adaptable to engine
use than some other types of fuel oils. The chief requirements for diesel-
engine service are good cetane number, freedom from impurities, and a
fairly high flash point.
The cetane number of diesel fuel is not so critical for compression-
ignition engines as is the octane number of gasoline for gasoline engines.
Still it is desirable to have a fuel whose cetane number is around 40 or
better for most modern engine designs. Water, sediment, and ash con
tents of diesel fuels must be kept low since dirty fuel is one of the chief
causes of trouble in compression-ignition engines. The sulfur content
should be low to prevent excessive cylinder and valve wear and corrosion
in the exhaust manifold. Table 5-7 gives diesel fuel properties for eight
selected test fuels. These fuels were chosen by the SAE as representative
of fuels of widely different characteristics, and considerable research
involving the combustion characteristics and general suitability of such
fuels is being conducted in a number of laboratories.
Viscosity is not an important criteria for a good diesel fuel because the
fuel can be heated to reduce the viscosity so it can be forced through the
fuel system. Fuel oils of the No. 6 grade, sometimes called Bunker C oil,
are being used as diesel fuels in a number of applications with excellent
success, although they are not recommended for most modern diesels,
particularly for the small high-speed engines. Oils similar to the No. 1 or
No. 2 fuel oils are found most satisfactory for high-speed compression-
ignition engines.
GASEOUS FUELS
Fewer physical tests are required to determine the properties of gaseous

fuels than for coal and liquid fuels. The chemical analysis of the gas, the
density, specific gravity, viscosity, diffusibility, and heating value are
some of the tests which may be made. The specific gravity of a gas is the
ratio of the weight of a given volume of gas to the weight of an equal
volume of dry air measured at the same temperature and pressure. The
AG A has established 60 F and 30 in. Hg abs pressure as the standard for
density and heating value determinations for gases. Both heating
values and densities may be computed to any other temperature and
pressure by applying the fundamental gas laws.
Table 5-7. Laboratory Analysis of SAE Diesel Test Fuels*
Fuel number and description
dieser
cracking
cracking
is."
naph.
primary
%
Fischer-Tr.psch
N..
Straight"run
8traight-run
£8-£8
n-Heptane,
fuer (mid-C)
.f
Catarytic
Cetane,
catarytic
Brend,
3. Brend
8,
.ctane
brend
7888
st.ck
&
8
7. 8.
8.
£.
8.
£.
8.
Cetane No., ASTM 41.0 54.0 46.0 40.0 40.5 29.0 52.0 83.7
Gravity, API 33.8 39.5 36.7 72.9 21.2 26.4 37.5 51.7
Specific gravity 60/60 F. . . 0.8560 0.8230 0.8413 0.6922 9230 0.8915 0.8341 . 7684
0
0
.
Viscosity, SUS at 100 F. . . 41 .0 34.0 36.0 30.0 33.3 35.6 35.4 33.8
Viscosity, kin at 100 F, cs. 4.594 2.393 3.00 0.5353 2. 160 2.875 2.812 335
- 10
2
Cloud point, F -10 40 -35 8' 12 32
8'
0
Pour point, F -10 -5 -10 40 -35 10 +30
8'
Flash point, F 210 150 185 225 190 164 172
Distillation:
Initial F, 417 350 372 205 458 419 337 352
10 %, F 453 412 425 206 466 453 432 423
20 %, F 464 432 446 207 467 466 454 448
30 %, F 474 448 460 207 468 476 473 463
40%, F 482 464 476 207 470 486 487 476
50%, F 492 481 489 207 473 496 503 489
60 %, F 502 499 504 207 476 509 518 505
70 %, F 515 519 526 207 482 525 535 523
80%, F 534 548 545 207 492 544 565 546
00%. F 556 587 584 207 514 579 597 586
End point, F 631 660 636 222 541 586 634 622
% recovery 99.0 99.0 99.0
% residue 1.0 4.5 2.2 2.4
Gum, nig/100 cc 43.4 6.2 52.4 1.0 13.8
Conradson carbon, % res. in 10 *
residuum None 0.011 None 058 0.08 0.025 11
0
0
Ash, % in residuum None 0.006 0.02

Neutralization No 0.08 0.09 0.45
Corrosion, Cu strip, 212 F No. OK No. OK OK Pass Pass Pass
2
1
strip strip
Sulfur, total %. 0.219 0.25 0.199 0.0097 0.87 0.20 0.12 0.04
Aniline point, F. . . . 134.6 150.2 145.0 153.8 96.4 108.5 154.8 189.7
Bromine No 8.2 3.7 0.0 5.2 6.9 2.6 7.2
Dispersion, specific. 155.6 118.5 104.0
Refractive index. . . . 1.4819 4586 . 5035 1.4651 1.4315
1
1
.
Aromatic C atoms, %. . . 24.0 13.0 31 16.0 0.0

0
Naphthenic C atoms, %. 17.0 22.2 32.0 24.2 0.0

Olefinic C atoms, % 10.5 4.5 8.5 3.5 9.5
Paraffin C atoms, %. . . . 48.5 60.3 28 56.3 90.5
5
E. W. Landen, Combustion Characteristics of Diesel Fuels, SAE Trans., 1949.

*
Problems
1. The proximate and ultimate analyses of a certain coal "as received" are given
below. From these, calculate the proximate and ultimate analyses on the "moisture-
free" or "dry" basis.
Proximate Ultimate
Per Cent Per Cent
Moisture 4 . 33 Carbon 68 . 30
Volatile matter 40 . 21 Hydrogen 5 . 37
Fixed carbon 45 . 07 Oxygen 10.69
Ash 10.39 Nitrogen 1 . 50
100.00 Sulfur 3 . 75
Ash 10.39
Sulfur. 3.75 iooToo
Heating value = 12,492 Btu/Ib
2. What would be the ultimate analysis of the coal in Prob. 1 "as fired," if the
moisture content changed to 1.8 per cent by the time the fuel was fired?
3. Find the ASTM rank of the coal listed in Prob. 1.
4. Coal used during a boiler efficiency test came from the same bed as the coal
listed in Prob. 1. The proximate analysis of the coal sample taken during this test
plus the percentage of sulfur is given below on the "as fired" basis. Calculate the
ultimate analysis of this coal on the "as fired" basis.
Proximate "as fired"

Per Cent
Moisture 6.92
Volatile matter 36 .44
Fixed carbon 42.09
Ash 14.55
Sulfur 3.12
5. A coal sample has the following analysis "as received":
Proximate Ultimate
Per Cent Per Cent
Moisture 6.1 Carbon 65.0
Volatile matter 45 . 5 Hydrogen 5.2
Fixed carbon 36.2 Oxygen 14.3
Ash 12.2 Nitrogen 1.3
Sulfur 2.0
Ash 12.2
By the time the coal was fired, the moisture content had changed to 3.8 per cent in the
"as fired" proximate analysis.
Calculate the percentage of carbon and of hydrogen in the ultimate on (o) the
"moisture-free" basis, (6) the "as fired" basis.
6. The proximate and ultimate analyses of a coal "as received" are as follows:
Per Cent Per Cent

Fixed carbon 45.53 Hydrogen 5.28
Volatile matter 37.42 Oxygen 14.26
Sulfur 2 07
Ash 9.87
When the coal was fired during a test, the moisture of the sample "as fired" was
redetermined and showed moisture equal to 4.62 per cent.
Calculate the percentage of carbon and of hydrogen in the fuel "as fired."
7. The U.S. Bureau of Mines has sampled the coal from a certain mine and found
the proximate and ultimate analyses "as sampled" from that mine to be as follows:
Proximate Ultimate
Per Cent Per Cent
Volatile matter 36 . 70 Hydrogen 5 . 30
Fixed carbon 45 . 96 Oxygen 14 . 36
'
Sulfur 2.20
Sulfur 2.20 Ash' 10.55
Heating value = 13,480 Btu/lb
A power plant buys a trainload of coal from this same mine, and the proximate
sample of this coal "as received" showed:
analysis of a representative
Per Cent
Moisture 8.76
Volatile matter 34.45
Fixed carbon 42 . 49
Ash 14.30
Sulfur 3.03
Three months later at the time this coal was fired in a boiler efficiency test, the moisture
of the coal "as fired" was determined to be 9.82 per cent. Calculate the ultimate
analysis of this coal "as fired."
8. Write the expression for the ASTM rank and grade of the coal in Prob. 7 "as
fired," if the size of the coal is J£ to 2 in. and the ash-softening temperature is 2260 F.
9. What would be the ASTM grade and rank designation for the coal of Prob. 6
"as received" if it was sized 1J.£ X J4 m- and had an ash-softening temperature of
2180 F?
10. Give the correct ASTM rank and grade designation for the high-volatile A
bituminous coal listed in Table 1-6 if the coal size is 4 X % in. and the ash-softening
temperature is 2610 F.
11. A petroleum fuel shows an API gravity of 64.5 at 78 F. Calculate the API
gravity at 60 F, the specific gravity at 60/60 F, and the pounds per gallon of fuel.
12. The specific gravity 60/60 F of octane is 0.707. What should be the API
gravity at 83 F?
13. List the coal tests which might be of significance to the purchaser for a large
power plant; for a small retail dealer.
14. Compare the various fuels listed in Fig. 5-3 for tendency to vapor-lock, relative
safety in handling and storage, loss in storage, tendency to leave a deposit in the
combustion chamber.
16. How many petroleum tests are related to a measure of fuel volatility? Discuss
the specific significance of each of these tests.
16. What fuel tests measure the cleanliness and clean burning properties of liquid
fuels?
17. What is meant by a 92 octane fuel? What other terms are related to the anti
knock characteristic of a fuel for use in spark-ignition engines?
18. Why would a wide-cut gasoline be undesirable for automobiles? Why can it be
tolerated in a jet engine?
References
1. ASTM, Standards, Current.

2. "Aviation Fuels and Their Effects on Engine Performance," U.S. A. A. F., T.O.
06-5-4, December, 1946.
3. Blade, 0. C, U S. Bur. Mines, Rept. Invest. 4248, January, 1948.
4. Johnson, A. J., and G. H. Auth, "Fuels and Combustion Handbook," McGraw-
Hill Book Company, Inc., New York, 1951.
5. Landen, E. W., Combustion Characteristics of Diesel Fuels, SAE Trans., vol. 3,
p. 200, 1949.
6. Murray, H. A., Jet Fuels for Aircraft, Oil Gas J., Jan. 8, 1948, p. 53.
7. Shnidman, L., "Gaseous Fuels," American Gas Association, 1948.
8. Boord, C. E., K. W. Greenlee, and J. M. Derfer, "Oxidation Reactions as
Related to Hydrocarbon Structure and Engine Knock," paper presented before
the ACS, Division of Petroleum Chemistry, April, 1951.
CHAPTER 6
GAS AND OIL BURNERS
In this and the following chapters, an attempt is made to explain

present combustion equipment and techniques in the light of the com
bustion phenomena and physical properties of the fuels discussed previ
ously. Such an explanation cannot be complete in every detail because
there is a definite gap between the techniques and the theories proposed
to date. In other words, the combustion theories have not yet caught up
with practice in the field. Regardless of this, the influence of certain
basic combustion phenomena on the design and operation of burners will
be evident, and equipment performance in general seems to conform to
the same fundamental principles which appear significant in the experi
mental results discussed in Chap. 4. Thus, the art of burning fuels
successfully approaches a science as the results of fundamental research,
and the ensuing theories permit an enlightened explanation of the process.
Although the techniques employed to burn the many fuels are extremely
varied, it will be noted that there is a definite pattern in all cases. At
some stage of the process, each fuel must be mixed with the proper
amount of air and subjected to an adequate ignition source.
Modern gas and oil burners are presented together in this chapter, not
particularly because they operate in the same manner, although there
are many similarities, but because they are both generally small compact
units designed to mix air and fuel effectively in the desired proportions.
Also, many industrial furnaces are equipped to burn gas or oil, either
separately or together, in any proportions. Hence, the design of one
burner is often built around the design of the other.
GAS BURNERS
Many designs and styles of gas burners are available today, but they all
operate on the common principle that a part of or all the air for combus
tion must be mixed with the gas before ignition. Gas burners are
generally classified as being of either the atmospheric or the high-pressure
type, depending upon the pressure of the gas admitted to the burner.
The latter units use gas at pressures from Yi to 40 psi and create high
rates of heat release in the large-volume combustion chambers of indus
trial furnaces, kilns, etc. Atmospheric gas burners use gas supplied at
pressures of about 2 to 12 in. water gage. Practically all domestic and
197
commercial gas appliances, as well as some industrial units, employ

atmospheric burners. Because of their simplicity and widespread appli
cation, the use and design of atmospheric burners will be dealt with in
some detail here.
The AGA has been the leading agency to sponsor research on, and
publish literature pertaining to, all types of gas burners. It has also dis
seminated information of general interest to the gas industry. Many
new design guides have been published by this organization to further the
Secondary
air
Port
Fixed orifice
Burner head Gas
,
Throat valve
Mixing tube
Primary air
Bell-
Primary -air shutter-*
Mixture pressure Gas distribution
in burner head pressure
Atmospheric
I Pressure at
flame cone
Mixture pressure
at throat
pressure
Fig. 6-1. Cross section of a typical atmospheric gas burner, and a diagram showing the
static pressure at different positions in the burner.
scientific design and safe operation of gas appliances. References 1

through 6 at the end of this chapter constitute a select list of this litera
ture. The Association also has sponsored codes which have been adopted
as American Standard requirements for all gas-burning appliances built
under AGA approval. The codes stipulate the design features of burners
and combustion chambers to ensure substantial and durable construction
and safe, acceptable performance.
Atmospheric Gas Burners. Natural gas is usually distributed in city
piping systems at about 7 in. water gage, manufactured gas at about 5 in.
water, and liquefied petroleum gases at about 11 in. A multitude of
different sizes and types of burners are built for use at these pressures. A
typical atmospheric gas burner is shown in Fig. 6-1. All atmospheric gas
GAS AND OIL BURNERS 199
burners operate on the long-established bunsen principle. Gas is emitted

from the orifice with a velocity proportional to the square root of the gas
pressure differential across the orifice. Actually, the pressure at the gas
orifice may be slightly less than the distribution pressure owing to pressure
losses in the piping and fittings immediately ahead of the burners. These
losses should be kept to a minimum to secure the best burner performance.
The ejection velocity is great enough to create a low static pressure in
the vicinity of the throat of the venturi-shaped mixing tube. This low
static pressure, a result of the Bernoulli effect, gives the necessary pres-
140, ,
Input rate, 1000 Btu/(sq in. port area) (hr)

Fig. 6-2. Characteristic limit curves for an atmospheric gas burner.
sure difference to induce primary air from the atmosphere to flow through
the openings in the air shutter and into the bell of the mixing tube. As
the primary air and gas continue through the diverging section of the
mixing tube, they are uniformly mixed and the pressure is increased as
velocity head is converted back into pressure head. A static pressure
diagram appears in Fig. 6-1. The primary-air /fuel mixture issues from
the burner ports where a small bunsen-type flame burns at the tip of each
individual port. Secondary air enters the flame from above the ports to
complete combustion in the outer envelope, or mantle, of the flame. The
combustion reaction of the premixed air and gas reaches chemical equi
librium at the surface of the inner cone as discussed in Chap. 4.
Curves showing the burner operating characteristics appear in Fig. 6-2.
A different set of curves of this type would be obtained for each gas
burner depending upon the burner design, the operating temperature, and
the type of gas used. However, the relative shapes of the curves are
always the same, and it is their position only which is changed. The
operating point at any instant can be plotted in terms of the gas input
rate and the amount of primary air in percentage of the theoretical air
required for complete combustion. (For natural gas about 9.5 cu ft of
air is required stoichiometrically per cubic foot of gas. If 4.75 cu ft of
air per cubic foot of gas is used as primary air, the primary air used is 50
per cent of theoretical.) For operating conditions above line AB, the
flames tend to lift from the burner ports and, under extreme conditions,
may be extinguished. This condition is termed blowoff or lifting. When
the air/gas mixture issues from the ports at such a high rate that the
transformation velocity (Fig. 4-10) of the fuel cannot permit burning to
occur in a compact, well-defined inner cone, the flame will tend to lift
from the ports until enough increased cone area is created to permit burn
ing at the transformation velocity of the mixture.
Below line CD so little primary air is used that a long, wavy, yellow,
dispersion-type flame results. Dispersion flames may cause incomplete
combustion with formation of soot or carbon monoxide. Hence, for any
given gas flow, the percentage of primary air should be kept between the
blowoff and the yellow-flame limits. It will be noted that as the gas flow
rate increases, the range of percentages of primary air for satisfactory
combustion narrows to the point where it may be difficult to achieve good
control. To the right of the line marked CO, increasing amounts of
carbon monoxide appear in the combustion products, usually due to
flame impingement on combustion chamber walls. Inadequacy of either
draft or secondary-air supply will likewise promote the formation of
carbon monoxide.
The gas flow may be so low that the flame may go out, or it may not
wait for the gas to issue from the port but may strike back through the
port and mixing tube to burn at the gas orifice. The latter difficulty is
termed flashback. It occurs when the transformation velocity of the
primary-air/ gas mixture is greater than the velocity of the mixture leaving
the port. Flashback occurs any place to the left of line EF and causes
incomplete combustion, sooting, clogging, and overheating of orifices and
burners. In addition, a very objectionable noise accompanies flashback.
The noise of extinction heard on some burners is actually flashback which
results as the burner is turned off and the gas flow rate is reduced to zero.
The tendency to flash back is influenced by the composition of the gas,
proportion of air in the mixture, mixture temperature, port diameter, and
other factors. Figure 6-3 shows the variation in flashback limits as
established by several of these factors.1 To the left of any one curve,
flashback would occur while normal burning would be experienced to the
right of the curve.
Turndown is the ratio of maximum to minimum gas flow obtainable

with a given burner. In practice the maximum gas rate is limited by
incomplete combustion, blowoff, or the inability to inject sufficient
primary air. The minimum is limited by the flashback propensity or the
extinction of the flames. For most gas appliances requiring a variable
heating rate, a turndown ratio of 5 to 1 is desired.
1 1 1 I6O1
— Natural gas,No 26 ports
\
o
o — Coke-oven gas,Na36 ports y
£ i V 1 1401 sis. - Coke-oven g8as,800 F-
x "X
\ *\
\s
8V OS)/-,C
v.
s
80) 1*80F
OF S
JSC S
\
/> i
10 12
input rate, 1000 Btu/(sq in.port area)(hr) input rate,1000 BtuAsq in.port areaKhr)
Fig. 6 -3. Effect of gas composition, temperature, and port size on the propensity for flash-
back. (AGA Bull. 10, 1940.)
The AGA American Standard test codes require that burners be tested
at the normal pressures shown in the accompanying table. The burners
Gas Pressure, Inches Water Gage
Liquefied
Application Manufactured Mixed Natural
petroleum
Central heating 6 7 11
Range, water heater, etc 6 6 7 11

Pressure range 8' 50% 8' 50% ±50% 8-11
are adjusted to operate at these normal pressures, and they must be

capable, without further adjustment, of good operation throughout the
pressure range prescribed. Operation should be steady with no lifting of
flames or formation of yellow tips. Most important, no carbon monoxide
should be formed under the test conditions. AGA codes specify that
not more than 0.04 per cent carbon monoxide may be present in the flue
gases from any domestic burner and that not more than 0.01 per cent
should remain in the combustion products from a gas-range burner. The
combustion space is mentioned in the code requirements, but the volume
of the combustion space is not a critical factor, provided there is no direct
flame impingment against cold walls.
To adjust an atmospheric gas burner, the orifice size or the manifold
pressure is adjusted so that the gas rate conforms to the rated heat input
of the burner. The air shutter is then adjusted to give the proper pro
portion of primary air for the type of flame desired. The pictures in
Fig. 6-4 show the appearance of natural gas and butane flames with vary
ing percentages of primary air. With high primary air, a short hard
flame is obtained, while with less air the flame tends to be longer and a
little softer.
Even in such a simple device as an atmospheric gas burner, good opera
tion is very much dependent upon several interrelated details of construc
tion. In spite of this, great flexibility of design is permissible and burners
are built in almost numberless sizes and shapes. Figure 6-5 shows several
designs of atmospheric gas burners currently used in gas appliances.
Butane Natural Gas

Fig. 6-4. Appearance of natural gas and butane flames at varying primary-air injections.
{Courtesy of American Gas Association.)
Burner Design. For years, gas burners were designed from a few
empirical relationships observed in successfully operating burners.
Recent fundamental research with burners has led to better correlation of
design data, and modern gas-burner design has become much more
precise. In AGA Bull. 10 design criteria and limitations are set forth.
This report is the result of years of research and study at the testing
laboratories of the AGA and other research institutions. Some of the
design factors suggested in that bulletin and in "Gaseous Fuels," pub
lished by the AGA, will be presented here along with other recently
published data on the subject.
Orifice. Gas orifices for atmospheric gas burners are of the fixed,
adjustable, and primary-air-control types. Fixed orifices yield higher
primary-air injection but do not make provision for variable gas flow as
readily as do the other designs. The primary-air-control type of orifice
has the added advantage that it regulates the quantity of primary-air
injection as well as the gas flow. Orifice sizes are designated by the
twist drill size used in drilling the hole. Table A-l in the Appendix lists
the standard drill manufacturers' sizes (DMS) along, with other com
mercially used drill sizes sometimes needed to fill wide gaps in the DMS
series.
The quantity of gas which will flow through a given orifice may be
determined from the following equation, which is derived from the basic
Fig. 6-5. Some contemporary atmospheric gas burners. {Courtesy American Gas
Association.)
flow relationships (q = AV and V = -\/2gH):
(1) q = 6880 GA E
where q gas flow rate, cu ft/hr at 60 F, 30 in. Hg abs
C coefficient of discharge
A area of orifice, sq in.
V pressure differential across orifice, in. water gage
d specific gravity of gas (air = 1.0)
P initial gas pressure, in. Hg abs
T temperature, R
This type of equation is used for flow calculations throughout the gas
industry because it gives the cubic feet per hour of gas flow at the indus
try standard temperature and pressure regardless of the existing
temperature and pressure.
Commercial fixed orifices, as used in gas appliances where variable gas
flow is not required, are generally of the type shown in Fig. 6-6. The
coefficient of discharge for this type of orifice varies little with variation
of angle of approach or of drill size. For general design calculations, dis
charge coefficients of the following magnitude
ê used with reasonable accuracy: No.
/^~"\\ f*4^^28)~^~/,5''
VV /
" "* / BSZL
'
"~ V'
^ orifices, 0.77, increasing gradually to 0.815
for a No. 71, 0.815 for No. 71 through 51,
Fig. 6-6. A common design of and increasing gradually to 0.85 for No. 45
fixed orifice.
and larger DMS sizes. The above values
are for orifices having angles of approach ranging from 12 to 60°; for
greater angles of approach, the coefficient of discharge is decreased con
siderably. Adjustable orifices of the type shown in Fig. 6-7 are used on
range-top burners and many other applications where considerable flow
variation is desirable. There are a number of such designs on the
market, but in general they all operate on the principle that a variable
orifice area is obtained by turning the needle relative to the hood or vice
versa. Coefficients of discharge for adjustable orifices vary from 0.60
to 0.77 but average about 0.70.
Adequate primary air is obtainable
with this type of orifice, although
it does not give so great a pos
sibility for air injection as a fixed
orifice. Primary-air-control ori
fices are used in certain commer
cial applications where it is desir
able to control the amount of
primary injection as well as the gas
flow. This gives greater burner
flexibility, especially at high heat Fig. 6-7. An adjustable orifice.
input rates. Figure 6-8 shows two
different types of air-control orifices. The first is the angle orifice which
varies primary air by injecting the gas at different positions into the
throat of the mixing tube. The second is the pressure-reducing and
double orifice which varies the pressure of gas within the orifice, thereby
adjusting both the gas flow and the gas-injection velocity. The latter
controls the percentage of primary air entrained.
Air Injection. Gas appliances are usually designed to be very flexible
in their operation, and the possibility of using almost any type of gas
under extremes of distribution pressure should be considered. Design
pressures as low as 3 in. of water for natural gas and 2 in. of water for
manufactured gases are frequently used even though the test codes of
the AGA do not require burners to operate at these extremely low pres
sures. Since gas burners should perform satisfactorily over a wide range
of operating conditions, the air-injection capacity of a burner must be
sufficient to ensure a generous supply of primary air under the least
favorable conditions. Hence, for some operating conditions blowoff
might result because the burner may be capable of supplying more pri
mary air than is needed. An air shutter or an air-control orifice is pro
vided to regulate the percentage of primary air to secure the flame charac
teristics desired at each setting.
Angle orifice Double orifice

Fig. 6-8. Two types of air-control orifices. (Courtesy of American Gas Association.)
To design burner for adequate aeration capacity, the effects of a

a
number of variables upon air injection must be understood. It has been
determined from tests on well-designed burners that the percentage of
primary air which can be injected follows closely the empirical equation:2
(2)
where Pa primary air, % of theoretical required for complete combus

=
tion
p = gas pressure at orifice, in. water
d = specific gravity of gas
H„ = heating value of gas, Btu/cu ft at 60 F and 30 in. Hg
Am = average area of throat and mixing-tube outlet, sq in
Ap = total port area, sq in.
/= input rate, Btu/hr
T = average temperature of gas mixture in burner head, R

The first term in Eq. (2) is a function of the gas used and the pressure
at the orifice. The numerator of this term reflects a relationship of the
injection phenomenon. The heating value is included in the term
because approximately the same amount of air is required for combustion
per Btu of heat obtained regardless of the gas used. Analysis reveals
that for general design purposes a value of 0.036 may be used for this
entire term. For different gases and distribution pressures there is a
variation in the numerical value of the term, but the suggested value is a
conservative figure based on 1100 Btu/cu ft of natural gas at in.
water pressure, or 530 Btu coke-oven gas at lj^ in. water. For more
favorable distribution pressures, more than adequate primary air would
then be available.
The second term in the equation suggests that the ratio of mixing-tube
area to port area and the total heat input rate also affect the percentage
of primary air entrained. Actually, the ratio of the area at the throat of
the mixing tube to the port area is the more significant term, although the
relationship lends itself to better correlation of data if the average mixing-
tube area is used in the equation. The use of Eq. (2) is limited to burners
having throat-to-port area ratios of 0.2 to 1.0. An interesting relation
ship between mixing-tube area and port area can be obtained by rewriting
the first two terms of Eq. (2) and substituting the fixed value of 0.036 for
the first term.
2-5 X W#AmA,
PA = 0.036 X
Letting K = Am/Ap and rearranging,
Note that the left-hand side of the equation is the Btu per hour per
square inch of the port area, which is shown to be a function of the per
centage of primary air and the ratio Am/Ap. Table 6-1 gives values of
gas input rates for different values of primary air and throat-to-port area
ratio. The latter, of course, is proportional to Am/Ap and is a more
easily visualized term. As can be noted from the table, if high input
rates are desired with high-percentage primary air, it is advantageous to
design for a high throat-to-port area ratio. With low throat-to-port
area ratios, low unit burning rates are obtained and, for a given heat
release per hour, the flames must be spread out over a relatively large
area. Conversely, with larger throat-to-port area ratios, high burning
rates per square inch of port area are obtainable giving longer flames and
greater concentration of heat. Thus, the adaptation of the burner to the
appliance must be considered in choosing a throat-to-port area ratio.
Research with various mixing tubes has shown that mixers machined
smooth give better aeration than rough-surface mixers such as would be
obtained with an ordinary cored sand casting. Smooth mixing tubes
achieve their maximum air entrainment at a given input rate with throat-
to-port area ratios of 0.35 to 0.45. Rough-surface mixers, while they do
not permit quite so high aeration for the same diameter of tubes, achieve
optimum injection at throat-to-port area ratios of 0.45 to 0.60.
Table 6-1. Maximum Gas Input Rates for Various Primary-air Percentages
and Throat-to-port Area Ratios*
(Btu per hour per square inch of port area)
Throat-to-port area ratio

Primary
air, %
0.25 0.50 0.75 1.00
30 45,000 63,600 78,000 90,000

40 25,300 35,800 43,800 50,600
50 16,200 22,900 28,000 32,400
60 11,250 15,900 19,500 22,500
70 8,250 11,700 14,300 16,500
80 6,300 8,950 10,900 12,600
90 5,000 7,070 8,660 10,000
100 4,050 5,730 7,100 8,100
* "Gaseous Fuels," American Gas Association, 1948. Room temperature, ^pd/y/Hl — 0.036.
The third term in Eq. (2) shows the effect of mixture temperature on
primary-air entrainment. In many burners, it is assumed that the
air/gas mixture temperature reaches 400 F before the gas issues from the
ports to burn. . In rare cases, mixing tubes may operate at red heat, and
the gas mixture may reach 800 F. To reduce rapid heating in the mixing
tube, it may be desirable to interpose a shield between the mixing tube
and the burner to protect the former from radiation from the flames and
combustion chamber walls.
Other aspects of the mixing tube and related parts affect the primary-
air injection. The shutter area should be approximately 1.25 to 2.25
times the total port area to ensure proper admittance of air. The air
shutter should be designed in such a way that accidental blocking of the
primary-air opening is unlikely, even when the shutter is partially closed.
The bell should be made with a flare radius large enough to provide the
necessary primary-air openings in the face of the burner and to give a
smooth and gradual approach to the throat.
The distance from the gas orifice to the throat of the mixing tube is
fairly critical in most designs. From general practice it may be con
cluded that if the diameter of the orifice hood or spud is equal to or greater
than the throat diameter, the orifice should be located about 1.5 throat
diameters away. If the orifice hood or spud is smaller than the throat
diameter, best primary-air injection is obtained with the orifice located
0.5 to throat diameter ahead of the throat.

1 For orifice hoods less than
0.5 the size of the throat, the orifice-to-throat distance is not critical and
may be varied from 0.5 to 2 throat diameters with little appreciable effect
on air injection.
The mixing tube should initiate mixing of the primary air and gas and
convert most of the velocity head at the throat to static head at the
burner manifold. The length of the mixing tube has no critical effect
upon operation except that about six throat diameters of tube length are
required to complete the mixing of the air and gas. Long mixing tubes
virtually eliminate noise of extinction but are generally avoided because
of the high manufacturing cost. A slope of 0 to 3° on the diverging side
of the mixing tube gives approximately constant performance. Tests
have shown that if the slope exceeds 3J^°, giving a total angle of diver
gence of more than 7°, undue pressure losses are obtained. Hence, the
slope of the mixing tube should usually be less than 33^°. Equation (2)
holds only for burners having a mixing-tube slope of 3° or less. A
straight-sided mixing tube can be used for certain types of burners, but it
does not lend itself to uniform operation over a wide turndown range.
Table 6-2. Recommended Maximum Port Sizes*
Slotted ports, width, in.
Drilled ports,
Burner for Where flames do not Where
DMS No.
coalesce, i.e., do not flames
impinge on each other coalesce
Natural gas 30-32 732 He

Mixed gas 34-36 %*
Manufactured gas 38-40 H4 %*
Natural and manufactured gases 34-36 He
Butane or propane 32 He
All gases 34 Hi He
* AQA Bull. 10, 1940.
Burner Head. The shape and size of any burner head must be adapt
able to the combustion space available and should provide for uniform
heat distribution. The number and size of ports to be used are deter
mined from the heat input rate and a knowledge of the type of gas to be
used. The port size is governed by the tendencies for flashback and
blowoff. The two extremes are to be avoided. The smaller the port
size, the greater the tendency for the lifting of flames; the larger the port
size, the greater will be the flashback propensity. The AGA recom
mends the port sizes listed in Table 6-2 as maximum.
Many modern burners are using either ribbon-type or slotted ports

instead of the more common drilled ports. The ribbon ports have the
advantage of giving high concentration of heat release with much greater
resistance to flashback. To resist flashback, a burner port should be at
least % in. deep. Greater depths have little advantage, but with more
shallow ports there is a pronounced increase in the tendency for the
flames to strike back through the ports.
Most burner heads are made of cast iron. An increasing number are
now being made of sheet metal and stainless steel, with nichrome alloy
being used for ribbon and slotted port burners. Port spacing influences
the minimum burning rate which can be tolerated without flame extinc
tion or flashback. Also the lifting limit, spread of flame from port to port
during ignition, and the secondary-air accessibility are affected by the
port spacing. Extensive tests have shown the following data to be
satisfactory guides for laying out the burner head. For ports placed on a
straight line, they should be spaced % to %2 in- center to center; for
those on radii of 2 in. or more, K.to ^6 m- center to center; for those on
radii of about 1 in., %2 to %<i in. center to center. The maximum
distances listed are for DMS No. 36 or larger, and the shorter center
distances are for ports smaller than No. 36. It is advisable to provide
openings for secondary air around the ports if more than three or four
rows of ports are placed side by side. About 1 sq in. of secondary-air
opening per 5000 Btu/hr heat release should be provided for closed com
bustion chambers.
In designing the burner head and manifold, it is well to keep in mind
that any abrupt change in area between the outlet of the mixing tube and
the inlet to the head may result in an abnormal pressure drop. A head
cross-sectional area of about 13^ to 2 times the area of the ports to be fed
is considered satisfactory.
Optimum Aeration and Burning Rates. The percentage of primary
air for which a burner should be designed or adjusted depends upon the
burner application, the design of various burner parts, the input rating,
and the flexibility desired. Water heaters, furnaces, ovens, etc., gen
erally have ample headroom over the burner ports to permit using a long,
wavy, soft flame without getting flame impingement on the combustion-
chamber walls. For these applications about 35 to 40 per cent primary
air with natural gas and most manufactured gases is required to yield a
satisfactory flame. With liquefied petroleum gases, at least 55 per cent
aeration is recommended for the same appliances.
Range-top burners and radiant-type space heaters must maintain
short, sharp flames to avoid impingement and yield quick, complete
combustion with no carbon monoxide formation. Such applications
require higher percentages of primary air and, consequently, are some

what limited in their maximum input rates. For range-top burners 55 to
60 per cent primary air is considered reasonable aeration while radiant-
type heaters require at least 65 per cent premixed air.
A study of Fig. 6-9 shows that the maximum burning rates which can
be obtained with an atmospheric burner are influenced by the type of
gas, distribution pressure, port size, and the amount of primary air which
a gas is capable of injecting. A number of other burner design factors
also have an effect on performance, but the variables listed are the
120 r
01 I 1 I i i I I
0 10 20 30 40 50 60 70
Input rate, 1000 Btu/(sq in. port area)(hr)
Fig. 6-9. Characteristic lifting and yellow-tip limits for natural gas, manufactured gas,
and butane. (Data from AG A Bull. 10, 1940.)
primary ones which establish the burner capacity. Figure 6-9 shows
that the positions of the limit curves for gas burners vary appreciably
with the composition of the gas used. Manufactured gas, being high in
hydrogen, has a high transformation velocity and hence does not tend to
blow off easily. Butane burns more slowly and may lift at relatively low
port input rates. This heavier hydrocarbon gas also requires more
primary air to burn with clean, sharp flames than the other gases. Mixed
gases burn with characteristics between those of manufactured gases and
the hydrocarbon gases.
There is some variation in the limiting curves with port size, as shown
in Fig. 6-10, but this factor is not so important as the gas composition.
The port sizes for which the curves are plotted, Nos. 29 and 51 DMS,
represent about the largest and smallest drill sizes currently used in
burners for domestic appliances.
Line AB of Fig. 6-10 shows the change in operating characteristic to be
expected in the normal operating range for a fixed-orifice burner with a
change of distribution pressure. If a burner is adjusted to operate at
setting A and the gas pressure is later increased to B, the quantity of gas
flow through the orifice increases while the primary aeration remains
about constant. As a result, the burner operation becomes unstable and

Fig. 6-10. Effect of port size on lifting and yellow-tip curves for natural gas. Line AB
shows the performance of a fixed-orifice burner adj usted at setting A when the gas distribu
tion pressure is increased. (Data from AGA Bull. 10, 1940.)
the flames lift from the ports. The burner could have maintained normal
flame characteristics with this same pressure rise if the initial setting
had been made at point C. The upper curve represents the total air-
injection capacity of this burner with a constant gas distribution pressure
of 4 in. water. The burner input rate is varied by changing the orifice
size. From the curves, it is apparent that in this burner natural gas has
ample primary aeration capacity for practically all conditions and that
the air shutter opening must be reduced to prevent blowoff .
Figure 6-11 compares the air-injecting capacity of a manufactured gas
at different pressures with its limiting curves. This gas would encounter
little difficulty from blowoff because for most conditions it could not
inject enough primary air to obtain lifting.
Extensive studies of lifting and yellow-tip limits are summarized in a
correlation of primary aeration and maximum burner input rates in
Table 6-3. Generally, the maximum input rate for natural gas is limited
by blowoff. In comparison, the burning rates for manufactured gases
are limited by their primary-air-injection ability, and liquefied petroleum
gases encounter either blowoff or incomplete combustion at extreme heat
input rates.
120 r
00
J 80
60
40
e
a.
20
Ye Ilow -tip limit
0 10 20 30 40 50 60 70

Fig. 6-11. Comparison of normal injection with limit curves for a coke-oven gas.
The AGA points out that the input rates shown in Table 6-3 are
maximum and can be decreased about 20 per cent to advantage in order
to give increased burner flexibility during periods of emergency when the
gas pressure or composition may vary. Burning rates for high-capacity,
heavy-duty burners may be increased above those given in the table, but
burner flexibility generally must be sacrificed to permit high capacity.
Lower input rates give greater burner flexibility. As can be seen from
the preceding discussion, the choice of the primary-air and maximum
input rates is quite flexible and depends on a number of closely inter
related factors. Input rates of 10,000 Btu/(hr) (sq in.) of port area for
range-top burners and 35,000 for water heaters, etc., are common, but
such appliances need not be limited to these capacities.
Table 6-3. Maximum Gas Rate at Normal Adjustment Pressure*

(Thousands of Btu per hour per square inch of port area for No. 36 DMS ports)
Appliance with Appliance subjected to Appliance subjected to

regulator 25% pressure increase 50% pressure increase
Max
primary- Lique Lique Lique
Manu Manu Manu
air injec Nat fied Nat fied Nat fied
fac fac fac
tion, % ural petro ural petro ural petro
tured tured tured
gas leum gas leum gas leum
gas gas gas
gas gas gas
70 12 19 13 10 19 12 9.5 19 11
65 13.5 22 15 12 22 14 11 22 13
60 15.5 25 18 14 25 16 13 25 15
55 18 30 21 16 30 19 15 30 18
50 21 35 19 35 17.5 35
45 25 42 22 42 21 42
40 31 52 27 52 25 52
35 37 33 31
30 45 41 38
*
AGA Bull. 10, 1940.
Interchangeability. Most gas appliances are designed to perform

satisfactorily with all types of fuel gases distributed in the United States
and Canada, provided they are properly adjusted on the particular gas
with which they are supplied. In some cases it is necessary to vary the
entire burner element to secure proper adjustment, but with many
appliances a wide range of gases may be used by varying only the orifice
and the primary-air shutter opening.
If one gas is substituted for another without readjustment of the
burner, the input rate, primary air injected, and the flame characteristics
may all be changed. The primary-air factor,
1000 y/pd
Hr
where p = gas pressure, in. water
d — specific gravity of gas
Hv heating value of gas, Btu/cu ft
=
is an empirical relationship often used to compare the theoretical air-

injecting ability of different gases.* The primary-air factor is approxi-
*
The expression 1000 y/ pd/Hv indicates the relative injection capacity of the gases.
Generally somewhat less air is injected. Equation (2) which is also empirical in
origin, likewise may be used to compare primary aeration capacities. Following
performance more closely, it shows slightly lower percentages of primary air than
the so-called primary-air factor.
Table 6-4. Primary-air Factors for Typical Gases*
Normal dis Primary-air

Heating
Specific tribution factor,
Type of gas value,
gravity pressure, 1,000 Vpd
Btu/cu ft
in. water Hr
Coke oven 552 0.416 3.5 2.18

820 0.528 6.0 2.17
,
Natural 1116 0.632 7.0 1.88
Butane 3207 2.00 11.0 1.46
*AGA Bull. 10. 1940.
mately inversely proportional to the gas input rate. Table 6-4 compares
primary-air factors of four representative gases at their normal distribu
tion pressures. Butane, typical of liquefied petroleum gases, has less
capacity for injecting its primary air than other gases. Hence, burners
adjusted for use with natural gas may produce yellow, smoky flames when
burning butane.
0 10 20 30 40 50 60 70 80
Input rate, 1000 Btu/(sq
port area) (hr)
in.
Fig. 6-12. Effect of interchangeability on performance of burners. (Courtesy of American
Gas Association.)
Some problems with interchangeability of gases are shown

associated
graphically in Fig. 6-12. If a burner is adjusted at setting A on coke-
oven gas and if carbureted water gas is substituted at the same distribu
tion pressure, the setting shifts to B. The limit curves show that this
setting would be satisfactory unless - the distribution pressure were
increased. The pressure change would shift the setting to C and blowoff
would result. If the burner were originally adjusted at point A on

carbureted water gas, substitution of coke-oven gas would cause carbon
monoxide formation at point D.
Had the burner been adjusted at a lower initial input rate, as at point
E, the above substitutions and pressure changes would have resulted in
satisfactory flame characteristics at all conditions, as evidenced by points
F, G, and H. This bears out the fact that burner operation is much more
flexible when the design gas input rate is conservative. Interchange-
ability indexes have been developed to correlate the effect on burner
performance of wide variations in the gas composition.11
More refined aspects of the design of gas burners include the analysis of
flame height, noise of extinction, height of inner cone, burner noise level,
etc., but these analyses are beyond the scope of this book.
Example 1. Design an atmospheric gas burner with a capacity of 40,000 Btu/hr

input for a space heater. Natural gas of 1007 Btu/cu ft, 0.56 specific gravity, at a
normal distribution pressure of 7 in. water is to be used. The heater combustion
chamber is rectangular and has ample clearance above the burner head to permit soft
flames without impingement.
Solution. For this application, 35 per cent primary air with resulting soft, high
flames will be satisfactory. Table 6-3 shows that the maximum input rate which
should be used for natural gas at 35 per cent aeration is 31,000 Btu/(hr)(sq in.) port
area if the gas pressure is subject to 50 per cent fluctuation from normal. An input
rate of 25,000 will be selected for design purposes to permit greater flexibility.
40,000
Total port area = = 1.6 sq in.
25,000
Port size selected is DMS No. 30, the maximum recommended in Table 6-2. A
cast-iron head will be used with port depth of % in.
It is decided that three rows of ports will be used spaced % in. apart. Adjacent
ports will be placed % in. center to center to promote good flame travel at ignition
and yet minimize coalescence.
Burner head dimensions:
Three port diameters = 3 X 0.1285 0 .386 in.

Between two rows = 2 X 0.375 0 .750 in.
Allow % in. to edge of burner, each side 0.750 in.
Total width 1.886 in., say, l1^ in.
Space between 41 ports = 40 X % 15 .000 in.
y2 port diameter each end = 2(0.1285/2) 0. 129 in.
Allow % in. at each end 0 .750 in.
Total length 15.879 in., say, 15% in.

Figure 6-13 illustratesthe burner designed here. A cast-iron mixing tube with
a 2° slope is selected. For this rough-surface mixer, a throat-to-port area ratio of
0.55 is assumed for good flexibility and high aeration. Table 6-1 shows that, with this
ratio, the desired input rate can be met with ample primary-air injection.
Throat = 0.55 X 1.6 = 0.88 sq in. or diameter = 1.058 in., say, IKe in-
Orifice-to-throat distance = 1^6 >n- (one throat diameter)
Length of mixing tube (throat to center line of burner) = —

i-H.6 —
6% in.
(if the gas orifice is located beneath the burner end).
Fig. 6-13. Dimensions of the burner designed in Example 1.
The entrance to the burner head is arbitrarily placed 2J.£ in. above the center
line of the mixing tube. Carry the 2° slope around the elbow and up to the burner
head. Then,
Total length of mixing tube = 6% + 2\i = 9% in.
Diameter at burner-head entrance = 1^6 +9% (2 tan 2°) = lHe +0.654 = 1.717,
use 1% in.
j ^16
-fe
Average mixing tube area = = 1.55 sq in.
Depth of burner head: Each side of the burner head should have a flow area equal
to about l}-2 times the port area on that side.
y
6
1
Area on each side = 1.5 X = 1.20 sq in.
Internal width of burner head = 11!}{g — =

}i 1K6 in. (assuming in. cast walls)
20
1
Internal depth of burner = 7^7- = 0.84, use in.

}i
1/1
6
This depth arbitrarily reduced to about in. at each end of the burner head.
is
Primary-air opening = 1% X port area = 1J^ 1.6 = 2.4 sq in.

X
Gas orifice size:
= 6880C,A
yj^
q
For an assumed gas pressure and temperature of 30 in. Hg, 80 F,
40,000
1007
Area = 0.0082, or DMS No. 37 (Table A-l)
Primary-air injection:
=
\/pd 2.5 X 106 x/AnAp v . /540
Pa
y/1 X 0.56 ^ 2.5 X 105 y^l.55 X 1.6
Vl007 V40,000
= 69.5% at room temperature
This would be reduced to 55% at 400 F mixture temperature.

Comparing the primary aeration with the characteristic limit curves for natural
gas in Fig. 6-10, it appears that the air-injection capacity will be sufficient and that
the air shutter may need to be partially closed to prevent blowoff .
High-pressure Burners. Where high rates of heat release are required,

as in most industrial furnaces and process ovens, high-pressure burners
must be used. Feeding gas at high pressure through large burner ports
permits the combustion of large quantities of gas in a small space. Single
gas-burner units of this type are available in capacities of 50 million Btu
per hour or more. The same problems of flashback, blowoff, and incom
plete combustion, which influence the design of atmospheric gas burners,
must be contended with in high-pressure burners. Burner limit curves
for the higher pressure burners are similar to the type found for atmos
pheric burners (Fig. 6-2). Since industrial burners must operate at high
capacity, the range of primary air needed for good flame characteristics is
rather critical. Good gas regulators and primary- and secondary-air
controls are required. Flashback can be prevented only by maintaining
a gas velocity in the burner well in excess of the transformation velocity
of the mixture. A small percentage of carbon monoxide may be per
mitted in the flue gases from furnaces, but the heat release rate limit per
burner is often reached when incomplete combustion forms excessive
carbon monoxide.
Many designs of high-pressure gas burners are available commercially.
A few typical designs are presented here to show the principles employed
to mix gas and air and to control combustion. Many industrial gas
burners are designed as combination gas and oil burners. Some of these
units are discussed later. A single-stage gas orifice, injector, and mixer
tube, similar to the type employed in atmospheric gas burners, is satis
factory for certain applications using high-pressure gas. However, better
mixing in a shorter space and more accurate control of the air/fuel ratio
are obtained with a two-stage injector, such as that shown in Fig. 6-14.
218 FUELS A\D COMBUSTION
The two venturi elements of the injector act in series and serve to increase
the aspirating effect and the resulting turbulent mixing. To help control
blowoff under high gas-flow velocities, a flame holder, or by-pass, such as
that shown in the figure, may be employed. A flame holder operates on
the general principle that a flame can be maintained in a zone of relatively
Gas
Fig. 6-14. Burner designed for two-stage injection and employing a flame holder.
low gas velocity, whereas, it may be blown out if the gas velocity greatly
exceeds the transformation velocity. If a steady flame is maintained in
the quiescent zone of the flame holder, ready ignition of the fresh gas mix
ture in the main part of the burner is obtained even though the central
section of the flame may tend to blow off at times.
Fig. 6-15. An industrial tile port burner.
The>effect of incandescent ceramic surfaces in promoting combustion

is utilized in some types of burners.Figure 6-15 illustrates an industrial-
type burner wherein the gas and air mixture is projected into a refractory
tunnel. The ceramic surface, heated to incandescence under normal
conditions, apparently has a catalytic and radiant effect which promotes

very rapid and complete combustion even at high burning rates.
Many process furnaces and kilns employ the true surface combustion
phenomenon wherein gas premixed with more than 100 per cent primary
air is fired tangent to the incandescent refractory surfaces of the furnace.
Adjacent to the wall the remarkable surface combustion reaction occurs.
The brick is heated intensely, rapid and complete combustion is obtained,
and the flame at times may not even be visible.
The Fanmix type of burner, illustrated in Fig. 6-16, utilizes a somewhat
different mixing principle to obtain complete combustion at unusually
Fig. 6-16. Fanmix burner. (Courtesy of Coppus Engineering Corp.)
high burning rates, even with excess air as low as 5 per cent. Gas under
line pressure is admitted through the hollow shaft to ports located on one
side of the turbine blades. The reaction from the gas issuing from the
blade ports at high velocity causes the turbine and the adjacent fan to
rotate. The fan delivers the necessary combustion air at right angles to
the streams of gas discharge from the orifice. Very thorough mixing is
obtained close to the burner. An air shutter controls the quantity of air
admitted to the fan and, consequently, the air/fuel ratio used at the
flame. In units of this type it is possible to obtain excellent flame
characteristics over a wide turndown range. The flame front is so
violently turbulent that it is practically invisible.
Figure 6-17 shows a center-diffusion type of gas burner. This type of
burner is frequently referred to as a gun-type unit because of the long
straight pipe leading to the tip of the burner. The gas issues from the
Fig. 6-17.
L .
A center-diffusion type of gas burner. (Courtesy
J
of Forney Engineering Co.)
Fig. 6-18. Janitrol conversion burner showing hole port design and diffuser plate. 1.
Flame diffuser; 2. pilot; 3. burner tube; 4. air door; 5. venturi; 6. venturi clip; 7. bleed
tube; 8. gas orifice spud; 9. manifold; 10. electric gas valve; 11. supports. (Courtesy of
Surface Combustion Corp.)
burner tip in the shape of a wide-angle hollow cone. Air for combustion
enters from a wind box and passes through the controlling air louvers.
Mixing of gas and air occurs not in the burner but at the entrance to the
furnace. The air louvers initiate rotation of the incoming air and control
the quantity admitted. Such units give a short, intense, transparent
flame well distributed throughout the furnace. A new development of
the burner shown employs reversible louvers and provides even better
control of the flame pattern in the furnace. Figure 6-17 shows another
type of flame holder device in the form of a diffuser plate attached to the
burner tip. The diffuser ensures a low gas velocity in the central zone
Fig. 6-19. An industrial gas-burner installation. (Courtesy of Bryant Heater Co.)
giving steady and prompt ignition. A diffuser plate may also promote
rapid mixing of the air and fuel. This type of burner is well suited for
combination oil- and gas-burner units and has been used in installations
generating more than a million pounds of steam per hour.
Figure 6-18 shows another type of single-hole burner construction
employing a diffusion plate. This particular model is an atmospheric
conversion burner, but the principles of this nozzle and diffuser are
employed in a number of industrial burners. An industrial high-pressure
gas-burner installation showing the piping, regulators, and controls
appears in Fig. 6-19.
OIL BURNERS
Most liquid fuels are extremely difficult to burn from the liquid state.
In fact, a stream of many liquid fuels will extinguish an ignition source
and make the combustion of such liquids impossible. The combustion of
those same fuels in the vapor state is quite another matter. Hence, the
primary function of any liquid fuel burner is to produce vaporization of
the liquid before ignition, or to achieve the next thing to it — a mist of
finely atomized fuel droplets. The burner must also initiate thorough
mixing of fuel and air in the required proportions. To avoid carbon
formations, the fuel should be vaporized or dispersed in a fine mist before
the oil begins to undergo cracking from the high temperatures of the flame
and combustion chamber. No oil should impinge directly upon any
metal surface or cold area. Impingement almost always results in the
formation of carbon and soot.
Many of the more viscous liquid fuels are difficult to evaporate at
atmospheric pressure. An attempt to vaporize these fuels by subjecting
them to high temperatures often leads to thermal cracking of the heavy
hydrocarbons into lighter hydrocarbons and petroleum coke, similar to
the cracking process used in the refining of petroleum. The coke is
actually a rather pure form of carbon and forms deposits in the combus
tion chamber. To promote rapid evaporation of heavy fuel oils, burners
are designed to break the fuel up into minute droplets and project these
into the combustion chamber in the form of a fine mist. Although igni
tion of the oil occurs before all of the fuel is evaporated, it may be assumed
that fuel vapor entirely surrounds each of the independent droplets. As
the fuel vapor mixes with air and burns, the radiant heat from the flame
promotes evaporation of the remainder of the droplet and the entire
combustion process probably occurs between vapor fuel and air.7 The
finer the atomization, the more rapid the evaporation since the vapor
pressure increases as the droplet size decreases and the exposed surface
area per unit quantity of fuel increases rapidly as the droplet size is
reduced. It has been well established that fine atomization is one of the
prerequisites 'for good oil-burner operation.
All oil burners may be classified under one of the following headings
depending upon the method used to atomize the fuel:
1. Vaporizing burner
2. Rotating-cup burner
3. Mechanical or oil-pressure atomizing burner
4. Steam or high-pressure-air atomizing burner
5. Low-pressure-air atomizing burner
There are many different oil-burner designs on the market. The National
Bureau of Standards has established commercial standards describing
desirable features of oil-burner design and application,9 and the Under
writers Laboratories rates, tests, and lists oil burners. While there is no
compulsory legislation demanding that the commercial standards be
adhered to, most manufacturers follow them rather closely. The

principles of operation of each of the five classes of burners are discussed
in the following sections along with a few representative burners.
A complete oil-burner installation must make provisions for the ignition
of the fuel, the proper control of fuel and air, heating of the fuel when
viscous oils are burned, cleaning of the fuel, and maintenance of safe
operating conditions.
Vaporizing Burners. The gasoline blowtorch is a common example of
a vaporizing type of liquid fuel burner. The volatile gasoline or naphtha,
at a low pressure, is passed through a tube adjacent to the flame before
the fuel is released through an orifice. In flowing through the hot tube
Fig. 6-20. A pot-type vaporizing oil burner.
much of the fuel is vaporized so that the fluid ejected from the orifice is
largely vapor. Provisions are made for the vapor stream to entrain
primary air and burn in a cylindrical tube. Thus, this type of liquid
burner is essentially a gas burner in which the liquid is vaporized before
being sprayed into the atmosphere.
A very common type of burner for kerosene and No. 1 distillate is the
wick, or shell, type of burner. An asbestos wick raises the liquid by
capillary action. Radiant heat from the flame and nearby heated
surfaces evaporates the fuel from the upper part of the wick. Air is
admitted through holes in the surrounding walls. This type of burner is
most commonly used for cookstoves, hot-water heaters, and other
domestic applications wherein the heat release is rather low. Best
operation is obtained by using fuels with flash points not higher than
110 to 120 F and end points not over 540 F in the ASTM distillation
test.
Figure 6-20 is a sketch of another common type of vaporizing burner.

The pot type of burner illustrated is typical of the units now used almost
exclusively for small space heaters up to about 60,000 to 80,000 Btu/hr
capacity. This type of burner is well suited for burning No. 1 fuel oil or
a liquid having an end point not over 560 F. The liquid is metered to the
pot by a carburetor. As the fuel flows out over the bottom of the pot, it
is evaporated by radiant energy received from the combustion chamber
walls and the flame. The vapors rise in the pot and mix with primary
air flowing through the holes in the sides of the inner walls. When the
burner is operating normally, the fuel/ air mixture near the bottom of the
pot is too rich to support combustion. Consequently, the flame rises to a
cup Primary air

Fig. 6-21. Diagram of a horizontal rotary oil burner.
position just above the rim. At this point enough air is mixed with the
vapors to give a good burning mixture. With pot-type burners, control
of the air and fuel is difficult, and somewhat less efficient combustion
results along with larger formations of sludge and carbon than with other
types of burners. Some soot formation with this type of burner is
inevitable, and units should be taken apart and cleaned periodically.
Rotating Cup. The rotating-cup type of burner is used extensively
under steam boilers and, in a few cases, in domestic applications. Indi
vidual burners are built in capacities up to 200 gal/hr and can burn No. 5
fuel oil cold and No. 6 if the oil is heated. The rotating-cup type is the
one most commonly used in medium-size installations. Figure 6-21
illustrates the principle of this type of burner. These units are most fre
quently mounted horizontally in the furnace wall, although a few domestic
designs are built with the shaft in a vertical position. The latter are
capable of burning No. 3 or lighter fuel oils only. Oil flows through a
tube in the hollow shaft of the burner and into the cup at the furnace end.
The shaft and cup are rotated at speeds from 3600 up to 10,000 rpm by an
electric motor or an air turbine. The usual design has the motor arma
ture concentric with the shaft and an integral part of the burner, although
belt drives are also used. Atomization is obtained largely as a result of
the centrifugal force exerted on the droplets of fuel as they are projected
from the edge of the rapidly rotating cup. A centrifugal fan rotates
with the shaft to force primary air from a cone surrounding the rotating
cup. These fans are built with low capacities but large diameters to pro
vide a small, high-pressure supply of primary air. About 10 to 15 per
cent of the theoretical air is thus supplied as primary air. The angle at
which the air hits the fine oil mist may be adjusted by regulating the rela-
Fig. 6-22. Ace Uniflow belt-drive rotary burner. (Courtesy of Ace Burner Co.)
tive position of the cup and the air cone. An additional atomizing effect
is obtained as the air blasts the fuel mist. In some types of burners, fins
are placed near the end of the air cone to provide for rotation of the pri
mary air, although this is generally not considered necessary. The shape
of the flame is governed by the shape of the cup and the position of the
air nozzle. Moving the air nozzle back changes the combustion from a
long, soft flame to a short, bushy one. Secondary air is generally sup
plied by natural draft through air shutters in the furnace wall. Rotating-
cup burners are compact, efficient, and comparatively low in first cost.
Piping for the air is usually unnecessary, and power requirements are
relatively low. Figure 6-22 is a picture of a horizontal rotary burner
with most of the accessories needed for operation.
Mechanical, or Oil-pressure Atomizing Burner. The oil-pressure
atomizing burner is used for all applications from domestic to units having
a capacity of 700 gal/hr for large industrial boilers. The principle of
atomization in all units is the same and is shown }n Fig. 6-23. High-
pressure oil enters the burner tip and flows from the passage A into the
whirl chamber through a series of slots tangent to the whirl chamber.

About half of the initial oil pressure is consumed in generating rotational
energy in the liquid.10 The remainder of the pressure is utilized to force
the oil out of the whirl chamber and through the orifice at the nozzle tip.
The rapidly rotating oil continues to gyrate as it leaves the orifice, and
centrifugal force tends to throw the oil apart into very fine droplets form
ing a cone of oil mist. The angle of the cone varies from a solid stream up
to 80° depending upon the shape of the nozzle tip and the viscosity of the
oil. Nearly 70 per cent of all domestic oil burners employ the oil-pressure
atomizing principle. These units are designed to operate with No. 2 or
No. 3 fuel oil and are built in capacities up to about 10 gal/hr. The
capacity in gallons per hour of 35 SI'S viscosity oil at 100 psi pressure is
stamped on domestic nozzles.
The capacity of a nozzle varies
as the square root of the pressure.
Hence, for the usual pressure
range of 70 to 150 psi in domestic
burners, very little turndown is
possible. For the most part,
these burners operate either on or
off. By running oil pressures up
as high as 250 psi, somewhat more
flexible operation can be obtained
Fig. 6-23. Principle employed for oil-pressure
for industrial use, but even then
atomizing.
turndown is limited. A recent
nozzle design permitting recirculation of part of the oil reaching the whirl
chamber now permits turndown ratios as high as 10 to 1 for some indus
trial burners. In these units, an excess of oil is pumped to the nozzle tip
and, if it is not needed for burning in the furnace, much of it may be
recirculated in the oil system.
Any grade of fuel oil may be used with the mechanical type of atomizer,
provided the oil is heated sufficiently to permit it to flow through the
nozzle. This type of burner requires more preheat for the fuel oil than
any other type because of the close dependency of atomization on vis
cosity. Figure 6-24 illustrates in phantom an industrial-type mechanical
atomizing burner and Fig. 6-28 shows the installation of a combination
burner of a similar type in an integral furnace boiler. Combustion air
flows from the windbox through the air register and into the furnace
through the throat opening surrounding the nozzle tip. The refractory
burner throat opening radiates heat to promote rapid ignition of the oil
spray and promotes good mixing of the fuel and air. The throat should
be concentric with the burner nozzle but adjustment should be made so
that no oil impinges against its surface. The flame shape may be adjusted
by the relative position of the nozzle to the burner throat, the quantity of
air supplied, and the rotational effect provided by the air shutters. The
air flow through the burner throat may be evaluated from the relationship :
V = 4,000CS
VP
where V is the air velocity in feet per minute, p is the pressure difference
between the furnace and the windbox in inches of water, and Cq is the
orifice coefficient, which is generally about 0.6 for smooth circular orifices.
Fig. 6-24. Babcock and Wilcox oil-pressure atomizing burner. {Courtesy of Babcock and
Wilcox Co.)
A diffuser is placed at the end of the burner gun just ahead of the nozzle
to initiate turbulence and act as a flame holder in the vicinity of the
nozzle. The oil-pressure atomizing burner is very quiet and has a low
operating cost.
The principle of vortex air flow may be employed to permit high rates
of heat release with low excess air in a very compact space. An oil-
pressure atomizing nozzle sprays the fuel into the air, which enters the
combustion chamber tangentially, as shown in Fig. 6-25.
Steam or High-Pressure-air Atomizing Burner. For use in large
industrial furnaces and steam generating units, the steam or high-
pressure-air type of atomizer is commonly employed. These burners are
built in capacities up to 850 gal/hr. The general principles upon which

the many designs operate are shown in Fig. 6-26. Steam enters one part
of the burner and the oil another part. In the outside mixing type, the
oil is ejected through one set of holes and is blasted by a high-velocity jet
of steam issuing from other holes. Mixing occurs entirely outside the
burner. The result is a flat flame which provides good combustion effi
ciency at moderate capacities, provided there is little fluctuation in the
rate of fuel injection. At very high firing rates, the fuel stream becomes
so dense that good mixing of fuel and air in the furnace is difficult.
, Air to be Ignition electrodes

heated
Tangential air ports
Combustion air
Fig. 6-25. The "whirling flame" aircraft heater. (Courtesy of Surface Combustion Corp.)
The inside mixing type of burner, which is more commonly used, pro
vides high efficiency at the high firing rates, a wide turndown, and a
flexible flame shape (Fig. 6-27). In burners of this design, the steam and
oil are mixed inside the burner before the mixture is projected into the
furnace in either a flat spray or in a hollow cone. In most burners of this
type, high-pressure air may be used for atomization in place of steam.
The cost of compressing the air to the 80 to 100 psi required is consider
ably more expensive than using steam. Air atomization results in more
rapid, complete, and efficient combustion than steam, but air is used only
when a large supply of high-pressure air is required for other purposes.
From 0.7 to 5 per cent of the total steam generated in the boiler is used in
the burners for atomization at a pressure of 40 to 80 psi. On the average,
a well-operated installation uses about 1.25 per cent of the steam for
burner atomization. The oil pressure should be slightly higher than
either the steam or the air pressure. Less preheat of the oil is required
for steam atomizing nozzles.
Steam '-^—^ Steam-

Steam
OUTSIDE MIXING
Steam-.
Steam .
INSIDE MIXING
Fig. 6-26. Principles of atomization employed in steam and air atomizing nozzles.
Fig. 6-27. An industrial steam atomizing burner. (Courtesy of National Airoil Burner Co.)
The means for admitting combustion air to the steam/oil mixture is

about the same as that employed for industrial applications of the oil-
pressure atomizing burner discussed previously. Usually about 1 sq in.
of air opening is used per pound of fuel per hour. The air should not be
projected against the oil supply so violently that it tears the flame apart
and throws raw oil against the walls, since such action would surely
result in carbon deposits. Some rotation of the air supply is desirable.
This type of burner needs no diffuser plate. Variable flame shapes may
be obtained by interchanging nozzle tips.
Low -pressure-air Atomizing Burner. A type of burner which operates
on the same general principle as those previously described but which
requires air at only 0.5 to 2 psi is termed a low-pressure atomizing burner.
Fig. 6-28. An integral furnace boiler for gas, oil, and pulverized coal firing. (Courtesy of
Babcock and Wilcox Co.)
These units are probably the simplest and most versatile of the atomizing
types of burners. They are built in domestic sizes for burning No. 2 and
No. 3 fuel oil. If the oil is heated, grades up to No. 5 may be burned
industrially although the applications are generally restricted to the use of
lighter fuels. Industrial designs capable of burning 500 gal/hr of oil are
made. Again, the many designs manufactured include both inside and
outside mixing units, but the inside mixers are the more popular because
they give better flame control. The outside mixing units are simpler,
give less trouble from clogging, and are more easily cleaned than the
others.
Combination Burners. It is often desirable to install firing equipment
capable of handling gas, oil, or pulverized coal, either separately or in any
combination. Many plants wish to take the economic advantage of
seasonal and annual fuel price fluctuations. Also the added reliability of
combination burners justifies their increased first cost.
Fig. 6-29. A combination gas and oil burner. Oil is atomized by the mechanical nozzle in
burner, while the gas issues from the circular ring. (Courtesy of Babcock
the center of the
and Wilcox Co.)
Figure 6-29 illustrates a burner of the type shown in Fig. 6-24 equipped
with a gas ring to permit combustion of gas and oil simultaneously. The
burners appearing in Fig. 6-28, which are for pulverized coal firing as
shown, may also be converted to combination units by adding mechanical
atomizing oil guns and gas rings. The cross section of another type of
combination burner depicting the combustion of three fuels at once is
presented in Fig. 6-30.
Burner Auxiliaries. An oil burner must be supplied with clean oil at a
uniform rate and pressure if it is to function properly. In addition, the
fuel oil should be of the proper viscosity. In vaporizing burners, com
paratively high-grade volatile fuels are used, and the control problem is
Fig. 6-30. A combination oil, gas, and pulverized coal burner. (Courtesy of Riley Stoker
Corp.)
Fig. 6-31. An oil-heating and pumping unit, equipped for both steam engine and electric
motor drive. Units of this type frequently contain all the fuel strainers and filters, pressure
regulators, by-pass piping, etc., required as auxiliary equipment for oil-burner installations.
232
relatively easy. For industrial and commercial burners utilizing heavy,

cheap fuel oils, several pieces of auxiliary apparatus are essential if effi
cient, trouble-free performance is to be achieved. Frequently a compact
package installation of burner auxiliaries, of the type shown in Fig. 6.-31,
is located somewhere near the burner to fulfill these requirements.
Oil generally is supplied through motor-driven gear or vane-type pumps,

although reciprocating pumps driven from small steam engines are some-
2,000
1,900
1,800
1,700
1,600
0g 1,500
§1,400
<y
"1.300
5 1.200
lltoo
1 1,000
±
I
900
800
>, 700
§ 600
|j 500
400
300
200
100
0
80 100 120 140 160 180200 220 240 260 280300 320 340 360
Temperature, Degrees F.
Fia. 6-32. Effect of temperature on viscosity for five heavy fuel oils. (Steiner, "Oil
Burners," 2d ed., McGraw-Hill Book Company, Inc., New York, 1950.)
times installed to meet power failure emergencies. By-pass relief valves

and pressure-regulating valves are included in the piping layout to help
maintain steady, uniform flow to the burner.
Poor atomization and performance are obtained when heavy fuel oils
are burned cold. Figure 6-32 shows that the viscosity of even the
heaviest oils drops quickly as the oil is heated. The lowest oil preheat
temperature which will permit good operation is considered the best,
but preheat temperatures up to 250 F are sometimes required. Most
heaters are of the steam-heated, tube type, although electric heaters may
be employed in smaller installations. Oil heaters are placed between the
oil storage tanks and the burner, so that the entire storage tank need
not be heated. Thermostats to control the preheat temperature, by-pass

valves, drain valves, wire mesh strainers, thermometers, pressure gages,
etc., are also provided to permit easy and accurate control.
Ignition for oil burners is achieved with an electric arc or with a gas
torch. Most domestic and commercial burners are equipped with a
transformer which supplies a secondary voltage of 5000 to 15,000 volts.
The steady arcing of this high voltage across the electrodes provides
ignition of the oil spray. Many high-duty burners and all rotary-cup
burners require a gas flame to ignite the oil. Both electric and gas
igniters may be made to operate either manually or automatically.
Safety devices to shut off the oil supply automatically in case the flame
is extinguished are often included in the installation. Photoelectric cells
aimed at the flame, or thermally responsive elements, are commonly used
as safety control instruments.
Gas burners also use a number of the auxiliaries listed for oil burners.
Problems
1. What is the difference between atmospheric and high-pressure gas burners?
2. Draw typical burner operating characteristic curves for burners operating on
hydrogen, carbon monoxide, methane, and butane.
3. Discuss why the limit lines shift position for the various fuels in the preceding
problem.
4. Determine the orifice size required to furnish 20,000 Btu/hr if gas No. 3 of
Table 1-9 is supplied at 7 in. water gage pressure.
6. What orifice size should be used if 100,000 Btu/hr are desired when gas No. 22
of Table 1-9 is furnished at 3 in. water?
6. What percentage of primary air could be injected if 40,000 Btu/hr of gas No. 2
of Table 1-9 is supplied at 5 in. water pressure to the burner in Fig. 6-13 when the
temperature of the gas in the burner head is (a) 500 F and (6) 100 F?
7. Show, on a burner-operating characteristic graph for butane, the proper setting
for primary aeration if the gas distribution pressure might increase or decrease by 30%.
8. Compare primary-air factors for gases Nos. 1, 5, 6, 9, 12, 17, 20, and 24 of
Table 1-9 if these fuels are supplied at their normal distribution pressures.
9. Design an atmospheric gas burner for a domestic furnace having a heat input
rate of 120,000 Btu/hr if gas No. 1 of Table 1-9 is to be used.
10. Design a range-top burner of 10,000 Btu/hr capacityfor gas No. 17of Table 1-9.
The burner should also be able to burn gas No. 2 satisfactorily with a change of the
air shutter only.
11. Discuss which general principles of combustion presented in Chap. 4 are applied
to the design and proper operation of gas and oil burners.
12. Why are flame holders required for certain burners?
13. List the range of usual application and the types of fuels used with the different
types of oil burners.
14. Discuss the advantages and disadvantages of steam atomization vs. mechanical
atomization for large industrial oil burners.
15. Discuss the merits of various types of ignition systems, safety devices, automatic
controls, and auxiliaries for industrial and domestic oil and gas burners.
References
1. ShnidMan, L., "Gaseous Fuels," American Gas Association, 1948.

2. Research in Fundamentals of Atmospheric Gas Burner Design, AGA Bull. 10,
1940.
3. Fundamentals of Design of Atmospheric Gas Burner Ports, AGA Bull. 13, 1942.
4. Primary Air Control Devices for Atmospheric Gas Burners, AGA Bull. 22, 1944.
5. Primary Air Injection Characteristics of Atmospheric Gas Burners, AGA Bull. 37,
1945.
6. Fundamental Data for Design of Totally Aerated Atmospheric Gas Burners,
AGA Bull. 38, 1946.
7. Heiple, H. R., and W. A. Sullivan, Mechanisms of Combustion and Their
Relation to Oil-burner Design, Trans. ASME, vol. 70, p. 343, 1948.
8. National Bureau of Standards, "Automatic Mechanical Draft Oil Burners
Designed for Domestic Installations," 2d ed., CS-75-42.
9. National Bureau of Standards, "Flue-connected Oil-burning Space Heaters
Equipped with Vaporizing Pot-type Burners," CS-101-43.
10. Steiner, K., "Oil Burners," 2d ed., McGraw-Hill Book Company, Inc., New
York, 1950.
11. Weaver, E. R., Formulas and Graphs for Representing the Interchangeability of
Fuel Gases, J. Research Nat. Bur. Standards, /fP2193, 1951.
CHAPTER 7
COAL-BURNING EQUIPMENT
Coal retains its position in this country as the king of fuels not only
because it represents nearly one-half of the total fuel burned annually,
but also because the known reserves of coal far exceed those of petroleum
and natural gas. Unfortunately, even though coal is the most used and
plentiful fuel, it is also the least desirable because of the difficulty of
handling and the necessity for removal of the ash. Also it takes more
elaborate equipment and care to burn coal completely and smokelessly
than petroleum liquid or gaseous fuels. However, because of the almost
limitless supply, it has in the past enjoyed a certain economic advantage
over other fuels and thus has been widely used. The economic aspects of
the fuel industry have shifted considerably during the past 10 years, but
in most cases coal still has the edge for such customers as industrials and
electric utilities who can purchase it in large quantities.
In recent years, coal has been the fuel used to develop 75 per cent of the
power generated by the electric utilities in the United States, excluding
hydroelectric stations, at an average fuel rate of about 1.18 lb coal per
kilowatthour. Coal has been utilized almost as widely for the generation
of industrial power. Thus, while oil and gas are also widely used for
power generation in steam plants, coal is the fuel most generally associ
ated with such installations. Therefore, a short discussion of the equip
ment required for modern steam generation is presented in this chapter
for the benefit of those not already familiar with the subject.
COMPONENTS OF A STEAM GENERATOR
A modern large-capacity steam generator consists of many fundamental

units, each of which serves a specific function in over-all successful opera
tion. Figure 7-1 is an example of a pulverized-coal-fired, waterwall boiler
with economizer and air preheater. The coal for combustion enters the
pulverizer and is ground to a fine dust. Primary air carries the powdered
coal to the burners where the mixture is blown into the furnace and
ignited.
Air for combustion enters the system through a forced-draft fan. The
air then flows through a large heat exchanger, called' an air preheater,
which removes heat from the hot flue gases and warms the air before it
236
COAL-BURNING EQUIPMENT 237
enters the furnace. In the air preheater, air passes over one side of the
metal surfaces while the flue gases flow over the opposite side to prevent
mixing of the two gases. The air preheater must be of sufficient size to
heat the combustion air to the temperature desired. The air for com
bustion then separates; the primary air goes to the pulverizers to pick
up the coal dust and carry it to the burners while the secondary air goes
directly to the burners. The fuel/air mixture ignites and burns in the
furnace, which provides the volume necessary for the complete combus
tion of the gases. The hot combustion products then flow across the
many boiler tubes which provide the surface required to transfer much of
the heat in the gases to the water and steam inside the tubes. Baffles
direct the flow of hot gases across several banks of tubes before permitting
them to-leave the boiler. The flue gases may still be at a temperature of
500 to 600 F at this point. Hence, they pass through another heat
exchanger, called an economizer, where the energy derived from further
cooling of the gases is used to heat the boiler feedwater before admitting
it to the boiler. The flue gases, which still contain some available heat,
pass through the air preheater where the combustion products are cooled
to the final stack temperature. They are then put through a dust col
lector to remove most of the fly ash. The flue gases finally leave the
unit through a duct called the breeching and are dispersed high into the
atmosphere over the plant by a stack.
The resistance to gas flow through the combustion equipment and over
all the heat-transfer surfaces may be too great for natural draft to provide
the large quantities of air and gas flow required for high-capacity opera
tion. For that reason forced-draft fans are used to push air into the
system, and induced-draft fans are installed at the exit to suck the com
bustion products out at the rate desired. The draft created by the
chimney effect of a tall stack may also be utilized. The variation in draft
throughout a typical large steam-generating system appears in Fig. 7-2.
Former practice has been to regulate the fans to provide a draft in the
furnace about 0.05 in. water pressure below atmospheric. Thus, any
small leakage which might occur through the furnace walls would be air
leakage into, the combustion space. Some recent installations are using
tightly sealed walls to prevent any gas leakage ; the combustion chamber
may then be pressurized to 20 in. water gage pressure or higher. With
this arrangement, larger forced-draft fans can be employed, and the
induced-draft fans can be eliminated.
Oil or gas burners, discussed in Chap. 6, or one of the several types of
combustion equipment for burning coal discussed in the remainder of this
chapter, may be installed in place of the pulverized-coal-firing equipment
in the steam generator of Fig. 7-1. That the fuel-burning equipment
is,
is just one of the basic components and any of

of a steam generator,
several types may be used effectively. Likewise, there are several good
designs of air preheaters, economizers, boilers, fans, etc., but a discussion
of the merits of the various types of construction of these components is
beyond the scope of this book.
Since the furnace, which encloses the combustion volume, is definitely
related to the problems associated with burning, it will be dealt with more
12
Slack
-Economize!
- + 10 " /
> : > ii ! 1) 11 1
1
-10
Fig. 7-2. Draft variation in a large steam-generating unit at rated capacity. {Skrotski
and Vopat, "Applied Energy Conversion," McGraw-Hill Book Company, Inc., New York,
1945.)
in detail herethan most of the other steam generator components. The

furnace should be designed to promote complete combustion of the fuel
fired in as small a space as practicable, to permit adequate transfer of
heat to the steam boiler, and to operate continuously over long periode
with a minimum of time lost for furnace maintenance. One method of
furnace construction employs firebrick refractory to line the combustion
chamber walls. High rates of heat release and the use of preheated air
for combustion both result in high flame temperatures which in turn con
tribute to undue slag, or ash, deposits, high localized temperatures, and
melting or softening of the refractory. Certain deposits of molten slag

cause a chemical decomposition of the refractory at temperatures below
normal furnace-operating temperatures and lead to excessive erosion and
furnace maintenance. Some refractory-type furnaces employ air-cooled
walls by passing the air used for combustion through holes parallel to and
near the inner surface. This air is thus preheated for combustion by the
Fig. 7-3. Example of a waterwall furnace construction. (Courtesy of Babcock and. Wilcox
Co.)
time it reaches the interior of the furnace. Air-cooled walls reduce

furnace maintenance and permit higher firing rates.
Most present high-capacity units are constructed with waterwalls. A
section of a waterwall showing one type of design appears in Fig. 7-3.
The water and steam inside the steel boiler tubes which line the furnace
wall keep the refractory temperatures low by absorbing much of the local
heat. Thus, the boiler heat-transfer area is increased and the furnace
wall temperatures are held low, both of which permit greater capacity in
less space. Waterwalls may be used on any or all furnace surfaces.
Figure 7-1 shows a furnace with complete waterwall construction. In
modern installations more than 50 per cent of all the heat is transmitted
to the water through the waterwall, much of it by direct radiation from
the flame to the relatively cool wall.
COMBUSTION OF SOLID FUELS

Solid fuels are oxidized either in suspension in the air or on a grate. In
each of the several methods for burning coal on grates, fresh fuel is fed to
the burning coal either from over the fuel bed or from beneath it. Sus
pension firing of coal is accomplished by blowing fine fuel particles into
the furnace and promoting rapid oxidation while the particles are sus
pended in and mixed with air.
(VM+CO) +
02+C02
Secondary air
> > >
lover fire)
Green -coal Distillation

zone
Reduction -2 CO
zone
Oxidation
zone
c+o?-+co.
Ash zone
Grate
Primary air —
Fig. 7-4. Diagrammatic cross section showing zones of reaction in an overfeed fuel bed.
The successful combustion of coal, and other solid fuels, involves four
stages; (1) distillation of the volatile matter from the coal, (2) combustion
of the fixed carbon to carbon monoxide and carbon dioxide, (3) comple
tion of combustion by the oxidation of the gases formed, (4) satisfactory
disposal of the ash. The sequence and rapidity with which the fuel
enters and leaves each of these phases vary with different firing methods.
The principles of combustion and some advantages and disadvantages of
the various methods for burning solid fuels are discussed in the sections
which follow.
Overfeed Firing. In overfeed firing, raw fuel is admitted to the fuel
bed from above the fire. If the fuel is fed in small quantities at frequent
intervals, the appearance of an overfeed fuel bed becomes similar to the
cross-section diagram of Fig. 7-4, which shows also the principal phases of
combustion existing with bituminous coal.
The freshly fired coal, often termed green coal, implying that it is not
yet conditioned, or "ripe," for complete combustion, is dropped onto the

fuel bed. As the green coal is heated from the intense heat below, the
moisture is vaporized and the volatile matter emanates from the coal
particles. The top area of an overfeed fuel bed consequently is termed
the distillation zone.
After distillation, the remaining coal consists only of fixed carbon and
ash. In other words, coke is formed in the process of burning bituminous
coal, and the coke settles to form the lower zones of the fuel bed. The
layer of ash which settles to the bottom as the coke is consumed protects
the cast-iron grate from the high temperature of the combustion zone.
Air for combustion enters as primary air (or combustion air) below the
fuel bed and secondary air (or overfire air) from above. The primary air
passes upward through the grate and the ash zone. The porous layer of
ash over the grate serves to help control the flow of primary air into the
combustion zones.
As the air flows through the bed of incandescent coke, part of the
carbon is oxidized. According to present combustion theories, it is likely
that the carbon burns to carbon monoxide as the primary product and
that the excess of oxygen in the vicinity of the lower part of the fuel bed
quickly oxidizes the newly formed carbon monoxide to carbon dioxide.
Thus, the lower combustion area is termed the oxidation zone. Recent
experimental evidence1 shows that the initial oxidation takes place in a
thickness of fuel bed equivalent to about one to three average diameters
of coal particle.
When the oxygen becomes depleted from the hot gases arising through
the fuel bed, the coke begins to reduce a part of the carbon dioxide back
to carbon monoxide according to the equation, C + CO2 — » 2CO. In
other words, a reduction zone exists in the upper portion of a coke bed
several particle diameters in thickness. The combustion gases, consisting
of carbon monoxide, carbon dioxide, and nitrogen, then pass through the
distillation zone and mix with the volatile matter and moisture being
vaporized from the fresh coal. Figure 7-5 illustrates the trend in gas
composition at various points in an overfeed fuel bed, as concluded from
early fundamental combustion experiments2 and substantiated by more
recent investigations.3
The volatile matter and carbon monoxide rising from the fuel bed
represent potentially about 60 per cent of the original heating value of
the coal fired. Hence, it is essential that these combustion gases be
properly mixed with oxygen and ignited in order to achieve good com
bustion efficiency as well as to avoid excessive smoke. Turbulent
secondary air over the fuel bed is needed to complete the burning. Open
spaces, or blowholes, even in a thick coke bed may admit great quantities
of unused primary air through the fuel bed. Similarly, oxygen may pass
around the edges of the coke bed without reacting with the hot coke.
Under such conditions, less secondary air is required, since there is still
oxygen available in the combustion gases above the fuel bed. It is still
necessary to provide turbulence for these gases, however, to prevent
stratification, or incomplete mixing. The volatile matter, containing
much heat, is composed of tarry hydrocarbons and carbon particles which
readily form a black soot if burned incompletely. Hence, a dirty black
smoke from a fire represents a heat loss from incomplete combustion as
well as a nuisance to the neighborhood.
Ash Oxidation Reduction

zone zone zone
24 L
!
«-20
D
V
1
— f
E
o
^*
i
<5 8
o /
/
a. 4 1 #
//
\J s
8 10
Distance above grate, in.

Fig. 7-5. Trend in gas composition for combustion gases at various locations in an overfeed
fuel bed. {After Haslam and Russell, "Fuels and Their Combustion," McGraw-Hill Book
Company, Inc., New York, 1925.)
The combustion rate in the coke bed is limited only by the speed with
which oxygen is diffused to the surface of the hot carbon. Tests show
that burning rates are proportional to primary-air flow. The latter is
a function of the pressure difference across the fuel bed and may be
increased by raising the pressure beneath the fuel bed with a forced-draft
fan or by opening the damper on the stack draft and increasing the draft
above the fire. Since oxidation occurs in only a thin bottom zone of the
coke bed, the fuel bed need be only thick enough to prevent blowholes in
the fire and provide the draft loss across the bed which may be desired to
help control the burning rate.
The sections which follow discuss present methods employed in over
feed firing practice.
Hand Firing. The oldest means of feeding fuel to a fire is by hand.
This method is still used extensively for small boilers and for domestic
heaters of not more than 1 2 sq f t grate area.
Proper hand-firing techniques depend upon the type of fuel fired.

Anthracite coal should be fired by spreading it evenly over a thin glowing
fuel bed, since there is little volatile matter to burn and an intense heat is
required for ignition. Noncaking bituminous coals may be fired success
fully by placing the fresh coal in strips on the fuel bed between alternate
strips of incandescent coke. In this way the volatile matter comes into
close proximity with the hot coke and has a good chance of igniting. If
high-volatile bituminous coal is spread over a hot fuel bed, such as is good
V Ashpit J_
Fig. 7-6. Caking method for burning bituminous coal in a domestic warm-air furnace.
practice for anthracite, the volatile matter is simply expelled and passes
out the stack as thick, unburned smoke because the newly fired coal
covers up the coke, which is the only ignition source.
Heavily caking bituminous coals are burned by the caking method. A
diagrammatic sketch depicting the use of this method for firing a domestic
furnace appears in Fig. 7-6. The hot coke bed is pushed to one side or to
the back of the grate before a fresh charge of high-volatile, caking coal is
fired to the other side or front of the firebox. It is expedient also to throw
a little slack over the green coal. As the heat from the still hot coke bed
slowly penetrates the mass of freshly fired coal, the entire mass cakes into
a dense agglomeration. The slack fuses into a solid cap. As the volatile
matter is released from the green coal, it is forced to flow over the bed of
hot coke. Thus, a high-temperature ignition source is available to burn

most of the volatile matter and yield a comparatively clean flue gas.
In hand-fired furnaces, the ash is disposed of by shaking or dumping the
grates occasionally. To reduce clinker formation, the fuel bed should be
disturbed as little as possible to avoid mixing the ash with the hot coke.
Good firing techniques go far toward producing satisfactory combustion
on hand-fired grates. Nevertheless, most hand-fired furnaces are oper
ated with excessive smoke, high excess air, and high carbon losses in the
ashpit.
Traveling-grate and Chain-grate Stokers. A very popular type of
mechanical firing device for many years has been the traveling-grate, or
Gases to main
furnace volume
Over -fire
air
Rear firing
arch ^Hopper
Chain grate
Fig. 7-7. Traveling-grate stoker with rear firing arch.
chain-grate, stoker. Coal is fed from a hopper onto a grate moving

through the combustion chamber (Fig. 7-7). The grate is composed of a
number of cast-iron links or bars, so interlocked that they form a grate
surface. The grate turns over sprockets at each end like a conveyor belt.
In the terms of the industry, a chain-grate stoker differs from a traveling-
grate chiefly in the manner of construction of the grate. The traveling-
grate type has more closely interlocking bars and is more adaptable to the
use of anthracite fines. Stokers of this type can be built in sizes up to
about 25 ft wide by 40 ft long. Firing rates of 15 to 30 lb of coal per
square foot of grate per hour are common. The moving grate is driven
by a motor or engine through a gearbox to permit varying the rate of
travel of the grate.
Air for combustion enters from a plenum chamber beneath the grate.
Grates for forced-draft units are constructed with about 5 to 10 per cent
of the total area in holes to permit adequate air flow without permitting a
high loss of fine coal. For natural draft, nearly 20 per cent of the grate
area may be needed for air flow. The design of the grate varies also for
use with different types of fuels. The air space beneath it is divided into
a number of compartments or zones to enable the air flow to be controlled
at different positions on the grate, as illustrated in Fig. 7-8. The stokers
are built with interlocking side plates to reduce air leakage around the
£20 1 I I II II
GAS ANALYSIS OVER FUEL BED
PRESSURE UNDER THE GRATE

Fig. 7-8. Variation of primary-air pressure and analysis of overfire gases at the points
shown for a traveling-grate stoker burning bituminous coal. The furnace design includes
a front firing arch. (Courtesy of Combustion Engineering-Superheater, Inc.)
sides to a minimum. The ashpits of most modern traveling-grate stokers

are sealed to prevent large amounts of excess air from flowing through the
ashpit, around the end of the grate, and into the furnace.
Traveling-grate and chain-grate stokers are well adapted for burning
anthracite, semianthracite, noncaking bituminous, subbituminous, lig
nite, and coke breeze.* In other words, almost any fuel but a caking
*
Coke breeze consists of the fines remaining after the more valuable coke has been
collected in the coking process.
bituminous coal can be handled. Caking coals form lumps of coke which
resist the flow of air. Since there is no means for agitating the fuel bed,
these lumps do not burn before they come to the end of the grate and are
dumped over into the refuse pit. Hence, the carbon loss is high for such
fuels. Also, big blowholes may form in the fuel bed with caking coals
which give higher excess air than necessary. Sizing of coal is important
for well-controlled combustion. A large percentage of fines leads to high
carbon loss. Best results are obtained with %-in. nut coal and smaller
with the fines removed. Occasionally it is necessary to temper a bitumi
nous coal by spraying it with water about 12 hr before firing to bring the
moisture content to 12 to 14 per cent. This helps keep the carbon loss
down and regulates the burning rate.
Since the fuel is dropped onto the grate from the hopper, the traveling
grate constitutes an overfeed method of firing. Ignition of the fresh fuel
is accomplished by a combination of three factors, the first of which is
predominant: (1) radiation from an incandescent firebrick arch, (2) hot
gases blowing over the green coal, (3) spray of fine glowing coke particles
which settle on the new fuel.
Figure 7-7 shows a rear-arch furnace construction above the stoker.
The arch radiates heat to the coal on the grate and serves as a powerful
factor in promoting rapid ignition and complete combustion. This arch
may be located at the front of the grate, or a combination of front and
rear arches may be employed. The rear arch is preferable with most fuels
because it helps to eliminate stratification of the combustion gases and
lowers carbon loss and fly ash, although the front arch is best for high-
volatile bituminous coal.
In addition to radiating heat to the fuel bed, the shape and location of
the arch do much to control velocities and direction of gas flow. A rear-
arch furnace, for example, has comparatively high gas velocities near the
rear of the furnace. These tend to pick fine incandescent coke particles
from the rear of the fuel bed, throw them into the combustion space, and
deposit them on the fresh coal near the front of the grate. Likewise, the
hot combustion gases are directed toward the green coal.
Near the front of the furnace, the gases are primarily freshly expelled
volatile matter; near the rear, where the fuel bed is thin, they are mostly
excess air. High turbulence and good mixing of these gases are vital.
This may be partially accomplished with good arch design, but the use of
overfire steam and air jets located in the furnace arch is usually required
to get smoke-free, complete, efficient combustion.
The ash from the burning coal settles to the bottom of the fuel bed and
protects the cast grates. A high-ash fuel is desirable because it leaves a
thicker layer for grate protection. Little trouble is encountered from
clinkering with any ash because the fuel bed is not agitated and the ash
is cooled below the fusion temperature by the primary air.
Inclined -grate Stoker. An older type of stoker and grate arrangement
still used extensively for burning refuse fuels is illustrated in Fig. 7-9,
which shows an overfeed stoker which pushes fresh fuel to a horizontal
platform where it ignites from the heat of the burning fuel below. When
the next charge of fuel enters, the ignited charge on the platform is
pushed onto the inclined grate and works its way down the grate to the
bottom. The refuse is dumped at the bottom. Lugs help hold the fuel
Refuse to pit
Fig. 7-9. Inclined-grate stoker.
to the grate to prevent dumping the fire into the refuse pit before the fuel
is well burned.
The inclined grate is limited to small units and is used principally to
burn such trash fuels as wood bark, bagasse, lignite, and wood pulp.
The fuel bed is disturbed frequently as fuel slides down the grate. Thus,
the lumps of the refuse fuels are broken up and burned. Such stokers
generally operate with rather high excess air and considerable fuel loss in
the ashpit. This is really no great disadvantage when the fuel costs are
negligible.
Spreader Stokers. The spreader stoker provides a means for feeding
fuel by the overfire method in such a way that much of the fuel is oxidized
while suspended in the air and the remainder is burned on a nonagitated
fuel bed. A diagrammatic cross section of such a unit appears in Fig.
7-10. The spreader stoker consists of a hopper, feeder, distributor, and
horizontal grate. Coal from the hopper is metered by the feeder to the
distributor at the desired steady rate. In most designs, the distributors
consist of paddle wheels with blades which alternately angle to the right
and the left. Thus, the coal is thrown toward the rear of the furnace and
spread evenly from side to side. The larger particles tend to be thrown
to the rear while the fines drop toward the front of the grate. A few
designs distribute the coal with pneumatic feeding nozzles instead of
rotors.
Ignition of the coal particles occurs almost immediately from the
intense radiation of the fuel bed and from the high temperature of the
Fig. 7-10. Diagram of a spreader stoker and an overfire steam-air jet.
combustion gases. Volatile matter is released, ignited, and burned

almost concurrently. The fixed carbon is ignited and partially burned
before the hot coke particles settle to the fuel bed (Fig. 7-11). Most of
the coal is burned within a minute from the time it enters the furnace.
The grates are made in several designs and may be of the stationary,
power dumping, or traveling type. The space beneath the grate acts as
a refuse pit. The coke zone on the fuel bed is very thin, and after the
carbon is burned completely, the ash which remains may build up to a
depth of about 4 in. before it is dumped. The air spaces in the grate
compose only about 5 per cent of the grate area. In this way the carbon
losses in the refuse are kept low, and the resistance to primary-air flow
through the grate is increased. Hence, even though the layer of ash on
the fuel bed may vary considerably from time to time, the air flow is con
trolled quite accurately because the grate resistance is the dominant
factor in flow. The ash layer protects the grate enough so that preheating
of primary air to 350 F is permissible without undue grate maintenance.
Secondary air may be added over the fire through openings above and
below the fuel spreader. In many cases, better efficiency is obtained by
forcing most of the air through the fuel bed. Sufficient air will pass
through holes in the fire to supply the necessary overfire air for burning
2» in the combustion space. In either
"o case, turbulence in the overfire gases
is essential. Steam or air jets, lo
cated about 3 ft above the grate,
may furnish 5 to 15 per cent overfire
air and provide the turbulence so
necessary for quick, complete com
bustion with low excess air. A
sketch of a well-designed steam-air
jet is included in Fig. 7-10.
The spreader stoker was first tried
in England in 1822 but has had a
number of design weaknesses until
the past 10 to 15 years. It is the
most popular new installation today
for coal-fired boilers from about the
Combustion rate, lb/(sq ft)(hr)
Fig. 7-11. Portion of coal burned in sus 150 boiler horsepower size up to
pension in a spreader-type stoker. Ohio 200,000 lb steam per hour units.
coal was high-volatile, caking; Indiana,
The capacity of a spreader unit may
medium-volatile, free-burning; Pocahon
tas, low-volatile, free-burning. (Courtesy be increased by adding more feeder
of Combustion Engineering-Superheater,
units in parallel and making the
Inc.)
grate and furnace wider. This gives
the advantage of permitting good part-load economy, since one or more of
the feeder sections may be out of service at part load or for cleaning.
Any grade of coal from semianthracite to lignite can be handled suc
cessfully. Size of coal is one of the important considerations in achieving
good combustion. If all particles of fuel are about the same size, they
tend to pile up at one point on the grate. The best performance is
obtained when the coal ranges in size from % in. to slack, with not more
than 50 per cent slack. Large lumps fail to burn completely and con
tribute to high carbon loss in the ashpit, while excessive fines may con
tribute to carbon loss up the stack. Caking tendencies have little effect
on performance because most of the coal tars are burned in suspension.
If high-moisture caking coals are burned, the moisture prevents rapid
ignition and some caking may exist on the grates; however, high-volatile
coals ignite quickly. Since there is no agitation of the fuel bed, clinker
formation is rare even with high-ash, low-ash-fusion temperature coal.
The spreader stoker of modern design is efficient, operates with low
maintenance costs, and has high capacity and great fuel flexibility. It is
relatively simple and low in first cost. The greatest disadvantage of such
units is that a part of the ash and fine carbon particles flow through the
furnace and out the stack as fly ash. This condition is especially bad
under high firing loads. Hence, a fly-ash separating system is usually
required to meet atmospheric pollution ordinances. Carbon losses nor
mally run between 2 and 12 per cent of the fly ash. Frequently, the fly
Fig. 7-12. Cross section of the fuel bed of a single-retort underfeed stoker.
ash collected in the last pass of the boiler is reinjected into the furnace
just above the fuel bed. The carbon thus reclaimed may increase over-all
furnace efficiencies as much as 3 to 5 per cent.
Underfeed Firing. A sketch of an underfeed fuel bed appears in Fig.
7-12. Fresh coal is forced into the fire from below by a power ram or
screw. The green coal beneath the combustion zone is heated by the
intense heat from the burning fuel bed. The volatile products are
expelled as the coal is pushed into the active burning zone. In rising
through the bed of incandescent coke, the volatile products are heated to
a sufficiently high temperature to ensure ignition. Since the flow of fuel
is upward, ignition occurs from above. The need for brickwork ignition
arches is thus eliminated. In commercial stokers, the grate slopes
toward the sides or back and the ash, which is above the bed of coke, is
worked to the edge and removed.
Air for combustion enters the fuel bed through openings, called tuyeres,
in the cast-iron supports of the fuel bed. Much of the volatile matter
mixes with primary air and burns on the way through the coke bed. The
lower part of the coke bed serves as an oxidation zone. In the upper
portion, reduction of the newly formed carbon dioxide occurs just as in
overfeed firing. Secondary air admitted over the fire, or air which has
passed through blowholes, completes the burning of combustible gases
rising above the fuel bed. With a little turbulence of the overfire gases,
it is possible to achieve complete combustion and to burn high-volatile
coals almost smokelessly.
In principle the continuous agitation of the fuel bed makes it possible to
burn caking coals because the agglomerations of coke are somewhat
broken up. The same factor of agitation makes it difficult to prevent
bad clinker formation, however, unless a coal having a high-ash-fusion
temperature is used. Fuels with a high sulfur content or a high iron
oxide content in the ash have an especially bad tendency to form clinkers.
Underfeed Stokers. Commercial underfeed stokers employ either a
power ram or a screw to force coal into the fuel bed from below. Air is
supplied from a forced-draft fan. With the usual thick fuel bed carried,
about 1.0 in. water pressure per 10 lb of coal per square foot per hour is
required to supply combustion air. The pressure of the underfire air
may be controlled at various zones to compensate for variations in thick
ness of the bed. Generally, sufficient air escapes through the fuel bed
unburned so that little overfire air is required. By the nature of the
underfeed principles, overfire jets are required only for high firing rates, if
at all.
Underfeed stokers operate most satisfactorily with free-burning to
moderately caking coals, although anthracite and strongly caking, high-
volatile bituminous fuels may be burned with some success. The size of
coal is an important characteristic. For most installations, 1% X 0 in.,
nut and slack give the best results. Coals having high-ash-fusion temper
ature give less trouble with clinkers.
Large sizes of underfeed stokers are built as multiple-retort units.
Several coal hoppers and power rams are operated side by side, feeding
coal to a wide grate area. Figure 7-13 shows a multiple-retort underfeed
stoker. The principles of operation are the same as for smaller underfeed
stokers. Multiple retorts employ secondary rams to agitate the upper
levels of the fuel bed, thus equalizing the distribution of fuel on the grate.
Units of this type are commonly operated at firing rates up to 40 lb coal
per square foot per hour continuously, and up to 60 for peak loads of not
more than 2 hr duration.
Domestic underfeed stokers are similar in principle of operation to the
single-retort stoker, with the exception that most domestic models are
fed with a screw instead of a ram. While domestic anthracite stokers

commonly have an automatic ash-removing grate, many domestic stokers
for bituminous coals have no provision for the dumping of ashes from the
firebox. The ash is most expediently removed by agitating an accumula
tion of ash into a hot coke zone and forming a clinker, which can be pulled
from the fuel bed with tongs. For this application, a coal with a low-
ash-fusion temperature is desirable.
Section A-A
Fig. 7-13. A multiple-retort underfeed stoker.
The BCR smokeless domestic heater, shown in Fig. 7-14, was developed
at the Battelle Memorial Institute under the sponsorship of the Bitumi
nous Coal Research and Stove Manufacturers. As the figure shows, it is
a magazine-fired, underfeed type of heater engineered to give efficient,
smokeless combustion on both high- and low-volatile bituminous coal.
Nut size coal, ^ X 3 in., is poured into the magazine from above. Burn
ing occurs on the ash layer just above the grate. Volatile matter released
from the green coal above the burning fuel bed is mixed with the primary,
or cross-feed, air and is carried across and downward through the hot
combustion zone. A small quantity of magazine air is used to carry dis

tilled gases from the upper part of the heater and to prevent formation of
large masses of coke in the magazine. A constant quantity of secondary
air is admitted through the hot refractory secondary-air arch to assure
complete combustion of the gases leaving the combustion chamber. The
burning rate is regulated by the cross-feed air damper and the draft,
TH
Fig. 7-14. The smokeless heater. (Courtesy of Bituminous Coal Research.)
which can be adjusted with either the twist damper or the automatic
limit-control check damper. By providing thorough mixing of air with
the slowly emitted volatile matter and by maintaining high temperatures
for ignition in the combustion zone, this design of heater or furnace has
proved to be capable of burning practically any type of coal easily, effi
ciently, and smokelessly.
Pulverized-coal Firing. The most frequently used method for burn
ing coal in large quantities today is by pulverized firing. The coal is
powdered in a pulverizing mill into extremely fine particles, and the coal
dust is mixed with heated primary air. The primary-air coal-dust mix
ture is blown into the furnace through a burner port directly into the
flame. Once inside, the coal is quickly ignited from the high temperature
of the combustion reaction. Almost simultaneously the volatile matter
and the moisture are released, and the volatile gases and fixed carbon
start to burn. To obtain quick ignition and complete combustion, the
coal must be in the form of a very fine dust. The size of pulverized-coal
particles is measured by the percentage of coal which will not pass a
50-mesh screen and the percentage which will pass a 200-mesh. A' 200-
mesh screen has 200 wires per linear inch or 40,000 openings per square
inch and will not pass a particle larger than 0.0029 in. in diameter.
Powdered coal and

"""
primary air out
Fig. 7-15. Diagram of a ball-mill pulverizer.
Pulverizers. Pulverization is accomplished in a pulverizer of the ball-

mill, impact, roller, or fluid type. A diagrammatic sketch of a ball mill
appears in Fig. 7-15. The unit consists of a horizontal cylinder mounted
on rollers and partially filled with 1- to 2-in. -diameter steel balls. A feeder
device for admitting fresh coal and air is located at one end of the cylinder.
As the cylinder is rotated, the balls cascade down and smash the inter
mingled pieces of coal. Heated air flowing through the mill picks up the
powdered coal and carries it out the discharge end of the cylinder. A
classifier permits the primary air and the sufficiently fine coal to leave the
mill while the oversize particles are returned for further pulverizing.
The ball mill provides low maintenance, reliability, and a quick response
to changes of fuel demand, although the power consumption per ton of
coal pulverized is relatively high.
A mill for comparatively small installations capable of handling unusu

ally moist coal is the impact mill. Crushing is accomplished from impact
action of coal smashing against lugs on a rapidly rotating impeller.
Impact mills are low in first cost, but maintenance and power require
ments are relatively high.
In the roller type of mill, the coal is powdered by passing it between two
surfaces, one rolling over the other. Pulverizers of this type have low
Fig. 7-16. Raymond bowl mill. {Courtesy of Combustion Engineering-Superheater, Inc.)
power consumption per ton of coal pulverized. They can be built in

compact but high-capacity units, are quiet in operation, and have low
maintenance costs. Figure 7-16 shows a bowl-mill crusher, typical of
the roller type. Grinding occurs between the revolving grinding ring
and the stationary rolls. An air stream carries the fine coal into the
classifier, where the oversize particles are separated and fall back into
the grinding bowl. The exhauster removes the fine coal and primary
air from the mill and discharges it to the burners at the furnace. Much
the same grinding job is accomplished by a ball-and-race type of machine
wherein large balls several inches in diameter crush coal against upper and
lower steel races.
The capacity of any pulverizer is affected by the moisture content of
the coal and the fineness of grind desired as well as the grindability index
Per cent capacity, corrected for moisture

60 70 80 90 100 IIP 120 130 140 150 160
60 70 80 90 100 110 120 130 140 150 160

Per cent capacity, corrected for grind
Fig. 7-17. Pulverizer capacity as affected by grindability and moisture of coal and fineness
of grind. This is a typical performance chart for one type of pulverizer. Capacity of
100 per cent based on 55 grindability coal with 8 per cent moisture ground to 70 per cent
through 200 mesh. Dotted lines show how to find the mill capacity for 75 grindability
coal, with 75 per cent through 200 mesh and 10 per cent moisture. (Courtesy of Combustion
Engineering-Superheater, Inc.)
of the coal. Figure 7-17 shows the relationships between these factors
for one typical mill. The size of coal fed to the pulverizer also has an
effect on the capacity of the mill.
Since mill capacity is reduced with moist coal, pulverizers are designed
to dry the fuel while crushing it. The dried coal also ignites more rapidly
in the furnace. Primary air, heated to 250 to 600 F, is passed over the
coal during crushing. The temperature of the coal/air mixture leaving

the pulverizer for the burners is usually held between 150 and 180 F. If
insufficient heated air is supplied to the mill, the pulverizer capacity may
be limited by its drying capacity.
The capacity of a pulveri zed-fuel-burning plant is limited by the ash
content and the heating value of the fuel since, for a given Btu output,
more tons of high-ash and low-heating-value fuel must pass through the
pulverizers than for a more favorable coal.
Engineering-Superheater, Inc.)
Early designs of pulverized-coal-fired plants employed several pulver

izers operating in parallel to supply powdered coal to a storage bunker,
from where it was fed to the individual burners at the furnaces. Present
practice allows for direct firing of coal from the pulverizers to the burners
with two or more pulverizers serving each boiler installation.
Burners. Figure 7-18 is a diagrammatic cross section of a burner
typical of the many designs built. Primary air and pulverized coal are
delivered to the burner through the coal pipe and are projected from the
central coal nozzle. The small directing ribs, or rifling, in the coal nozzle
spirals the primary air/coal mixture leaving the burner and assures a
uniform mixture at the burner outlet. Secondary air from the air
chamber flows past the regulating dampers and is given a spinning action
as it passes the guide vanes on its way to form a turbulent envelope over
the mixture leaving the nozzle. Turbulence and flame shape in the
furnace can adjusted by regulating the velocity and angle of flow from
be
the guide vanes. With such a burner, both long diffusion and short
turbulent flames can be obtained. Ignition for "lighting off" a pulver-
ized-fuel furnace is secured from a gas or oil torch located within the
central nozzle. By varying the design of the pulverized-coal nozzle to
include a gas or an oil burner, or both, within the central tube, it is
possible to operate a burner on pulverized coal, oil, or gas with any
individual fuel or in any combination as desired.
— 'i
. —
40
'Sj,
8 30
£
a-
o
o [/-
20 o—
10
f (/
/
o
r
<n
0 1 2 3 4 5 6 7 8 9 10 II 12 13 14 15 16
lb air /lb coal
Fig. 7-19. Inflammability curves for mixtures of pulverized coal and air. (Courtesy of
Combustion Engineering-Superheater, Inc.)
The effects of the volatile and ash contents of the pulverized coal on
ignition rates and inflammability limits are shown in Fig. 7-19 for various
air/fuel ratios. It will be noted that the quickest ignition is obtained
with a rather rich mixture of coal and air and that the ignition rates
increase as the volatile content increases and the ash content decreases.
Also the high-volatile fuels are inflammable over a much greater range of
air/fuel ratios than the higher rank coals.
The complex nature of the pulverized-coal combustion reaction makes
quantitative data showing the effect of individual variables difficult to
obtain under actual firing conditions. However, extensive studies of the
subject4-7 show the following trends to exist :
1. Ignition occurs within a few hundredths of a second. Ignition

rates vary inversely with particle diameter.
2. High-volatile coals ignite quicker and at lower temperatures than
low-volatile fuels.
3. Distillation takes place in about 0.005 sec. During this period
the particle may swell two to eight times its original volume.
4. In commercial furnaces, one-half the coal particle is burned by the
time it has been in the furnace 0.05 sec. Within 0.1 to 0.3 sec, 95 per
cent of it is consumed. Combustion of the last 5 per cent of the coke
particle proceeds much more slowly and is dependent on the type of fuel,
furnace temperature, excess air, and turbulence.
5. Burning rates are greater at higher furnace temperatures.
6. Small particles burn faster, probably because more coal area is
exposed to oxygen. The time for complete burning is proportional to the
square of the particle diameter.
7. Swelling leads to more rapid burning. Swelled particles also are
less dense and tend to be carried away with the combustion gases.
Hence, a coal which swells greatly may contribute to high carbon loss
with the fly ash.
8. With incomplete combustion, little or no carbon monoxide is pres
ent in the flue gases, but solid carbon may be carried up the stack.
9. To achieve nearly complete combustion, either intense turbulence
of coal and air must be created near the burner, or the coal must be
permitted to remain for a relatively long time in the combustion chamber.
10. Good firing practice dictates that at least 65 to 85 per cent of the
coal must pass a 200-mesh screen, and not more than 2 per cent should
be larger than 50 mesh. The lignites and subbituminous fuels need not
be ground to the same degree of fineness as low-volatile bituminous and
anthracite. The fineness of grind becomes increasingly more important
as combustion space is reduced.
Ignition of the pulverized coal should start about 3 to 12 in. from the
nozzle tip. If ignition occurs too close to the burner, high localized
temperatures will cause overheating of the burner and result in high
burner maintenance. Primary- and secondary-air velocities, turbulence,
etc., influence the point of ignition, rate of combustion, and flame shape.
The velocity of air and fuel is regulated to a value slightly greater than
the ignition rate. Too low an exit velocity from the burner invites flash
back and burning in the central cone, while exit velocities greatly in
excess of the ignition rate for the fuel and air/fuel ratio used result in
unsteady flame conditions and waste valuable furnace volume.
With all the advantages of pulverized firing comes the disadvantage
that up to 85 per cent of all the ash in the coal starts through the boiler
with the combustion products. At the high gas temperatures desirable

for good efficiency and heat transfer, much of the ash softens and deposits
on the furnace walls, boiler tubes, baffles, etc., as a slag deposit. Before
the gases leave the furnace, as much as half of the ash is deposited in the
furnace or boiler. As the ash deposits accumulate, the heat-transfer
capacity of the boiler heating surfaces is drastically reduced and boiler
efficiency suffers. In addition, the slag formations may block the flow of
Fig. 7-20. A vertically fired pulverized-coal furnace with continuous slag tap.
gases in the narrow boiler passages, increasing the draft loss through the
installation and making it difficult if not impossible to keep the combus
tion rate up to capacity. Also, the slag tends to erode or corrode and
weaken some surfaces. Hence, adequate slag-removing equipment is
required to knock the ash back down into the furnace where it can be
removed. The amount of ash that passes through the boiler with the
flue gases is usually too high to be tolerated with present atmospheric
pollution ordinances, so that expensive fly-ash collectors are also required.
Furnaces. Furnaces are classified according to the method used for
ash removal as being either of the slag-tap or dry-bottom type. Figure
7-20 shows a furnace designed for continuous molten slag tapping. Part
of the ash settles to the bottom of the furnace and mixes with the pool
of already molten ash. The pool is kept fluid by playing a part of the
flame across it. As the molten ash accumulates on the furnace hearth,
part of it flows over a water-cooled slag dam, which surrounds the slag
drip opening, and into a water bath. Quenching shatters the molten ash
into a granular form which may be removed from the furnace readily.
Fig. 7-21. A package steam generator employing horizontal firing and water-screen bottom
construction. {Courtesy of Combustion Engineering-Superheater, Inc.)
Another design of wet-bottom furnace provides for intermittent instead of

continuous slag tappings.
A series of water tubes, called a water screen, can be placed across the
lower part of the furnace to absorb heat rapidly from the hot combustion
gases in that zone (Fig. 7-21). As a result, the space beneath the water
screen is kept relatively cool and the ash settles there as a dry, dusty
powder. About the same effect as a water screen can be achieved by
sloping the bottom waterwalls of the furnace to form a hopper, as shown
in Fig. 7-1. The dry ash, kept cool by the lower waterwalls, slides down
the steep sides of the hopper bottom and into a dry ashpit.
Furnaces are classified according to firing method as vertical, hori
zontal, and tangential. Figure 7-20 illustrates diagrammatically a

vertically fired furnace. The primary air and pulverized fuel are fired
directly downward and into the combustion space through a series of long,
narrow rectangular burners. These "fantail" burners produce a wide,
thin sheet of flame. The long axis of the rectangle is placed perpendicular
to the front of the furnace so that alternate lanes of flame and air space
exist across the width of the combustion volume. In vertical firing,
secondary air is admitted through small ports down the front of the
furnace wall. By admitting much secondary air high up the furnace wall
near the burner, the flame produced can be made short and intense. A
long, diffusion type of flame may be created by admitting secondary air
progressively down the entire height of the furnace. The fantail burner
also employs tertiary air which enters through the burner to form an
envelope around the primary air and fuel to help regulate the point of
ignition and provide better mixing. Both primary and tertiary air may
be adjusted to help regulate the depth of flame travel.
Vertical firing provides a uniform temperature distribution throughout
the furnace and is especially good for high, narrow furnaces. Low-
volatile coal, which requires a long flame travel to allow time for complete
combustion, can be fired vertically better than by any other method,
although this design is equally good for high-volatile fuels.
A horizontally fired integral furnace boiler appears in Fig. 7-21.
Burners of the type pictured in Fig. 7-18 are used, and they may be aimed
horizontally, or slightly up or down. By proper burner adjustment,
either a long, penetrating flame, or a short, sharp turbulent flame can be
produced. The latter flame yields a very high rate of heat release with
high combustion efficiency. If the burners are placed in both the front
and the back furnace walls to give opposed horizontal firing, extremely
high turbulence is produced in the zone where the flames impinge, per
mitting extra-high firing rates. As a general rule, the flame that creates
the most turbulence with the least amount of slag deposit is the best.
Horizontal burners are capable of unusually high capacities, up to 900 lb
of coal per hour per burner; hence fewer burners may be required for a
given installation. Horizontal firing is the method used in the majority
of United States pulverized-coal-fired plants.
Figure 7-22 shows a plan view of a tangentially fired installation. The
burners located in the four corners of the furnace are aimed tangent to a
small imaginary circle at the center. Ignition at each burner is aided by
the flame from the preceding one. An intensive, turbulent, rotative
motion is developed where the flames impinge, which spreads the flame
out to fill the furnace area and promotes rapid, complete combustion.
Since the whole furnace acts as the burner to promote mixing and ignition,
the air and fuel from each burner need not be so accurately controlled as
in the horizontal burner. Tangential firing permits extra-high rates of
heat release and gives high combustion efficiency, even with very little
Primary air and coal
Fig. 7-22. A plan view of a tangentially fired furnace.
excess air. Figure 7-23 shows the carbon loss in the fly ash as it is
affected by the different firing methods and the coal particle size.
Because there is never much unburned fuel in the furnace at one time,
pulverized-coal-fired boilers are capable of meeting wide fluctuations in
Per cent excess air supplied Fineness, per cent through

through burners 200 mesh
Fig. 7-23. Effect of turbulence and particle size on unburned carbon loss for pulverized
firing. (Courtesy of Combustion Engineering-Superheater, Inc.)
load quickly. In modern furnaces, the temperature of the superheated

steam leaving the boiler can be held constant throughout the operating
range by tilting the burners up or down a few degrees. If a burner is
shooting slightly up, the gases reach the boiler heating surfaces sooner
and have less time to radiate heat to the waterwalls than if the burners
are tilted down.
Because of the complexity and high first cost of the equipment required
to operate a pulverized-fuel system, pulverized firing is used only in large
furnaces burning at least 2000 lb of coal per hour where decreased fuel
costs and fuel flexibility can justify the expense.
Cyclone Furnace. A new method of burning coal in suspension which
utilizes many of the advantages and avoids a number of the disadvantages
of other combustion systems has been developed by the Babcock and
Wilcox Co.8 The cyclone furnace (Fig. 7-24) was first put into field
service in 1944 and is now being used in a number of new and proposed
Fig. 7-24. Diagram of the cyclone furnace. (Courtesy of Babcock and Wilcox Co.)
installations. Primary air and coal crushed finer than in. in diameter
are fired tangentially into a compact cylindrical combustion chamber
about 8 ft in diameter. About 75 to 80 per cent of the total air used also
enters the cylinder tangentially as secondary air and provides thorough
mixing of fuel, air, and combustion products within the horizontal, water-
cooled combustion chamber. The combustion products leave through a
reentrant water-cooled throat at the end of the cylinder and flow into the
main furnace where they are exposed to waterwalls and other boiler heat
ing surfaces. Approximately 5 per cent of the total air is supplied as
tertiary air at the axis of the burner to provide a velocity component in
the direction of the secondary furnace and to ensure burning of any fine
coal particles suspended near the center of the vortex flow pattern.
Combustion chamber temperatures close to 3000 F are achieved by
using only 5 to 15 per cent excess air. At this temperature, the ash is
melted to liquid slag. The centrifugal force resulting from admitting the
primary and secondary air at 20,000 to 30,000 ft/min throws the molten
ash to the periphery of the cylinder. Fuel particles are trapped and
embedded in the film of sticky slag and are subjected to an intense scrub
bing action air sweeps across the coal. Heat release rates in excess
as the
of 500,000 Btu/ (cu ft) (hr) within the cyclone cylinder maintain the ash
in the molten state until it flows by gravity down the slope of the furnace
and through a slag taphole at the rear.
The cyclone principle affords high rates of heat release and at the same
time permits only about 10 per cent of the ash of the coal to enter the main
furnace. The low ash carry-over reduces slag deposits, lowers main
tenance costs, and simplifies any dust-collection apparatus required or
eliminates it entirely. The low excess air, high turbulence, and high
furnace temperatures permit high combustion efficiency. In addition,
the furnace can be operated with any grade of coal with a minimum of
fuel preparation, and air preheat temperatures up to 850 F may be
permitted.
HEAT BALANCE
A heat balance is an account of all the energy entering and leaving a
process. Tests are often made to determine heat balances for engines,
heaters, and boilers because they show the fraction of the heat input that
is being used, and reveal what happens to the remainder. Once the
nature of the losses is understood, operation or design changes which
would result in increased efficiency may be obvious.
Heat-balance calculations for various pieces of equipment differ only in
their details. Since the steam-boiler heat balance is typical of this type
of problem, the method for computing the heat balance for a steam boiler
is presented first. It is felt that once the principles of a heat balance are
understood for one type of equipment, the method of approach may be
adapted readily to other apparatus.
Steam-boiler Heat Balance. The heat input to a boiler is equal to the
number of pounds of coal fired during the test multiplied by the heating
value per pound of coal "as fired." The heat turned into useful energy is
equal to the number of pounds of water evaporated during the period
times the change in enthalpy per pound of water in passing through the
boiler. According to the method proposed in the ASME Boiler Test
Code, the heat losses are attributed to
1. Combustible material in the refuse

2. Heating the dry flue gases
3. Incomplete combustion to carbon monoxide
4. Heating and evaporating the moisture in the coal
5. Heating and evaporating the water formed by combustion of the
hydrogen in the coal
6. Miscellaneous and unaccounted-for factors
Derivations of formulas which are useful in calculating and analyzing

the individual heat losses are given in the following paragraphs.
Determination of Percentage of Excess Air from Flue-gas Analysis. Let
02, CO, and 2V2 represent the percentages by volume in the dry flue gas of
oxygen, carbon monoxide, and nitrogen, respectively.
The nitrogen in the flue gas is the nitrogen which was with the total air
supplied (neglecting the small amount in the coal). The oxygen in the
flue gas is the oxygen which was with the air not used for combustion.
If carbon monoxide is present, a volume of oxygen equal to one-half the
volume of carbon monoxide should unite with it in order to have complete
combustion, because }4 mole of oxygen is required to burn 1 mole of
carbon monoxide. Excess oxygen is that portion which would have
remained if combustion had been complete, and would equal 02 — CO/2.
This excess oxygen was accompanied by 3.76 vol of nitrogen, or
—
3.76(02 CO/2) is the volume of nitrogen which came in with the excess
air. Nitrogen with theoretical air would equal the total nitrogen present
minus that in the excess air. Then, N2 — 3.76(02 — CO/2) is the volume
of nitrogen which came in with the air theoretically needed for complete
combustion. Therefore,
Nitrogen with excess air

Nitrogen with air theoretically needed for combustion
-
N2
3.76(02
- CO/2)
- 3.76(02 CO/2)
Hence, the percentage of excess air is
-
100 X
N2 - 3.76(02 CO/2)
3.76(02
- CO/2)
It should be noted that this equation is not accurate for use with the
combustion products of fuels containing large amounts of nitrogen, such
as producer gas.
Determination of Weight of Carbon Burned per Pound of Coal Fired. Let
A equal the percentage of ash in the coal fired; a, the percentage of ash in
the refuse; R, the pounds of refuse per pound of coal fired. Then,
assuming that the weight of ash in the refuse equals the weight of ash in
the coal fired,
aR = A
and the pounds of refuse formed per pound of coal fired is

If the analysis of the refuse is given on the dry basis, the refuse consists
only of ash and combustible material. The latter lost in the refuse is
practically all carbon and is usually assumed to be pure carbon. Hence,
the percentage of carbon lost in the refuse is (100 — a), and the pounds of
carbon lost in the ashpit per pound of coal fired is (100 — a)A/100a.
The weight of carbon burned per pound of coal fired is the weight of
carbon in each pound of coal fired (as determined from the ultimate
analysis of the coal) minus the weight of carbon in the refuse per pound of
coal fired. In equation form, the weight of carbon burned per pound of
- a) A
coal fired is
C - (100
100a
where C is the weight of carbon per pound of coal fired, from the ultimate
analysis of the coal.
Determination of the Loss Due to Combustible Material in the Refuse. Let
hc equal the heating value of the combustible material in the refuse, Btu
per pound. This is usually assumed to be the same as that for carbon,
14,600; for very accurate work it may be determined with a calorimeter.
Then, since (100 — a)A/100a is the weight of combustible in the refuse
per pound of coal fired, as shown above, the heat loss due to combustible
in the refuse, in Btu per pound of coal fired, is
(100
- a) A
K 100a
Determination of Loss Due to Heating the Dry Flue Gases. A sizable heat
loss results from the dry flue gases passing out the stack at a relatively
high temperature when the air for combustion enters the furnace at a
much lower temperature. In Chap. 3, it was shown that the heat carried
away with the combustion products was the energy required to heat the
products from the initial temperature of the air to the exit temperature.
Hence, the heat loss to the dry flue gas may be computed as the change in
enthalpy of the combustion products from the initial to the final tempera
This loss is
ture.
QL = wCp(Tc
- Ta)
where QL = heat loss to dry flue gases, Btu/lb coal fired
w = weight of dry flue gases formed per pound coal fired
Cp = specific heat at constant pressure of flue gases (a value of
0.24 Btu/lb may be used here with sufficient accuracy for
most work)
Tc = exit temperature of dry flue gases
Ta = temperature of air entering the furnace
The weight of dry flue gases may be computed from stoichiometric

analysis once the percentage of excess air used is determined, or it may be
calculated from the flue-gas analysis directly. To calculate the weight of
dry gases from the Orsat analysis, let the percentages by volume as
determined by the flue-gas analysis be represented by C02, CO, O2, and
N2- These symbols will, therefore, also represent the number of moles of
carbon dioxide, carbon monoxide, etc., in 100 moles of dry flue gas. The
weight of each constituent in 100 moles of dry gas may be determined by
multiplying the number of moles by its molecular weight. The total
weight of the 100 moles will be
44C02 + 3202 + 28N2 + 28CO
The weight of carbon in the 100 moles will be 12(C02 + CO).

Therefore, the weight of dry gas per pound of carbon burned will be
UCO2 + 32Q2 + 28 (CO + N,) _ UCOi + 8Q2 + 7(CO + N2)

12 (CO2 + CO) 3(C02 + CO)
This expression may be simplified by remembering that
CO2 + 02 + CO + N2 = 100,
from which 7(CO + N2) = 700 - 7C02 - 702. If the latter expression
for 7 (CO + N2) is substituted in the preceding equation, the weight of
dry gas per pound of carbon burned reduces to
4C02 + O2 + 700
3(C02 + CO)
The product of the above expression with the weight of carbon burned
per pound of fuel gives the weight of dry flue gas per pound of fuel fired.
Determination of Loss Due to Carbon Monoxide Formation. Again let
the percentages of carbon dioxide and carbon monoxide in the Orsat
analysis be represented by CO2 and CO, respectively. Then the weight
of carbon in 100 moles of the dry flue gas would be 12(C02 + CO).
Likewise, the weight of carbon in the carbon monoxide formed would be
12CO. Therefore,
Weight of carbon burned to C0 12CO

=
CO
Total weight of carbon burned 12 (C02 + CO) CO2 + CO
—
g^
Since C is the weight of carbon burned per pound of coal
CO
is the weight of carbon burned to
CO2 + CO
CO per pound of coal fired.
The heating value of 1 lb of carbon if burned to C02 is 14,600 Btu, but

if burned to CO is 4440 Btu. Therefore, the loss per pound of carbon
burned to CO instead of C02 is (14,600 -
4440), or 10,160 Btu. Thus,
the loss due to carbon burning to carbon monoxide, in Btu per pound of
coal fired, is
Determination of Loss Due to Moisture in the Coal. When coal is fired,

it may be assumed that the moisture in it is heated to a boiling tempera
ture of 212 F, evaporated into steam, and finally superheated to flue-gas
temperature. The heat required to change the moisture into super
heated water vapor is carried up the stack and lost. The heat loss due to
heating the moisture in the coal, in Btu per pound of coal fired, is
w[(212
- Ta) + 970.3 + 0Ai)(Tc - 212)] = m(1084.8 - Ta + 0AQTc)
where m = weight of moisture in each pound of coal, lb
Ta = temperature of coal when fired, F
Tc = temperature of flue gas leaving boiler, F
The first term in the brackets represents the heat added to 1 lb of water
to bring it up to the boiling point; that
is,
the temperature change times

an assumed specific heat of water of Btu/lb. The next term, 970.3,
is
1
the latent heat of vaporization of water at 212 while the last term
F,
represents the heat required to superheat the water vapor from 212 to
F
the final temperature. A value of 0.46 Btu/lb of steam used as the
is
mean specific heat in this temperature range.

Thus, the expression within the brackets represents the heat added to
lb of water to convert from liquid to superheated vapor at constant
it
1
atmospheric pressure. This equation based on the assumption that

is
no vaporization takes place until 212 reached and that all the
is
F
moisture in the coal evaporated at that temperature. Although this

is
assumption not strictly accurate, the error introduced by very

is
it
is
slight.
Determination Loss Due to Moisture Formed by Burning Hydrogen in
of
the Fuel. When hydrogen in any fuel burns, unites with oxygen to
it
form water. Each pound of hydrogen unites with lb of oxygen to form

8
lb of water. In the ultimate analysis of coal, the hydrogen usually

is
9
reported as the total hydrogen, including that which was already com
bined with oxygen to form water. Therefore, the weight of water formed
m) lb, where H
by burning the hydrogen in the coal —
is
(9H
is
the
weight of hydrogen in each pound of coal from the ultimate analysis, and
m the weight of moisture in each pound of coal from the proximate
is
analysis.Again referring to the derivation on page 93 of Chap. 3, it

may be assumed that the hydrogen is converted to H20 liquid at the
temperature of the incoming fuel and air. The heat loss to this moisture
formed from the combustion of hydrogen is the energy required to con
vert the liquid water to vapor superheated to stack temperature. Thus,
the loss due to moisture formed by burning the hydrogen in the coal, in
Btu per pound of coal fired, is
(9H
- m) (1084.8
- Ta + 0.46rc)
where H =
pound hydrogen per lb of coal
m = pound moisture per lb of coal
Ta = temperature of fuel and air supplied to furnace, F
Tc = flue-gas temperature, F
It follows that the heat loss due to moisture formed from the burning of
hydrogen in any fuel is M(1084.8 — Ta + 0.46rc), where M represents
the pounds of water formed from the combustion of hydrogen.
Example 1. The following data were observed during an efficiency test of a boiler,
furnace, and underfeed stoker. Compute a heat balance for the installation.
1. Proximate analysis of fuel, as fired:

Moisture 6.83%
Volatile matter 34 . 22 %
Fixed carbon 43.06%
Ash 15.89%
2. Ultimate analysis of fuel, as fired:
Carbon 62.74%
Hydrogen 4.85%
Oxygen 14.14%
Nitrogen 1.20%
Sulfur 1.18%
Ash 15.89%
3. Heating value of coal,as fired 11,040 Btu
4. Combustible in the analysis of dry refuse 21 .66%
(a) Ash in dry refuse 78 . 34 %
5. Orsat analysis of flue gases:
CO, 9.06%
()a 10.13%
CO 0.11%
N2 80.70%
6. Pressure of steam leaving boiler 144 . 0 psia
7. Temperature of air used for combustion 71. 4F
8. Temperature of flue gases leaving boiler 497 F
9. Temperature of feedwater entering boiler 174 F
10. Temperature of steam leaving throttling calorimeter (calorimeter
used to sample steam leaving boiler) 284 F
11. Fuel fired during test 584 . 2 lb
12. Water evaporated during test 4015.4 1b
Solution:
13. Per cent excess air used for combustion is
- - (0.11/2)]
N, -
3.76[Q2
3.76[02
- (CO/2)]
(CO/2)]
80.70 -
3.76110.13
3.76[10.13
- (0.11/2)] = gg 4%
14. Weight of combustible in refuse per pound of coal fired is
—
(100
~ a)A 100 - 78.34 15.89
=a04391b
lOOa 100 78^4
1 5. Weight of carbon burned per pound of coal fired equals the weight of C in coal,
from ultimate, minus the weight of combustible in refuse per pound of coal
(assuming all the combustible in the refuse is carbon). This is
0.6274 - 0.0439 = 0.5835 lb
16. Weight of dry flue gases formed per pound of C burned is
4C02 + 02 + 700 = 4 X 9.06 + 10.13 + 700

=
„
3(C02 + CO) 3(9.06 + 0.11)
17. Weight of dry flue gases formed per pound of coal fired is
Lb dry flue gas

Lb C burned JbCburned
lb coal fired
= ^ ^ ^ Q = ^
18. Heat absorbed by water and steam in the boiler per pound of coal fired is
... ,, TT lb H20 evaporated

(Ah per lb H2O)
lb coal fired
From the throttling calorimeter data:
h2 = 1185.2 Btu/lb steam (from superheated-steam tables at 284 F and 14.7 psia)*
Since hi = h2 in the throttling process
hi = hf + hfg = h2 = 327.13 + x(866.3) = 1185.2

Hence,
x = 99.1% = quality of steam leaving boiler
and the enthalpy of the steam leaving the boiler is 1185.2 Btu/lb H20.
The enthalpy of the feedwater entering, ht at 174 F, is 141.9 Btu/lb H2O*
Aft = heat added per lb H20 in the boiler = 1185.2 - 141.9 = 1043.3
Heat absorbed by the water and steam in the boiler per pound of coal fired is
4015 4
1043.3 X = 7641.8 Btu 69.2%
19. Heat loss (per pound of coal fired) due to uncon-

sumed carbon in the refuse:
0.0439 X 14,600 = 640.9 Btu 5.8%

*
From Keenan and Keyes, "Thermodynamic Properties of Steam," John Wiley &
Sons, Inc., New York, 1936.
20. Heat loss (per pound of coal fired) due to dry gas
leaving the boiler :
wcp{Tt - To) = 15.83 X 0.24(497 - 71.4) = 1616.9 Btu 14.7%

21. Heat loss (per pound of coal fired) due to carbon
monoxide formed:
Lb C burned CO
A
Lb coal fired C02 + CO
'
= 0.5835 X X =' 71.0 Btu 0.6%

9 Qg^p u 10,160
22. Heat loss (per pound of coal fired) due to moisture in

the fuel:
- Ta + 0.467V)
to(1084.8
= 0.0683(1084.8 - 71.4 + 0.46 X 497) = 84.8 Btu 0.8%
23. Heat loss (per pound of coal fired) due to water
formed from the combustion of hydrogen:
- -
(9ff to) (1084.8
-
Ta + 0.467\.)
-
= (9 X 0.0485 0.0683)
(1084.8 71.4 + 0.46 X 497) = 457.3 Btu 4.1%
24. Heat loss (per pound of coal fired) due to miscellane
ous and unaccounted for:
Heating value of fuel —

Z(losses + heat to steam) = 527.3 Btu 4.8%
11,040 Bki 100.0%
The various forms of heat loss from the combustion of any type of fuel
may be computed by a combination of the special formulas developed in
the preceding section and a stoichiometric analysis of the combustion
process. These losses are applicable to other types of combustion equip
ment. Examples of heat loss calculations with other fuels and equipment
follow.
Example 2. Calculate the percentage of heat loss to the dry gases up the flue
when charcoal burns in a fireplace with 800 per cent excess air. Analysis of the
charcoal shows: carbon =85%, moisture = 12%, ash =3%, heating value =
12,400 Btu /lb. Air enters the fireplace at 80 F and the stack temperature is 250 F.
Solution. The weight of dry combustion products may be determined by stoichio
metric analysis. Carbon is the only constituent in this charcoal which will burn.
85 lb C = sYi2, or 7.08 moles of C
Thus, 7.08 moles of C02 would be formed in the combustion products, and 7.08
molesof 02 would be required for perfect combustion.
Dry flue gases Moles Pounds

C02 7.08 7.08 X 44 = 312
Excess 02 7.08 X 8.0 = 56.64 56 . 64 X 32 = 1812
Total N2 7.08(8 + 1) X 3.76 = 239.6 239.6 X 28 = 6709
8833
The heat loss to the dry gases up the stack would be, in Btu per pound of charcoal :
Qt = wcp(Tc - Ta) = 88.33 X 0.24(250 - 80) = 3604 Btu
The loss to the dry flue gases up the chimney represents 3604/12,400, or 29.0 per
cent of the heating value of the charcoal.
Example 3. The following data were obtained from the heat balance of an oil-fired
boiler:
Firing rate 25 gal oil /hr
Specific gravity of oil 0 . 730
H/C (by weight) of oil 0. 172
Temperature of air and oil for combustion 80 F
Temperature of flue gases leaving boiler 580 F
Calculate the heat loss, in Btu per hour, due to the water formed by the combustion
of hydrogen in the fuel.
Solution:
Weight of oil burned per hour is
25 X 62.4 X 0.730 X 23K?28 = 152 lb
Per cent of hydrogen in the fuel is
0-172 lb H
H
x
X 100 = 14-b&%
14 65%
1.000 lb C + 0.172 lb
Weight of hydrogen burned per hour is
152 lb oil X 0.1465 = 22.3 lb
Weight of water formed per hour from the combustion of hydrogen is
22.3 X 9 = 200.7 lb
Heat loss per hour to moisture is
200.7(1084.8 - 80 + 0.46 X 580) = 255,000 Btu/hr
OPERATION AND DESIGN CONSIDERATIONS

Upon analyzing the heat losses in Example 1, it is apparent that some
of them could be changed more readily than others so as to yield higher
efficiencies. The losses might be classified as controllable and uncon
trollable. For instance, some loss due to combustion of hydrogen has to
be accepted if the fuel contains hydrogen. Likewise, "miscellaneous and
unaccounted for" losses often are mostly radiation losses (Fig. A-7),
which have to be accepted. By the same token, while fuel may be dried
before burning, with most types of combustion apparatus this would
involve extra equipment and expense ; in fact, the loss due to moisture in
the fuel is more or less uncontrollable.
Analysis of Heat Losses. The losses due to unconsumed combustible
material, carbon monoxide formation, and the dry flue gases remain
then as the controllable ones. Both unconsumed combustible and carbon
monoxide can be reduced by proper selection of firing equipment and

furnace so as to provide improved (1) mixing of fuel and air by turbulence,
(2) ignition temperatures, (3) furnace volume to give the gases time to
complete combustion in the furnace, or grate area to complete combustion
on the grate. Here are the three vital T's of combustion: turbulence,
temperature, and time. To these three should be added proper air /fuel
control to satisfy the fundamental requirements for complete, efficient
burning. At the same time intelligent operation is necessary if efficient
firing of even a well-designed installation is to be gained under all firing
conditions. Generally it is at high firing loads that undue losses to
carbon monoxide and uhconsumed combustible are experienced, because
the above-named criteria for good combustion are not met.
The loss due to the dry flue gases is often the greatest loss, and fortu
nately it usually is the one which is most easily controlled. From the
equation,
(To dry flue gases) QL = wCp(Tc - Ta)
it is readily apparent there are two avenues of approach for reducing the
loss: (1) to reduce the temperature difference between the stack gases
leaving and the air entering for combustion, (2) to limit the weight, w, of
flue gases formed per pound of fuel fired. To accomplish the former,
additional heat-exchange surface must be added to the boiler and furnace
installation to cool the combustion gases leaving the boiler, or the existing
heat-exchange surface must be used more efficiently. Thus, an econo
mizer may be added to absorb some of the heat from the flue gases and
heat the feedwater nearer to evaporation temperature before forcing it
into the boiler. Or an air preheater may be added to warm the combus
tion air with heat from the flue gases. The air preheater does double
duty in reducing gas losses, since the stack temperature is reduced and the
inlet-air temperature is raised in the same process.
Often the flue-gas temperature may be lowered by better operation of
the existing installation. For instance, thorough cleaning of the boiler
tubes frequently, both inside and outside, increases the rate of heat
transfer, and a greater portion of the heat of the combustion products is
transferred to the water and steam. Decreasing the exit temperature of
the flue gases also reduces the losses to moisture and combustion of
hydrogen slightly. The lowest temperature to which the flue gases may
be cooled is limited in practice to a value considerably above the dew
point to avoid condensation on heating surfaces at all operating condi
tions. Exit flue-gas temperatures from boilers having no air preheaters
or economizers must be expected to be rather high if these boilers are
heating steam to several hundred degrees. Either an economizer, an air
preheater, or both may be installed in a furnace to achieve high effi

ciencies. Because of the high first cost of these auxiliaries, they are used
only in medium- and large-sized plants where the increased efficiency
justifies the additional expense.
The second operating variable affecting the loss to the dry flue gases is
the weight of combustion products formed per pound of fuel fired.
Analysis shows that reducing the amount of excess air used for combus
tion cuts down the weight of dry products formed. While in theory it
would be desirable to use only stoichiometric air for combustion, there
600 F
500 F
400 F
100 120 140 160 180 200 220

Total air used, per cent of theoretical
Fig 7-25. Variation of heat lost to the flue gases with total air used for a stoker-fired
furnace with a medium combustion space. Flue-gas temperature rises as excess air
increases because of poorer heat-transfer conditions. (Courtesy of Bailey Meter Co.)
are several practical limitations which determine the minimum per

centage of excess air allowable. Some of these are
1. Incomplete combustion, giving carbon monoxide and unconsumed
combustible, becomes more prevalent as excess air is reduced.

2. Low excess air with little turbulence and poor mixing usually results
in excessive smoke.
3. Extra-high flame temperatures are obtained with low excess-air
mixtures which may result in undue damage to furnace walls, high rates
of burner or stoker wear, excessive slag deposits, and localized overheating
of the furnace.
Each of these conditions contributes to high maintenance costs.
The graph in Fig. 7-25 shows the relationship between the total air used
for combustion and the heat lost up the stack for a stoker-fired furnace
having a medium-sized combustion chamber. For this type of furnace,
optimum operation is achieved with about 25 to 30 per cent excess air and
a slight trace of carbon monoxide in the products. The results of many
tests by field engineers showing the excess air used in a large number of
plants with different types of equipment appear in Fig. 7-26. The upper
and lower mark for each type of equipment show the maximum and
minimum percentages of excess air found in the plants contacted while
the curves pass through the averages of the values reported.
r Averages for
240 i 0
Refractory walls
— o— Air-cooled walls
tx Water walls
220 IVIUA. VUIUC CI llsUUIIICI VU
^
Min. /alue enc write red

0•=200
EL 180
o
1
v
N—
160
o
>
140 i
I
i\\
120
100
is 1 9>
Be g in
.00
Fio. 7-26. Percentage of theoretical air supplied when furnaces are adjusted for the best
operating conditions. This graph is plotted from tests of nearly 7000 installations.
(Courtesy of Bailey Meter Co.)
The chief reasons why the percentage of excess air was held as high as it
was for the different installations and the relative importance of these
reasons are given in Fig. 7-27. Thus, for pulverized-coal-fired units the
lower excess-air limit is governed principally by the desire to reduce
furnace maintenance and eliminate smoke, while for oil-fired units it is
smoke, for gas burners it is carbon monoxide formation, etc. The per
centages shown in Fig. 7-27 merely point out the relative importance of
each factor in arriving at the excess air required for each unit.
Since a low percentage of excess air is so important for good operation
and economy, a continuous check of the amount used is very necessary.
As Fig. 7-28 shows, the percentage of excess air can be found if either the
carbon dioxide or the oxygen content of the dry flue gases is known.
Thus, an operator must adjust his air supply to hold the carbon dioxide in
the flue gases to 10.5 per cent if he is firing natural gases and wants to
maintain 10 per cent excess air. As shown in the chart, the percentage of
oxygen is a more uniform index of excess air for all types of fuels. How
ever, carbon dioxide meters are the more common instruments used for
combustion control.
LEGEND:
R — Refractory furnace walls
A — *Air- cooled " "
H — Furnace maintenance
0 — Smoke
W- * Water-cooled " "

0— CO □— Ashpit loss
* — Includes partially cooled furnaces
Fig. 7-27. Relative importance of operating factors which limit the reduction of the per
centage of excess air. For example, with all the natural-draft chain-grate stokers tested,
15 per cent ran into trouble with furnace maintenance if the excess air was reduced
below the best adjustments; another 38 — 15, or 23 per cent, encountered excessive CO
with less air; 12 per cent got smoke; while 50 per cent of the plants tested could not reduce
the percentage of excess air further without getting high ashpit losses. (Courtesy of Bailey
Meter Co.)
With operating factors taken into account, typical efficiencies obtain

able in different furnaces are listed in Table 7-1, along with typical heat
balances for these installations.
The more efficient installations employ either air preheaters or econo
mizers, or both, while the poor units probably have neither. It might be
noted that natural-gas-fired units cannot achieve quite the same high
efficiency as coal and oil because the high ratio of hydrogen to carbon in
methane results in an extra-high uncontrollable loss due to combustion of

hydrogen in the fuel.
Heat Release Rate. The rate of heat release per cubic foot of com
bustion volume is the index used to compare the intensity of heat libera
tion within a furnace. The furnace is considered to be all combustion
0 20 40 60 80
EXCESS AIR, PER CENT
Fio. 7-28. Relation between oxygen, carbon dioxide, and excess air for various fuels.
(Courtesy of Bailey Meter Co.)
volume from the floor up to the first heat-transfer surface, which assumes
that no burning exists after the gases contact the relatively cool boiler
heating surfaces. Since size and first cost of equipment are related to
the volume of the combustion chamber, the desirability of obtaining high
capacity in a small space is obvious. On the other hand too compact an
installation may give incomplete, smoky combustion and high localized
temperatures which may cause softening of the furnace walls, undue
stoker and burner wear, and excessive deposits of slag. Thus, a small
furnace may be more costly over a period of years than a more generously
designed one. The factor of reliability versus conservation of space is
also involved. The formation of slag deposits in the furnace is the most
important factor limiting heat release rates in coal-fired installations.
Typical rates which are generally considered good practice today for
various types of steam-generating installations are presented in Table 7-2
for comparison.9 It has been found good practice to provide plenty of
combustion space in which to burn the combustible gases, especially
Table 7-1. Boiler Heat Balances for Typical Installations*
(In per cent of heating value fired)
'
Coal Fuel oil Natural gas
Good Fair Poor Good Fair Poor Good Fair Poor
Heat used in generating

steam 89.8 84.0 61.9 87.6 82.7 63.4 84.0 79.1 60.8
Heat loss to dry flue gases . . . 5.5 8.4 20.5 5.1 8.1 20.0 4.8 7.6 19.3
Heat loss to unburned coal . . 0.4 1.5 5.0
Heat loss to unburned gas
(CO, etc.) 0.2 0.4 5.0 0.2 0.6 6.0 0.2 0.6 5.0
Heat loss to water vapor in
products 2.9 3.2 3.6 5.9 6.1 6.6 9.8 10.2 10.9
Miscellaneous and unac
counted-for losses 1.2 2.5 4.0 1.2 2.5 4.0 1.2 2.5 4.0
Based on flue-gas tempera
ture, F 300 400 600 300 400 600 300 400 600
Based on excess air, % 20 30 100 15 30 100 15 30 100
* Bailey Meter Company.
with pulverized coal, oil, and gas where the fuels are completely burned
in suspension. With pulverized coal especially, considerable time is
required for complete burning, and the gases must be retained in the
furnace long enough to achieve this. For short periods of time, firing
rates may substantially exceed those listed in the table. As already
pointed out, a design heat release rate is a compromise, and for certain
conditions some factors take priority. For instance, in marine practice
the need for keeping the volume and weight of the furnace low is more
influential in design than very low maintenance.
Data pertinent to the design of combustion chambers for single-retort
underfeed stoker-fired boilers are presented in the "Technical Manual,"
2d ed., Stoker Manufacturer's Association, Chicago, 1947.
Atmospheric Pollution. During the past 15 years an increased interest
has been shown by municipalities toward decreasing the atmospheric
pollution which commonly abounds in congested urban areas. The

problems of atmospheric pollution are complicated by their relationship
to topography of the area, prevailing and local wind and weather condi
tions, formation of "smog," etc. The latter is smoke-induced fog result
ing from condensation formed with smoke and dust particles as nuclei.10
Thus, there are many factors to be considered by those engineers attempt
ing to bring about a clean, clear, atmosphere in an industrial community.
Contributing factors to atmospheric pollution include excessive dust and
odors from a variety of manufacturing processes, but the ash-laden
products of incomplete combustion are the leading offenders.
Table 7-2. Typical Heat Release Rates*
Type of combustion Heat release,

Remarks
equipment Btu/(hr)(cu ft)
Gas burner 25,000- 30,000 Conservative

Oil burner 25,000- 30,000 Lower figure for natural draft
50,000-100,000 Merchant marine boilers, higher in naval
boilers
13,000 With refractory walls
Vertical firing 8,000- 10,000 With natural draft
15,000- 21,000 Depending on ash-fusion temperature
Horizontal firing 30,000 With high ash-fusion-temperature coal
Tangential firing 35,000 With high ash-fusion-temperature coal
Cyclone furnace 50,000 Computed on basis of entire furnace
volume
500,000 Computed on volume of cyclone chamber
Spreader stoker 20,000- 30,000 Conservative to reduce fly ash
60,000 In Scotch marine boilers
Traveling-grate stoker . . . 40,000
Underfeed stoker 35,000 Multiple retort
45,000- 55,000 With refractory walls, single retort
* Figures given are those considered to be reasonable design practice for continuous operation of
steam-generating furnaces using forced draft and waterwall cooling, unless otherwise noted.
There are two primary sources of atmospheric pollution from furnaces :

(1) from smoke, resulting from unburned carbon and volatile matter, and
(2) from ash and other dust particles which rush out the stack with the
combustion gases. The degree of blackness of the combustion gases
from a chimney can be measured with the Ringelmann type of chart,
illustrated and explained in Fig. 7-29. Many smoke ordinances state
that it is illegal to produce smoke denser than a No. 2 on the smoke-
density chart with the additional provision that smoke corresponding to
a No. 3 may be permitted twice a day for periods not to exceed 30 min.
The latter provision gives the operator a little leeway during the times
when the ash and soot are blown down and the fires are cleaned of clinkers.
These operations tend to produce extra smoke.
Smoke-free combustion is now a recognized goal of operating and com
bustion engineers, not only to meet the legal requirements of smoke-abate
ment ordinances but to gain optimum operating efficiencies. Black
smoke can be obtained while fuel oil, gas, or coal is burned; but coal, and
especially high-volatile bituminous, is the most difficult to burn smoke-
Fig. 7-29. Ringelmann type of chart for grading smoke density. The observer should
stand about 100 to 1000 ft from the stack and hold the chart at arm.s length. The'jet of
flue gases issuing from the stack is viewed through the hole in the chart. The observer
determines the shade in the chart most nearly corresponding to the shade or density of the
smoke. Care must be taken to prevent having either bright sunlight or dark buildings in
the background. (Courtesy of PHbrico Jointless Firebrick Co.)
lessly. However, with properly designed and operated equipment,

essentially smoke-free combustion can be attained with any fuel if the
principles of proper control are followed: adequate turbulence of combus
tion gases with overfire air, sufficient ignition temperatures for these gases,
sufficient furnace volume to permit time for burning, and proper propor
tioning of air and fuel.
The removal of fly ash and dust from combustion gases may be more
difficult and costly than the elimination of smoke. All types of coal com
bustion equipment emit more or less fly ash, but the problem is especially
bad with spreader-type stokers and pulverized-fuel furnaces. In the
latter as much as 85 per cent of the ash may rise with the combustion
gases through the furnace and boiler. Much of it settles out as slag on
the boiler heating surfaces, necessitating frequent cleaning with soot
blowers, but the finer particles continue on through the boiler and out the
stack with the flue gases. As the flue gases rise, are cooled, and lose their
velocity, the ash particles begin to settle out and blanket the vicinity of
the plant with a layer of gray-black fly ash. Such a plant is a nuisance.
The recent trend is to clean up such offenders.11 An ASME "Model
Smoke Law" recommends that fly ash be reduced to 0.85 lb of fly ash per
1000 lb of flue gas, which corresponds to 0.257 grain of ash per cubic foot
of flue gas at 500 F.12
Plants may do much to eliminate the fly-ash problem by
1. Changes in operating conditions,
a. Achieving proper adjustment of firing equipment
b. Reducing firing rates
c. Using a higher grade coal having a smaller percentage of fines

and a lower ash content
2. Changes in firing equipment
3. Installation of a fly-ash collector
Figure 7-30 shows typical data on fly-ash carry-over from a multiple-
retort underfeed stoker at various firing rates. In general practice,
underfeed stokers when operated at firing rates not exceeding 25 to 30 lb
of coal per square foot of grate per hour stay within the ASME sample
ordinance. The same may be said for traveling-grate stokers and hand-
fired furnaces. Thus, for conservatively designed equipment of these
types, little nuisance from fly ash should be experienced. When higher
firing rates are used frequently, some type of fly-ash collector may be
required. With spreader stokers operated at the usual design rating of
30 to 40 lb of coal per square foot of grate per hour, fly-ash emission
averages 1.0 to 2.0 grains per cubic foot of flue gas.13 This constitutes a
definite nuisance. Furthermore, even at lower firing rates, the ash emis
sion is reduced but little, because of the method of burning much of the
coal in suspension. Thus, most spreader-type stokers should be installed
with high-efficiency ash collectors to meet reasonable atmospheric pollu
tion ordinances.
Fly-ash Collectors. Dust and fly-ash collectors are built in many
designs, but all operate on some combination of the following principles:
1. A quick change in direction past a baffle separates the heavier solid
particles from the gases.

2. Centrifugal force from a vortex flow separates the heavy dust.
3. A sudden reduction of gas velocity permits the solid matter to settle.
4. Electrostatic precipitators (of the Cottrell patent type) collect ash
on charged plates.
The efficiency of any collector is measured in terms of the percentage of

solid matter removed in one pass
through the separator to the total
amount originally present. The
OUtLEt HEAD
various designs of collectors are
more or less efficient, depending
upon the relative amount of cen
trifugal action, direction changes,
sudden increases and decreases of
gas velocity, etc., that the gases
experience in flowing through the
separators. In general with me
chanical separators, the higher the
UPPER CONE
percentage of collection, the greater
the draft loss.
A low-draft-loss type of collector
requires about 0.1 to 0.2 in. of
1.4,
1
., 1 1
UPPER CYLiNDER MiDDLE CONE
V4 x 0 Pocahontas coal
1.2
Fly ash size
1
1
o Burning Passing
o rate 325 mesh
m MIDDLE CYLiNDER
10; 23.6 35.8
o
39.0 28.9
I 0.8
- 35.5
41 6
30.9
I* T
DUST OUTLEt LOWER CONE
i0.6
.0.4
, 0.2 BLAST GATE RECEPtACLE
20 25 30 35 40 45
Burning rate,lb/sqft/hr
Fig. 7-30. Fly-ash carry-over from a mul Fia. 7-31. A higli-draft-loss dust collector.
tiple-retort underfeed stoker. (Courtesy of (Courtesy of American Blower Corp.)
Bituminous Coal Research.)
water draft and will ordinarily give about 50 to 60 per cent efficiency on
the type of fly ash normally encountered. Sufficient natural draft is
often available to operate such units. Low-draft-loss collectors are com
monly used with underfeed and traveling-grate stokers where the fly-ash
problem is not severe.
Collection efficiencies of 85 to 90 per cent may be obtained with high
draft-loss collectors, but induced-draft fans are usually required to over

come the 1.5 to 2.5 in. of water draft loss created. Figure 7-31 shows the
principle of operation of one of the many designs of high-draft-loss col
lectors on the market. These units are generally used with pulverized-
fuel-fired furnaces and spreader stokers, and occasionally with underfeed
and traveling-grate stokers which are operated at high firing rates.
The electrostatic precipitator is capable of efficiencies of 90 to 95 per
cent, but it is much higher in first cost than the mechanical type of
separators. Where an unusually bad fly-ash problem exists or unusually
stringent removal requirements are to be met, an electrostatic precipitator
may be installed in series with a mechanical type. If the mechanical
collector removes 85 per cent of the fly ash, the size of electrostatic unit
required to achieve a given over-all collection efficiency is much smaller
than if the electrostatic type were used alone.
Problems
1. Which components of a steam generator would be needed in the simplest unit
possible and which are added to increase efficiency, capacity, and control?
2. Make a diagrammatic sketch of a complete modern steam generator showing
the direction of flow of gases, fuel, water, and steam in each element and showing
reasonable temperatures and pressures throughout the unit.
3. Discuss the advantages and disadvantages of overfeed, underfeed, and pulver-
ized-coal firing.
4. What is stratification and what may be done to reduce its effects?
5. Compare the merits of the various methods of overfeed firing.
6. What purposes does the firing arch serve with a traveling-grate stoker?
7. How important is "fuel flexibility" for coal-burning equipment?
8. What basic principles does the BCR smokeless heater apply to achieve cleaner
combustion than former heaters of comparable size?
9. List the advantages and disadvantages of the three firing methods for pulverized
coal.
10. List the advantages of the cyclone firing principle over pulverized firing.
11. List, in the order in which they might practically be considered, the changes
which might be made to improve the efficiency of an existing steam generator which
tests 68 per cent efficiency.
12. What would be the advantage of an O2 meter versus a CO2 meter for combustion
control of a furnace utilizing gas, oil, and coal in varying percentages?
.13. Discuss the principles of furnace design and operation which permit some types
of combustion equipment to operate with a lower percentage of optimum excess air
than others.
14. What is the significance of the heat release rate?
16. What are the two primary sources of atmospheric pollution from combustion
equipment?
16. Discuss the ramifications of the three T's of good combustion so far as they can
be employed to reduce atmospheric pollution.
17. Discuss the merits of the different pieces of firing equipment from the standpoint
of clean, smoke-free operation.
18. The following data were observed during four different efficiency tests of steam
generators. Compute the heat balance for each.
Test A Test B Test C Test D

1. Proximate analysis of fuel, as fired,
CI •
%.
7.93 8.2 12.0 3.3
Volatile matter 36.28 36.1 4.2 20.5
Fixed carbon 45.21 48.7 65.8 70.0
Ash 10.58 7.0 18.0 6.2
2. Ultimate analysis of fuel, as fired,
% •
65.96 68.4 66.8 80.7
5.28 5.6 2.5 4.9
Oxygen 14.78 16.4 11.2 5.3
1.26 1.4 0.9 1.1
Sulfur 2.14 1.2 0.6 1.8
Ash 10.58 7.0 18.0 6.2
3. Heating value of coal, as fired , Btu. 11,689 12,160 10,200 14,310
4. Analysis of dry refuse, % :
Combustible 28.58 23.1 14.2 3.0
Ash 71.42 76.9 85.8 97.0
5. Orsat analysis of flue gases, % :
C02 12.21 11.7 14.1 16.3

(h 6.82 7.6 5.8 2.1
*
CO 0.07 0.2 0.1 0.5
N, 80.90 80.5 80.0 81.1
6. Pressure of steam leaving unit, psig 141.3 721 190 2400
7. Temperature, air and coal initial,
F 73.9 92 77 85
8. Temperature, flue gases leaving, F. 547 491 425 280
9. Temperature, feedwater entering,
F 191 239 172 427
10. Temperature, steam leaving unit,
F.. . 620 1050
Steam leaving throttling calorim
eter, F 281 291
11. Pounds water evaporated per hr. . 4611.2 119,000 59,140 1 . 121 X 106
12. Pounds coal fired per hr 563.1 13,560 7,560 96,500
Calculate (per pound of coal fired):

Ans. for Test A
a. Per cent excess air supplied 46 . 1
6. Weight of combustible in refuse, lb 0 . 0423
c. Weight of carbon burned, lb 0 . 6173
d. Weight dry flue gases formed, lb 12.66
e. Heat absorbed by water and steam, Btu 8393
/. Heat loss due to unconsumed combustible, Btu 618
g. Heat loss to dry flue gases, Btu 1437
h. Heat loss to carbon monoxide, Btu 35
i. Heat loss to moisture in fuel, Btu 100
j. Heat loss to water from combustion of hydrogen, Btu 500
k. Miscellaneous and unaccounted-for losses, Btu 606
I. Heating value of fuel, Btu (as given) 11 ,689

Compute the heat balance for an oil-fired boiler, given the following:
19.
Firing rate is 200 gal oil per hour having an API gravity of 21.0 and H/C = 0.172
(by weight).
Air and oil are fired at 80 F, flue gases leave at 450 F, C02 meter averages 11.2%
during the test. (Refer to Fig. 7-28 and assume complete combustion.) Feedwater
enters at 210 F, steam leaves at 700 F, 410 psig, 19,800 lb of water are evaporated per
hour.
20. Ethyl alcohol is burned with 15 per cent excess air. Compute the percentage of
the higher heating value lost to the dry flue gases if the air and fuel are ignited at 77 F
and the flue gases are at 480 F.
21. Compute the heat lost to the dry flue gases per cubic foot of fuel fired when
methane burns with 18 per cent excess air. Gas and air are initially at 82 F, 14.8 psia;
flue gases leave at 340 F.
22. Same as above except calculate the heat loss to the water formed from the
combustion of hydrogen. What percentage of the higher heating value is lost?
23. Plot a curve showing the heat loss due to the formation of carbon monoxide
versus the percentage of theoretical air, through the range of 70 to 100 per cent
theoretical air, for the combustion of C16H34. Assume complete mixing of fuel and
air and disregard any dissociation effects.
24. Compute the percentage of heat loss due to the formation of carbon monoxide
if the Orsat analysis of the combustion products of butane reads: CO2 = 13.4%,
02 = 0.2%, CO = 0.4%.
References
1. Arthur, J. R., D. H. Bangham, and M. W. Thring, Combustion in Fuel Beds,

J. Soc. Chem. Ind., vol. 68, p. 1, January, 1949.
2. Kreisinger, Ovitz, and Augustine, U.S. Bur. Mines, Bull. 137, 1917.
3. Hiles, J., and R. A. Mott, The Mode of Combustion of Coke, Fuel, vol. 23, 1944.
4. Hottel, H. C, and I. M. Stewart, Space Requirements for the Combustion of
Pulverized Coal, Ind. Eng. Chem., vol. 32, p. 719, 1940.
5. Sherman, R. A., Burning Characteristics of Pulverized Coals and the Radiation
from Their Flames, Trans. ASME, vol. 56, p. 401, 1934.
6. Sherman, R. A., Proc. Third Int. Conf. on Bituminous Coal, vol. II, p. 510, 1931.
7. Orning, A. A., The Mechanism of the Combustion of Pulverized Fuel, Carnegie
Inst. Technol., Coal Research Lab., Contrib. 137, 1947.
8. Fairmont Coal Bureau, The Cyclone Furnace, Reference Bull. 12, 1949.
9. de Lorenzi, O., "Combustion Engineering," Combustion Engineering — Super
heater, Inc., 1949.
10. Newberger, H., Condensation Nuclei, Mech. Eng., vol. 70, p. 221, March, 1948.
11. Barkley, J. F., Air Pollution Prevention in the United States, Mech. Eng., vol. 73,
p. 284, April, 1951.
12. "Model Smoke Law Committee. Example Sections for a Smoke Regulation
Ordinance," paper presented before the ASME, P'uels Division, May, 1949.
13. Miller, C. E., What Can the Small Plant Do about Fly Ash? Bituminous Coal
Research, April, 1949.
14. Rowley, L. N., J. C. McCabe, and B. G. A. Skrotzki, Fuels and Firing, Power,
vol. 92, p. 735, December, 1948.
CHAPTER 8
COMBUSTION IN ENGINES
Liquid and gaseous fuels burn with an explosive flame with approxi
mately a constant-volume reaction in spark-ignition or compression-
ignition engines; in gas turbines or rockets they burn with a stationary
type of flame at approximately constant pressure.
SPARK-IGNITION ENGINES
The operation of a spark-ignition engine is on either the so-called four
cycle or the two-cycle principle. The four-cycle engine requires four
strokes of the piston to complete the cycle: (1) intake, (2) compression,
Spark plug
I. Intake 2. Compression 3. Power 4. Exhaust

Fig. 8-1. Four-stroke-cycle spark-ignition engine.
(3) power, and (4) exhaust. The cross section of such an engine is
shown diagramatically in Fig. 8-1 with the intake and exhaust valves in
their correct positions for each stroke. The intake valve is open during
the intake stroke, while a new charge of fuel and air is drawn, into the
cylinder, and remains closed during the remainder of the cycle, while the
exhaust valve is open only on the exhaust stroke during the expulsion of
the exhaust gases. The two-cycle engine completes the same cycle in
two strokes of the piston. In Fig. 8-2 a simple type of two-cycle engine
is shown. During each upward stroke of the piston a mixture of fuel
288
COMBUSTION IN ENGINES 289
and air is pulled into the crankcase. On the downward stroke it is

compressed so that, when the piston uncovers a transfer port leading to
the cylinder above, the mixture is forced up into the cylinder. When this
port is covered as the piston moves up, compression occurs and the power
stroke follows. Exhaust occurs when the exhaust port is uncovered and
while a new charge is being forced in from the crankcase. Each time the
piston moves up on a compression stroke, a mixture of fuel and air is
pulled into the crankcase, while on each power stroke this new charge
is being compressed ready to enter the cylinder when the ports are
uncovered. In both the two- and the four-cycle engines, the compressed
charge of fuel and air is ignited by an electric spark near the end of the
compression stroke, and the nearly homogeneous mixture is burned
approximately at constant volume.
Mixing Fuel and Air. Liquid fuels, such as gasoline, are generally
mixed with air in a carburetor, and this mixture is carried through a mani
fold to the intake valve of each cylinder. A simple carburetor is shown
in Fig. 8-3. It consists of & float chamber, where a constant supply of fuel
is maintained, and a jet, connected to the float chamber, through which
L Intake 2.Compression 3. Ignition and power 4. Exhaust

Fig. 8-2. Three-port type of two-cycle engine.
the fuel flows and mixes with the air passing through the venturi and the
mixing chamber. The air/fuel ratio is controlled by the relative sizes of
the jet and the air passage. In some modern designs, a fuel injection
system is used in place of a carburetor. In this case, the fuel/air mixture
is obtained by injecting the desired amount of fuel into each intake port
near the intake valve, or directly into the cylinder, during the intake
stroke. Gaseous fuels may be mixed with the air in a mixing valve out
side the cylinders, or they may be injected directly into each cylinder.
The air/fuel ratio is controlled by adjusting the gas flow rate.
Normal Combustion Process. The actual ignition of the fuel/air mix

ture taken into a spark-ignition engine on the intake stroke' does not
occur until 10 to 40° of crank-angle rotation before the end of the compres
sion stroke. During the compression stroke the pressure and the tem
perature of the air/fuel charge are increasing. This increase in tempera
ture may initiate very marked prereactions prior to the spark ignition.
After these reactions have begun, the additional increase in temperature
promotes more reactions, such as cracking, dehydrogenation, and poly
merization. 1 A portion of the fuel is changed to oxides of carbon, water,
Fuel supply line
Venturi
Fuel \>\ Float pivot pin

'discharge v>
nozzle
Float chamber
Mixing —
chamber
-Throttle valve
Fig. 8-3. Schematic diagram of a simple float-type carburetor.
and oxygenated products during this prereaction period, thereby reducing

the thermal value of the charge in the cylinder. The extent of these pre
reactions is a function of the fuel and also of the engine operating vari
ables. All paraffinic fuels undergo these prereactions to a certain extent,
but isooctane is much more resistant than normal heptane. Tetraethyl-
lead appears to have little effect on the tendency of paraffinic fuels to
react during compression, but it delays the beginning of autoignition.23
Aromatic fuels are fairly stable and resist prereaction. These prereac
tions increase with the severity of engine operation, increased intake air
temperatures, and higher compression ratios. It seems logical to con
clude from these facts that precombustion reactions are related to and
may promote the undesirable phenomenon of detonation.
Spark ignition of the compressed fuel/air mixture forms a number of
chain carriers through heating and molecular excitation. Following a

short ignition delay the length of which is dependent upon the fuel,
temperature, pressure, and rate of compression, a small sphere of gas is
ignited. After the gases are ignited, the flame travels through the com
bustion chamber somewhat radially from the spark plug. The spatial
velocity of the flame front is dependent upon the combustion chamber
design and size, temperature, fuel, air/fuel ratio, turbulence, humidity,
etc., as discussed in Chap. 4.
Fuel Volatility. Liquid fuel must be vaporized and mixed with air
before combustion can be initiated by a spark. Fuels with low distilla
tion temperatures, therefore, tend to burn more readily than those
requiring higher temperatures for evaporation. This is particularly true
when a cold engine is being started. The temperature at which an engine
may be started is generally considered to depend upon the temperature
at which the first 10 per cent of the fuel is evaporated. A fuel which has
50 per cent evaporated at a relatively low temperature tends to give
improved general engine performance, while the 90 per cent evaporation
temperature is a direct indication of the amount of fuel likely to be carried
over into the crankcase lubricating oil. These factors have already been
discussed at length. When kerosene is used in a spark-ignition engine, it
is often found necessary to start the engine on gasoline and change to
kerosene after the cylinders and manifold are heated to their operating
temperatures.
Turbulence. The speed of combustion under extremely quiescent
conditions in a glass tube is shown in Fig. 4-14, where the maximum speed
obtained when burning methane and air in a 6-in. -diameter tube was less
than 4 ft/sec. Such a flame speed would never permit the operation of
an internal-combustion engine. In a high-speed engine, the flame front
must traverse the combustion chamber in from 0.001 to 0.003 sec to per
mit economical operation. For example, if an engine is running at 4200
rpm, the crankshaft turns 25° per 0.001 sec, and much of the combustion
should occur within about 30° of crank-angle rotation to obtain maximum
power and economy. Thus, in an engine of average size, the flame must
traverse the combustion chamber at speeds up to 100 ft/sec. Turbulence
is the primary factor in accelerating combustion to these velocities. This
is accomplished by increasing the rate of diffusion of heat and active reac
tion centers into the unburned gases. Turbulence is obtained to a great
extent from the shape of the combustion chamber and also from the
velocity and direction of flow of the gases entering the cylinder.
As was shown in Fig. 4-22, flame velocity is proportional to the intake
velocity, however excessive intake velocities materially reduce the weight
of the fuel and air charge taken into the cylinder. It was also shown that
approximately the same crank angle is required for inflammation through

out a wide range of engine speeds. So as to obtain high turbulence with
out high intake velocities, combustion chamber designs which produce
turbulence near the end of the compression stroke are generally used.
In Fig. 8-4, three types of combustion chambers are shown. A valve-
in-head type is shown at A with a flat head and a flat piston. In this
type little turbulence is created. At B the cylinder head is nearly flat,
but the piston is quite unsymmetrical and, as the piston approaches the
head, the gases are forced to the left at a high velocity and thus create
turbulence. Even higher velocities and turbulence are produced in the
L-head construction shown at C since the clearance between the piston
ABC
Fig. 8-4. Turbulent and nonturbulent combustion chambers.
and the cylinder head is generally much less than in B. Also, some
turbulence is created by the relative direction of flow into the cylinder.
The inlet ports in two-cycle engines may direct the gases in a tangential
flow pattern, rather than radial, and thus produce turbulence.
Detonation. As was stated in Chap. 3, detonation is a special type of
explosion characterized by a tremendously increased reaction rate accom
panied by an ultra-high velocity percussion wave within the cylinder and
an abnormal rate of pressure rise. Detonation is accompanied by
increased peak pressures and very high localized temperatures. This
knock reaction also results in a definite loss of chemical energy. Detona
tion may vary from the incipient stage to a very high intensity, and con
tinued operation under conditions which promote high-intensity knock
may prove very detrimental to an engine. The effects of knock on engine
operation and parts are shown in Table 8-1.
Autoignition Theory. For several years a simple detonation theory
has been generally accepted that knock is caused by the autoignition of
the end gases. According to this theory after the charge is ignited, the
flame front travels somewhat radially through the combustion chamber,
Table 8-1. Effects of Detonation on Engine Operation
Brake mean effective Knock Engine Effect on

pressure, or load intensity operation engine
High
Very Scored cylinders
rough
Heavy Cracked pistons and
cylinder heads
Fairly
rough Ring scuffing
Medium
knock Burning of alumi
num pistons
Light
knock Smooth
Incipient None
knock
No knock
Low
raising the temperature of the unburned gases by adiabatic compression

due to the expansion of the burning gases behind the flame front. When
the temperature of the unburned gases is well above their autoignition
temperature, all of them burn with an explosive reaction. This is shown
diagrammatically in Fig. 8-5. The spark has occurred, and the flame
Fig. 8-5.
ABC
Progression of combustion in a spark-ignition engine, ending in autoignition.
front has just started to progress radially from the spark plug in (a).
The flame has traveled more or less radially, but conforming to the shape
of the combustion chamber across a large portion of the chamber in (b).
Finally, (c) shows the last part of the charge starting to autoignite at a
great many points before the flame front has reached it. The violent
combustion of this last part of the charge sets up the pressure differential
which results in knock. Autoignition as it occurs here and in a compres

sion-ignition engine is sometimes confused with preignition, but they are
not the same. Preignition is the inflammation of the gases in a spark-
ignition engine before the spark occurs. It may occur in overheated
engines where sufficient heat is supplied by local hot spots, such as spark
plug points, exhaust valves, or incandescent carbon deposits, to cause
ignition. Preignition may occur without causing detonation.2 How
ever, it is very detrimental to the engine as well as to power output. Pre
ignition tests on an aircraft engine showed that cylinder-head tempera
tures increased as much as 200 F, the maximum cylinder pressures
increased from 10 to 30 per cent, while the power decreased from 65 to
100 per cent. With all air/fuel ratios, except those richer than 11 to 1,
the preignition caused back firing in the intake manifold.
Detonation Wave Theory. The autoignition of the end gases in a com
bustion chamber requires from 2 to 5 crank-angle deg (400 to 1000 micro
seconds),4 while the detonation wave reaction requires but 25 to 50
microseconds. 2 These detonation waves travel at 3000 to 6500 ft/sec, or
one to two times the speed of sound. These high velocities have been
confirmed with pictures taken at 40,000 and 200,000 per second.2 The
abnormal noise or roughness of operation of an engine, many times called
knock, has different forms:
1. Knock which has a comparatively low pitch is caused by simple
autoignition of the end gases at a rate too slow to produce audible gas
vibration.
2. Knock that produces both low- and high-pitched tones may be
caused by autoignition followed by the formation of a detonation wave
in the autoignited gases.
3. Knock of a high pitch may be caused by a detonation wave origi
nating in the afterburning gases behind the flame front, and may pass
through the unignited end gases.2 This detonation-wave reaction is
evidently of the chain-branching variety which results in an extremely
violent explosion.
Most of the research on engine knock or detonation indicates that the
so-called high-pitch metallic sound is the result of a detonation wave (3,
above), or a combination of autoignition and the formation of a detonation
wave (2, above), and may occur in either form.
Factors Influencing Detonation. There are many factors which affect
the knocking, or detonating, characteristics of a spark-ignition engine.
However, they influence detonation only in so far as they vary either the
fuel, the temperature of the mixture, or the pressure of the mixture. Table
8-2 lists most of the influencing factors.
The effect of the fuel on detonation depends wholly upon its octane
number; the higher the octane number, the less the detonation. The
knocking tendencies of various liquid fuels were discussed in Chap. 5.
The intensity of detonation in an engine is directly dependent upon the
mixture temperature, and any factor in engine design or operation which
will cause the mixture temperature to increase will increase the severity
of detonation. All combustion reactions are accelerated by an increase in
temperature, as was pointed out in Chap. 4.
Table 8-2. Factors Influencing Detonation

Fuel Temperature of the mixture Pressure of the mixture
Octane number Air /fuel ratio Air /fuel ratio
Initial air temperature Initial air temperature
Engine temperatures Engine temperatures
Load Load
Ignition timing Ignition timing
Compression ratio Compression ratio
Turbulence
Engine speed
Spark-plug location
Combustion chamber, size and shape
Humidity
Higher cylinder pressure will also cause an increase in the intensity of

detonation. As the pressure increases, the temperature of the mixture
rises and the speed of the reactions may be increased by both a thermal
effect and a chain-branching effect. Also, an increase in pressure
will lower the autoignition temperature of the fuel, and the chain reaction
path followed during the combustion reaction may be changed at the high
pressures.
The temperature and pressure of the burning mixture are both influ
enced by the air/fuel ratio and will have their peak values with the most
powerful mixture, which is a slightly rich mixture. Either richer or
leaner mixtures will be slower burning, will thus cause a decrease in
temperature and pressure, and, therefore, will lower the intensity of
detonation. Very rich mixtures suppress detonation, apparently because
of their influence on the reaction paths followed.
An increase in temperature of the incoming air, or of the engine, will
also cause an increase in both the compression temperature and the pres
sure. An increase of engine load at a given speed, measured by the
change in brake mean effective pressure (bmep), has the same effect on
the mixture temperature and pressure. This increase might be obtained
by an increase in throttle opening or by an increase in supercharger
pressure.
Ignition timing also affects the mixture temperature and pressure.
Advancing the ignition timing causes an increase in both the temperature
and the pressure because more of the mixture is burned before top dead
center, which increases peak temperatures and pressures. The power
output, however, reaches an optimum at an intermediate spark advance.
Increase of the compression ratio, while increasing the power output and
the thermal efficiency of an engine, also causes an increase in mixture
temperature and pressure.
The temperature of the end gas, or the last of the unburned mixture, is
dependent upon the rate of burning. When the rate of burning is slow,
the end gases have more time to absorb heat from hot surfaces and radia
tion from the approaching flame front. One of the chief factors in pro-
All factors increase *~

Fig. 8-6. Influence of various factors upon the intensity of detonation.
moting fast burning is turbulence. Turbulence tends to break up a slow-

moving, uniform flame front, by carrying the flame quickly through the
combustion chamber to the unburned mixture (Fig. 4-21). The shorter
the time required for the completion of combustion, the less the tendency
for detonation. As was pointed out under Turbulence above, an increase
in engine speed causes a proportional increase in turbulence with a result
ing proportional decrease in the time required for flame front travel
across the combustion chamber. This resultant decrease in combustion
time causes lower temperatures to exist in the end gases and a correspond
ing reduction in detonation tendency at higher engine speeds.
When the spark plug is located in the position which will result in the
lowest temperature of the end gases, there will be the least tendency for
detonation to occur. The shape and size of the combustion chamber have
a decided effect on knock intensity. The size influences the distance the
flame front must travel, while the distance and the degree of turbulence
are dependent upon the shape of the combustion chamber.
The trends of most of the above factors upon the intensity of detona
tion are shown in Fig. 8-6.
Injection for Cooling and Detonation Suppression. Water injection
has been used in kerosene engines for over 50 years as a cooling agent
I I I I J—51 I I I I i 1 I I I I I 186
.04 .05 .06 .07 .08 .09 .10 .11 .12
Fuel/Air Ratio ( Measured )

Fig. 8-7. Exhaust-gas composition versus measured fuel/air ratio for a supercharged
aircraft engine burning octane. (Gerrish and Meem, NACA Rept. 757, 1943.)
for suppressing detonation and producing smooth engine operation. In

automotive engines, the addition of water to the fuel/air mixture is not a
direct means of increasing either power or economy. The water lowers
the maximum cylinder pressure and temperature, resulting in smoother
operation, but does not produce more power or better economy. In cases
of severe detonation, water will cool the combustion chamber and reduce
detonation, thus the heat released from combustion increases slightly, but
the latent heat of the water may cause an over-all reduction of the net
power.
Water mixed with alcohol, ammonia, etc., is commonly injected into
aircraft engines. These mixtures tend to reduce detonation but are used
primarily as cooling agents to permit the operatioa of the engine at abnor
mal manifold pressures (emergency power output) without overheating.
As was shown in the examples in Chap. 2, the air/fuel ratio may be
calculated if the analysis of the exhaust gases for a spark-ignition engine
is known. Figure 8-7 shows the composition of the exhaust gases of an
aircraft engine for a wide variation of fuel/air ratios. It is seen that the
CO2 peaks approximately at the stoichiometric ratio, but some oxygen is
present when rich mixtures are used and some combustible gases still
exist when the fuel/air mixture is lean.
COMPRESSION-IGNITION ENGINES
There are several marked differences between compression-ignition and

spark-ignition engines. The compression-ignition engine (1) takes only
air into the cylinder on the intake stroke in place of fuel and air, (2) uses
compression ratios of about 15 to 1 instead of 7 tol, thereby producing
compression temperatures above the autoignition temperature of the
fuel, (3) has the fuel injected into the cylinder at approximately the same
time in the cycle that the spark occurs in a spark-ignition engine, and
(4) uses no carburetor or ignition system.
The two- and the four-stroke-cycle principles of operation are used
in compression-ignition engines. The two-cycle compression-ignition
engine may use the crankcase for the partial compression of the air,
although usually the air is supplied by a low-pressure compressor, as
shown in Fig. 8-8. The air from the compressor enters the cylinder
through ports uncovered by the piston and forces the burned gases out
past the exhaust valve at the top of the cylinder. The exhaust valve
opens on the power stroke before the intake ports are uncovered, and
closes about the same time that the intake ports are closed on the com
pression stroke. Fuel is injected a few degrees before the piston reaches
top dead center.
The four-cycle compression-ignition engine employs the conventional
four strokes per cycle of intake, compression, power, and exhaust, as
shown in Fig. 8-1. The air inducted on the intake stroke is either nor
mally aspirated or forced in by a supercharger, while the fuel is injected
into the cylinder near the end of the compression stroke. In most
present-day engines, the combustion chamber temperature at the end of
the stroke is approximately HOOF.6

compression This temperature is
dependent upon the compression ratio and the initial air temperature.
Near the end of the compression stroke, fuel is sprayed into the com
bustion chamber at pressures varying from about 1200 to over 30,000 psi.
The injection pressure is governed by the engine speed and size and by
the type of combustion chamber and injection system used.
Fig. 8-8. Cross section of a two-cycle compression-ignition engine using a blower to supply
scavenging air. (Courtesy of Detroit Diesel Engine Division, General Motors Corp.)
Combustion Process. With the beginning of fuel injection, the process

of combustion is under way. Each charge of fuel experiences several
phases in the reaction as follows :
1. A delay period which covers the time from the initial injection to the
actual ignition of the fuel and air
2. A period of rapid combustion of the fuel accumulated during the
delay period
3. A period of combustion of the remainder of the fuel charge as it is
injected
4. An afterburning period in which the unburned fuel may find oxygen
and burn
In order to study the paths which may be followed by any fuel particle,
an outline6 is shown in Fig. 8-9. In following the paths of the fuel
particles, it should be kept in mind that, after ignition has occurred, many
of these steps will be proceeding at the same instant. This is true since
Mixture -
Ignition - 0 Final combustion -
formation
Delay period -
-Physical delay-»j*Chemical delay*]
Disintegration
of stream of —
1
I Inflammation
injected fuel
1
Mixing of Preflame Oxidization Thermal
liquid fuel oxidation of fuel-air decomposition
with air of fuel mixtures of fuel
Vaporization Local Mixture of

of fuel ignition products of
partial oxidation
Mixing of fuel or of thermal
,vapor with air decomposition
with air
Temperature Temperature
and oxygen and oxygen
concentration concentration
favorable unfavorable
for complete for complete
combustion combustion
(chilling,
overlean,
and overrich
mixtures)
1 *
Products of Products of
complete incomplete
combustion combustion
Fig. 8-9. The combustion process in a compression-ignition engine. (Elliott; SAE Trans.,
July, 1949.)
the mixture is very heterogeneous — consisting of air, precombustion and

combustion reaction products in all stages, along with the continuing fuel
spray.
Delay Period. The delay period consists of a physical and a chemical
delay. The physical delay is required to atomize the fuel, mix it with the
it,
air, vaporize and produce mixture of fuel vapor and air. During the
a
chemical delay, preflame oxidation reactions occur in localized regions

with a temperature increase of from 1000 to 2000 F.6 These preflame

reactions are the side reactions initiated by the catalytic effect of wall
surfaces, high temperatures, and miscellaneous particles, which form the
active chain carriers prior to rapid combustion. As the local tempera
tures increase, the fuel vapors begin to crack at an accelerating rate and
produce materials with high percentages of carbon which become heated
to incandescence as local ignition is initiated.
General inflammation develops quickly and proceeds either by rapid
and complete oxidation of the fuel and air or by the oxidation of the
intermediate products of the chain reactions of the fuel. The latter are
the aldehydes, peracids, and peroxides which are present in the chain
reactions and may result in either complete or incomplete combustion.
Factors Affecting Delay Period. The temperature of the air at the end
of compression has a major effect upon the delay period. If it is low, the
precombustion reactions are slowed materially and the delay is mostly
chemical. If the temperature is high, chemical reactions are greatly
accelerated, and the delay is chiefly physical.7 The physical delay is
practically constant and does not vary materially with temperature
changes.
An increase in compression pressure, or air density, reduces the delay
period not only by increasing the air temperature but also by lowering the
autoignition temperature of the fuel. The effects of temperature and
pressure upon ignition lag are shown in Fig. 8-10 for a straight-run mid-
continent gas oil.20,22
Increased air density, resulting from a higher compression pressure,
increases the degree of atomization of the fuel spray, and any factor which
aids in producing smaller fuel droplets will shorten the physical delay
period. Atomization is also dependent upon the spray nozzle, the
average size of the droplets being proportional to the nozzle diameter.
Injection pressure has a decided effect upon droplet size, the size decreas
ing with increases of the injection pressure. More viscous fuels result in
larger droplets.
As explained in Chap. 5, the cetanenumber of a fuel is a relative meas
ure of its ignition lag tendency. High-cetane fuels have low autoignition
temperatures and short delay periods, while the low-cetane fuels have the
opposite.
The molecular structure of the fuel has a very marked influence upon the
delay period, as shown in Fig. 8-11. The greater the number of carbon
atoms, the higher the cetane number for the various hydrocarbon groups.
For a specific number of carbon atoms, the normal paraffin gives the
highest cetane number while the corresponding olefin, isoparaffin, naph-
thene, and naphthalene give a continued decrease in rating.20 It should
be noted that the naphthalenes, which are dicyclic aromatics, in all cases
give very low cetane ratings.
Figure 8-12 shows the variation of ignition lag of three gas oils as meas
ured by the Bureau of Mines bomb test.19'22 These oils were produced
from widely different crudes. One gas oil was from the paramnic Brad
ford crude, another with a high percentage of dicyclic naphthenes was
from Hastings crude, while the third with a high percentage of dicyclic
aromatics was from Conroe crude. The aromatic portion of each oil was
850 900 950 1000 1050
Temperature, F
Fig. 8-10.Effect of pressure and temperature on ignition delay in a bomb. The fuel was
straight-run mid-continent gas oil. (Hurn and Hughes, SAE Trans., January, 1952.)
separated from the remaining paraffin-naphthene portion for individual

testing. The results of these tests are shown along with those for the
original gas oil. The aromatics are seen to have at least double, the
ignition lag of the total diesel cut, and their presence in the fuel causes
approximately 25 per cent increase in ignition lag over that shown for
the paraffin-naphthene portion.
It has already been shown in Fig. 1-19 that a diesel fuel is composed of a
large number of hydrocarbons as evidenced by the variation of boiling
temperatures. A curve of cetane numbers of the various fractions is
also shown to illustrate its variation with the varying proportions of
paraffin, naphthene, and aromatic hydrocarbons present in successive

fractions.
Turbulence was shown be of prime importance in spark-ignition
to
combustion, and it wasto vary directly with engine speed with the
seen
result that the crank angle required for combustion of the charge was
independent of engine speed. Figure 8-13 shows the same to be true for a
compression-ignition engine using several fuels having widely varying
cetane numbers.8 The engine was operated at constant throttle and a
104 r
8 10 12 14 16 18 20 22 24 26 28
Carbon atoms in molecule
Fig. 8-11. Relationship of cetane number and hydrocarbon type. (Ward and Smith,
" Why Diesel Fuels Behave As They Do,"
paper -presented before the Western Petroleum
Refiners. Association, San Antonio, April, 1951.)
constant injection advance of 20°. It is very interesting to note that the

crank angle required for each of the three combustion stages shown was
approximately constant for all engine speeds. It is also shown that
cetane number does not affect the total combustion time, but lower
cetane numbers cause longer delay coupled with increased speed of com
bustion after autoignition. The faster rate of burning of the low-cetane
fuels results in an undesirable engine roughness.9
The curves in Fig. 8-13 seem to indicate that very high engine speeds
are equally as permissible for compression-ignition engines as they are
for spark-ignition engines. Higher speed engines presuppose reductions
of piston displacements, which in turn causes an increase in the ratio of

combustion chamber surface area to combustion chamber volume. The
latter results in a decrease in temperature of the compressed air. This
disadvantage may be partially offset by the use of high-cetane fuel, a
very hot catalyzing surface in a portion of the combustion chamber to
reduce the delay period, and very high turbulence to shorten the inflamma-
850 900 950 1000 1050

Bomb temperature, F
Fig. 8-12. Influence of hydrocarbon types on ignition lag. (Hurn and Hughes, SAE Trans.,
January, 1952.)
tion period. A further means that has been used to reduce the delay
period in very small high-speed compression-ignition engines is an increase
in compression ratio to values of more than 20 to 1.
The type and shape of the combustion chamber determine to a great
extent the degree of turbulence created. It should be remembered that
any turbulence generated by the combustion chamber shape or any other
factor will be further increased in direct proportion to the engine speed.
Figure 8-14 shows representative types of combustion chambers used in
compression-ignition engines.
Open combustion chambers are shown at A and B. Here all of the air is
compressed between piston
the and cylinder head.
the In A little
turbulence is created, but, for a piston shape such as B, very high veloc
ities toward the center are created as the piston approaches closely to the
head. Sketch C is a quiescent precombustion chamber in which some of the
air is compressed through a small opening into the chamber. The air
enters at high velocity but does not tend to create a specific direction of
swirl, and the resultant turbulence is not very great. After the injection
occurs and autoignition has taken place, the increased pressure forces the
roc
4.0 3.0 2.0 1.0 0 i.O 2.0 3.0

Time, millisec
! Fig. 8-13. Effect of rpm and fuel cetane rating on the crank position at the beginning of
ignition, at the point of maximum pressure, and at the end of combustion in a compression-
ignition engine operating with a constant injection advance angle of 20 deg. {Adapted
I from Shoemaker and Gadebusch, SAE Trans., 1946.)
fuel and burning mixture down into the main chamber at a high velocity,
thus promoting intimate mixing with the compressed air over the piston,
i and the quick completion of combustion. A turbulent precombustion
chamber is shown at D. The principal difference between C and D is that
in D the air is directed in a path as shown and continues in this path at
very high velocity. When fuel is injected, this high turbulence materially
aids in mixing the fuel and air. A hot catalyzing surface is employed in
both C and D to promote combustion. In each case the lower part of the
I
precombustion chamber is so located in the cylinder head that it is not
cooled by the cooling liquid. These hot surfaces are a very decided aid
in promoting the chain reactions of combustion. One of the popular
types of combustion chamber designs is shown at E. It is generally
called an energy cell, and the auxiliary chamber may be somewhat similar
to C or D except that the fuel is injected directly into the precombustion
chambers in C and D, while the fuel is injected into the main combustion
chamber in E and directed toward the air or energy cell. Some fuel is
supposed to enter the energy cell, the combustion is supposed to begin in
the cell, and then the partially burned mixture blasts back into the main
D E
Fig. 8-14. Compression-ignition combustion chambers.
chamber aiding in the completion of combustion and creating very high

turbulence in the burning mixture.
A normal intake valve permits the air to enter the cylinder radially
from the valve stem and thus does not create a definite flow pattern for
the air. If approximately half of the valve carries a cylindrically shaped
shroud to prevent air entering around that section, the air will then be
directed in a helical path as it comes into the cylinder, and a circulatory
motion will still remain at the end of compression.
To achieve efficient and smooth engine operation, the fuel injection

must begin a sufficient crank angle before dead center to cause the high
rate of pressure rise to be in progress when the piston reaches top dead
center. A large of injection may cause rough operation similar
advance
to a low-cetane fuel. This is because the air is too cool, too low in
density, and lacking in turbulence at the beginning of injection. It has
Fuel /air ratio, pound of fuel per pound of air

Fig. 8-15. Effect of fuel/air ratio on concentration of carbon dioxide, carbon monoxide,
aldehydes, and smoke in the exhaust gases of a CFR compression-ignition engine. The
extinction coefficient represents the relative concentration of smoke. (Elliott, SAE Trans.,
July, 1949.)
been shown9 that a rough-operating engine may be smoothed by reducing

the injection advance, injecting the fuel at a higher rate, and ending
injection at the original position.
Exhaust-gas Composition. The products of combustion from a com
pression-ignition engine are much different from those found in the
exhaust of a spark-ignition engine. This results from the fact that a
spark-ignition engine operates with air/fuel ratios between 11 and 18 —
corresponding to fuel/air ratios of 0.09 to 0.05 — while a compression-

ignition engine always operates with an excess of air with fuel/air ratios
ranging from below 0.005, when idling, to 0.06.
Incomplete combustion occurs in a compression-ignition engine when
(1) the temperature at the end of compression is too low, (2) the oxygen
concentration is not sufficient to permit complete combustion, and (3) the
fuel concentration is too low for the preflame reactions to produce suffi
cient heat to promote quick and complete combustion.6 Exhaust gas
from a compression-ignition engine is composed of CO, aldehydes,
unburned or partly burned fuel, carbon, polymerized products, and
oxides of nitrogen.
In Fig. 8-15 the percentages of CO2, CO, aldehydes, and smoke inten
sity are shown for a compression-ignition engine.6 The percentage of
CO2 is very much as would be expected. However, with decreasing
values of fuel/air ratio, the increasing proportion of excess air might be
expected to produce decreasing amounts of CO, aldehydes, and smoke.
But the curves show that the percentages of CO and the aldehydes
(although very small) increase as the fuel/air ratio becomes less. Also
there is an appreciable amount of smoke between 0.003 and 0.01 fuel/air
ratios which consists of unburned or partially oxidized fuel and is seen as
a fog. When the fuel/air ratio is near 0.01, the average combustion
temperature is comparatively low, and the CO, aldehydes, and smoke are
formed in locally overlean regions from chilling of the preflame oxidation
6
reactions.
As the fuel/ air ratio is increased, causing higher temperatures, the per
centages of CO, aldehydes, and smoke decrease rapidly. When the
fuel/air ratio gets over 0.05, there is a tendency to form locally over-
rich regions resulting in an increase in CO and smoke. This smoke is
unburned carbon, formed by thermal decomposition, which did not find
oxygen to complete the reaction.
HIGH-COMPRESSION GAS ENGINE
The so-called gas-diesel, or dual-fuel, engine is increasing in popularity

for stationary installations because of the low cost of gas in comparison to
fuel oil and because the thermal efficiency of these engines is approxi
mately the same as the regular compression-ignition engine.
These engines are built either two- or four-cycle and use conventional
compression-ignition compression ratios. The gas is mixed with the air
before compression, and fuel oil is injected to initiate combustion. Only
3 to 10 per cent of the total heat is supplied by the fuel oil, and the gas/air
mixture provides the remainder of the energy. Where the compression-
ignition engine requires at least 20 per cent excess air, the dual-fuel engine
operates with approximately an over-all theoretical air/fuel ratio at full

power. This results in a corresponding increase in maximum power out
put. The exhaust is much cleaner and carries less smoke than the con
ventional compression-ignition engine.
Continued development of the dual-fuel engine has resulted in a new
type of engine wherein spark ignition has replaced the fuel-oil injection
system. Thus it is strictly a gas engine employing the same compression
ratio as the compression-ignition engine. Autoignition is prevented by
careful control of the air/fuel ratio by maintaining a slight excess of air
at all times, and by the use of gases consisting principally of the lighter
hydrocarbons such as methane and ethane.
GAS TURBINES
Combustion takes place in a gas turbine with a stationary type of

flame at approximately constant pressure. Fundamentally, a gas tur
bine consists of an air compressor, one or more burners or combustors,
Fig. 8-16. Simple can type of combustor.
and a turbine. Air is delivered from the compressor to the combustors

where the fuel is sprayed into the air and ignited, combustion takes place,
and the gases pass out through the turbine.
Many of the present aircraft gas turbines use fuels ranging from high-
grade gasolines to light fuel oils, while stationary gas turbines generally
use the heavy fuel oils because of their low cost. During the past few
years much development work has been done on the use of pulverized coal
as a fuel for stationary and locomotive gas turbines.
Kerosene and light fuel oil may have 0.015 per cent water absorbed.
In aircraft use at low temperatures, water crystals freeze out and clog the
filters. This trouble is more prevalent with fuels containing higher pro
portions of wax since the water crystals are found to form on the small
wax crystals from the fuel.
Combustors. The types of gas turbine combustors or burners most
commonly used are
1. The multiple-basket or can type (Fig. 8-16) which are mounted sym
metrically about the turbine axis
A-*-1 Section AtA

Fig. 8-17. Multiple-injection annular type of combustor. (Westing house Type.)
^Fuelnozzle
Fig. 8-18. Elbow type of combustor. (Elliott Type.)
2. The multiple-injection annular type (Fig. 8-17) which surround the

turbine axis
3. The elbow type (Fig. 8-18) where one or more fuel nozzles supply all
of the fuel to a single combustion chamber
The heat release in a gas turbine is continuous and at a rate comparable
to the combustion intensities found in modern supercharged reciprocating
aircraft engines. The rate of heat release in a turbine combustion
chamber may be as high as 30,000,000 Btu/(hr) (cu ft) of combustion

space10 which might be contrasted with 250,000 Btu/(hr) (cu ft) in a
naval boiler, or with stationary boilers as listed in Table 7-2.
The air flow into the burners must be separated into primary and
secondary streams. The primary air, at low velocity, mixes with the
fuel spray, and burning is initiated. As the burning continues and the
combustion products move downstream, secondary air is admitted to the
burner at increasing velocities to provide air for the completion of com
bustion and to reduce the temperature of the combustion products to a
value permissible in the turbine.' In some burners the air supply is
-About 40 ft /sec About 40 ft/sec

lin. H20 head. 1 in. H20 head
Tangential
feed holes
o jfAbout 215 ft/sec o o

through these .About 15 ft/sec
holes. 2 'A in. HG '/e in. H20 head
head.
o o o
^"
About 15 ft/sec About 75 ft/sec
'/sin. H20 head.
A B
Fig. 8-19. Burners with graduated air velocities. (Mock, SAE Trans., 1946.)
admitted through several series of circumferential holes, and the air

velocities increase with the distance from the fuel spray nozzle as shown
in Fig. 8-19, where the primary air enters at 15 ft/sec while the last of the
secondary air enters at 215 ft/sec.12
The primary air must mix with the fuel very vigorously to produce a
quickly combustible mixture. Thorough mixing is accomplished by using
a good fuel-atomizing nozzle along with a high degree of turbulence. The
latter is created by feeding the primary air through many small holes, by
deflecting it off screens or deflectors placed in the air stream, or by
employing counterflow of the fuel and air. Fuel nozzles for gas turbines
are mostly of the oil-pressure-atomizing type, but some stationary tur
bines utilize air-atomizing nozzles. The principles of operation of the
fuel nozzles for gas turbines are the same as for the corresponding oil-
burner nozzles discussed in Chap. 6.
Combustion. The gases entering the turbine are produced in a com
bustion process consisting of four steps:
1. Formation of the combustible mixture

2. Ignition of the mixture by the existing flame
3. Flame movement, or the propagation of combustion
4. Addition of excess air to the products of combustion to reduce the
exhaust temperature
The first three steps are common to all combustion processes, but the
fourth is seldom found except in gas turbines.
The fuel and air must be highly homogeneous if there is to be a com
plete chemical reaction. Before such mixing can occur, the liquid fuel
must be vaporized. Even though some heat may be added to the fuel
before it enters the burner, it remains for the turbulence in the compara
tively low-velocity primary air to aid materially in the vaporization of the
fuel. The vaporization is also aided by the radiant heat from the flame
front, though little heat is obtained from the burner walls.
Any specific combustor will have its own limiting fuel/primary-air
mixture ratios which will support combustion. With lean mixtures it
has been found that the flame-extinction limit is leaner than the ignition
limit, that the absolute pressure has little effect on either, that an increase
in air temperature broadens the combustible range, while improved
atomization promotes a better flame-extinction limit.13
The flame propagation rate will be accelerated as the Btu per pound of
mixture is raised, as the difference between the initial temperature of the
fuel and air and the ignition temperature decreases, and as the specific
heat of the fuel/air mixture is lowered. The propagation rate will be
slower as the weight of inert diluents (such as nitrogen) is greater, as
the specific heat of the inert diluents is increased, and as the heat of
vaporization of the liquid fuel is raised. 12 These performance trends con
form with the discussion of flame propagation in Chap. 4.
For octane, the theoretical fuel/air ratio by weight is 0.0662 while
fuel/primary-air ratios between 0.25 and 0.04 can be burned. In gas
turbines, for the maximum reaction speed, the ratio should be approxi
mately 0.073, and for most complete combustion about 0.060. The com
bustion of gasoline with fuel/primary-air ratios less than about 0.055 is
slow and erratic.10
With the high rates of heat release demanded of gas turbine combustors,
their design presents some specific problems. One primary need is to
provide stable combustion over the complete range of power demands.
Another is to provide quick ignition. These problems are much more
difficult than in stationary oil burners owing to the high burning rates,
the high air velocities, and the absence of radiation surfaces. One of the
flame-stabilizing methods employed in several burners is to shelter the
conical fuel spray in the initial zone from the high-velocity secondary
air, as shown in Fig. 8-19. This type of design improves the stability of
combustion at idling and aids ignition when starting.11
The gases leaving the burners are a mixture of the air supplied by the
compressor plus 0.6 to 6.0 per cent water vapor, plus 1.5 to 15 per cent
carbon dioxide, less the oxygen consumed from the air. Fuel /air ratios
required to maintain a given gas temperature leaving the combustor may
100 r
40l__J i i i
4 6 8 10
Grades of altitude, ft/constant
Fig. 8-20. Variation of combustion efficiency with increasing altitude at a constant engine
speed of 88 per cent rated rpm operating an annular combustor with two fuels. (Gibbons
andJonash, " Effect of Fuel Properties on the Performance of the Turbine Engine Combustor,"
paper presented before the ASME, November, 1948.)
be computed by the methods presented in Chap. 3. The degree of com

pleteness of the combustion reaction is expressed in terms of combustion
efficiency, which is usually defined as
—
Ve TTr
fc*" complete
where j/c = combustion efficiency

Aff.otuai =
actual increase in enthalpy in combustion chamber
t,. = increase of enthalpy for complete combustion
Generally combustion efficiencies for good operation are slightly less than
100 per cent. If incomplete combustion occurs, there also will be
unburned fuel, a small amount of cracked fuel vapors, carbon monoxide
(usually less than 0.1 per cent), solid carbon particles, and other undesir
able products, such as aldehydes, present in the combustion gases.10
The combustion efficiency of the gas turbine operating at sea-level
static conditions appears to be independent of the molecular structure or
the boiling range of the fuel.21 This fact does not hold, however, when it
is operating at high altitudes, as shown in Fig. 8-20. Tests were made
by the NACA comparing the combustion efficiencies of aviation gasoline
and a kerosene-type fuel. It can be seen from the curves that the latter
400 600 800 1000 1200 1400 1600 1800

Mean temperature rise through combustor, F
Fig. 8-21. Variation of combustion efficiency with mean temperature rise through a
tubular combustor with fuels of different boiling ranges at simulated altitude of 45,000 ft
and engine speed of 76 per cent rated rpm. (Gibbons and Jonash, "Effect of Fuel Properties
on the Performance of Turbine Engine Combustor," paper presented before the ASME, Novem
ber, 1948.)
has a lower combustion efficiency, and that the difference between the
performance of the two fuels increases with altitude.
The results of study by the NACA of the combustion efficiencies of a
a
series of fuels at high altitudes are shown in Fig. 8-21. A single tubular
combustor was used at a simulated altitude of 45,000 ft at 76 per cent
rated engine rotor speed. This represents very severe operating condi
tions and serves to increase the spread of combustion efficiencies as well
as to lower them from the corresponding values which would result from
operation at 100 per cent rated engine rotor speed. All fuels showed a
decrease in combustion efficiency with a decrease in the mean tempera
ture rise through the combustor, which is dependent upon the quantity of
fuel injected. These curves also show a very definite decrease in com
bustion efficiency with the less volatile fuels.21
ROCKETS
There are two general types of rocket units : the solid-propellant rocket
and the liquid- or bipropellant rocket. Rockets differ from other prime
movers in that not only the fuel but also the oxygen for complete com
bustion is carried with the rocket. The oxygen supply may be com
pressed oxygen or it may be contained in the so-called oxidizer, which is
a chemical reagent that releases the oxygen during the chemical reactions
of combustion. Atmospheric conditions have no bearing on the combus
tion reactions, since the oxygen supply as well as the fuel is provided.
Perfect combustion may be obtained even beyond the earth's atmosphere.
Only a general discussion of combustion in rockets is given here, since a
more complete treatment of this subject
is beyond the scope of this text. propellant^ rlgnitor
Solid-propellant Rockets. This clas
sification is applied to rockets which use
a solid propellant, that
is,
self-combus
a
tible mixture. Black powder, typical Fig. 8-22. Solid-propellant rocket

a
sectl0n-
solid propellant, contains about 75 per
cent saltpeter, 15 per cent charcoal, and 10 per cent sulfur. An electric
source used for ignition, and the reaction may be symbolized as
is
KNO3 -» CO N2 + SO2 K2S K2O

+
C
+
+
+
+
CO2
•
S
•
,
Gas Smoke
The general construction of such unit shown in Fig. 8-22. The charge
is
a
placed in the rocket in layers or in sticks in such manner that will

it
is
burn at uniform rate. The products of combustion are discharged

a
through De Laval type of nozzle at very high velocity, thus creating the
a
rocket thrust. The power duration short, being from 12 to 45 sec,

is
but pressures from 200 to 2800 psi are developed during this time. Some
1
the solid-propellant rockets may be stopped at will, but they cannot be

of
restarted.
Liquid or Bipropellant Rockets. An oxidizer and fuel are carried in
a
tanks in liquid or bipropellant rocket. They are delivered from the

a
tanks to the rocket combustion chamber where many of the fuels will
ignite spontaneously on contact with the oxidizer. When the thrust
duration matter of seconds, the oxidizer and the fuel tanks are pres
is
a
surized with an inert gas such as nitrogen (Fig. 8-23). When longer
thrust duration desired, the liquids are forced to the combustion cham
is
ber by turbine-driven pumps. separate combustion chamber or steam

A
generator furnishes the gas or steam to operate the turbine (Fig. 8-24).
The duration of operation limited chiefly by the supply of fuel and
is
Fig. 8-24. Pump-pressurized liquid-propellant rocket with jacketed combustion chamber.

oxidizer and by the method of pressurizing. If long duration is expected,

some method of cooling the combustion chamber and nozzle is needed.
One method of doing this is to pass the fuel through a cooling jacket
around the combustion chamber, as shown in Fig. 8-24. This method is
regenerative since the fuel temperature is raised about 200 F, and thus the
heat losses are reduced.
Table 8-3. Principal Rocket Propellants*
Propellants Symbol Jet speeds, ft /sec
Nitric acid HN03

7,300
Aniline C,HsNH2
Gasoline
8,070
Liquid oxygen . . . o2
Ethyl alcohol .... C2HsOH
8,100
Liquid oxygen . . . 02
Hydrogen Hs
8,800
Liquid oxygen . . . 02
Hydrazine N2H4
9,740
Liquid fluorine. . . F,
Liquid hydrogen . H,
12,000
Liquid oxygen. . . 0,
Liquid hydrogen . H„
13,150
Liquid ozone O,
*F. H. Clauser, "Flight beyond the Earth's Atmosphere," paper presented before the SAE, Detroit
Section, November, 1947.
The exit gas velocities derived from various fuel and oxidizer combina
tions are shown in Table 8-3. 17 018 These velocities are a function of the
temperature and pressure created in the combustion chamber. The maxi
mum amount of useful work for each pound of propellant depends upon
the thermochemical energy of the propellant and the efficiency with which
it is utilized. The higher the thermochemical energy per pound, the
greater the net propulsion work per pound even if the efficiency of
utilization is less.16
The chemical reaction during combustion in a rocket may be symbolized
with a fuel and an oxidizer from Table 8-3, gasoline and liquid oxygen,
where
C8H18 + 12.502 -» 8CO2 + 9H2O
This reaction is identical with the complete combustion of octane in any

other type of combustion chamber except for the absence of nitrogen.
Problems
1. List the factors which influence precombustion reactions in a spark-ignition
engine.
2. What effect does tetraethyllead have on these reactions?

3. What is the relation between detonation and these precombustion reactions?
4. What are the factors which influence ignition delay in a spark-ignition engine?
5. Explain how three points on the ASTM distillation curve of a gasoline indicate
its relative performance in a spark-ignition engine.
6. Why is it generally necessary to start a kerosene engine with gasoline?
7. What is the advantage of high turbulence in a spark-ignition engine?
8. What are the possible sources of this turbulence?
9. How does the flame velocity in a spark-ignition engine vary with engine speed?
10. Explain the difference between autoignition and preignition.
11. What are the fundamental factors (a) in engine design and (6) in engine opera
tion which affect detonation in an engine?
12. What is the effect of water or water-alcohol injection (a) on detonation and
(6) on maximum power when detonation is not present?
13. How does ignition timing affect detonation in an engine?
14. What is the approximate temperature at the end of the compression stroke in a
compression-ignition engine? What two factors chiefly control it?
16. Explain the terms physical delay and chemical delay.
16. Rate the following hydrocarbons in their order of preference in diesel fuels:
naphthenes, aromatics, and paraffins.
17. List the types of combustion chambers in their order of preference for high
speed compression-ignition engines.
18. What is the function of a very hot surface in a precombustion chamber?
19. What are the primary factors which permit the operation of the dual-fuel
engine?
20. What are the primary factors which permit the operation of a spark-ignition
gas engine at 15-to-l compression ratio?
21. Explain the difference in flame type found in a spark-ignition engine from that
found in a gas turbine.
22. Sketch a can-type combustor showing the fuel nozzle and the paths of the
primary and secondary air.
23. How does the heat release rate of a gas turbine compare with that of a modern
steam plant?
24. What air-fuel ratio produces the maximum combustion reaction rate? the most
nearly complete combustion?
25. Explain two methods of construction and operation of bipropellant rockets.
26. What are the approximate exit gas velocities attainable from rockets?
References
1. Retailliau, E. R., H. A. Ricards, Jr.,

and M. C. K. Jones, "Precombustion
Reaction in the Spark Ignition Engine," SAE Quart. Trans., vol. 4, No. 3, p.
438, 1950.
2. Miller, C. D., Roles of Detonation Waves and Auto Ignition in Spark-ignition
Engine Knock as Shown by Photographs Taken at 40,000 and 200,000 Frames
per Second,SAE Quart. Trans., vol. 1, No. 1, p. 98, 1947.
3. Corrington, L. C, and W. F. Fisher, Effect of Pre-ignition on Cylinder Tem
peratures, Pressures, Power Output and Piston Failures, NACA, R.M. E6L10.
4. Withrow, L., and T. A. Boyd, Photographic Flame Studies in Gasoline Engines,
Ind. Eng. Chem., vol. 28, May, 1931.
5. Landen, E. W., Combustion Studies of the Diesel Engine, SAE Trans., vol. 54,
p. 270, 1946.
6. Elliott, M. A., Combustion of Diesel Fuel, SAE Quart. Trans., vol. 3, No. 3,
p. 490, July, 1949.
7. Elliott, M. A., and R. F. Davis, "The Composition of Diesel Exhaust Gas,"
paper presented before the SAE, Diesel Division, November, 1949.
8. Shoemaker, F. G., and C. D. Gadebusch, Effect of Fuel Properties on Diesel
Engine Performance, SAE Trans., vol. 54, p. 339, 1946.
9. Barth, H. W., F. A. Robbins, and H. C. Lafferty, "Railroad Diesel Engine
Improvement Based on Study of Combustion Phenomena and Diesel Fuel Proper
ties," paper presented before the SAE, Tractor and Diesel Division, September,
1948.
10. Godsey, A. W., and L. A. Young, "Gas Turbines for Aircraft," McGraw-Hill
Book Company, Inc., New York, 1949.
11. Zucrow, M. J., "Jet Propulsion and Gas Turbines," John Wiley & Sons, Inc.,
New York, 1948.
12. Mock, F. C, Engineering Development of the Jet Engine and Gas Turbine
Burner, SAE Trans., vol. 54, p. 218, 1946.
13. Samaras, D. J., SAE Trans., vol. 54, p. 227, 1946.
14. Summerfield, M., "Applications of Rocket Propulsion," paper presented before
the SAE, Southern California Section, May, 1945.
15. Zucrow, M. J., The Rocket Powerplant, SAE Trans., vol. 54, p. 375, 1946.
16. Cole, R. A., "German Rocket Aircraft and Their Powerplants," paper presented
before the SAE, Metropolitan Section, March, 1946.
17. Katz, Israel, "Principles of Aircraft Propulsion Machinery," Pitman Publishing
Corp., New York, 1949.
18. Clauser, F. H., "Flight beyond the Earth's Atmosphere," paper presented before
the SAE, Detroit Section, November, 1947.
19. Hurn, R. W., and H. M. Smith, "Combustion Studies of Hydrocarbons in the
Diesel Boiling Range," paper presented before the ACS, Division of Petroleum
Chemistry, Cleveland, April, 1951.
20. Ward, C. C, and H. M. Smith, "Why Diesel Fuels Behave As They Do," paper
presented before the Western Petroleum Refiners' Association, San Antonio,
April, 1951.
21. Gibbons, L. C, and E. R. Jonash, "Effect of Fuel Properties on the Performance
of the Turbine Engine Combustor," paper presented before the ASME, November
1948.
22. Hurn, R. W., and K. J. Hughes, Combustion Characteristics of Diesel Fuels as
Measured in a Constant-volume Bomb, SAE Quart. Trans., vol. 6, No. 1, p. 24,
January, 1952.
23. Rifkin, E. B., C. Walcutt, and G. W. Betker, Jr., "Early Combustion
Reactions in Engine Operation," paper presented before the SAP., Fuels and
Lubricants Division, November, 1951.
APPENDIX
APPENDIX 323
Table A-l. Standard Twist Drill Sizes
Designa Diam, Area, Designa Diam, Area, Designa Diam, Area,

tion in. sq in. tion in. sq in. tion in. sq in.
0.5000 0.1963 3 0.213 0.03563 Hi 0.0938 0.00690

0.4844 0. 1843 4 0.209 0.03431 42 0.0935 0.00687
0.4688 0.1726 5 0.2055 0.03317 43 0.0390 0.00622
'Hi. 0.4531 0. 1613 6 0.204 0.03269 44 0.0860 0.00581
H« 0.4375 0.1503 0.2031 0.03241 45 0.0820 0.00528
"A* 0.4219 0.1398 7 0.201 0.03173 46 0.0810 0.00515

z 0.413 0.1340 8 0.199 0.03110 47 0.0785 0.00484
0.4063 0.1296 9 0.196 0.03017 H* 0.0781 0.00479
Y 0.404 0.1282 10 0.1935 0.02940 48 0.0760 0.00454
X 0.397 0.1238 11 0.191 0.02865 49 0.0730 0.00419
0.3906 0.1198 12 0.189 0.02806 50 0.0700 0.00385

w 0.386 0.1170 K« 0.1875 0.02761 51 0.0670 0.00353
V 0.377 0.1116 13 0.185 0.02688 52 0.0635 0.00317
0.375 0.1104 14 0.182 0.02602 Ht 0.0625 0.00307
u 0.368 0.1064 15 0.1800 0.02545 53 0.0595 0.00278
0.3594 0.1014 16 0.1770 0.02461 54 0.0550 0.00238

T 0.358 0. 1006 17 0.1730 0.02351 55 0.0520 0.00212
S 0.348 0.09511 0. 1719 0.02320 H* 0.0473 0.00173
0.3438 0.09281 18 0.1695 0.02256 56 0.0405 0.001698
R 0.339 0.09026 19 0.1660 0.02164 57 0.0430 0.001452
Q 0.332 0.08657 20 0.1610 0.02036 58 0.0420 0.001385

"<■* 0.8281 0.08456 21 0.1590 0.01986 59 0.0410 0.001320
P 0.323 0.08194 22 0.1570 0.01936 60 0.0400 0.001257
O 0.316 0.07843 %2 0.1563 0.01917 61 0.039 0.001195
H. 0.3125 0.07670 23 0.1540 0.01863 62 0.038 0.001134
N 0.302 0.07163 24 0.1520 0.01815 63 0.037 0.001075

0.2969 0.06922 25 0.1495 0.01755 64 0.036 0.001018
M 0.295 0.06835 26 0.1470 0.01697 65 0.035 0.000962
L 0.29 0.06605 27 0.1440 0.01629 66 0.033 0.000855
Hi 0.2813 0.06213 'A* 0.1406 0.01553 67 0.032 0.000804
K 0.281 0.06202 28 0.1405 0.01549 Ht 0.0313 0.000765

J 0.277 0.06026 29 0.1360 0.01453 68 0.031 0.000755
I 0.272 0.05811 30 0.1285 0.01296 69 0.0292 0.00067Q
H 0.266 0.05557 0.1250 0.01227 70 0.028 0.000616
l%* 0.2656 0.05542 31 0.1200 0.01131 71 0.026 0.000531
G 0.261 0.05350 32 0.1160 0.01057 72 0.025 0.000491

F 0.257 0.05187 33 0.1130 0.01003 73 0.024 0.000452
EM 0.2500 0.04909 34 0.1110 0.00968 74 0.0225 0.000398
D 0.246 0.04753 35 0.1100 0.00950 75 0.021 0.000346
C 0.242 0.04600 K« 0.1094 0.00940 76 0.020 0.000314
B 0.238 0.04449 36 0.1065 0.00891 77 0.018 0.000254

0.2344 0.04314 37 0.1040 0.00849 78 0.016 0.000201
A 0.234 0.04301 38 0.1015 0.00809 X* 0.0156 0.000191
1 0.228 0.04083 39 0.0995 0.00778 79 0.0145 0.000165
2 0.221 0.03836 40 0.0980 0.00754 80 0.0135 0.000143
0.2188 0.03758 41 0.0960 0.00724
Note: Designations are in fractions of an inch, in standard twist drill letters, or in standard drill
manufacturers' numbers, the latter being the same as steel wire gauge numbers. (Courtesy of AG A.)
APPENDIX 325
1300
200
400
600
=
-^800
o
1000
1200
1400
T3
600
1800
2000
2200
2400
2600
Fig. A-3. Chart for determining the loss due to heat in the dry flue gas. (Courtesy of
Combustion Engineering-Superheater , Inc.)
APPENDIX 327
Heat loss, Btu per lb fuel

900 800 700 600 500 400 300 200 100
1400
0 2 .4 .6 .8 1.0 1.2
Carbon monoxide in flue gases, per cent by volume

Fig. A-5. Chart for determining the loss due to the presence of carbon monoxide in the flue
gas. (Courtesy of Combustion Engineering-Superheater, Inc.)
1100
1000
. 900
o
o
800
8R 700
600
o 500 h*—
400
300
200
100
1100 1200 1300 1400 1500
Heat in 1 lb of moisture evaporated and superheated to flue -gas temperature, Btu

Fig. A-6. Chart for determining the losses due to hydrogen and to moisture in fuels.
(Courtesy of Combustion Engineering-Superheater, Inc.)
of
walls
Ns
cooled furnace
4
3
2
10
25
1000
9
400 500
3
4
5
6
7
8
10 20 30 40 50 60 7080 100 200 300
actual evaporation, 1000 Ib/hr (specified or required)

gloSSS rc^d-^Tacta
A
Fig. A-7. Radiation loss from furnace walls. furnace wall must have at least one-third its projected surface covered by water-cooled
if
is
surface before reduction in radiation loss
permitted. Air through air-cooled walls must be used for combustion reduction in radiation
is
loss to be made. Example: Unit guaranteed f5 maximum continuous capacity of 050,000 lb/hr with three water-cooled walls. Loss at
W 5 CD
of
150,000 = 0.65 per cent; loss at 75,000 = 0.30 the American Boiler
per cent. (Courtesy Manufacturer's Association.)
Mc Adams.)
INDEX
A Atmosphere, composition of, 46
moisture in, 47, 324, 331
Absorption process, 23 Atmospheric gas burners, 198
Acetylene, flame speeds of, 149 design, 202-217
ignition temperature, 157 Atmospheric pollution, 280
inflammability limits, 155 Atomic energy, 6
physical properties, 18 Atomic weights, 40
Acids, structure, 131 Autoignition, by adiabatic compression,
Active centers, 140 161
Afterburning, 115, 147 theory of, 292, 293
Air, composition of, 46 Avogadro's law, 39, 42
enthalpy, 78
moisture in, 47, 331 B
specific heat, 73, 74, 79
theoretical, 48 Bagasse, 13
Alcohol injection, 298 Balances, carbon, 59
Alcohols, for fuels, 36 carbon-hydrogen, 62
ignition temperature, 157 material, 39
inflammability limits, 155 oxygen, 57
physical properties, 18 Ball-mill pulverizers, 255
structure, 131 Benzene, ignition lag, 161, 162
Aldehyde degradation mechanism, 132 (See also Aromatics)
Aldehydes, inflammability limits, 155 Bipropellant rockets, 315-317
structure, 131 Bituminous coal (see Coal)
Alkyl radical, 131 Blast-furnace gas, 31
Alkylation process, 24 composition, 28
Amagat's law, 39, 44 Blue gas, 28, 32
Ammonia inflammability limits, 155 , Boiler heat balance, calculations, 266-
Anthracite (see Coal) 271
API gravity, conversion to specific grav of typical units, 280
ity, 177 Boiling points of hydrocarbons, 18
heating values from, 92 Bomb calorimeter, 81
of hydrocarbons, 18 Bomb method for flame speeds, 146
of liquid fuels, 177, 187, 324 Bowl-mill pulverizers, 256
temperature correction, 177 Bubble point, 175
Aromatics, characteristics, 20 Bunsen-burner method for flame speeds,
inflammability limits, 155 141
mechanism of combustion, 134 Burner head for gas burners, 208
physical properties, 18 Burners Gas burners; Oil burners;
(see
Arrhenius equation, 138 Pulverized coal, burners for)
Ash content, of liquid fuels, 180 Burning (see Combustion)
of solid fuels, 13 Burning rates of gas burners, 210-215
composition, 171 (See also Combustion rates)
331
Butane, 24, 29 Coal, agglomerating and agglutinating

composition of, 28 tests, 172
ignition lag, 162 anthracite, use of, 4
(See also Paraffins) ash-deformation and ash-fluid tempera
ture, 171
c ash-softening temperature, 170, 174
bases for analysis, 166
Calorific value, 81 bituminous, markets for, 4, 5
(See also Heating value) combustion of, 241
Calorimeter, bomb, 81 combustion calculations, 55
constant-pressure, 82 composition, 11, 13
sulfur detection in, 14 consumption of, 2
Carbon, combustion mechanism, 129 drop-shatter tests, 172
heat of reaction, 90 grades, 173, 174
Carbon balance, 59 grindability index, 171
Carbon dioxide, enthalpy, 78 heat loss in, due to moisture, 270
heat of formation, 90 heating value, 13
specific heat, 73, 74, 79 computations, 92
Carbon-hydrogen balance, 62 mining of, 8-10
Carbon monoxide, combustion mecha strip, 8, 9
nism, 128 origin, 7
enthalpy, 78 overfeed firing, 241-244
heat loss, 269 petrographic types, 10
heats of reaction, 90 physical properties, 170-173
ignition temperature, 157 production by states, 8, 9
inflammability limits, 155 proximate analysis, 12, 13
specific heat, 73, 74, 79 calculations, 167-169
transformation velocity, 143, 144, 147 ranks, 7, 172, 173
Carbon residue, Conradson, 179 reserves, 2, 3
Carbonyl radical, 132 by states, 8, 9
Carbureted water gas, 28, 32 synthesis, 34
Carburetor, 290 ultimate analysis, 12, 13
Casing-head gasoline, 27 calculations, 167-169
Cetane number, 162, 183 underground gasification, 36
of gas oil, 25 weathering or slacking index, 171
Chain-branching chain-breaking reaction, Coal gas, 28, 29
117 flame speeds, 149
Chain carriers, 117 Coal pulverizers, 255-258
effect on delay, 160 Coalification process, 8
Chain-grate stoker, 245-247 Coke, bases for analysis, 166
Chain reaction theory, 117 composition of, and heating value, 13
Charcoal, 13 physical properties, 170-173
Chemical definitions, 16 Coke-oven gas, 28, 30
Chemical energy, 84, 108 Combining weights, law of, 39, 40
Chemical-equation significance, 46 Combustible material in refuse, heat loss
Chemical equilibrium, 96 due to, 268
Chemical kinetics, 138 Combustion, of coal, 241
effect of branching, 120 definition of, 114
Chemical nomenclature, organic, 131 in engines, 288, 299
Chemically correct mixture, 49 gas turbines, 311-314
INDEX 333
Combustion, in engines, gas turbines, Critical compression ratio, 182

altitude effect on efficiency, 313 Cyclic compounds, 19, 20
flame-extinction limit, 312 Cyclone furnaces, 265
ignition limit, 312 heat release rates, 281
homogeneous, 136, 137
intermediate products of, 117 D
propagation, 136
slow, 128 Dalton's law, 39, 42
surface, 151, 219 Delay period, 160, 161
Combustion calculations, charts, 104-109 in compression-ignition engines, 200
of coal, 55 factors affecting, 301-307
of gases, 54 Density of liquid fuels, 324
with insufficient air, 57 Detonation, 116
of liquids, 48-50 effect on engine operation, 293
from Orsat, 59 factors influencing, 294-296
Combustion chambers, compression-igni in spark-ignition engines, 292-298
tion, 305, 306 suppression by injection, 297
furnaces, 239, 263, 265 Detonation wave theory, 294
gas appliances, 201, 209 Dew point, 50, 175
gas turbines, 309, 310 Dialkyl peroxides, structure, 131
rockets, 315, 316 Diesel fuel oil, 192, 193
spark-ignition, 292 analysis of, 193
Combustion charts, 104-109 heating value, 324
Combustion efficiency in gas turbines, 313 Diffusion effect on flame speed, 138, 139
Combustion mechanism, 116 Diffusion flames, 115
carbon, 129 Diffusion theory of flame propagation,
carbon monoxide, 128 140
hydrocarbons, 131 Dissociation, 96
hydrogen, 118-128 in engines, 103
methane, 129 pressure effect, 102
Combustion process, types, 115 temperature effect, 102-104
Combustion products, analysis, 52 Dissociation effects, 58
dry, 55 Distillation, fractional, 22
at equilibrium, 103 Distillation curves, ASTM, 176, 188
Combustion rates, effect of diffusion on, EAD, 188
129 Distillation test, ASTM, 175
in engines, 152 EAD, 177
factors affecting, 138 Domestic heater, 244
measurement, 141-147 smokeless, 254
Combustion zone, turbulent, diagram, Draft variation in steam plant, 239
152 Drill sizes, 323
Combustion zones, diagram, 137 Dry flue gas, heat loss, 325
Combustors for gas turbines, 309-311 Dual-fuel engines, 308
Compression-ignition engines, combus Dulong's formula, 92
tion in, 299 Dust collectors, 284, 285
Compression ratio, critical, 182
Conradson carbon residue, 179 E
Cool flames, 136, 137
Corrosion test of liquid fuels, 180 Electric power production, 4, 6
Cracking process, 23 Endothermic reaction, 89
Energy, atomic, 6 Flame propagation, factors affecting, 138

conservation of, 38, 39 rates, measurement of, 141-147
Energy consumption in United States, 1, thermal theories, 139
2 Flame speeds, effect of heat transfer on,
Energy equations, 70 138, 139
Engines, combustion, in compression- in engines, 152
ignition, 299 measurement of, 141-147
in spark-ignition, 288 methods for, Bunsen-burner, 141
four-cycle, 288, 298 glass-tube, 144
two-cycle, 289, 298 pressure effect on, 149
Enthalpy, sensible, 85 values of, 143, 144
total, 69 Flame temperature, adiabatic, 102, 104
Enthalpy values, table, 78 calculations, 93
Equilibrium, in engines, 103 from charts, 104-109
of flame temperatures, 102, 104 equilibrium of, 102, 104
frozen, 102, 104 Flame zone, turbulent, diagram, 152
Equilibrium constant, 97, 98 Flames, cool, 136, 137
Equilibrium equations, 96 diffusion, 115
Ethane, heat of reaction, 90 explosion, 116
specific heat, 73, 74 stationary, 115
transformation velocity, 143 Flash points of liquid fuels, 178, 187, 191,
(See also Paraffins) 193
Excess air, 49 Flashback in gas burners, 199, 200
determination of, from flue-gas analy Flue gas, analysis of, 52
sis, 267 dry, 55
percentage of, 50 heat loss in, 268
Exhaust-gas analysis, compression-igni Fly-ash collectors, 283
tion engine, 307 Formation, heat of, 89
furnace, 279 Fractional distillation, 22
spark-ignition engine, 297 Free radicals, 117
Exothermic reaction, 89 Freezing points, hydrocarbons, 18
Explosion flames, 116 Front-arch furnace, 246, 247
Explosive limits, 153 Frozen equilibrium, 102, 104
of hydrogen, 121, 125-127 Fuel consumption in United States, 1, 2
Fuel oil, 26
F analysis, 193
aniline point, 184
Finishing processes, 25 cetane rating, 25, 183, 193
Fire points of liquid fuels, 178, 187, 191, distillation temperatures, 176
193 heating value, 324
Firing arch, 245-247 computations, 92
Fixed carbon, definition of, 14 preheating, 233
Flame, definition of, 114 specifications for, 187
Flame front, definition of, 114 viscosity, 233
diagram of, 137 Fuel reserves, in United States, 1-3
in tubes, 144 of petroleum, 3, 4
effect of vessel shape, 146 in world, 4
Flame holder, 218 Fuel systems for rockets, 316
Flame propagation, 136, 137 Furnaces, 261-266
diffusion theories, 140 cyclone, 265
INDEX 335
Purnaces, firing methods, pulverized fuels, Gasoline, casing-head, 27

263, 264 distillation temperatures, 176
flue-gas analysis, 279 heating value, 189, 324
front- and rear-arch, 245, 247 computations, 92
wall construction, 239, 240 ignition temperature, 157
water screen, 262 natural, 24
waterwall, 240 octane rating, 181
performance number, 183, 189
G representative formula, 26
specifications for, 189, 190
Gas burners, 197 tetraethyllead in, 183, 189, 190
atmospheric, 198 volatility, 187-190
burner head for, 208 Glass-tube method for flame speeds, 144
burning rates of, 210-215 Gross heating value, 86
center-diffusion-type, 220 (See also Heating value)
characteristic curves, 199, 201 Gum content of liquid fuels, 180, 187-
combination, 231, 232 191, 193
design, 202-217
H
Fanmix-type, 219
flames, 202 Heat, of combustion, 81
flashback in, 199, 200 of evaporation, of hydrocarbons, 18
heat release rates in, 281 table, 89
high-pressure, 217-221 of water, 86
Janitrol conversion, 220 of formation, 89, 90
test code, AG A, 201 of reaction, 81
tile port, 218 table, 90
turndown of, 201 (See also Heating value)
two-stage injection, 218 Heat-balance calculations, 266-274
Gas constant, specific and universal, 41 Heat balances, typical, for boiler, 280
Gas-diesel engines, 308 Heat capacity (see Specific heat)
Gas oil, 22 Heat loss, analysis of, 274-282
composition, 25 due to, carbon monoxide, 269, 327
Gas producer, 31 combustible material in refuse, 268
Gas turbines, 309-314
dry flue gas, 268, 275, 325
furnace walls, 329
combustion, 311-314
combustors for, 309-311 hydrogen in fuel, 270, 328
fuel specifications for, 190, 191 moisture in coal, 270
unburned carbon, 326
Gas velocity, 137
Heat release rates, in engines, 310
Gaseous fuels, 26-33
in furnaces, 281
combustion calculations, 54
Heat transfer, effect of, on flame speed,
composition, 28
138, 139
heating value, 27, 28
Heating value, 81
computations, 91
computations, 91
properties of, 192 constant-pressure and constant-vol
specific gravity, 27-28 ume, 83-85
Gaseous mixtures, inflammability limits, of gases, 27, 28
154 gross and net, 86
spark ignition, 159 higher and lower, 86
specific heat, 80 of hydrocarbons, 90, 92
Heating value, of liquid fuels, 174, 324 Ignition delay, 160, 161
of solid fuels, 13 Ignition energy, 158, 159
table, 90 Ignition lag, 135, 160, 161
temperature effect, 87 effect of compression ratio on, 162
(See also specific fuels) (See also Delay period)
Heptane ignition lag, 161, 162 Ignition temperature, definition of, 154
(See also Paraffins) natural gas, 156
Hexane, ignition of, 135 table, 157
(See also Paraffins) test methods, 154
Higher heating value, 86 Illuminants, 30
(See also Heating value) Inclined-grate stoker, 248
Homogeneous combustion, 136, 137 Inflammability limits, 152-155
Humidity chart, 331 of pulverized coal, 259
Hydrocarbon fuels, representative for table, 155
mula for, 26, 50 (See also specific fuels)
Hydrocarbon radicals, 131 Insufficient air, 48
Hydrocarbons, combustion equations, calculations, 57
48-50 Intermediate products, 1 17
flame speeds in bombs, 147 Internal energy, sensible and total, 84
heat of formation and heating values, Isooctane ignition lag, 161, 162
90 Isoparaffins, characteristics, 17 .
computations, 92 inflammability limits, 155
ignition temperatures, 157 mechanism of combustion, 134
limits of inflammability, 155 physical properties, 18
mechanism of combustion, 131
physical properties, 18 J
Hydrogen, combustion mechanism, 1 18—
128 Jet speeds of rockets, 317
enthalpy, 78
explosion limits, 121, 125-127 K
in fuel, heat loss due to, 270
heat of combustion, 90 Kerosene, 22
ignition temperature, 157 distillation temperatures, 176
inflammability limits, 155 heating value, 324
mechanism, wall effect, 123 computations, 92
reaction rates, 127 ignition temperature, 157
specific heat, 73, 74, 79 representative formula, 26
transformation velocity, 143 Ketones, structure, 131
Hydrogenation process, 24 Kinetics, 138
Hydroperoxides, structure, 131
I L
Ideal-gas law, 39, 41 Lean mixture, 152

Ignition, by adiabatic compression, 161, Lifting of flames, 199, 200
162 Lignite (see Coal)
in oil and gas burners, 234 Limits of inflammability, 152-155
by radiation, 130 table, 155
by spark, 158, 159 (See aho specific fuels)
in methane, 130 Liquefied petroleum gases, 24, 29
suppression of, 130 composition, 28
INDEX 337
Liquid fuels, ash content, 180 Mixtures of gases (see Gaseous mixtures)
combustion equations, 48-50 Moisture in coal, heat loss due to, 270
corrosion test, 180 Mole, definition of, 40
density, 324 Mole fraction, 43
distillation curves, 176 Molecular velocity, 120
flash and fire points, 178, 187, 191 Molecular weight, air, 47
heating value, 324 definition of, 40
computations, 92 hydrocarbons, 18
heating-value test, 174 mixtures, 47
physical properties, 174-192, 324
pour point, 179 N
representative formula for, 26, 50
specific gravity of, 177, 187, 324 Naphthenes, characteristics, 19
sulfur in, 180, 187, 193 ignition temperature, 157

viscosity, 178, 187, 191, 193 inflammability limits, 155
volatility, 175, 176 mechanism of combustion, 134
water and sediment test, 179 physical properties, 18
Liquid-propellant rockets, 315-317 Natural gas, 27
Lower heating value, 86 composition, 28
(See also Heating value) consumers of, 5
Luminous zone, 114 consumption of, 2
diagram, 137 flame speeds, 148
synthesis, 35
M Natural gasoline, 24, 27
Net heating value, 86
Manufactured gases, 28, 30 (See also Heating value)
Mass, conservation of, 38, 39 Nitrogen, enthalpy, 78
Material balance, 39 specific heat, 73, 74, 79
Mechanism, of aldehyde degradation, 132 Nomenclature, organic chemical, 131
of combustion, 116
carbon, 129 O
carbon monoxide, 128
effect on flame speed, 138 Octane, enthalpy, 78
hydrocarbons, 131 heats of reaction, 90
hydrogen, 118-128 specific heat, 74
intermediate products, 117 (See also Paraffins)
methane, 129 Octane number, 181, 189, 190
propane, 133 Oil-burner auxiliaries, 232-234
structure effect on, 134 Oil burners, 221-234
Methane, enthalpy, 78 air-pressure atomizing, 227-230
heat of reaction, 90 ignition, 234
mechanism of combustion, 129 oil-pressure atomizing, 225-227
spark ignition, 159 rotary-cup, 224
specific heat, 73, 74, 79 steam atomizing, 227-230
transformation velocity, 143 vaporizing, 223
Mining, coal, 8-10 Oil pools, petroleum, 15
oil-shale, 34 Oil shale, 33
Mixture composition, effect of, on flame reserves of, 3
speed, 150 Olefins, characteristics, 19
on ignition lag, 162 ignition temperature, 157

Olefins, inflammability limits, 155 Power production, electric utilities, 4, 6

physical properties, 18 water, 2, 4, 6
Oxsat analysis, 52 Preignition, 294
Overfeed firing, 241-244 Pressure, effect of, on delay, 160
caking method, 244 on flame speed, 149
gas composition, 243, 243 Primary air, definition of, 141
Oxidizers, rocket, 317 Primary-air factors, gas burners, 214
Oxygen, enthalpy, 78 Primary fractions, 22
specific heat, 73, 74, 79 Producer gas, 28, 30
Oxygen balance, 57 Propagating center, 130
Propagation of flames, 136, 137
P diffusion theories, 140
factors affecting, 138
Paraffins, characteristics, 16 thermal theories, 139
ignition temperature, 157 Propane, 24, 29
inflammability limits, 155 composition, 28
mechanism of combustion, 134 mechanism of combustion, 133
physical properties, 18 transformation velocity, 143
Partial pressures, 42 (See also Paraffins)
Peat, 7 Propellants, rocket, 317
composition and heating value of, 13 Proximate analysis, 12
Pentane reaction rates, 150 bases for, 166
Peracids, structure, 131 calculations, 167-169
Perfect-gas law, 39, 41 of solid fuels, 13
Peroxides, structure, 131 Psychrometric chart, 331
Petroleum, chemistry of, 16-20 Pulverized coal, burners for, 258-266
combustion equations, 48-50 heat release rates, 281
composition, 20 inflammability curves, 259
gas-oil, 25 Pulverized-coal furnace, 237
consumption, 2 Pulverizers, 255-258
oil pools, 15 ball-mill, 255
origin, 14 bowl-mill, 256
refinery diagram, 21 capacity of, 257
refinery yields, 5
refining processes, 21-25 R
reserves of, 4
Petroleum products, composition, 25, 26 Radiation loss from furnace walls, 329
heating values, 90 Radicals, free, 117
ignition temperature, 157 Reaction rate, chain-branching and ther
representative formula for, 26, 50 mal effects, 120
(See also Liquid fuels) Reaction zone, 114
Phlogiston, 114 diagram, 137
Physical properties, of liquid fuels, 174- thickness, 115
192, 324 Reactions, of carbon, 129
of solid fuels, 166 of carbon monoxide, 128
Poisoning of walls, 121, 123, 124 chain-branching and chain-breaking,
Pollution, atmospheric, 280 117
Polymerization process, 23 of hydrocarbons, 131
Port sizes of gas burners, 208 of hydrogen, 127
Pour point of liquid fuels, 179 kinetics of, 138
INDEX 339
Reactions, of methane, 129 Specific heat, 71

of propane, 133 equations, 74
third body in, 122, 128 instantaneous, 72
Rear-arch furnace, 245, 247 mean, 76, 79
Refinery, flow diagram, 21 of mixtures, 80
picture of, 24 pressure effect on, 75
processes, 21-25 variation with temperature, 73
production, 5 Spectroscopic analysis, 118
Refinery oil gas, 28, 29 Spreader stoker, 248-251
Refining, coal synthesis, 35 combustion in, 250
oil-shale, 34 Stationary flames, 115
Reid vapor pressure, 179, 189-191 Steam generator components, 236-240
Relative humidity, 47 Stoichiometric calculations, definition of,
Rich mixture, 150 38
Ringelmann smoke-density chart, 282 Stoichiometric mixture, 49

Rocket fuels, 315, 317 Stoker, chain- or traveling-grate, 245-247
Rockets, 315-317 heat release rates, 281
jet speeds of, 317 inclined-grate, 248
solid-propellant, 315 spreader, 248-251
Rotary-cup oil burner, 224 underfeed, 251-253
Straight-run fractions, 22
S composition, 25
Strip mining, 8, 9
Safety lamps, 130 Sulfur, effect on dew point, 51
in gases, 27
Saturation temperatures and pressures,
in liquid fuels, 180, 187, 193
water, 51
in Orsat, 57
Saybolt furol and universal viscosity, 178
Surface combustion, 151, 219
Sewage gas, 28, 33
Synthetic fuels, 33-36
Slow combustion, 128, 136
Smoke-density chart, 282
Smokeless domestic heater, 254
T
Soap bubbles for flame measurements,
Temperature, effect of, on delay, 160,
143
161
Solid fuels, analysis of, 13
on flame speed, 148
physical properties, 166
flame, 93, 104-109
Solid-propellant rockets, 315 static and total, 69
Spark ignition, 158, 159 Tetraethyllead, effect of, on prereactions,
Spark-ignition engines, combustion in, 290
288 on transformation velocity, 147
exhaust gas analysis, 297 in gasoline, 183, 189, 190
Spatial velocity, 136 Theoretical air, 48
in tubes, 144 Thermal theory of flame propagation, 139
Specific gas constant, 41 Thermodynamic principles, 69-75
Specific gravity, conversion to API grav energy equations, 70
ity, 177 equilibrium, 96
of gases, 27, 28 heat of formation, 89, 90
heating values from, 92 heating values, 83-89
of hydrocarbons, 18 specific heat, 71-76
of liquid fuels, 177, 187, 324 temperatures, 69
Thermodynamic tables, 77, 78 Viscosity, of liquid fuels, 178, 187, 191,

Third body, 122, 128 193
Transformation velocity, 136 Saybolt, 178
benzene, 147 Volatile matter, 14
carbon monoxide, 144 Volatility, effect on engine operation, 291
effect of physical variables on, 147-151 of gasoline, 187-190
of gases, 143 of liquid fuels, 175, 176
heptane and isooctane, 147 Volumetric composition, 45
natural gas, 148
W
in tubes, 144
Traveling-grate stoker, 245-247 Wall effect, on combustion, 117
Turbulence, effect on flame speed, 151, on flame speeds, 150
152 on ignition, 156
in spark-ignition engines, 291, 292 Wall-surface poisoning, 121, 123, 124
Turbulent combustion zone, diagram, 152 Water and sediment test, 179
Turndown of gas burners, 201 Water gas, 32
Turpentine, ignition temperature, 157 carbureted, 28, 32
inflammability limits, 155 composition, 28
Water-gas generator, 32
U Water-gas reaction, 30
Water injection for detonation suppres
Ultimate analysis, 12 sion, 297
bases for, 166 Water power, 2, 4, 6
calculations, 167-169 Water screen, furnace, 262
heating value from, 92 Water vapor, enthalpy, 78
of solid fuels, 13 heat of formation, 90
Underfeed stoker, 251-253 saturation temperatures, 51
Universal gas constant, 41 specific heat, 73, 74, 79
Waterwall furnace, 240
V Weights, atomic, 40
combining, law of, 39, 40
Vapor pressure, Reid, 179, 189-191 Wood, 13
Vaporizing oil burner, 223
Y
Vegetable oils for fuels, 36
Velocity, molecular, 120 Yellow-tip flame, 199, 200
7 354 A A 30 r.
I.
THE UNIVERSITY OF MICHIGAN
DATE DUE
JAN 011993
DATE DUE
JAN 01
1993

Fuels and Combustion Marion L. Smith, Karl W. Stinson.

Uploaded by

Copyright:

Available Formats

Fuels and Combustion Marion L. Smith, Karl W. Stinson.

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Fuels and Combustion Marion L. Smith, Karl W. Stinson.

Uploaded by

Copyright:

Available Formats

What is the book about?

What is the book about?

What are some of the topics covered in the book?

What are some of the topics covered in the book?

»*OPERTT Of

New York Toronto London

Library of Congress Catalog Card Number: 51-12949

THE MAPLE PRESS COMPANY, YORK, PA.

4. The Process of Combustion 114

5. Physical Properties of Fuels 166

6. Gas and Oil Burners 197

7. Coal-burning Equipment 236

8. Combustion in Engines 288

As an introduction to a study of fuels and their combustion, it is only

fine luc ting wc ter po 1

1899 1904 1909

Petroleum 0.4% Oil shale 1.6%

0Anthracite coal 4% Anthracite coal 0.5%^ Natural gas 0.5%

SOURCES OF 1949 PROPORTIONS OF ESTIMATED

Table 1-1. Estimated Mineral-fuel Reserves of the United States, Jan. 1,

Equivalent billions of net

Coal, billions of net tons:

* Carl E. Miller, The General Fuel Situation, Combustion, October, 1948.

Figure depicts the increase in proved reserves of petroleum as com

greatest. It is interesting to observe that the United States also enjoys

Table 1-2. Bituminous Coal Consumption by Users, 1950*

Millions of Per cent of

Total 453 8 100.0

* Bituminous Coal Institute.

Table 1-3. Percentage Yields from January, 1951, Petroleum Production*

Table 1-4. Natural-gas Consumption by Users in 1949*

products from the January, 1951, refinery production in the United

shown recently in the development of synthetic liquid and gaseous fuels.

1920 22 24 26 28 1930 32 34 36 38 1940 42 44 46 48 1950

the generation of electricity in large central stations and of providing

nated. Further increases of pressure and temperature were then respon

Wood Peat Lignite Subbi- Semi Anthracite

Production of Coal. In the preceding sections, the extent of the coal

Table 1-5. Coal Production by States, 1949*

Millions of net tons

which determine what type of mining will be most advantageous in a

coal is not a homogeneous substance, but consists of alternate layers or

head types. Figure 1-9 is a picture of a lump of cannel coal. It is uni

Ultimate Analysis. The ultimate analysis of a coal is a precise chemical

potassium hydroxide for weighing, while the water vapor is collected in a

Eventually the natural pressure decreases to the point where it will no

Fig. 1-11. Molecular structures of three normal paraffins.

Isoparaffins. Any compound with four or more carbon atoms may

carbon and 18 hydrogen atoms present in the molecule, hence this is an

CMOCO'f^COCONWHHHHHOOOfllOlOO SNNKNSS Tj)©lOi-t

'2 H e.2 i« c« s4n sn" s e e e OO&

Olefins.Olefins are unsaturated paraffin compounds. The maximum

CieH34 — ► CsHi8 + CsHi6

The and the still more unsaturated diolefins CnH2n-t, are

Naphthenes. Naphthenes are saturated, stable compounds with a ring

Table 1-8. Typical Analyses of Representative Crude Oils

Per cent, by weight

Petroleum from the Pennsylvania fields is composed principally of

paraffin family, but the number of naphthenes present is much greater

Fig. 1-16. Flow diagram of a petroleum refinery. (After API.)