HEAT FLUX RBQUIREMENTS FOR FAST PYROLYSIS

AND A NEW METHOD FOR GENERATING BIOMASS VAPOR

Thomas B . Reed & Craig D . Cowdery
Colorado School of Mines, Golden CO 80401
ABSTRACT
The term "fast pyro1,ysis" has been used to describe a pyrolysis
regime in which vapor produbtion is enhanced and char mimimized by
rapid heating. It has been found in the last decade that high yields
of primary pyrolysis oil can be achieved using fast pyrolysis. More
recently it has been found that the pyrolysis vapors can be converted
to high grade fuels using a catalyst. This makes fast pyrolysis of
biomass desirable for synthetic fuel manufacture
We present here results derived from the Diebold Integrated
Kinetic Model (DKM) that predict the time, the temperature, and the
products of pyrolysis and the heat for pyrolysis of cellulose as a
function of heating rates between 0.01 and 105aC/min. This range
covers very slow pyrolysis requiring days to fast pyrolysis occuring
in fractions of a second. The predictions are in good qualitative
agreement with experimental measurements.
We then compare the heat flux required for slow and fast
pyrolysis for particles with that which can be obtained with practical
heating devices. The comparison shows that convective and radiative
heat transfer is adequate for fast pyrolysis of small particles, but
not for large particles, due to conduction to the interior. We derive
the heat flow requirements for large bodies, the time for onset of
pyrolysis, and the depth of heat penetration in that time. We compare
the heat flux from various practical devices wi'th those observed in
"contact pyrolysis" experiments of Diebold and Lede on large
particles. The comparison shows that higher heat flux methods are
required for fast pyrolysis of larger particles
We have designed a "heat flux concentrator", based on the
experiments of Diebold and Lede to generate wood vapors for upgrading
to gasoline-like liquids. A rotating birch dowel is fed into a heated
copper block. Vapors emerge from the bottom and are condensed and
collected, or passed over a catalyst to establish optimum conversion
conditions. The pyrolysis rates and results of catalytic conversion in
these experiments will be described.
INTRODUCTION
Pyrolysis of biomass is a very old and complex process producing
variable quantities of charcoal, pyrolysis liquids and gases from
biomass, peat or coals (1). Pyrolysis was the principle source of
chemicals in Western society for about a century and could gain become
a major source, particularly if the products can be tailored to modern
needs through better understanding of the pyrolysis process or
improved upgrading of the products.
The relative amounts of charcoal, liquid and gas obtained from
pyrolysis depend on the time-temperature-pressure history of the
sample in a way which may never be completely understood. This has
led to apparent controversy over the magnitudes of kinetic factors,
energies and products when investigators compare results from
different experiments. Nevertheless great progress has been made in
the last decade in understanding the role of these variables in
controlling the nature and quantity of the products.
The term "fast pyrolysis" has been used to describe a pyrolysis
regime in which vapor production is maximized and the formation of
char is minimized by rapid heating. It has been found in the last
decade that it is possible to obtain high yields of pyrolysis oils or
68
gas using high heat flux ( 1 - 3 ) . It has more recently been found that
pyrolysis oil can be converted catalytically to high grade, high
octane motor fuel. Thus there is a strong motivation to understand
these relations.
While there is no hard boundary between "slow" and "fast"
pyrolysis, it is necessary to understand the relation between the time
and energy required for each and the different chemistry and the
possible mechanisms available for supplying this heat. We hope this
paper will help draw together the apparently disparate results using a
time-temperature-pressure kinetic model and lead toward more effective
methods of pyrolysis.

CELLULOSE AND BIOMASS PYROLYSIS KINETICS

Biomass is a composite plastic, consisting of cellulose, hemi-
cellulose and lignin with cellulose constituting more than half the
tot.al and giving most of the mechanical strength. Cellulose has the
best defined structure and thus has been studied more than lignin or
hemi-cellulose. Furthermore it constitutes more than half the
substance of most biomass and contributes most of the strength. Much
of the followi.ng discussion applies primarily to cellulose, but
qualitatively to all biomass.
The rates of pyrolysis reactions are usually represented in the
form of a first order Arhennius rate equation
dX/dt. = -X A exp(-E/RT) 1)
where X represents the quantity of any reactant X at time t and
temperature T. (A is the pre-exponential and E is the activation
energy. R is the gas constant). If the biomass is heated at a constant
rate of R QC/sec, simple substitution leads to
dX/dT (-AX/R) exp(-E/RT) 2)
The values for A and E in Eqs. 1 and 2 are usually measured by
thermogravimetric analysis (TGA) and describe the global pyrolysis
(including all solid state reactions) in terms of a single A-E pair
(2,3). While this oversimplfification may be justified over a narrow
temperature range for engineering purposes, the global approach
ignores preliminary solid state reactions and and thus is not able to
predict the change of products with heating rate, the time required
for pyrolysis or the temperature of pyrolysis.
Several reaction schemes for cellulose pyrolysis have been
proposed involving competing parallel and consecutive reactions
(2,3,5,6). Cellulose pyrolysis now appears to involve an initial
drying step (not discussed here), followed by two parallel reactions
which for instance differentiate between two subsequent paths as shown
in Table 1. The first reaction involves cross linking (trans-
glycolization) of the cellulose and leads to the formation of
charcoal. The second reaction is a depolymerization leading to a low
molecular weight liquid or solid depending on temperature and rate of
formation. This was called the "active state" by Bradbury and
Shafizadeh ( 4 ) . These more volatile components may either immediately
evaporate at low pressure or evaporate subsequently at higher
pressures.
AN INTEGRATED MODEL OF CELLULOSE BIOMASS PYROLYSIS KINETICS
The justification for studying the various pyrolysis reactions is
to predict the detailed course of pyrolysis under a wide variety of
conditions. Diebold has recently collected previous work on various
aspects of cellulose pyrolysis into a single integrated model of
cellulose pyrolysis ( 5 , 6 ) . The six reactions included are shown in
Table 1. The temperature at which the reaction rate reaches 1 0 - 6 / s and
l/s is listed for convenience in evaluating the relative strength of
each reaction.ote that the reaction producing char reaches a rate of
69
10-6/s at 208 o C , while the depolymerization reaction does not reach
10-6/s until 225 OC. Thus slow pyrolysis favors charing. However the
depolymerization reaction reaches a rate of l/s at 379 06, while the
charing reaction does not reach a rate of l/s until 703 O C . Thus
rapid heating favors depolymerization and volatilization over charing.
The Diebold Kinetic Model (DKM) permits calculation of the
relative amounts o f the products of pyrolysis as a function of time-
temperature history. The kinetic equation 1) using values in Table 1
are integrated using the Runge-Kutta technique on a microcomputer. A
second-derivative test is used to determine the time increment. The
model permits any type of heating history, but the results given here
are those for a constant heating rate as given in Eq. 2).
-_--_-__------__--------------------------------------------------
----_--_------_--__-----------------------------------------------
Table 1 - Simplified Cellulose Pyrolysis Reaction Scheme
kvc+ SECONDARY CAS
FEED -ka j
f- -
ACTIVE >
kvg
PRIMARY -kvT+ SECONDARY
VAPORS TAR ( v )

ka+ CHAR(d)tHzO k c 4 CHAR(a) t HzO

REACTlON k A E T(R=10-6) T(R 1)

sec-1 kJ O C O C

Cellulose to char t H 2 O ka 6.69Et05 109 2oa 703

Cellulose to k t i v e ke 2.80Et19 243 225 379
Active to Primary Vapor kv 6.79Et09 140 188 470,
at 1 atm
Active to primary vapor kv Y 3.20Et 14 198 231 440
vacuum
Active to Char(b1 t H20 kc 1.30Et10 153 224 518
Active to gas kv e 3.57Etll 204 335 65 1
Primary vapors to gas kv e 3.57Etll 204 335 651
Primary vapors to tars kv t 1.81Et03 61 70 700

iu’otes: Kinetic constants used in Diebold kinetic model. T(R = 1 0 - 6 )

is the temperature where the rate contsant R = 1 0 - 6 ; T(R = 1) is the
temperature at which R = 1.
------------_____-_-____________________-----------------------------
---------_______________________________--------------------------
While this model may not. explain all aspects of cellulose
pyrolysis, it goes a long wag toward predicting the changes observed
with time-temperature-pressure. Other values for kinetic factors and
other pathways should be substituted as they become known. In
particular, an improved model should include the effect of pressure
explicitly in the vaporization rates given by k, and k,, The nature .
and role of t.he “active” state needs to be better defined, since it
plays a key role in this and other schemes. We will use this model
here to predict heat flux requirements for slow and fast pyrolysis of
cellulose.
DEPENDENCE OF TIME, TEMPERATURE AND PRODUCTS ON HEATING RATE
The dependence of the pyrolysis time, tp and pyrolysis
temperature, TP on heating rate predicted by the DKM are shown in
Table 2 and Fig. 1 . The dependence of products on heating rate is
shown in Fig. 2 . (It is the nature of an exponential decay that it is
never complete. We have therefore arbitrarily taken the time and
temperature of pyrolysis as that time and temperature where the
reaction is more than 99.9% complete.)
70
 0.01 1650000 275 23.9 46.2 29.9
0.1 181200 302 : 12.4 73.2 14.9
1 19800 330 I 6.9 84.6 8.7
10 2142 357 : 4.2 90.3 5.4
100 234.6 391 : 2.6 94.3 3.1
1000 25.5 425 : 1.2 97.2 1.5
10000 2.778 463 ; 0.005 98.7 0.007
100000 0.3042 507 I 0.003 99.4 0.003
Notes: Values calculated from Diebold model.
---------_______________________________-----------------------
---------------------------------------------------------------------
Here it is seen that pyrolysis temperatures vary between 280 and
over 500oC as heating rates increase from 10-2 to 105 O/min. In slow
"commercial" pyrolysis, char yields are still increasing with
pyrolysis times of over a month, and pyrolysis is complete at
temperatures below 300oC. These correspond to the conditions that
have been used classically for the manufacture of charcoal (1). Most
of the kinetic data of Table 1 were acquired in experiments using
convenient "laboratory" heating rates of 1-1000C/min. The predicted
char yields of 2.6-6.9% and temperatures of 330-39OOC correspond to
those observed experimentally in this range.
4t even higher "heroic" heating rates, gas and oilyields are
still increasing with rates of over 10' oC/min where pyrolysis occurs
at about 460OC. These conditions correspond to the experimental
conditions of Diebold, Lede and Reed (7-9) produced by contact
pyrolysis (see below). Thus it is clear that heating rate is a primary
variable for controling products in pyrolysis.
While the calculations in this paper are for cellulose, we
believe the conclusions are qualitatively valid for all biomass.
HEAT OF PYROLYSIS AND HEAT FOR PYROLYSIS:
The energy required for pyrolysis, A h p has long been a subject
of interest to those involved in pyrolysis; yet no commonly accepted
value is available to an engineer wishing to design a pyrolysis
reactor, and even now it is n o t clear from the literature whether
pyrolysis is endothermic or exothermic. It has been reported to vary
from t370 J/g (endothermic) to -1700 J/g (exothermic)(5). One reason
there is no commonly accepted value is that it is a difficult
measurement to make under many of the conditions of pyrolysis (very
slow or very rapid heating, large quantities of condensible vapors
present, large temperature gradients in the sample). A more
fundamental reason however is that the value depends on the particular
form of biomass, the conditions of the experiment and the products
formed, so that there is a unique value for each experiment. We will
give here a method of calculating the heat of pyrolysis for known
heating rates and products which, while approximate, underlines the
factors that must be taken into account.
We define the "heat of pyrolysis" of biomass, a h p , as that heat
required to effect the phase change from biomass to char-liquid-gas at
71
the temperature where pyrolysis occurs. (We define the “heat of
pyrolysis as the heat required for the phase change plus the heat
required to reach this temperature - see below. ) A thermodynamic
calculation of A h p can be made for cellulose based on the idealized
reactions shown in Table 1 . The idealized charing react,ion for
cellulose can be written as Rx I:

Rx I l/n (CsH1005).---> 6 C t 5 HzO AH. = - 4 1 5 . 1 kJ/mole

Rx I1 l/n ( C S H I O O ~--->
) ~ C S H I O O S(v) ~ H = v6 0 . 0 kJ/mole
(Here we use the notation l/n (CsHl00s)~for cellulose to emphasize
its polymeric nature. The values above are calculated from the heats
of combustion or formation of the products and reactants.) The second
reaction represents the depolymerization of cellulose to levoglucosan,
with simultaneourvaporization. (Levoglucosan is the principal
1
decomposition product of cellulose.)
These two reactions can be combined in the appropriate ratio to
give a value f o r the heat of pyrolysis depending on the relative a-
mount of carbon formed, eg
AHp = 5 9 . 9 6 - 4 7 5 . 1 F, kJ/mole 3)
or Ahp 3 7 0 - 6 6 0 3 Fc J/g 4)

where Fc is the weight fraction of carbon produced in the charing

reaction. These values of the heat of pyrolysis should be considered
as idealized because pyrolysis of cellulose gives charcoal (not
exactly carbon) and other products beside levoglucosan. Nevertheless
they illustrate the necessity for knowing the products in measuring the
heat of pyrolysis.
bntal recently used a high pressure scanning calorimeter to !
measure the A h p of cellulose ( 2 ) . He varied the amount of char formed
by changing the pressure of the experiment. He found an approximately
linear change in the A h p of cellulose from - 1 7 0 kJ/g with a char \
I
production of 2 3 % (typical of slow pyrolysis) to t270 kJ/g with a char
production of 9% (typical of fast pyrolysis)(5). I
These values can be used in a linear equation to predict the heat
of pyrolysis on the basis of char content F c , ie
i
Ah, = 553 - 3142 F c kJ/g
where Fc is the fraction of charcoal produced. Note that the values
for the coeficients resulting from the experimental values are similar
in magnitude to those predicted from thermodynamic calculation in
Reaction 4 ) . The heats of pyrolysis calculated from the predicted
char production and pyrolysis temperatures are listed in Table 3 .
A quantity of more interest to the engineer is the heat for
pyrolysis. Wc define the “heat for pyrolysis”, hp as the sensible heat
required to raise a biomass particle to pyrolysis temperature, ( = c T p
where c is the heat capacity of the biomass) plus the heat required to
pyrolyse it, A h p . (If the products of pyrolysis are then heated above
this temperature, this heat must also be included.) Table 3 also shows
values of hp for pyrolysis of cellulose calculated from 2 ) and the
heat capacity of cellulose, 1 . 3 1 J/g-OC ( 0 . 3 2 Btu/lb-F). Note that the
variation of hp with heating rate is small compared with the variation
of Ahp.

72
 hp
J/g
0.01 275 23.9 -198 162
0.1 302 12.4 163 559
1 330 6.9 336 769
10 357 4.2 421 8R9
100. 39 1 2.6 47 1 084
1 no0 425 1.2 51 5 072
10000 463 0.005 553 159
1noooo 507 0. 003 5.53 21 7

Notes: Char yield calculated from Diebold Model. A h p calculated

- 553-3142Fc : hp calculated from
hp A h p =-c(TP - T o ) with c = 1.3 J/g-oC for cellulose
________________________________________----------------
________-__--_-_________________________----------------------
HEAT FLUX REQUIRED FOR SLOW AND FAST PYROLYSIS
A. Small Particles & Slow Heating
The average heat flux required to pyrolyse a particle is given by
6 = AhppV/tpA 6 )

where 4 is the rate of heat supply per unit area, is the density, V
is the volume and A is the heated surface area of the particle. If
the particle is a cube, this simplifies to

4= h P P L / 6 tp 7)

where L i s the length of the cube edge.

The data in Table 2 and 3 can be combined with this equation to
predict. the magnitude of heat flux required for slow and fast
pyrolysis. Table 4 shows the heat flux required to pyrolyse a 1 cm
cube of cellulose (or biomass). The values shown are calculated for
heating the particle just to pyrolysis temperature. They assume that
the particle is approximately isothermal and pyrolysis occurs
everywhere at once.
The above calculation assumes that there is no resistance to heat
transfer, and the particle will be essentially isothermal. However
most forms of wood and biomass are relatively poor conductors of heat
and at higher heating rates the results shown above will not be valid.
The validity of these assumptions is tested by the Biot number, given
by
Ne = Heat flux to surface/Heat flux to interior
Ne H L/K 8)

(where H is the heat transfer coefficient, characteristic of whatever

method of heating is used; L is a characteristic length of the
particle, typically the cube root of the volume: and K is the thermal
conductivitY.)
Unfortunately the heat transfer coefficient H is generally
applied to convection or radiation heating with small temperature
differences, where heat .transfer is a strong function of A T . However
for high temperature radiation and contact pyrolysis sources, the
73
change in AT is relatively small and the Biot number is approsimated
by
N~ = &/KAT

0.01 275 162 : 8.208-06 7.1E-06

0.1 302 559 2.57B-04 2.2E-04
1 330 769 : 3.23E-03 2.88-03
10 357 889 : 0.035 0.030
100 391 984 : 0.35 0.304
1000 425 1072 : 3.50 3.0
10000 463 1159 : 34.78 30.2
100000 507 1217 : 333.41 289.9

Note: Heat f l u s calculated from 6 = hp L/6 tp

Biot number calculated from Ns = h P p L / 6 t p K A T P
X = 0.0023 J/s-cm-OC
.- .- .- ._.-.-. -. -. -. _. ._. .- .- .- .- ._.-.-. _. ._. -. -. -. .- .- .- .- .-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-. -. -. -. -. -. -. -. .- .- .- .- .-.-.-.-.-.-.-.-.- - - - - -
F o r Biot numbers less than 1 , the particle will be nearly
isothermal and will dry, then pyrolyse in sequence. F o r Biot numbers
larger than 1 , the resistance to heat transfer within the particle
becomes large compared to that of the heat source and steep gradients
esist. in the particle. In this case the pyrolysis wave travels from
the outside of the particle to the inside, producing simultaneous
drying and pyrolysis.
The Biot number for a 1 cm cube is also shown in Table 4 ,
calculated for the particle heating rates shown and assuming an
average temperature difference of 500 ii in. Here i t can be seen that
the assumption o f an isothermal particle is valid for slow heating
rates and small particles. However for a heating rate larger than 100 \
OC/min and a heating rate of 0 . 3 5 W/cma there will be steep gradients
in the particle, so that the heat flus will be altered and the drying
and pyrolysis will occur simultaneously.
B. Large Particles IL Rapid Heating

For high heat flux and larger particles, when the Biot number
esceeds 1 , it is necessary to calculate the non-steady state heat
transfer for the particular particle geometry and surface temperature.
While this can be quite complex for most cases, it is relatively
simple for the one dimensional steady state case esperiment described
by Lede in contact pyrolysis ( 8 ) . In t.his case a heated disk supplies
sufficient heat to a beech dowel to pyrolyse and vaporize it at a rate
V. The steady state temperature distribution in the rod is given by

T(s) = To t (q/pcV)exp(-Vx/B() = (Td - To)exp(-Vx/*) t To 10)

where T(r) is the temperature at a distance x from the heat source of
strength q , TD is the temperature at the pyrolysing interface and TO
is the initial temperature of the rod. The density of' the wood is p , c
14
 is the heat capacity, r& the thermal diffusivity and V the rate of
pyrolysis. IHere we have taken the following values used by Lede for
consistancy; heat c a p a c i t y , ~:: 2 . 8 0 J / g - " C ; density = p = 0.70 g/cm*. ;
thermal conductivity K = 0 . 0 0 2 3 J / s - c m - o C ; thermal diffusivity = (II =
0.0012cmz/s
__-______________-______________________--------------------
________________________________________--------------------
Table 5 - Steady State Fusion-Pyrolysis of Birch Rod

HEAT F L U X q 10.00 100 1000 W/cmz

Velocj ty (b) V 0.01 0.11 1.14 cm/s
Penetration ( c ) Xc 0.10 0.01 0.001 cm
Heat stored Q 89.76 8.98 0.90 J/cmz
Induction Time (e) ti 9.0 ' 0.09 0.0009 S

Notes and assumpt.ions: (a) Tp = 4 6 6 O C , To = 2 0 O C

( t i ) The pyrolysis velocity, V was calculated from V = q/(pc)iTp-To)
( c ) 'The heat penetration X(l/e) was taken to be the distance at
which the t,emperature had fallen to l/e of T p , V X / u = I , S = W V .
(d)The heat stored in the rod, Q , was calculated as the integral
of the sensible heat between x = O and s = o o , = e c p (Tp-Ta)/V
(e) The induction time ti is the time required to establish the
steady state temperature gradient, t. = Q / G . This assumes the
heat transfer intensity is constant before steady state is
reached.
( f ) Temperature Distribution T = T. esp(-Vs/aO + To
d _ ___- -_______________ _
_ __ _ _______- -_-_- _
__ - -_-_- -_-_- -_-_- _
- -_-_- -_-_- -_-_- _
- -_- _- -_-_- -_---------------- -- -- -- -- -- --------- - - -
Note in Table 5 that the velocity and heat penetration increase
linearly with heat flus. However, the induction time required to
reach steady state varies inversely as the square of heat flux. At
low flus, considerable char may build at the interface s o that the
steady state condition may never be reached.
I HEAT TRANSFER MECHANISMS FOR FAST PYROLYSIS
The magnitudes of heating rates which can be obtained from
various methods of heat transfer are shown in Table 6 (Heed, 1 9 8 1 C u
Xtn.1. Comparing the fluxes shown in Table 4 to the values in Table 5,
it can be seen that the high heating rates required for fast pyrolysis
of small particles can be achieved with convection, radiation or
conduction.
Convection is the least satisfactory heat transfer mechanism for
fast pyrolysis because the water and pyrolysis vapors produced during
pyrolysis interfere with heat transfer. A l s o convection from gas
sources with temperatures above 6 0 0 O C is unsuitable for oil
production because they crack the pyrolysis oils which are only stable
to about 6 0 0 O C . Low temperature radiation sources (below 1 0 0 0 0 C are
also unsatisfactory for fast pyrolysis of larger particles.
The production of pyrolysis oil is favored by radiation from a
high temperature source or by "contact pyrolysis" (see below). In the
case of radiation, the solid is heated rapidly, but the vapors are
largely transparent and transient and s o are not overheated ( 1 2 ) .
Unfortunately black body sources with temperatures above 2 0 0 0 O C are
expensive and difficult to use.
CONTACT PYROLYSIS
A new method of heat transfer, "contact pyrolysis", has been
developed in the 1980's and appears to be especially suited for the
production of pyrolysis oil vapor.
75
 Typical Maximum Maximum
Temp diff Ht T$ansf Ht tr coef
AT 9 H
K W/cm2 W/cma -K
CONVECTION
Gas Free Convection 500 10 0.02
Gas forced convection 500 300 0.6
Air-Gas Flame 1500 200 0.13
Oxy-Acetylene flame 3000 3000 1 .oo
CONVECTION, ELECTRONIC
Electric arc 10,000 20,000 2

RADIATION Surface T
O K

Black Body 773 20 0.026

1,273 150 0.12
2,273 1,500 0.66
5,273 44,000 8.3
Focused C 0 2 laser 100,000 NA
CONTACT PYROLYSIS 873 3,000 3.2
-------_--------------------------------------------------------------
-_---_-------_--_---------------------------------------------------
In 1980 Diebold showed that a moving hot wire would cut through a
piece of wood at rates of several cm/sec with apparently no production
of charcoal (7). Led& et a1 have since shown that a wood dowel can be
consumed at rates up to 3 cm/sec corresponding to heat transfer rates
of 3000 J/cmZ by pressing it against a heated disk (8). Furthermore
the heat transfer rate is directly proportional to pressure, and a
pressure of 30 atmospheres was used to attain the above heat transfer
rate. The heat transfer is also proportional to the difference
between the disk teaperature and 466 O C . This was interpreted to show
that the wood had a "fusion behavior" above 4 6 6 O C (8,13). Reed has
now developed a "heat flux concentrator" (see below) using a copper
block to produce vapors for catalytic conversion to fuel ( 9 ) .
Contact pyrolysis has several interesting and non-obvious
features :
o The thermal conductivity of metals is typically 3-4 orders
of magnitude higher than that of hot gases or biomass, and so the heat
transfer produced by direct contact is proportionally high.
o The use of pressure at the heated interface retards the
vaporization of the pyrolysis products so that they do not
interfere with heat transfer, and instead the biomass pyrolyses
to an oii or foam
o When the resulting oil o r foam is squeezed out of the interface
region, it is immediately in a low pressure region and can
vaporize very rapidly
o The rubbing contact removes char or ash which would otherwise
interfere with heat transfer
A HEAT FLUX CONCENTRATOR FOR CONTACT PYROLYSIS PRODUCTION OF
PYROLYSIS O I L VAPOR
In order to produce vapors for catalytic conversion to
hydrocarbon fuels, we have developed a "heat flux concentrator" shown
76
 i n F i g . 5 a n d 6. I n o p e r a t i o n , a wood d o w e l i s r o f a t e d i n a d r i l l
p r e s s a n d f o r c e d i n t o a 1.2 c m d i a m e t e r t a p e r e d h o l e i n t h e h e a t e d
c o p p e r b l o c k shown i n F i g . 5. T h e r o d i s f e d a t a r a t e o f 0.2 - 0.25
cm/sec. The v a p o r s e s c a p e t h r o u g h 1 2 h o l e s t h r o u g h t h e bot.tom o f t h e
block and a r e t h e n captured in traps and in a gas burette.
. A l t e r n a t i v e l y , t h e vapors t r a v e l w i t h o u t c o o l i n g t o a c a t a l y s t t e s t
f u r n a c e a s shown i n F i g . 6 w h e r e t h e p r o d u c t s a r e c a p t u r e d ( 9 1 .
The h e a t f l u s c o n c e n t r a t o r h a s b e e n o p e r a t e d t o produce p y r o l y s i s
v a p o r as a n e n d p r o d u c t , , o r t o t e s t t h e o p e r a t i o n o f v a r i o u s
~ ~ . t a l y s t . s .4 p r e l i m i n a r y m a s s b a l a n c e o n p y r o l y s i s o i l p r o d u c t i o n is
s h o w i n Table 7 f o r pyrolysis runs. W e o b s e r v e t h a t t h e w a l l s of t h e
p y r o l y s e r a n d t h e e x i t h o l e s hecome c o a t e d w i t h R h a r d , s h i n y f o r m o f
c a r b o n Like p e t r o l e u m c o k e . W e now b e l i e v e t h a t t h e r e i s a h i g h
r e s i R t a n r e t o vRporizat.ion which c a u s e s plugging of t h e p y r o l y s e r
3 f i . e r t h e f e e d i . n g o f a h o u t 5 g . We a r e c u r r e n t l y m o d i f y i n g t h e d e s i g n
t o i m p r o v e \ v a p o r i z a t i o n . .A few c a t a l y s t . r u n s h a v e a l s o b e e n made a n d
t h e y x i 1 1 b e d e s c r i b e d i n more d e t a i l l a t e r .
_______-_-______________________________-------------------
_ _______________________________________--------------_-----
- P r e l i m i n a r y Mass B a l a n c e o n C o n t a c t P y r o l y s i s E s p e r i m e n t s
Table 7

Pyrulysis TP Wood g liq/ g gas/ g coke/ Mass out/

Run O C Consumed g wood g wood g wood Mass i n

5P 550 2.56 0.61 0.10 0.27 0.98

8P 600 5.59 0.57 0.10 0.33 1.00
J 9P 500 6.10 0.52 0.12 0.25 0.89
1I P 7 00 4.21 0.57 0.13 0.14 0.85
I
___-__----______________________________----------------------
__---_----______________________________-----------------_-------
Acknowledgements
T h e a u t h o r s would l i k e t o thank t h e S o l a r Energy Research
I n s t i t u t e f o r s u p p o r t o f t h i s work u n d e r C o n t r a c t HK-6-06059-1 a n d t h e
Nidwesl. R e s e a r c h I n s t i t u t e for t h i e r s u p p o r t , The a u t h o r s w o u l d
e s p e c i a l l y l i k e t o t h a n k t h e i r c o l l e a g u e s a t SERI a n d e l s e w h e r e whose
r w r k h a s s u p p l i e d t h e b a s e on w h i c h t h i s p a p e r h a s b e e n w r i t t e n .

77
 REFERENCES

1. Stern, and Harris, E. E., "Chemica Processing of Wood"

Chemical Publishing Co., New York, 1953
2. Antal, M. J . , Jr., "Biomass Pyrolysis: A Review of the
Literature, Part I-Carbohydrate Pyrolysis", in Advance8 in Solar
Energy, ed K. W. Boer and J . A . Duffie, American Solar Energy
Society, Boulder, CO, Vol 1 , pp 61-112.

3. Antal, M. J., Jr., "Biomass Pyrolysis: A Review of the

Literature, Part 11-Lignocellulose Pyrolysis, in Advances in Solar
Energy, ed K. W. Boer and J . A. Duffie, American Solar Energy and
Plenum Press, New York, Vol 2 , pp 175-255.
4 Bradbury, A. G . , Sakai, Yoshio and Shafizadeh, F., "A Kinetic
Model f o r Pyrolysis of Cellulose", J. Appl. Polymer Science, 23
3271 (1979)
5. Diebold, J. P . , "The Cracking Kinetics of Depolymerized Biomass
Vapors in a Continuous Tubular Reactor", Thesis T-3007, Colorado
School of Mines. Golden, CO 1985.
6. Diebold, J. P., Scahill, "Ablative, Entrained Flow Fast Pyrolysis
of Biomass", in Proceedings of the 16th Biomass Thermochemical
Contractors' Meeting, Portland, OR May 8-9, 1984, PNL-SA-12403,
1984. See also, Diebold, J . P. "A unified Global Model for the
Pyolsis of Cellulose", (in preparation).
7. Diebold, J. P., "Ablative Pyrolysis of Macroparticles of Biomass,"
Proceedings of the Specialists Workshop on the F a s t Pyrolysis of
Biomass, Copper Mountain, CO., Oct. 12 1980, Solar Energy Research
Institute, Golden, CO. 80401, SERI/CP-622-1096.
8. Lede, J., Panagopoulos, J., Li, H. 2. and
Villermaux, J . , "Fast Pyrolysis of Wood: Direct Measurement and
Study of Ablation Rate", Fuel 64, 1514 (1985)
9. Reed, T. E., Diebold, J. P., Chum, H. L., Evans, R. J.
Milne, T. A. and Scahill, 3. W., "Overview of Biomass Fast
Pyrolysis and Catalytic Upgrading to Liquid Fuels," in
Proceedings of the American Solar Energy Society. K. W.
Boer, Ed., Boulder, CO June 11, 1986.
10. Roberts, A. F., "The Heat of Reaction During the Pyrolysis of
Wood", Combustion and Flame u, 79 (1971).
11. Reed, T. B., Diebold, J. P. and Desrosiers, R., "Perspectives in
Heat Transfer for Pyrolysis", Proceedings of the Specialists
Workshop on the Fast Pyrolysis of Biomass, Copper Mountain, CO.,
Oct. 12 1980, Solar Energy Research Institute, Golden, CO. 80401,
SERI/CP-622-1096.
12. Chan, W. -C. R., Kelbon, M. and Krieger, B. B., "Product Formation
in the Pyrolysis of Large Wood Particles", Fundamentals of
Thermochemical Biomass Conversion, ed. R. P. Overend et al,
Elsevier, New York, 1985.
78
 500

V
0

al
L -
5 400 -
Io
L
al -
n
E
t Pyrolysis
Temp - O C

300

-2 0 2 4

Log h e a t i n g r a t e -
'C/rnin
Figure 1 - Time and t e m p e r a t u r e f o r p y r o l y s l r o f c e l l u l o s e as
p r e d i c t e d b y D i e b o l d Kinetl'c Model

1OD
V
I
WATER ____c_3__

---
60
/
OIL/GAS
y.
0

bp

, Ot--l---
-. 0 d
Log h e a t i n g r a t e - C/min

Figure 2 - Products o f c e l l u l o s e p y r o l y s i s as p r e d i c t e d by
D i e b o l d K i n e t i c Model
79
 -2 0 2 4

Log heating r a t e - OC/min

F i gure 3 - Heat of p y r o l y s i s a n d heat f o r p y r o l y s i s as predicted
from char y i e l d s a n d p y r o l y s i s temperature

N
E ,
u
\
2
I

al
c,
0
L
L
al
u-
VI
E
-’ 1
0
L
c, NB = 1
c,
0
al
L + isothermal high g r a d i e n t s -+
m
0
-I

-2 0 2 4

Log heating r a t e - ‘C/min

Figure 4 - Heat t r a n s f e r r a t e required t o support heating r a t e

80
 c

t
I