m e Catalytic Decomposition of Isopropyl Alcohol Vapor on Manganese (11) Oxide

D. G. K ~ r s s m s e r ,E.
~ F. MCCAFFREY,
AND R. A. ROSS
Department of Chemistry, Lakehead University, Thunder Bay, Ontario
Received June 17, 1971

The vapor phase decomposition of isopropyl alcohol has been studied from 210 to 365 "C on man-
ganese(I1) oxide. Rates of conversion to acetone and propylene were measured with partial pressures of
isopropyl alcohol from 0.3 to 22 mm. The catalyst selectivity towards dehydrogenation was around 0.80
Can. J. Chem. Downloaded from www.nrcresearchpress.com by 190.67.178.174 on 09/15/19

between 320 and 365 "C while at 306 "C it was somewhat less and depended on the alcohol partial
pressure. Propylene and hydrogen additions had no effect on the reaction rate while the reaction order
with respect to isopropyl alcohol partial pressure was 0.4 to 0.6 except at 306 "C when it decreased with
increase in the alcohol partial pressure. The effects of acetone and water additions were shown respec-
tively to be exerted mainly on the dehydrogenation and dehydration reactions while at acetone pressures
above 4 mm, considerable amounts of by-products including mesityl oxide and heavier ketones were
clearly detected. The apparent activation energy of the overall decomposition as calculated from the
+
experimental rates was 26 0.5 kcal mol-'. The results have been interpreted to be in general agreement
with the concept that the rate-limiting step is linked to the direct interactions on the catalyst surface and
it has been established that the reaction is significantly more complex than previously recognized. The
physical properties of the manganese(I1) oxide were determined by X-ray and nitrogen adsorption
techniques while infrared, n.m.r., and gas chromatographic methods were used to analyze the products
of catalysis.
On a etudie la dCcomposition de l'alcool isopropylique en phase vapeur, entre 210 et 365 "C, sur
I'oxyde de mangankse(I1). Les taux de conversion en acetone et propylene furent mesures au moyen des
For personal use only.

pressions partielles d'alcool isopropylique entre 0.3 et 22 mm. La stlectivite du catalyseur pour la deshy-
drogenation se situait autour de 0.80 entre 320 et 365 "C, alors qu'a 306 "C elle etait 16gerement infkrieure
et dtpendait de la pression partielle de l'alcool. Des additions de propylene et d'hydrogkne n'ont eu
aucun effet sur la vitesse de la reaction, alors que I'ordre de la reaction par rapport a la pression partielle
d'alcool isopropylique etait de 0.4 a 0.6, excepte a 306 "C ou il diminuait avec l'augmentation de la
pression partielle d'alcool. On a constate que les effets de I'addition d'acetone et d'eau s'exer~aient
principalement sur les reactions de deshydrogenation et de deshydratation, alors qu'a des pressions
d'acetone de plus de 4 mm, des quantitks considerables de produits secondaires incluant l'oxyde de
mtsityle et des cetones plus lourdes etaient clairement detecties. L'energie d'activation apparente de la
+
dkcomposition totale, telle que calculCe a partir des vitesses experimentales etait de 26 0.5 kcal mol-'.
Les resultats ont ete interpretes comme etant en accord general avec le concept que 1'Ctape limitante
est reliee aux interactions directes a la surface du catalyseur et il a ete etabli que la reaction est singuliere-
ment plus complexe qu'on I'avait prevu. Les proprietks physiques de I'oxyde de manganese(I1) ont 6tB dB-
terminees par les techniques de rayons-X et d'adsorption d'azote, alors qu'on a utilise l'infrarouge, la
r.m.n. et la chromatographie en phase gazeuse pour analyser les produits de catalyse.
Canadian Journal of Chemistry, 49, 3778 (1971)

htroduetion in two directions: (a) dehydrogenation and (b)

Few comprehensive kinetic studies of the dehydration.
heterogeneous catalytic properties of manga- According to current theories, catalyst activity
nese(I1) oxide or its mixed oxides have been and selectivity in this reaction may be related to
reported and as a first step in an experimental lattice parameters, type of conductivity, and
assessment of these properties an examination of width of forbidden zone as well as to the acidic
the catalytic decomposition of isopropyl alcohol properties of the surface (2-5). For manganese(I1)
vapor on manganese(I1) oxide has been carried oxide in particular, a relatively high activity in the
out. This reaction is frequently used as a model dehydrogenation of isopropyl alcohol has been
in studies of variations in catalytic activity and reported (6) but unfortunately no data on the
selectivity involving nonmetallic catalysts (1). surface area were given and hence the specific
At temperatures below 500 "C it proceeds mainly activity cannot be estimated. At the same time,
the reaction order with respect to isopropyl
'On leave from Academy of Sciences of Bulgaria, alcohol or any of the reaction products has not
Sofia 13. been studied and yet it is generally accepted that
KLISSURSKI ET AL.: DECOMPOSITI(3N OF ISOPROPYL ALCOHOL VAPOR 3779

only dehydrogenation proceeds on the surface of SP800A, were the same as those given on their accom-
this catalyst (6-8). panying certificates. All gaseous reagents and the nitrogen
carrier (Canadian Liquid Air Ltd.) were 99.99% pure.
In the present work the activity and selectivity
of manganese(1I) oxide in the decomposition of
isopropyl alcohol vapor have been examined Results
from 300 to 365 "C.Reaction orders and activa- The Eflect of Temperature on the Reaction Rate
tion energy have been measured and attention and Products
given to the more complex character of the reac- Experiments were carried out with 9.72 mm
tion. partial pressure isopropyl alcohol at various
Can. J. Chem. Downloaded from www.nrcresearchpress.com by 190.67.178.174 on 09/15/19

Experimental temperatures from 325 to 363 "C in which the

Apparatus rates of conversion of the alcohol to acetone (r,)
The all-glass flow system has been described previously and propylene (r,) were measured along with the
(9, 10). A quartz integral reactor was used and tests from
210 to 363 "C showed that isopropyl alcohol did not de- overall rate of conversion (r,) and the selectivity
compose to any detectable extent in the absence of the ratio r,/r,. Within this temperature range, several
catalyst. High-purity nitrogen at atmospheric pressure other products were formed which accounted for
was used as a carrier gas and the catalyst bed temperature approximately 0.1% of the total conversion.
was controlled to + 0.1" from 300 to 400 "C by a platinum
resistance thermometer connected to an expanded-scale Isopropyl ether and mesityl oxide were positively
"Sigma" ~rovortionalcontroller. identified while i.r. and n.m.r. analyses both
&action products were analyzed at 80 "C by a Beck- indicated the presence of an acetylenic compound.
man GC5 gas chromatograph using stainless steel Although the dehydrogenation reaction was
columns, 72 in. x 0.125 in. o.d., packed with "Carbowax predominant, dehydration also took place simul-
1540" on "Teflon 6".
To establish the experimental conditions for which the taneously to a significant extent. The average
For personal use only.

reaction rate was not influenced by external or internal selectivity for dehydrogenation was close to 0.80
diffusion, the influence of space velocity, catalyst mass, at all temperatures with no obvious trend. Earlier
and particle size was studied in the usual manner (9). As a it had been accepted (6) that manganese(I1) oxide
result of these preliminary studies the kinetic measure-
ments were performed at flow rates from 220 to 360 ml catalyzes the dehydrogenation reaction with a
min-' (N.T.P.) (space velocities at 342.5 "C, 300 to 500 cc selectivity of 99-loo%, a conclusion which is
gas (cc catalyst)-I m i x i ) . evidently not in agreement with the present
All reaction rate measurements were made after the results. However, the disagreement may be ex-
establishment of the steady catalyst activity level. plained simply by the fact that the reaction rate
Materials previously had been measured only from the
( a ) Manganese(1I) Oxide pressure changes of the gaseous products (7) or by
Electrolytic grade metallic manganese, 99.999% pure, the quantity of hydrogen evolved (6).
was dissolved in nitric acid to give a 4 Msolution at p H 6. In Fig. 1, log,, r, is plotted against the
The solution was kept at 60 "C and 4 M ammonium
oxalate solution added with vigorous stirring. The pre- reciprocal catalyst temperature in the Arrhenius
cipitate, MnCZO4.2H2O, was washed with warm de- fashion. From the slope of this line the apparent
ionized-water until free of nitrate ions and then dried at activation energy, E', was calculated at 26 +
0.5
80 to 100 "C for 6 h. Calcination to manganese(II1) oxide kcal mol-l. Additional measurements in the
was carried out at 300 "C for 5 h followed by 16 h each interval 210 to 300 "Cconfirmed this value.
a t 400 and 600 "C (11-13). Subsequently, analysis by
X-ray powder photography confirmed the cr-Mn203 For a catalytk reaction inhibited by the
structure (14). Chemical analysis (15) showed 69.98% wlw products it has been shown (16) that the relation-
manganese (theory 69.59%) while the specific surface area ship between E ' and the apparent activation
was 26 _+ 2 m2 g-I (B.E.T.; N 2 ; - 196 "C). Finally the energy E, calculated from rate constants is given
a-Mn203 was heated for 12 h at 600 "C in a flow of
hydrogen (60 ml min-') and X-ray powder photography
by: E ' = E/(r + I), where r is the overall
of the product confirmed the MnO structure (14). reaction order with respect to the inhibiting
Chemical analysis gave 76.90% wlw manganese (theory products. Because of the complexity of rate con-
77.44%) and the surface area was 23 + 2 mZg-I. stant calculations for some reactions, it was
It was established that the specific surface area of the further shown (16) that E could be obtained by
manganese(I1) oxide samples did not change after use in
the catalytic measurements. plotting the temperature dependence of the flow
( 6 ) Reagents rate at which a given degree of conversion was
Isopropyl alcohol and acetone were "Spectrosol" achieved, since then: flow rate, V c c k, rate
quality and their U.V. absorption characteristics, Unicam constant.
3780 CANADIAN JOURNAL OF CHEMISTRY. VOL. 49, 1971
Can. J. Chem. Downloaded from www.nrcresearchpress.com by 190.67.178.174 on 09/15/19

FIG. 1. Arrhenius plot of overall experimental

reaction rate (r,). Partial pressure isopropyl alcohol,
9.72 rnm; flow rate, 300 ml min- ' (N.T.P.).
For personal use only.

Figure 2 shows the mole fraction of isopropyl

alcohol converted us. the reciprocal of the flow
l ,
rate at the temperatures indicated. From these 0.2 0.3 0.4 05 06 07
data four separate Arrhenius plots of log,, V lo2/v
against I / T ° K were drawn for four different FIG. 2. Changes in mole fraction of isopropyl alcohol
degrees of conversion between 15 and 30%. The converted with variation in flow rate, V, at temperatures
plots gave an apparent activation energy of shown.
30 5 2 kcal mol-I. The slightly higher value
found for E and E' indicates the level of the though no data were quoted. At 306 "C, the
product inhibition effects on the overall reaction dependence of the reaction order on isopropyl
rate. alcohol partial pressure was more complex.
There was a distinct tendency for the order to fall
The Effect of Isopropyl Alcohol Partial Pressure with increase in the alcohol partial pressure, e.g.
on the Reaction Rate from 0.3 to 10.0 mm the order decreased from
The partial pressure of isopropyl alcohol was 0.6 to about 0.2, Fig. 3. Similar effects have been
varied between 0.3 and 21 mm at 306, 323, and observed also with ZnO.Cr,O, catalysts (18).
356 "C at 300 ml min-I (N.T.P.). The results are
shown in Table 1. There is a slight trend for The Effect of Hydrogen
dehydration to become more favored with de- In the temperature range studied with hydrogen
crease in temperature, particularly so at the high- partial pressures from 1.5 to 40 mm and an
est isopropyl alcohol partial pressures at 306 "C. isopropyl alcohol partial pressure of 7.78 mm
Logarithmic plots of these data indicate a reaction the reaction rate was not influenced by hydrogen
order for the overall decomposition from 0.4 to partial pressure.
0.6 with respect to isopropyl alcohol partial
pressure at 323 and 356 "C. The order lay at the The Effect of Propylene
higher end of the range at 356 "C and at the lower At 323 "6with 7.78 mm partial pressure iso-
end at 323 "6. Previously it has been stated (17) propyl alcohol and propylene partial pressures
that the rate of reaction of isopropyl alcohol on from 0.3 to 33 mm no influence of the gas was
manganese(I1) oxide at "low" temperatures de- detected on the rates of either the dehydration or
pends little on the alcohol partial pressure al- dehydrogenation reactions.
KLISSURSKI ET AL.: DECOMPOSITION OF ISOPROPYL ALCOHOL VAPOR 3781

TABLE1. The effect of isopropyl alcohol partial pressure on the experimental reaction rates at flow rate 300 ml rnin-'
(N.T.P.)

Partial pressure Rate of conversion to Rate of conversion to Overall rate,

Temperature of isopropyl acetone, rl propylene, rz r3
("c) alcohol (rnm) (mol m-2 s-') x (mol m-' s-I) x lo-' (mol m-' S-') X lo-' rllra
Can. J. Chem. Downloaded from www.nrcresearchpress.com by 190.67.178.174 on 09/15/19
For personal use only.

2
0d4 08 1.2 16 20

[loglo Po, (mm ) ] +I

FIG.3. Variation of overall reaction rate with initial
isopropyl alcohol partial pressure at 306 "C.

The Effect of Water Vapor

The effect of changing the water vapor partial
pressure from 0.2 to 3 mm was studied at 323 and
343 "C with isopropyl alcohol partial pressures [logop, (mm ) ] + I
of 7.78 and 16.83 mm, respectively. At 323 "C,
rl cc (gW)-O.O6,r2 cc (Pw)-0.45,andv, oc (p,)-'.08 FIG.4. The effect of the total water vapor partial
pressure on r , , r,, and r3 at 323 "C. Partial pressure
and at 343 "C,r , cc (pw)-0.059r, cc ('J,)-'.~~, and isopropyl alcohol, 7.78 rnrn.
r, cc (pw)-O.ll. Figure 4 illustrates these effects
for the experiments at 323 "C. After the removal
356 "C at 300 ml min-l (N.T.P.), about 1%
of the added water vapor the catalyst activity
initially condensed to give mesityl oxide and six
recovered to its original value.
other heavier ketones. After 2 h exposure this
The Efect of Acetone activity had fallen to about 0.5%.
When 27.77 mm of acetone was passed alone At 323 "C and isopropyl alcohol partial
over newly prepared manganese(I1) oxide at pressure 7.78 mm, the acetone partial pressure
3782 CANADIAN JOURNAL OF CHEMISTRY. VOL. 49. 1971
Can. J. Chem. Downloaded from www.nrcresearchpress.com by 190.67.178.174 on 09/15/19

FIG.5. The effect of the total acetone partial pressure

on r , , r,, and r3 at 323 "C. Partial pressure isopropyl
alcohol. 7.78 mm.

was varied from 2 to 6 mm. Figure 5 shows the

results obtained with respect to r,, r,, and r,.
The rates r, and r , are respectively K (pa,)-a.06
and (p,,)-O.10 over the entire pressure range while
r , cc from 2.02 to 4.32 mm but at the
logio P,, ( mm
higher pressures this index becomes - 1.00.
Clear evidence of catalyst poisoning was ob- FIG. 6. Variation of the rates of ketone, r,,,, and
tained here. For example, when the lase increment mesityl oxide, r,,,, formation with acetone partial
For personal use only.

pressure at 323 "C. Partial pressure isopropyl alcohol,

of acetone added was entirely removed at the 7.78 rnm.
conclusion of the experiments, r, was reduced by
17% and r, by 13% at the steady level of catalyst
activity. mol-I applies to both dehydrogenation and de-
In this series of experiments a small trace of hydration and therefore it may be compared
isopropyl ether was detected which increased in directly with the value of Balandin et al. (6),
quantity with increase in acetone partial pressure 25.4 kcal mol-l, for the dehydrogenation of
but never exceeded 0.1% of the total products. isopropyl alcohol on manganese(I1) oxide in the
Considerable amounts of other by-products in- same temperature region. Good agreement is
cluding mesityl oxide principally, and two of the obtained also between the value of 30 f 2 kcal
heavier ketones were present. These increased in mol-I found here for E from the data of Fig. 2
amount sharply but smoothly with increase in and the earlier value of 31.9 kcal mol- cal- '
acetone partial pressure such that they accounted culated (6) from adsorption coefficients. The
for approximately 30% of the total products at difference between E and E' is consistent with
5.74 mm acetone. Their rate of formation was the effects of the products on the overall reaction
between first and second order with respect to the rate. Balandin et al, interpreted their results as
acetone partial pressure, Fig. 6. being consistent with desorption of acetone as the
rate-controlling step. However, the complexity
Discussion of the reaction was not then entirely recognized.
The decomposition of isopropyl alcohol on Although the decomposition of isopropyl
manganese(I1) oxide involves principally de- alcohol has been studied frequently using a
hydrogenation, dehydration, and condensation variety of catalysts (I), few papers have em-
reactions. In the absence of added acetone the phasized the kinetics of the reaction and informa-
overall rate can be equated to the sum of the tion on the kinetic equations is limited. In
dehydrogenation and dehydration rates as the dehydrogenation, the possible rate-limiting steps
contributions of the condensation reactions are are (a) adsorption of the alcohol, (6) interaction
then very small. Since the selectivity ratio is al- on the surface, and (c) desorption of products.
most constant at 0.80 from 320 to 365 "C, the These possibilities have been discussed recently
+
apparent activation energy, E', of 26 0.5 kcal (17) for catalysis on zinc(I1) oxide and a kinetic
KLISSURSKI ET AL.: DECOMPOSITION OF ISOPROPYL ALCOHOL VAPOR 3783

analysis carried out. Also, in a general form the

following equation has been proposed (17) for the
catalytic dehydrogenation
klk3Palc
C11 rate =
k1Pa1c + k2Pac + k3
where k,, k,, and k, are, respectively, the rate
constants for the decomposition of the alcohol,
adsorption of acetone, and desorption of acetone
Can. J. Chem. Downloaded from www.nrcresearchpress.com by 190.67.178.174 on 09/15/19

while pal, and pa, are the partial pressures of

isopropyl alcohol and acetone, respectively. The
equation is in agreement with experimental data
(18) and was derived on the assumption that at
the steady state the rate of desorption of acetone
is equal to its rate of adsorption plus the rate of 0
on
dehydrogenation of the alcohol. By making suit- 0 04 12 16

I I p s , (mm I
able deductions in the classical manner regarding
the relative weights of the component terms in the FIG. 7. Langmuir plots of r l , r 2 , and r , at 306 "C.
equation it can be modified (17) to explain the
results of experiments with various catalysts similar results would be found at the higher tem-
where either (b) (18) or (c) (6, 19) was concluded peratures if higher partial pressures of isopropyl
to be the rate-limiting step. For the dehydration alcohol could have been used but experimental
For personal use only.

of isopropyl alcohol on metallic oxides a similar limitations were prohibitive because of the
kinetic treatment (20) yields an equation for the possibility of cracking reactions occurring (21).
rate like eq. 1 with rate parameters for water The Langmuir pattern observed at 306 "C was
substituted for those of acetone. also noted with Zn0.Cr203 (18).
The present results show that the order of the The zero order rate dependence of r,, r,, and r ,
overall decomposition reaction with respect to on the partial pressure of hydrogen is interpreted
isopropyl alcohol partial pressure depends on in terms of a very rapid saturation of any surface
the temperature and on the pressure region, sites present which might be amenable to the
Table 1 and Fig. 3. Similar effects have been adsorption of the gas or alternatively that ex-
observed (18) for the reaction of 30-300 mm tremely rapid desorption is involved. A similar
isopropyl alcohol at 383 "C on a ZnO.Cr,O, interpretation may be applied to explain the zero
catalyst. All of these data are in agreement with order kinetic relationship found with propylene.
+
eq. 1, since if it is accepted (17) that k, k,p,,, >> The effect of water vapor partial pressure on the
k,p,, the equation can be transformed to reaction rates, Fig. 4, is indicative of the accepted
view (21, 22) that different surface sites are in-
rate = k1
volved in the dehydrogenation and dehydration
( 1 1 ~ ~+1kiIk3
~ of alcohols since r, is only barely related to p,
Hence, with increase in the partial pressure of while r, is substantially inhibited by increase of
isopropyl alcohol the dehydrogenation reaction this partial pressure.
rate will be increasingly less affected by the The influence of acetone partial pressure on the
isopropyl alcohol partial pressure. A similar mechanism and rate of the reaction on man-
conclusion can be obtained for the dehydration ganese(I1) oxide is of a significantly more complex
reaction. The present results show that r,, r,, and character than has been accepted (6). The rates r,
also r, are affected in this fashion by increase in and r, are only slightly inhibited by increase in
the alcohol partial pressure with the major change acetone partial pressure while r, is more signifi-
reflected in r,, the rate of dehydrogenation. These cantly affected, particularly at high pressures.
features are illustrated in Fig. 7 where Langmuir Indications are that inhibition of acetone forma-
plots of llr,, llr, and llr, are given for the data tion would become almost total at acetone partial
obtained at 306 "C. It would be expected that pressures around 10 mm at 323 "C. Since the rate
3784 CANADIAN JOURNAL OF CHEMISTRY. VOL. 49, 1971

retardation exerted by acetone is largely con- due to an increase in the inter-molecular reactions
centrated on the other aath of the reaction from which produce mesityl oxide, isopropyl ether, and
that of the water vapor, the results can be regarded the heavier ketones as was observed when the
as added confirmation of the hypothesis that acetone partial pressure was increased. Since r ,
different surface loci are involved in the two was only slightly affected in these experiments it
competitive reactions. At the same time it is may be concluded that no significant change in
important to note (i) that at the highest acetone the surface concentration of adsorbed alcohol
partial pressures a substantial increase takes place took place. Hence the effect of added acetone is
in the rates of formation of the observed second- most probably exerted on a sensitive acetone
Can. J. Chem. Downloaded from www.nrcresearchpress.com by 190.67.178.174 on 09/15/19

ary products such that a kinetic relationship adsorption/desorption equilibrium condition as

exists between these rates and the acetone partial the course of catalysis is considerably modified.
pressure, Fig. 6, (ii) that a measurable reduction in The secondary reactions and effects present a
overall catalyst activity is observed after removal special interest for further study.
of this acetone, and (iii) that acetone itself under-
goes condensation reactions on the manganese(I1) 1. 0. V. KRYLOV. Catalysis by nonmetals. Academic
oxide. Hence the intervention of additional sur- Press, New York and London. 1970. p. 115.
2. S. Z. ROGINSKII.Zh. Fiz. Khim. 6, 334 (1935).
face reactions in these conditions occurs with 3. S. 2. ROGINSKII.Dokl. Akad. Nauk. SSSR, 67, 97
complex consequences in regard to mechanistic (1949).
interpretations especially involving r,. 4. K. HAUFFE. Reactions on solids and their surfaces.
At present, we have been unable to develop Russ. IL (1962).
mechanisms which completely satisfy the kinetic 5. G. M. SCHWAB.Adv. Catal. 2, 229 (1952).
6. A. A. BALANDIN, A. A. TOLSTHOPYATOVA, and V. H.
results. Nevertheless, some useful conclusions can MATUSHEKO.Izvest. Akad. Nauk, SSSR, 1333
For personal use only.

be made with respect to general if not detailed (1960).

reaction sequences. Thus the results show fair 7. P. SABATIER and A. MAILHE. Ann. Chim. (Paris),
agreement with the concept (18) that the rate- 8 (20), 289 (1910).
8. P. MARS. The mechanism of heterogeneous cataly-
limiting step is linked to reaction on the catalyst sis. Edited by J. H . de Boer et al. Elsevier Publishing
surface. Figure 7 supports this deduction as it Co., New York. 1960. p. 49.
expresses a proportionality between the reaction 9. R. A. Ross and B. G. WALSH. J. Appl. Chem. 11,
rate and the surface coverage although, taken 469 (1961).
alone, it does not imply that the slow step in the 10. A. B. HART,J. MCFADYEN, and R. A. Ross. Trans.
Faraday Soc. 59, 1458 (1963).
mechanisms of dehydrogenation and dehydration 11. T. E. MOORE,E. MARYLINN, and P. W. SELWOOD.
is reaction of the alcohol. However, some further J. Am. Chem. Soc. 72, 856 (1950).
support for this deduction is evident in the small 12. D. DOLIMORE, J. DOLIMORE, and J. LITTLE. J. Chem.
effect of acetone and water on the overall reaction Soc. A, 2946 (1969).
rate. Thus r, oc p,-O.' and but the effect 13. 0. KADLECand V. DANES. Coll. Czech. Chem.
Comm. 32, 1871 (1967).
of these products on the separate competitive 14. American Society for Testing Materials, Card Index.
reactions is more pronounced, r, K p,-0.5 and 15. 6. SCHWARZENBACH and H. FLASHKA.Complexo-
-(0.3-1.0) depending on partial pressure. metric titrations. Methuen and Co. Ltd., London.
r1 cc P,,
The addition of water vapor hardly affects the 1969. p. 234.
16. S. L. KIPERMAN.Introduction to the kinetics of
rate of the dehydrogenation reaction and the heterogeneous catalytic reactions. (Russ.) Nauka,
overall relationship of r, withp, is mathematical- Moscow. 1964. p. 543.
ly consistent with the catalyst selectivity. Tso- 17. 0. V. KRYLOV. RUSS.J. Phys. Chem. (Eng.) 39,
propyl ether disappears as the water vapor 1554 (1965).
pressure is increased and the catalyst recovers its 18. J-E. GERMAIN, J. BIGOURD, J-P. BEAUFILS, B. GROS,
and L. PONSALLE.Bull. SOC.Chim. Fr. 1777 (1961).
original activity when the added water vapor is 19. Y. DECHATRE and S. J. TEICHYER.Bull. Soc. Chim.
removed. This indicates the reversible adsorption Fr. 2804 (1967).
of water on active dehydration sites and clearly 20. 0. V. KRYLOV.RUSS.J. Phys. Chem. (Eng.) 39,1422
shows that the surface equilibrium is very sensi- (1965).
21. D. E. R. BENNETT and R. A. Ross. J. Chem. Soc. A,
tive to the pressure of water vapor. 1542 (1968).
The order found with respect to acetone for r, 22. 0. V. KRYLOVand E. A. FOK~YA. Kinet. Catal.
when compared with that for r , is believed lo be (Russ.) 1, 421, 541 (1960).