Professional Documents
Culture Documents
Vinyl Acetate
Vinyl Acetate
1. Introduction
During the last 25 years, increasing attention has been given to the vapour-phase addition of
acetylene and acetic acid to give vinyl acetate.1-4 Among the various studies reported there is wide
disagreement in the reported data and the kinetics of the reaction are still not understood cIearly.5-25
Furukawa12 carried out kinetic measurements at 200C in a flow system on a zinc acetate-activated
carbon catalyst. The order of the reaction with respect to acetylene was 1.6 and 0.6 with respect to
acetic acid. From an investigation of the same reaction on a similar catalyst Janada and Vanko17
obtained a second-order equation. Also, from a similar study at reduced pressure it has been
suggested14.15that the rate proceeds according to the equation.
r= kpC2H2PCH3COOH
(1 + KPCH3COOH)2
Vasil'eva et al.25 have performed a detailed study of this reaction. The rate of reaction was
reported to be first order with respect to acetylene and independent of the partial pressure of acetic
acid or vinyl acetate above 180C. By analogy with the hydrochlorination of acetylene26 a rate
expression was presented. It has been well established that neither zinc acetate nor activated carbon
show any catalytic activity by themselves, even at relatively high temperatures (180-218C). 22.27
This distinguishes the reaction of acetylene and acetic acid very clearly from the hydrochlorination
" of acetylene, which is catalysed by one of the components.
The chemisorption of acetylene has been assumed to be the rate-determining step in the kinetics
of the acetylene-acetic acid reaction.9.13.19,21.22In addition to the heterogeneous mechanisms, a
homogeneous route involving free radicals is also believed to be involved.1
From the literature review it is evident that no systematic quantitative analysis of the experimental
data has been reported. Also, the values of the rate coefficients and the adsorption coefficients are
unknown. Accuracy of the experimental data is no guarantee of a successful kinetic analysis when
the experimental programme suffers from a lack of insight into the effect of different variables.
Recently, Froment28 has discussed in detail the estimation of parameters and has emphasised the
necessity of the statistical testing of the results.
a To whom correspondence should be addressed.
0375-9210/78/0800-0547 $02.00 ;) 1978 Society of Chemical Industry
547
.,
548 H. B. Goyal et af.
In this kinetic investigation, increased emphasis has been placed on model discrimination and
parameter estimation. Both tasks are based upon experimental data; therefore, reliable data were
obtained by studying this reaction using a differential reactor. The reaction was studied at atmo-
spheric pressure in the temperature range 150-200C, where the homogeneous reaction was negligible.
The purpose of this study was to obtain some understanding of the catalytic activity Zn(OAch-C
system and to examine the possibilities of interpreting the mechanism. The experimental results are
analysed on the basis of Langmuir-Hinshelwood kinetics with interpretation of statistical data to
show the real significance of mechanism determination with precise experimental data. In addition,
physicochemical characterisation was desired in order to amplify literature results for the Zn(OAch-
C system since no extensive determinations have been published on the physical properties and
characteristics of these catalysts.
the two peaks which appeared were those of C2H20 and H2O. Zn+ intensity could not be 0bserved even
up to 1000oe. This study clearly indicates that Zn(OAc)2-e catalysts decompose to ZnO(s), e2H20-
(g) and H20(g).
2.1.4. D.t. and t.g. analysis
A programmed heating at a rate of 15C min-I between 25 and 10000e was used. The d.t.a. thermo-
grams revealed three regions of thermal activity: (a) a weak endotherm peaking near 1000e very
similar in appearance to the 1000e endotherm observed for carbon alone, (b) a strong and broad
exotherm 400 and 800C, peaking near 508C, and (c) a medium and broad endotherm (depending
upon the composition of the catalyst) between 220 and 360C, peaking near 280C. The thermo-
gram for pure zinc acetate exhibited a sharp endotherm of medium strength between 220 and 340C,
peaking at 280 and 320C. The peaks at 280 and 310C are due to the decomposition of zinc acetate.
The results of the catalysts also agree with this; however, the intensity of the endotherms depends
upon the composition of the catalyst. The peak in the 500C region represents burning of the carbon.
T.g. analysis was performed with programmed temperature increase at a rate of lOoe min-I. The
weight loss obtained in the 200-320oe region was mainly due to the decomposition to zinc oxide.
The weight loss above 400C was mainly due to burning of carbon.
An analysis was also performed for pure zinc acetate. The thermogram was flat up to 180C.
Between 200 and 300C a sharp weight loss was observed. The observed weight loss of about 60 %
is in agreement with the calculated weight loss of 56 % (ZnO being the only solid product). The
results for the activated carbon showed no weight loss except above 400C.
The decomposition of zinc acetate has been studied by Edward and Hayward34 by t.g.a. who
reported the formation of ZnO at 350C as the sole product. The present results of d.t.a. and t.g.a.
which are in complete agreement by themselves confirm this temperature range of decomposition,
which was complete below 350C.
Table 1. Surface area, pore volume and average pore radius of various catalysts
3. Experimental considerations
3.1. Equipment
A tubular reactor of glass was used to obtain the data. The jacketed reactor was 1.25 cm in diameter
and 60 cm long and was provided with a flexible preheating zone of maximum length 16 cm. The
preheating zone contained stainless steel tubing packed with stainless steel foils to provide more heat
transfer area. The temperature in the reactor was maintained by condensing vapours of various
~
550 H. B. Goyal et al.
organic liquids. Temperatures at various points in the reactor, including the catalyst bed and the
heating liquid, were measured with the help of iron-constantan thermocouples. The thermocouples
were calculated against a 25 Q Leads and Northrup platinum resistance thermometer and a
Mueller bridge assembly. A schematic diagram of the experimental equipment is given in Figure 1.
Vent
6 8
3 3
4 4
10 Product
II 12
,.,
Figure 1. Schematic diagram of the experimental set-up: (I) Test reactor; (2) acetic acid vaporiser; (3) product
condensers; (4) soap bubble meters; (5) silica gel driers; (6) capillary flow meter; (7) manometer; (8) needle valve;
(9) acetylene purifier; (10) two-way stop cock; (II) acetylene gas cylinder; (12) nitrogen gas cylinder.
3.3. Procedure
Acetylenewas passed through a purifier and a series of driers. The purifier contained a mixture of
solutions of 30 %CuCb,30% FeCb and 35%HCl in a volumeratio of 3:7:1.The gasesfrom the
purifier were dried with indicating blue silica gel. The purity of the gas was then checked by gas
chromatography.
~
During the runs, different feed ratios (acetylene/acetic acid) were obtained by adjusting the acety-
lene flow rate in conjunction with the temperature of the vaporiser, while the reciprocal of the space
velocity was varied by changing the amount of catalyst and the total feed rate.
Preliminary runs with an empty reactor and a reactor filled with stainless steel foils indicated that
the homogeneous reaction and the catalytic action of the system were not important under the
experimental conditions.
The effect of various variables, acetylene to acetic acid in feed (R), reaction temperature (T), the
reciprocal of space velocity (W/F), on the conversion of acetic acid and the yield of vinyl acetate
(V A) were investigated. The conversion (X), the yield of VA (Y) and the rate of the reaction (r) are
defined as follows:
rr .
Kinetics of acetylene/acetic acid reaction 551
C
. 0/ - Moles of acetic acid reacted
onverslOn ( /0 - M 0 If'
) es 0 acetic aCI'd ~e d x 100
R ate .
0 f reactIOn, r-
- Moles of acetic acid reacted
(h) (g 0 f cata Iyst )
The % yield of vinyl acetate was maximum at a zinc acetate concentration of about 15%. This is in
agreement with Rostovskii and Arbuzova8 who have reported that the yield of vinyl acetate passed
through a maximum at a catalyst concentration of about 18 % of Zn(OAc)2. All the experiments
described in this paper were performed exclusively with the catalyst of 15 % Zn(OAch concentration.
The effects of diffusion were kept to a minimum by passing the gas at high velocity through the
catalyst. The fair constancy of conversion obtained by changing the feed ratio while keeping (WI F)
constant suggested that the diffusion of the gases'was not controlling the rate.
The role of internal diffusion was checked by varying the size of the catalyst particles while main-
taining the catalyst weight, (R), and (WI F) constant. The internal diffusional resistance in the
catalyst particles was practically negligible as observed by.insignificant change in the reaction rate
on varying the catalyst size between 2.8 to 1.4 mm.
In order to study the effect of external diffusion it was necessary to increase the bed height while
keeping (R) and (WIF) constant. The rate was unaffected by doubling the bed height. Thus, the
external diffusion in the catalyst bed was negligible.
The effect of the reciprocal of space velocity (WI F) on the conversion of acetic acid was examined.
The reaction conditions were as follows: reaction temperature, 180C; (R)=3.0. The results are
presented in Figure 2. It is seen that the % conversion increased rapidly with the (WI F) ratios
studied.
7,0
6.0
5.0
"
s
'"r> 4.0
c
0
U
:t 3.0
'"
2.0
0 70
Figure 2. Effect of reciprocal space velocity on % conversion of acetic acid at 180C and (N) = 3.0.
{\\'
552 H. B. Goyal et al.
A study of the reaction rate of acetic acid vs the feed ratio should give more precise information
on the role of feed ratio in the conversion of vinyl acetate. Prior to each data point, the activity of
the catalyst was established by running a standard and comparing it with the initial standard. There
was essentially no deactivation of the catalyst within 12 h. X-ray diffraction patterns showed that
the used catalyst retained its activity.
Among the by-products, acetone, acetaldehyde and crotonaldehyde were detectable by gas
chromatographic analysis, but the values of thermal conductivity were too small for accurate
measurements.
A minimum of four runs were performed at each datum point. The standard deviation of the rate
was found to be approximately 1.5 %. Data were collected between 155 and 200C. The average
reactor temperature was controlled to within :t O.5C of the desired value. Conversions were usually
kept below 8 % as based on the acetic acid present in the feed.
The measured reaction rate data obtained are summarised in Table 2. The treatment of the experi-
mental data is somewhat complicated by the number of independent variables, that is, the three
partial pressures and the temperature. The technique of non-linear estimation was applied to various
mathematical rate expressions as a method of correlating the experimental data. Based on Langmuir-
Hinshelwood kinetics, these models were examined with respect to experimental data, using the
approach suggested by Hougen and Watson.36 In this method, various mechanisms which might
control the reaction are postulated, based on a single site and two types of active sites.
It is assumed that the surface cDnsists of two types of active sites. On these active sites acetylene
and acetic acid are adsorbed selectively.
The heterogeneous catalytic reactions traditionally described as (a) diffusion (external and internal)
of reactants and products, (b) adsorption and desorption of reactants and products (physical
adsorption and weak chemisorption) and (c) surface reaction should be modified to take place as
follows: (i) electron transfer from or to the'reactants, (ii) homogeneous reaction (formation of
activated complex and rearrangement of the ionised species on the surface) and (iii) electron transfer
from or to the products.
This treatment is similar to those used by Mann and co-workers, 37,38for the oxidation of olefins
over bismuth molybdate where the surface was considered to consist of two types of active sites on
which olefins and oxygen are selectively adsorbed.
combinations of the various adsorption terms were examined in order to allow for various negligible
adsorption terms. Rate expressions were eliminated from further consideration when any of the
converged adsorption constants became negative. The converged values with errors for the constants
of the models are presented in Table 4.
Table 5 gives the rate models with all the positive constants which had to be retained statistically
after the regressions. Although model 2 had all the positive constants at all temperatures, the like-
f!.
lihood ratio, LR, was exceedingly high. The approximate 95 % confidence intervals for the various
parameters in this model were so large as to be meaningless.
The data were correlated most satisfactorily by model 12. The effect of temperature on the rate
expression was introduced by substituting the Arrhenius temperature dependency relation for each
of the constants.
16.0
In KA= 11.78- RT
13.3
InKB=8.71- RT
'"
428 (-0.1I8:t 12.4) x 10-2 (-0.635:t 63 .6) x 101 (-0.001:t5.04)x 104 54 0.25 x 10-7
438 ( - 0 . 825 :t 8 . 3) x 10- 2 (0.0:t0.519) x J01 (0.0511 :t25) x 102 26 0.15 x 10-7
453 (-0.407:t 115) x 10-2 (-0.082:t22.8)x 102 (-0.001 :t4.983) x 104 61 0.25 x 10-6
465 (-0. 754:t 191) x 10-2 (-0.0907:t21.3)x 102 (-0.01:t 15) x 103 52.6 0.33x 10-6
472 (33:t 29.8) x 10-4 (0.0197 :to.0958) x 101 (-0.031O:t0.388)x 102 1.5 0.27 x 10-9
Model no. 2
Temp. (K) K1 KR
428 (0.102:t 12.8) x 10-1 (0.1I2:t 3 .28) X 105 15.4 0.37 x 10-8
438 (2.45:t 748) x 10-1 (0.01:t 3 .03) X 107 13.8 0 . 66 x 10-8
453 (0.00648:t 13.4) X 102 (0.01 :t20. 7) x 107 31..6 0.79 x 10-7
465 (8.8:t23.3) x 10-3 (0.0462:t0.157) X 104 12.9 0.26 x 10-7
472 (2.9:t0.16) x 10-3 (0.0230:t 0.0464) X 102 1.2 0 . 34 X 10-9
I'
Model no. 3
Temp.
(K) K1 KA KR
428 (9. 8 + 16. I) X 10-5 (-0.151 :to.31) x 101 (-0.1:t 10.9) x 102 16.9 0.26x 10-8
438 (0.159 :to.245) x 10-3 (-0.148 :to. 359) x J01 ( - 0.1 :t 6.76) x 102 14.9 0.53 x 10-8
453 (2.9 :t 4 . 13) X 10-4 (-0.13 :to.232) x J01 ( - 0 . I :t 3 .080) x 102 9.4 0.69 X 10-8
465 (0.0232 :t247) x 10-2 (-0.1:t2.83) X 102 (0.21O:t8.81)x 103 31.6 0.63 x 10-6
472 (26:t 6.79) X 10-5 (-0.32:t0.258) x J01 (0.0462:t 0.695) X 102 6.4 0.56xlO-8
Model no. 4
Temp.
(K) K1 KR
Model no. 8
Temp.
(K) K1 KA KB KR
428 (9.9:t 781) X 10-2 (8.2:t430) X 102- (13.1 :t449) x 10-2 (0.773 :t4.268) x 101 0.28 0.68 x 10-12
438 (l2.5:t 173) X 10-2 (9.2:t964) X 10-2 (20.7:t 107) X 10-2 (5.631 :t 503) x J01 1.26 0.52 x 10-10
456 (l4.8:t161)x 10-2 (6.92:t1l)xI0-1 (67:t 695) x 10-2 (l0.75:t 7 .121) X 101 2.34 0.47xlO-9
465 (l3.7:t 247) X 10-2 (2.6:t156)x 10-1 (9.3:t135)x 10-2 (-0.918:t10.14)x J01 0.43 0.23 x 10-10
472 (l7:t0.0)x 10-2 (0.83:t0.0)x 10-1 (6.9:t0.0)x 10-2 (-1.000:t0.0)x 101 64.5 0.69xlO-6
36
f,~
556 H. B. Goyal et af.
Table 4. (continued)
Model no. 10
Temp.
(K) K, KB KR
428 (0.163 :!:26.2) x 10-1 (0.83:!: 305) x 10-1 (-63.2:!:IOI.I)x 10-1 15.9 0.13 x 10-8
438 (1.04:!:0.00) X 10-1 (0.22:!:0.00) X 10-1 (=- 13 .O:!: 0.00) X 10-1 13.8 0.13 x 10-8
453 0.044:!:22 0.0144:!:22 -9:!:41 7.4 O. 63 X 10-"
465 0.206:!:i 0.470:!:i 33.0:!:i 30.6 0.17 x 10-8
472 0.051 :!:472 0.0044:!:i 991:!:i 5.4 O. 50 x 10-8
428 (9.2:!: 394) x 10-2 (6.4:!:323) x 10-2 (16.4:!:82.2) x 10-2 (1.30:!:541) x 101 0.26 0.68xlO-12
f'
438 (11.7:!: 419) X 10-2 (9.3:!:413)x 10-2 (16.7:!:875) x 10-2 (6.57:!:48.7)x 101 1.47 0.76 X 10-10
435 (8.6:!: 323) X 10-2 (270.8:!: 206) X 10-2 (34:!: 1368) x 10-2 (35.4:!: 212) X 10' 2.39 O. 50 X 10-"
465 0.143:!:i 0.216:!:i O.III:!:i -0.381:!:i 1.3 0.33 X 10-"
472 (6.2:!:0.0)x 10-2 (11.3:!:0.0)x 10-2 (6.2:!:0.0)x 10-2 (-0.395:!:0.0)x 101 8.4 0.12 X 10-7
Model no. 12
Temp.
(K) KA KB
Model
no. Rate controlling step Rate equation
12
(1 + KRpR) ~
Irreversible ionised adsorption of acetylene and acetic acid KAPA
1+ (KA!KB)(PA!pB)
The plots of the experimental rate vs that predicted by the above expressions are shown in Figures
3 and 4 for the temperatures of 155 and 192C. The maximum deviations between the predicted and
experiment values were less than :t 4 %. The good agreement between calculated and experimental
rates indicates that the assumptions about the reaction mechanism are correct. However, the
assumptions which are made in the development of the particular rate expression must be verified
experimentally.
~
Kinetics of acetylene/acetic acid reaction 557
3.0
..
Q
" 2.0
1:
T
0U
'"I
~
,g
u
'"
U
U
'"
a.
'" 1.0
a
a::
7.5
6.5
..
Q
"
,..s;;;
~u 5.5
I
'"
c;
EI
!!!
~ 4.5
u
u
~
a.
'"
0
a::
3.5
3.5 4.5
Role observed (g-mol g-cot-I h-I x 104)
Nomenclature
C Carbon.
F Total flow rate, g mol h-l.
K Equilibrium constant, adsorption equilibrium constant.
Kl Rate constant.
' RSS of the model in consideratiOn n/2
1 00 d ratIO. --
L 1k e rh
LR
( RSS 0 f ot h er mo d eIs )
p Total pressure.
R Reactant mole ratio, acetylene to acetic acid.
T Temperature.
V Volume.
W Weight of the catalyst.
X Conversion of acetic acid.
Y Yield.
d Distance between each set of atomic planes of the crystal lattice.
g Gas.
Imaginary.
k Rate constant.
n Numbers of observations.
p
r
Partial pressure.
Experimental rate of reaction, pore radius.
~
f Calculated reaction rate.
i' Mean pore radius.
s Solid.
Subscript
A Acetic acid.
B Acetylene.
R Vinyl acetate.
References
1. Europ. Chem. News 1965, no. 27, 21.
2. Shetty, R. S.; Chandalia, S. B. Chem. Process Engng 1969, September, 17.
3. Stanford Research Institute, Process Economics Programme on Vinyl Acetate (private communication).
4. Stobaugh, R. B.; Allen, W. C, Jr.; Sternbergh, R. H. Hydrocarb. Process. 1972,51,153.
5. Furukawa, J. Chemy High Polym. 1952,9,240.
6. Takigawa, B. Kogyo Kagaku Zasshi 1952, 55, 254.
7. Leonard, E. C. Vinyl and Diene Monomers Part 1, Wiley Interscience, New York, 1970.
8. RostovskiI, E. N.; Arbuzova, 1. A. Zh. Prikl. Khim. 1959, 32, 2258.
9. Mitsutani, A.; Kominami, T. Nippon Kagaku Zasshi 1959, 80, 886.
10. Cristea, 1.; Mihai, F.; Pop, Gr. StudU Cere. Baya StUnte, Timisoara 1963, 10(1), 101. 18
II. Liverovskaya, N. V.; Terushkin, V. R.; Satarova, R. A.; Glushankova, Z. 1.; Golubev, V. N.; Kravchenko, Q
B. A.; Gel'bshtein, A. 1. Khim. Prom. 1972,48(2), 100.
12. Furukawa, J. Chemy High Polym. 1949,6, 116.
13. Furukawa, J.; Ozawa, H. Bull. Inst. Chem. Res., Kyoto Univ. 1951,26,93.
14. Yamada, N.; Kobayashi, Y. J. Chem. Soc. Japan, Pure Chem. Sect 1955, 76, 722.
15. Yamada, N. Nippon Kagaku Zasshi 1957, 78, 252.
16. Janda, J. Chemicke Zvesti 1957, 11,478.
17. Janda, J.; Yanko, A. Chemicke Zvesti 1958, 12, 657.
18. Arbuzova, A.; Ushakov, S. N.; RostovskiI, E. N. Zh. Prikl. Khim. 1958,31, 1704.
19. Mitsutani, A.; Matsumoto, M. Nippon Kagaku Zasshi 1958, 79, 948.
20. Flid, R. M.; Chirikova, A. V. Zh. Prikl. Khim. 1959,32,660.
21. RostovskiI, E. N.; Ushakov, S. N. Zh. Prikl. Khim. 1959, 32, 2060.
22. Mitsutani, A. Nippon Kagaku Zasshi 1960, 81, 298, 508.
23. Nakamura, A.; Kubota, S.; Kobayashi, J.; Higuchi, 1. Nippon Kagaku Zasshi 1966, 87, 1042.
8
Kinetics of acetylene/acetic acid reaction 559
24. Ho Sui Dao; Trofimova, I. V.; Flid, R. M. Zh. Prikl. Khim. 1972,45,700.
25. Vasil'eva, I. B.; Gel'bshtein, A. I.; Tolstikova, I. N.; Wangtiong, T. Kinet. KataI1964, 5, 144.
26. Gel'bshtein, A. I.; Scheglova, G. G.; Khomenko, A. A. Kinet. Katal. 1963, 4, 625.
27. Miller, S. A. Acetylene, Its Properties, Manufacture and Uses Vol. 2, Academic Press, New York, 1966.
28. Froment, G. F. A.I.Ch.E.J11975, 21, 1041.
29. Bristol, J. E.; Call, R. N. U.S. Patent 2 988 517, 1961
30. Agarwal, A. K.; Srivastava, R. D. J. Catalysis 1976,45, 1.
31. Farber, M.; Srivastava, R. D.; Uy, O. M. J. Chem. Soc. Faraday I 1972, 68, 249.
32. Farber, M.; Srivastava, R. D. J. Chem. Soc. Faraday I 1973, 69, 390.
33. Grigorev, A. I. Russ. J. Inorg. Chem. 1963, 8, 409.
34. Edward, D. A.; Hayward, R. N. Can; J. Chem. 1968,46,3443.
35. Ross, S. D. Inorganic Infrared and Raman Spectra McGraw Hill, London, 1972.
36. Hougen, O. A.; Watson, K. M. Chemical Principles Part III, John Wiley Inc., New York, 1947.
37. Mann, R. S.; Ko, D. W. J. Catalysis 1973, 30, 276.
38. Mann, R. S.; Yao, K. C. J. appl. Chem. Biotechnol. 1972,22,1915.
39. Marquardt, D. W. J. Soc. Ind. Appl. Math. 1963,11,431.
40. Draper, N. R.; Smith, H. Applied Regression Analysis John Wiley Inc., New York, 1966.
41. Yang, K. H.; Hougen, O. A. Chem. Engng Prog. 1950,46,140.
42. Kudchadker, S. A.; Goyal, H. B.; Kudchadker, A. P. Thermochimica Acta 1975, 11, 361.