July 5, 19G1 6
A L U M I N A CATALYZED
\’
I
1
DEHYDRATION OF ALCOHOLS
1 4
-1
1
I \
2347 I
‘
I
(0
I
-0 X
2 x
1
I 0.3
I
0.2
0.1
[D]
2
I 0.4
I
I
I
I
0.1
0.2
0.3
0.4
[D]
(rnoles/l.).
Fig. 3.-Plot of kob.d. at 605 mp us. hexamethylbenzene concentration. Open circles represent experimental values; solid line calculated from equation 21.
Differentiation of this expression with respect to time gives which, by equation 4, equals - k o b s d . ( A A A ) ~ . Thus, by combining equation 20 with the rate of disappearance of the complex (equation 15), the observed rate constant becomes Values of kobsd. calculated from this expression are compared in Figs. 3 and 4 (solid lines) with the experimentally determined values (open circles) a t 605 and 646 my, respectively. For the calculations, the more reliable value of K,from Fig. 1 (2.7 1. mole-’) was used for both wave lengths, although the particular value of used was that determined for the specific wave length.
(rnoles/l.).
Fig. 4.-Plot of kobsd. at 646 mp us. hexamethylbenzene concentration. Open circles represent experimental values; solid line calculated from equation 21.
It is clear that equation 21 correctly gives the observed qualitative behavior of k o b s d . with varying hexamethylbenzene concentration and appears to deviate badly only a t the lowest concentrations of D, where experimental errors are quite large. It is concluded, therefore, that this interpretation in terms of a very rapid attainment of equilibrium between iodine atoms and D molecules to form D I complexes, together with the several simultaneous combination processes leading to Is, provides a satisfactory and consistent picture of the observed kinetics. As pointed out, though, the assumption that all of the combination rate constants are the same can only be considered approximate, and more accurate measurements on k should show deviations from second order kinetics. Acknowledgment.--? he authors gratefully acknowledge support by the National Science Foundation through grant No. NSF G-9988.
[CONTRIBUTION FROX ’IH E I P a T I E F F HIGH PRESSURE AND CATALYTIC LABORATORY, DEPARTMENT OF CIIEXISTRY, ILLINOIS] NORTHWESTERN UNIVERSITY, EVANSTON,
Alumina : Catalyst and Support. IX.1 The Alumina Catalyzed Dehydration of Alcoh01~2~3 B Y HERMAN PINES
AND WERNER
0. HAAC4
RECEIVED DECEMBER 19, 1960 There is no agreement in the literature with regard to the mechanism of catalytic dehydration of alcohols over alumina and not even with rpspect t o the nature of olefinic hydrocarbons. It was demonstrated t h a t the discrepancies result from different catalytic properties of the alumina catalysts used. Alumina catalysts can vary widely in their activity for double bond shift and for skeletal isomerization of olefinic hydrocarbons. These differences also influence the product distribution in the dehydration of alcohols. Dehydration was studied with aluminas having a whole spectrum of isonierization properties. The following alcohols were used: cyclohexanol, 2-butanol, 2-pentanol, 3-pentanol, 3,3-dimethyl-2-butanol (pinacolyl alcohol). The mechanism of the dehydration and of the accompanying isomerization is discussed.
Alumina is an excellent and widely used catalyst for the dehydration of alcohol^.^ I n spite of this (1) For paper VI11 of these series see H. Pines a n d C. T. Chen, Proceeding of t h e 2 n d International Congress in Catalysis, Paris, July 4-9, 19GO. (2) Paper 11 uf t h e series “Dehydration of Alcohols.” For paper I, see H. Pines a n d C. S . Pillai, J. A m . Chem. Soc., 82,2401 (1960). (3) Presented in p a r t before t h e Division of Colloid Chemistry, American Society Meeting. S a n Francisco, April 13-18, 1958; a n d
fact there is no agreement in the literature with regard to the mechanism of this reaction or the nature of the olefinic products. For example, pure before t h e Gordon Research Conferences in Catalysis, J u n e 23-27, 1958, New London, N . H. (4) Predoctoral Felluw, Univrrsirl Oil I ’ r d u u t s C
