KINETICS OF THE CATALYZED AND

UNCATALYZED LIQUID-PHASE HYDRATION

OF PROPYLENE OXIDE

A. L. BENHAM and FRED KURATA

University of Kansas, Lawrence, Kansas

The kinetics of the catalytic and noncatalytic hydration of propylene oxide were studied in a continuous reactor. Both
the catalytic and noncatalytic reaction were studied over a temperature range of 100" to 300°F. The feed ratios ranged from
2.5 to 10 Ib. water/lb. propylene oxide. In the case of catalytic reaction, the catalyst concentration of sulfuric acid ranged up
to 0.25 wt. 70 of the total feed.
The study showed the noncatalytic reaction to be pseudo first order with respect to propylene oxide, and an equation was
found for the rate constant. The catalytic reaction was found t > be pseud3 second order with respect to propylene oxide, and
the effect of the catalyst Concentration is shown graphically.

Propylene glycol (1-, 2-propane-

diol), which was first prepared by
Wurtz(l0) in 1859 by hydrolysis
of propylene glycol diacetate, was
first produced on a commercial
scale by Carbide and Carbon Chem-
icals Co. in 1931 via the chlorhy- I / 650'. 1/4' CU TUBING
8" PIPE
6"PlPEi 1 CHILLER PRODUCT
drin process. Curme and Johns- 9 KW
ton(3) state that the production of PREHEATER
propylene glycol has become in-
creasingly important, and in 1950
85,000,000 lb. was produced. In P REHEATER DRAIN 9 K.W. MAKE UP
1951 propylene glycol was being HEATE WATER
produced in five plants located in ROTAMETERS
STEAM
three states(3). Although the ma- HEATER
j o r portion of propylene glycol is
manufactured by the chlorhydrin
process, there is increasing interest
in its production by the hydration If I II SURGE TANK TURBINE PUMP
of propylene oxide.
This investigation, made because
of the lack of information on this
reaction in t h e literature, was set
FIG.1. FLOWDIAGRAMOF REACTOR.
up f o r study of the kinetics of this
reaction. Some data on the distri-
A. L. Benham is a t present with the
University of Michigan, Ann Arbor, Michipan.

Page 118 A.1.Ch.E. Journal March, 1955

bution of products when propylene
and dipropylene glycols a r e formed
were also desired. The study was
limited to a liquid-phase, homoge-
neous reaction, and both the un-
catalyzed reaction and the reaction
catalyzed with sulfuric acid were
included in the study. The reaction
was studied over a temperature
range of 100" to 300" F.; the feed
ratio was varied from 2.5 to 10 lb.
waterllb. propylene oxide; and the
catalyst concentration ranged up to
0.25 wt. % of the feed.

FIG.2. REACTORCONTROLPANEL.-+

APPARATUS AND PROCEDURE

Apparatus. The reactor may be de-
scribed with reference to the flow
diagram shown in Figure 1 and the
photographs in Figures 2, 3, and 4.
In general the reactor consisted of a
650-ft. section of 1/4-in.' copper tub-
ing wound on the inside pipe of two
concentric pipes with water circulat-
ing in the annulus t o maintain the
desired temperature. The coil was
wound so that the water could circu-
late on both sides of the coil. The
reactants mixed immediately before
entering the reactor, and the reacted
mass was quenched in an ice bath im-
mediately after leaving the constant-
temperature reactor.

FIG.4. OVERHEAD VIEW OF REACTOR.?,

FIG.3. SAMPLINGTHE REACTOR

PRODUCT.

Vol. 1, No. 1 A.1.Ch.E. Journal . Page 119

 The two reactants were water and Thermocouples 1 through 4 were the circulation system. The reac-
propylene oxide. The water was con- soldered onto the outside of the re- tion temperature was easily con-
tained in a 20-gal. steel tank and the actor tube wall. Calculations showed trolled within 1"F. for any given
propylene oxide in a 5-gal. steel con- that for the conditions chosen for run. Feed water was then fed into
tainer, When catalytic runs were this study, these thermocouples would
made, the sulfuric acid was mixed indicate essentially the temperature the reactor system a t a specified
with the feed water. For these runs of the fluid in the tube. The other rate and the feed greheater was
the steel feed-water tank was re- thermocouples were placed in wells in used t o heat it a few degrees above
placed by two 9-gal. stainless steel the fluid stream. The temperatures in the desired reaction temperature.
tanks. A nitrogen cylinder was con- the various sections of the reactor Then cold propylene oxide was me-
nected to both the water and the never varied more than 1°F. from tered into the system and the water
propylene oxide containers and a one point to another. preheater was regulated to bring
pressure regulator maintained a con- The reactants for this project were the temperature of the reactants,
stant pressure on them. The feed distilled water and propylene oxide, after mixing, t o the desired reac-
water left the tank through 114411. which was obtained from Carbide and tion temperature. The individual
copper tubing which led to a needle Carbon Chemical Corp. The specifica-
valve, then t o a rotameter, a pre- flow rates were indicated by rota-
tions for the propylene oxide ( 2 ) fol-
heater consisting of a steam heat ex- low : meters and controlled by needle
changer, a 9-kw. electric immersion
preheater, and a check valve t o the Acidity ................................. 0.05 wt. 70 max. as acetic acid
point where it mixed with propylene Water ................................. 1.0 wt. YO max.
oxide. The propylene oxide was caused Residue ................................ 0.005 g./ml. max.
t o flow through a needle valve, a Color .................................. 15 pt-Co APHA units, max.
rotameter, and a check valve t o the Suspended matter ....................... Substantially free
point where it mixed with the feed Distillation a t 760 mm. .................. 1.b.p. 30.0"C. min.
water. The reactant solution then 95 ml. 40.0"C. max.
passed into the reactor proper. As a
safety precaution, a rupture disk was
placed in the mixing section. The re- Procedure. The circulating pump valves. Pressure was maintained in
actor proper was made of 650 ft. of was started first and the reactor the reactor by a pressure-relief
1/4-in. copper tubing which was was heated t o the desired tempera- valve downstream from the quench-
wound on an 8.5-ft. section of 6-in. ture by controlling the heater in ing zone so that the reactants were
pipe, which was then inserted coaxial-
ly in an 8-in. pipe. Water flowed in
the annular space between two pipes TABLE~.-UNCATALYZEDREACTION-RATE CONSTANTS
in order t o maintain the reaction Reaction-
mass a t a constant temperature. This Feed Fractional rate
water was kept a t a constant tem-
perature by circulating i t through a n ratio, conversion constant,
outside circuit where it flowed first Reaction Ri VRfF OE kl,
into a surge tank, then through a Run temp., O F . lb. H20 (cu ft.) (sec.) Pro, X lb. mole
. . --__
turbine pump, through a heater, and
back into the bath. Pressure was
Ib. P r o (lb. mole) (cu. ft.) (sec.)
maintained in the bath circuit t o 1 200 10.0 55.9 0.0357 9.000649
keep the water in the liquid phase. 2 200 10.0 111.7 0.0572 0.000526
After the reactant mass had passed 3 200 10.0 227.7 0.111 0.000511
17 200 10.0 219.2 0.120 0.000645
through the 650 ft. of tubing, it was 7 200 10.0 319.2 0.187 0.000593
immediately passed through an ice 18 200 10.0 312.7 0.173 0.000580
bath, where it was chilled in order 8 200 10.0 446.9 0.234 0.000604
effectively to stop the reaction. The Representative 0.000625
solution was then passed through a 14 200 5.0 217.0 0.081 0.000405
' pressure relief valve and was col- 15 200 5.0 310.0 0.102 o.000331
lected in a sample container. 16 200 5.0 867.7 0.232 0.000303
A pressure gauge was placed in the Representative 0.00037
line preceding the reactor section and 27 200 2.5 102.5 0.0454 0.000535
one was placed following the quench- 28 200 2.5 205.0 0.0550 0.000275
ing section. Nine thermocouples were 23 200 2.5 356.5 0.0760 0.000244
included in the system in order to 24 200 2.5 510.5 0.111 0.000242
measure temperatures. Their loca- Representative 0.00024
tions were as follows : 4 250 10.0 55.9 0.154 0.00298
1. Ten feet from the inlet in the 5 250 10.0 111.7 0.242 0.00247
reactor tubing 6 250 10.0 223.5 0.460 0.00275
2. Fifty feet from the inlet in the 9 250
~.. 10.0 319.2 0.536 0.00239
reactor tubing 19 250 10.0 319.2 0.502 0.00264
10 250 10.0 447.0 0.697 0.00216
3. One hundred and fifty feet from Representative 0.0025
the inlet in the reactor tubing
4. F o u r hundred feet from the in- 20 250 5.0 215.9 0.280 0.00151
21 250 5.0 309.9 0.372 0.00148
let in the reactor tubing 22 250 5.0 867.3 0.724 0.00147
5. Exit of circulating water from ReDresentative 0.00148
reactor 29 250 2.5 102.5 0.I04 0.00060
6. Inlet of circulating water to 30 250 2.5 205.0 0.139 0.00107
reactor 25 250 2.5 356.0 0.194 0.00073
7. Six hundred and fifty feet from Representative 0.00087
inlet in the reactor tubing just out- 11 300 10.0 55.9 0.392 0.00888
side reactor 31 300 10.0 219.2 0.802 0.00728
8. After preheater in feed water 32 300 10.0 319.2 0.916 0.00765
line Representative 0.0080
9. In the mixing section 12 300 15.0 63.5 0.548 0.01242

Page 120 A.1.Ch.E. Journal March, 1955

 --ORIGINAL
--CALCULATED

0 100 200 300 400 SO0 600 700 800

,VR/F, cu. ft.stc./lb.mol.

FIG.5. UNCATALYZED
DATA, CONVERSIONOF PROPYLENE
OXIDE TO FIG.6. ARRHENIUS
EQUATION
FOR
GLYCOL,250" AND 300°F. UNCATALYZED
REACTION.

TABLE2.-cONSTANTS CHARACTERISTIC O F UNCATALYZED REACTION

maintained in the liquid phase.
The reactor product was chilled Arrhenius constants
after leaving the constant tempera-
ture zone and was collected in a
sample bottle. A sample was not -E
kl=Ae-
taken until after the flow rates and RT
temperature had been maintained
over a period of time which would Feed Energy of
allow t h e reactants to pass through ratio, activation,
the reactor. The sample was ana- Ri A, E,
lyzed by the method used by Davis, lb. €I20 lb. mole B.t.u.
Von Waaden, and Kurata ( 4 ) . lb. Pro (cu. ft.) (sec.) lb. mole
When catalytic runs were made, 10.0 1.868 X 1C6 25.6 X lo3
the sulfuric acid catalyst was 5.0 1.125 x l a 6 25.6 X lo3
mixed with the feed water. In or- 2.5 0.668 X 1C' 25.6 x 103
der to stop the reaction effectively,
the product sample was immediate-
ly neutralized with sodium hydrox- Concentration function
ide.
Under certain conditions two liquid
phases are possible f o r this system.
The solubility data of Wickert, Tamp-
lin, and Shank(9) were extended f o r
use a t higher temperatures. Condi- Reaction a, h,
tions for study were then chosen temperature, OF. lb. mole unitless
which would ensure the reactant mass (cu. ft.) (sec.)
being in the single liquid phase.
200 1.41 x10-3
250 4.40 X
RESULTS
300 1.11x 10-4
Woncatalytic Reaction. The results
of thirty noncatalytic runs a r e
summarized in Table 1. The un-
catalyzed reaction was found to be measurable quantities (V12,F , X , A+B-+C
pseudo first order with respect to A') was further developed starting
the propylene oxide concentration. with the material balance over a or
The reaction-rate constants were differential section of the flow re-
calculated from the integrated actor, a s suggested by Hougen and
Watson(5).
vR 1
--=--[(I-A') In--+A'X]
1
equation below, which was derived F ki 1-x
from the basic assumption that the
rate of reaction is proportional to
the mole fraction of propylene ox-
CH3 - CzH30 + H2O+ where
A' = initial concentration of pro-
ide. The relationship between the CH3 - CHOH -CHzOH pylene oxide, mole fraction

Vol. 1, No. 1 A.1.Ch.E. Journal Page 121

 F =total feed rate, lb. moles/ F = total feed rate, lb. moles/
SW. SeC.
kl = first-order reaction - rate k z = second-order reaction-rate
constant, lb. mole/ (cu. ft.) constant, lb. mole/ (cu.ft.)
(see.) (SW.1
V R= reactor volume, cu. f t . N A = concentration of propylene
X = fractional conversion of oxide a t any time, mole
propylene oxide fraction
A representative reaction con- V , = reactor volume, cu.ft.
stant was found for each series of X = fraction conversion of pro-
runs at a given temperature and pylene oxide
feed ratio but with varying feed This expression was derived on
rhtes. The representative r a t e con- the assumption that the rate of re-
stant was found by fitting a calcu- action is proportional t o the square
lated curve to the experimental of the mole fraction of propylene
points as shown in Figure 5. Simi- oxide and that there is no compet-
lar procedures were carried out at ing reaction for propylene oxide.
other temperatures and feed ratios. This latter asumption is true ex-
The effect of temperature on the cept for twelve of the catalytic runs
reaction-rate constant at three dif- in which detectable amounts of di-
ferent feed ratios is shown in Fig- propylene glycol were formed. The
ure 6. results of these twelve runs are
The value of the reaction-rate given in Table 3. It is seen from
constant was found to change with this table that the conversion of
the initial propylene oxide concen- propylene oxide t o di-propylene gly- FIG. 7. CONCENTRATION EFFECTFOR
tration. This effect is correlated col did not exceed 2% and that the UNCATALYZED REACTION.
for three different reaction tem- conversion to propylene glycol
peratures, as shown in Figure 7. varied from 23 to 90%. The effect
The equation of the line is of the of the competing reaction upon the for other catalyst concentrations
form calculation of the rate constant was and feed ratios.
k l = uR? therefore negligible. The effect of temperature on the
where A representative r e a d ion-r at e reaction-rate constant is shown in
a = c o n s t a n t for a given tem- constant was found for each series Figure 9 f o r different catalyst con-
perature, Ib. moles/ (cu. ft.) of runs at a given temperature, centrations and feed ratios.
(see.) feed ratio, and catalyst concentra- For conditions of constant tem-
b = constant f o r uncatalyzed re- tion but a t different throughput perature and acid concentration,
action, unitless rates. The rate constant was the effect of the initial feed ratio
kl = reaction-rate constant, lb. found in the same manner as f o r on the rate constant was correlated
mole/ (cu.ft.) (see.) the uncatalyzed reaction. Figure 8 as shown in Figure 10. The curve
R,=feed ratio of reactants, lb. shows the computed and the ex- is of the form k, = uRab.
waterilb. propylene oxide perimental curves of fractional The effect of the catalyst concen-
The values of the constants for conversion vs. V,iF a t loo", 125", tration on the rate constant a t re-
the Arrhenius equation as obtained and 150°F. for a feed ratio of 10 action temperatures of loo", 125",
from Figure 6 and the constants lb. water/lb. propylene oxide when and 150°F. and with feed ratios of
obtained f o r the curves in Figure 7 the catalyst concentration is 0.05 5 and 10 is shown in Figures 11
are given in Table 2. wt. %. Similar curves were drawn and 12.
Catalytic Reaction. The summar-
ized results of the reaction cata-
lyzed with sulfuric acid are avail-
able elsewhere.* This reaction was
found t o be pseudo second order
with respect t o the mole fraction of
propylene oxide. The rate constants
were computed by use of the inte-
grated form of the second-order
rate expression f o r the reaction :

A' X + (A' - 1) In (1 - S)']

where --- ORIGINAL
I I Pr
R;= 10 b . w / b.

CALCULATED
0

A' = initial concentration of pro-

pylene oxide, mole fraction 0 80 LOO 150 200 250 300 350 400 450
*Deposited with the Photodwlication Service, VR/F, cu.ft. rec./lb.mol.
American Documentary Institute, Library of
Congress, Washington 25 I>.& as document FIG. 8. CATALYTIC DATA,CONVERSION OF PROPYLENE OXIDE TO GLYCOL,
4486, obtzinable for $1.25 for microfilm or
photoprints. 0.05 WT. c/b H,SO,, 10 LB. H 2 0 / ~Pro.
~.

Page 122 A.1.Ch.E. Journal March, 1955

 F o r the uncatalyzed reaction t h e
average percentage of deviation of
the correlation from the experi-
mental points was found to b e 9.6,
and the standard deviation was
0.024. The average percentage of
deviation for the catalyzzd reac-
tion was 3.8 and the standard devi-
ation was 0.026.
Since other experimental data
for this reaction a r e not available
in the literature, a direct compari-
son of this study cannot be made;
however, data a r e available for a
few other epoxide reactions.
Bronsted et a l . ( l ) and Lichten-
stein and Twigg(6) studied the
liquid-phase hydration of ethylene
oxide. Both groups f o m d t h a t the
uncatalyzed reaction was first order
1.60 1.64 1.68 1.72 1.76 180 1.84 with respect to ethylene oxide. This
INVERSE TEMPERATURE, I/T. I/.RxlOs
may be 'compared with the pseudo ACID C O N C E N T R A T I O N . U ~ . % H ~ S O ~
FIG.9. ARRHENIUS
RELATIONSHIP
FOR first-order rate found f o r the un-
catalyzed reaction of this study. FIG. 11. EFFECTO F CATALYST CON-
CATALYZED
REACTION. CENTRATION UPON REACTIONRATE.
Bronsted et a l . ( l ) studied the per-
chloric acid catalyzed hydration of
ethylene oxide and found this re-
action to be first order with respect
to ethylene oxide and the rate t o
be proportional to the hydrogen
ion concentration. I n the present
study such a simple dependence on
the hydrogen ion was not found,
and the catalytic reaction was
found to be second order with re-
spect to propylene oxide.
Lichtenstein and Twigg ( 6 )
studied the sodium hydroxide cata-
lyzed reaction and found t h a t the
reaction-velocity constant could be
expressed as follows: lc = a [OH-]
+ b [0H-l2. Smith, Wode, and
Widhe(8) caried out kinetic studies
on the liquid-phase hydration of
ethylene oxide with perchloric and
nitric acids as a catalyst. They
found that the rate of this reaction
was independent of the kind of
acid used. Percorini ( 7 ) , studying
the reaction of propylene oxide
with methanol to form l-methoxy-
F E E D R A l l O , ~ , l hHzO/lb.PrO 2-propanol by use of a sodium hy- '0
droxide catalyst, found t h e effect
FIG. 10. CONCENTRATION EFFECTFOR of the sodium hydroxide to be one
CATALYZEDREACTION. FIG. 12. EFFECTOF CATALYSTCON-
of direct proportionality. CENTRATION UPON REACTIONRATE.

The constants f o r t h e Arrhenius CONCLUSIONS

equation as obtained from Figure initial propylene oxide concentra-
9 and f o r the effect of t h e feed The kinetics of the catalyzed and tions have been expressed as a
ratio as taken from Figure 10 are uncatalyzed homogeneous liquid- function of the feed ratio of water
given in Table 4. phase hydration of propylene oxide to propylene oxide.
to form propylene glycol have been The reaction catalyzed with sul-
studied and the results correlated furic acid was found ;to be pseudo
DISCUSSION O F RESULTS in a form convenient for the design second order with respect to the
A comparison of the correlation of reactors. concentration of propylene oxide.
with the experimental data points The uncatalyzed reaction was The rate constant for this reaction
has been made in order to estimate found to be pseudo first order with has been expressed a s a function
the accuracy of the correlation, respect to the concentration of of the acid concentration, feed
which was used to calculate the propylene oxide. From the results ratio, and temperature.
experimental data points as shown of the uncatalyzed study, the re- In order to check the accuracy of
by dashed lines in Figures 5 and 8. action-rate constants for different the correlations obtained from this

Vol. 1, No. 1 A.1.Ch.E. Journal Page 123

 TABLE.3.-DISTRIBUTION O F PRODUCTS FOR RUNS WHERE of the reactor was contributed by
DIPROPYLENEGLYCOLWAS FORMED Stone and Webster Engineering Cor-
poration. Much of the credit for the
Feed Propylene oxide conversion, laboratory work of operating the re-
Catalyst ratio VRIF, fraction actor and analyzing the samples must
conc. Ri, CU. ft.1 to to be shared with Joseph Christy, Wil-
Reaction H2S0,, Ib. H 2 0 (sec.) glycol diglycol Total liam Behrmann, John - Gerety, and
George Daniels.
Run temp., “F. wt. % lb. Pro‘ (lb. mole) Xl x2 XT
72 125 0.10 5.0 108.5 0.633 0.012 0.645 NOTATION
70 125 0.10 5.0 216.9 0.829 0.007 0.836
71 125 0.10 5.0 309.9 0.887 0.017 0.904 A = proportionality factor defined
by t h e Arrhenius equation, Ib.
78 0.10 5.0 108.5 0.326 0.007 0.333 mole/ (cu.ft.) (see.)
76 100
loo 0.10 5.0 216.9 0.512 0.007 0.519
77 100 0.10 5.0 309.9 0.625 0.009 0.634 A’ = initial concentration of propy’
lene oxide, mole fraction
79 125 0.05 5.0 108.5 0.497 0.008 0.505 a = constant for a given tempera-
80 125 0.05 5.0 216.9 0.634 0.009 0.643 t u r e and catalyst concentra-
81 125 0.05 5.0 309.9 0.722 0.010 0.732
tion, units of k
82 0.10 2.5 102.5 0.228 0.004 0.232 b = constant for reaction, value
83 100
loo 0.10 2.5 205.0 0.321 0.006 0.327 depending only on whether a
84 100 0.10 2.5 356.5 0.451 0.007 0.458
reaction is catalyzed, unitless
E = molal energy of activation,
study, they were used to compute tion was found t o be 9.6. The cata- B.t.u./lb. mole
t h e conversion of propylene oxide. lytic data show a standard devia- F = feed rate, lb. mole/sec.
These calculated conversions were tion of 0.026 and a n average per- k , = specific reaction-rate constant
compared with the experimental centage deviation of 3.8. f o r first-order reaction, lb.
conversions, and standard devia- mole/ (cu.ft.) (see.)
tions and percentage errors were ACKNOWLEDGMENT k2 = specific reaction-rate constant
computed. The standard deviation Financial aid supplied by the Uni-
f o r second-order reaction, lb.
for the uncatalyzed runs was 0.024 versity of Kansas Research Project mole/ (cu.ft.) (see.)
and the average percentage devia- 84 made this work possible. The cost N , = propylene oxide concentra-
tion, mole fraction
R = gas-law constant, B.t.u./ (lb.
TABLE4.-cONSTANTS CHARACTERISTIC O F CATALYZED REACTION mole) (OR.)
Rj = initial feed ratio of water to
Arrhenius constants propylene oxide, lb./lb.
-E T = absolute temperature, OR.
k, = Ae-- V , = volume of reactor, cu.ft.
RT X = fractional conversion of pro-
Feed Energy of pylene oxide, lb. mole con-
ratio, Catalyst activation vertedilb. mole fed
Ri, concentration, A, E,
Pro = propylene oxide
lb. HzO HzS04, Ib. mole B. t. u.
LITERATURE CITED
lb. P r o wt. % (CU. ft.) (sec.,’ lb. mole
1. Bronsted, J. N., M. Kilpatrick,
10.0 0.025 1.09 xi013 36.9 X lo3 and M. Kilpatrick, J. Am. Chem.
10.0 0.050 2.42 xi013 36.9 x 103
10.0 0.075 4.04 xi013 36.9 x lo3 Soc., 51, 428 (1940).
10.0 0.100 6.82 X l O l j 36.9 x.-103
- -.- , . 2. Carbide and Carbon Company,
10.0 18.2- Xi013 36.9 x 103 Inc., “Ethers and Oxide” (1949).
10.0 50.5 xi013 36.9 x 103 3. Curme, G. O., and F. Johnston,
5.0 0.318 X IOl3 36.9 x 103 ‘‘Glycols,” Reinhold Publishing
5.0 0.909X1013 36.9X103 Corp., New York (1952).
5.0 2.10 xi013 36.9 x 103 4. Davis, P. C., C. E. Von Waaden,
2.5 0.100 0.555x 10’3 36.9 x 18 and F. Kurata, Chem. Eng. Progr.
Sumposium Series No. 4, 48, 98
Concentration function (i952).
5. Houcen. 0. A.. and K. M. Watson.
“Chgmical Process Principles,’;
k2= a RP vol. 3, John Wiley and Sons, Inc.,
New York (1949).
Catalyst 6. Lichtenstein, H. J., and G . H.
concentration, a, Twigg, Trans. Faraday SOC.,44,
Reaction HX4, lb. mole b, 905 (1948).
temperature. “F. wt. % (cu. ft.) (sec.) unitless 7. Pecorini, H. A., Ph.D. Thesis,
Univ. Michigan (1954).
1CO 0.10 0.00428 1.8
8. Smith, L., G. Wode, and T. Widhe,
100 0.05 0.00155 1.8
100 0.025 0.000682 1.8 2. Physik. Chem., 130, 154 (1927).
9. Wickert, J. N., W. S. Tamplin,
125 0.10 0.0174 1.8 and R. L. Shank, Chem. Eng.
125 0.050 0.00634 1.8 Progr. Symposium S e r i e s No. 2,
125 0.025 0.00285 1.8 48, 92 (1952).
150 0.10 0.0650 1.8 10. Wurtz, A., Ann. ehim. et phys.,
~-
150 0.050 0.0230 1.8 55, ( 3 ) , 438 (1859).
150 0.025 0.0106 1.8 (Presentcd a t A.I.Ch.E. N e w York meetiiig)

Page 124 A.1.Ch.E. Journal March, 1955