Mechanism of cyclohexanol dehydration catalysed by zirconium phosphate

Article abstract of DOI:10.1039/a801360h

Deuterium-labelling studies have shown that dehydration of cyclohexanol to cyclohexene over amorphous zirconium phosphate at 350°C proceeds through a carbocation mechanism and not through a concerted process.

Full text of DOI:10.1039/a801360h

View Article Online / Journal Homepage / Table of Contents for this issue
Mechanism of cyclohexanol dehydration catalysed by zirconium
phosphate
Robert A. W. Johnstone, Jun-Yao Liu and David Whittaker*
Department of Chemistry, University of Liverpool, Liverpool, UK L69 7ZD
O
Deuterium-labelling studies have shown that dehydration
of cyclohexanol to cyclohexene over amorphous zirconium
phosphate at 350 ЊC proceeds through a carbocation mech-
HO
H
H
D
D
D
D
D
D
D
D
D
D
D
anism and not through a concerted process.
1
2
The dehydration of cyclohexanol to cyclohexene is often used
as a model reaction to determine the catalytic efficiency of
metal phosphates1–3 even though the mechanism of the reac-
tion is unknown. Studies of the dehydration of the 2-, 3- and
4-methylcyclohexanols, over alumina4 and over metal phos-
phates5 suggest that there are two mechanisms; one a synchron-
ous mechanism, in which the OH and H are removed by the
catalyst, possibly simultaneously, so that the reaction proceeds
without any rearrangement, and the second a carbocation
reaction which permits rearrangement. Any modification of the
stereochemistry of the methylcyclohexanols which occurs
during reaction is indicative of a carbocation reaction. In the
absence of any such stereochemical change, the reaction is
assumed to be synchronous. Both mechanisms have been
observed4 (Scheme 1).
The symmetry of cyclohexanol precludes using a stereo-
chemical probe. Therefore, in this work, the reaction has been
studied by deuterium labelling. For a synchronous mechanism,
dehydration of 1 should give a mixture of 2 and its mirror
image in equal proportions. Dehydration of 1 via a carbocation
would give an identical result, only if it took place without
rearrangement. However, a shift of the carbocation centre from
its initial position at C-1 would start a process of scrambling
the deuterium label, so that a modified pattern of deuterium-
labelling would appear. Evidence that such shifts take place
comes from work on the dehydration of methylcyclohexanols
and from the behaviour of carbocations in superacids;
migration of charge round a cyclopentane ring is so fast that
the hydrogen atoms become equivalent on the NMR timescale.6
It is not known whether or not a cyclohexyl carbocation
behaves similarly, since in superacid it rearranges rapidly to a
methyl substituted cyclopentyl carbocation.7
Deuterium labelled cyclohexanol was prepared by exchange
of the four labile protons of cyclohexanone with deuterium
oxide, in the presence of an acid catalyst. Reduction of the
exchanged ketone with lithium aluminium hydride gave
cyclohexanol in which 94% of the hydrogens on C-2 and C-6
1
had been replaced by deuterium, as determined by H NMR.
13C NMR confirmed the positions of the labels. Dehydration
of the 2,2Ј,6,6Ј-[2H4]cyclohexanol over amorphous zirconium
phosphate at 350 ЊC in the gas phase8 gave cyclohexene, con-
1
taining 4.0% of methylcyclopentene. The H NMR spectrum
of the product contained three peaks, resulting from sets of
equivalent hydrogen atoms on carbon atoms 1 and 2, 3 and 6,
and 4 and 5. In unlabelled cyclohexene, these are in the ratio
1:2:2. If the reaction had been synchronous, the product
should have shown peaks in the ratio of 1:2:4. Similarly,
Concerted dehydration
H
H
H
O
+ H2O
H
H
+
O
H
H
H
H
H
O
O
O–
O
O
recycle
Non-concerted dehydration
+
H
H
H
O
+ H2O
H
H
+ H2O
+ H2O
H
+
O
H
H
H
H
H
H
O
O
O
O–
O–
O O
recycle
Scheme 1
J. Chem. Soc., Perkin Trans. 2, 1998
1287

View Article Online
carbocation reaction without rearrangement would produce
the same ratio, but migration of the carbocation centre would
cause the ratio to approach 1:2:2. For the cyclohexene
obtained in this work, the ratio was 0.93:2:2.48, which fits well
with a reaction involving a highly mobile carbocation, but one
which has not had sufficient time to randomise completely.
The process of randomisation would in any case be slowed by
isotope effects, which would cause hydrogen to move more
rapidly than deuterium.
tion cannot scramble the label, so the extent of scrambling is
irrelevant to this work. Any isotope effect on loss of H/D from
the carbocation would favour loss of hydrogen from the double
bond carbon atom, and hence reduce the observed scrambling;
since extensive scrambling is observed, this effect can be
ignored.
It is concluded that the dehydration of cyclohexanol over
amorphous zirconium phosphate at 350 ЊC proceeds via a carbo-
cation mechanism and involves charge migration around the
ring. The speed of the reaction expected depends on the
number of acidic sites in the catalyst, their acid strength, and
the availability of the site to the OH group of cyclohexanol.
Since this does not appear to be a concerted reaction, bonding
of the whole substrate to the catalyst is probably not important.
In this reaction, the catalyst is simply a solid acid.
As a further test of this randomisation, the work was
repeated except that the labelled cyclohexanone was reduced
with lithium aluminium deuteride. This led to cyclohexanol,
2
1
having 18% of H on C-1 and 82% of H, as determined by H
NMR spectroscopy, and the position confirmed by 13C NMR.
The dehydration over amorphous zirconium phosphate at
350 ЊC was repeated. Reaction without rearrangement should
give a ratio of peaks of 0.18:2:4, while a complete scrambling
of the label should give a ratio approaching 1:2:2. The spec-
trum obtained showed a ratio of 0.76:2:2.46, indicating again
that extensive scrambling of the label had taken place.
References
1 A. Clearfield and D. S. Thakur, J. Catal., 1980, 65, 185.
2 T. N. Frianeza and A. Clearfield, J. Catal., 1984, 85, 398.
3 J. L. Colón, D. S. Thakur, C.-Y. Yang, A. Clearfield and C. R.
Martin, J. Catal., 1990, 124, 148.
4 C. L. Kibby, S. S. Lande and W. K. Hall, J. Am. Chem. Soc., 1972, 94,
214.
It could be argued that labelling the 2 and 6 positions of
cyclohexanol with deuterium favours the carbocation reaction
since the concerted reaction is slowed by the isotope effect
introduced. This is true, but it is balanced by a similar slowing
of the shift of a deuterium atom to C-1 relative to a hydrogen
atom in the first step of migration of the carbocation from C-1.
It could also be argued that randomisation of the label may
have taken place after formation of cyclohexene; this is unlikely,
since we have found that 4-methylcyclohexene is only 4% con-
verted into 1-methylcyclohexene under our reaction conditions,
despite having a methyl substituent which should favour reac-
tion relative to the unsubstituted alkene.
5 L. F. Hodson, Ph.D. Thesis, Liverpool, 1996.
6 D. M. Brouwer and E. L. Mackor, Proc. Chem. Soc., 1964, 147; D. M.
Brouwer, Recl. Trav. Chim. Pays-Bas, 1968, 87, 210; M. Saunders
and J. Rosenfeld, J. Am. Chem. Soc., 1969, 91, 7756.
7 G. A. Olah and J. Lukas, J. Am. Chem. Soc., 1968, 90, 933.
8 M. C. Cruz-Costa, Ph.D. Thesis, Liverpool, 1991.
Paper 8/01360H
Received 17th February 1998
Accepted 7th April 1998
Detailed pathways of deuterium scrambling are difficult to
predict because of isotope effects. However, a concerted reac-
1288
J. Chem. Soc., Perkin Trans. 2, 1998