Acid-catalyzed trimerization of acetaldehyde: A highly selective and reversible transformation at ambient temperature in a zeolitic solid

Article abstract of DOI:10.1021/jp012440y

Acetaldehyde underwent a reversible Bronsted acid-catalyzed cyclotrimerization reaction, with 100% selectivity, at ambient temperature within the zeolites host material ferrierite. The cyclic trimer was the only product formed in the reaction. The equilibrium proportions of acetaldehyde and the cyclic trimer at ambient temperature corresponded to a conversion > 90%. On the contrary, a broad distribution of products was obtained in the corresponding acid-catalyzed transformation of acetaldehyde in the liquid state. The reversibility of the cyclotrimerization reaction in ferrierite was confirmed from the fact that, on adsorption of a pure sample of the cyclic trimer within ferrierite, a reaction occurred to produce acetaldehyde as the only product with the same equilibrium distribution of the cyclic trimer and acetaldehyde as that realized from the reaction of acetaldehyde in ferrierite. The fact that no reaction occurred on adsorption of acetaldehyde within sodium-exchanged ferrierite confirmed the role of Bronsted acid catalysis in the transformation between acetaldehyde and the cyclic trimer in ferrierite.

Full text of DOI:10.1021/jp012440y

1322
J. Phys. Chem. B 2002, 106, 1322-1326
Acid-Catalyzed Trimerization of Acetaldehyde: A Highly Selective and Reversible
Transformation at Ambient Temperature in a Zeolitic Solid
Sang-Ok Lee, Simon J. Kitchin, and Kenneth D. M. Harris*
School of Chemistry, UniVersity of Birmingham, Edgbaston, Birmingham B15 2TT, United Kingdom
Gopinathan Sankar, Markus Dugal, and John Meurig Thomas*
DaVy Faraday Research Laboratory, The Royal Institution, 21 Albemarle Street,
London W1X4BS, United Kingdom
ReceiVed: June 26, 2001; In Final Form: October 15, 2001
Acetaldehyde is shown to undergo a reversible Brønsted acid-catalyzed cyclotrimerization reaction, with 100%
selectivity, at ambient temperature within the zeolite host material ferrierite. It is shown by in situ solid-state
13C NMR spectroscopy and other techniques that the cyclic trimer is the only product formed in this reaction.
The equilibrium proportions of acetaldehyde and the cyclic trimer at ambient temperature correspond to a
conversion greater than 90%. In contrast, in the corresponding acid-catalyzed transformation of acetaldehyde
in the liquid state, a broad distribution of products is obtained. The reversibility of the cyclotrimerization
reaction in ferrierite is confirmed from the fact that, on adsorption of a pure sample of the cyclic trimer
within ferrierite, a reaction occurs to produce acetaldehyde as the only product with the same equilibrium
distribution of the cyclic trimer and acetaldehyde as that obtained from the reaction of acetaldehyde in ferrierite.
The role of Brønsted acid catalysis in the transformation between acetaldehyde and the cyclic trimer in ferrierite
is confirmed from the fact that no reaction occurs on adsorption of acetaldehyde within sodium-exchanged
ferrierite.
1. Introduction
Concepts of catalysis arise in many diverse areas of chemical,
biological, and materials sciences, and a worthwhile goal is to
seek to establish a fundamental understanding of the unifying
features that link different types of catalytic systems and
processes. As a starting point, we require to define systems and
Figure 1. Molecular structure of acetaldehyde.
processes that are chemically related, have sufficient structural
similarities, and can be carried out under comparable experi-
mental conditions. Zeolitic materials and other microporous
solids are utilized in a wide variety of catalytic applications,1-5
many of which rely on the Brønsted acidity of the framework
and/or the geometric constraints imposed upon the reaction by
the framework (i.e. shape selectivity). However, most catalytic
applications of these materials are investigated at elevated
temperatures (recalling that these framework structures may be
stable up to temperatures of the order of 1000 °C in some cases),
involving irreversible transformations to produce some desired
product. In this study, we focus instead on acid-catalyzed
reactions that occur at ambient temperature, with a view to
exploring possible parallels with the wide range of acid-
catalyzed transformations that are known to occur for organic
molecules in the liquid state around ambient temperature. In
addition, it is relevant to recall that the analogy3,4,6-11 between
the mode of action of enzymes and zeolites (as well as other
molecular sieve catalysts such as metal-substituted alumino-
phosphates) requires the study of chemical transformations close
to ambient temperature, corresponding to the temperatures at
which enzyme-catalyzed processes occur in living systems.
In pursuing these analogies, and focusing on acid-catalyzed
transformations that occur under ambient conditions, model
reaction systems include unimolecular rearrangements (follow-
ing protonation), interconversions between tautomeric forms,
and simple reactions involving nucleophilic substitution (either
for the protonated reactant or the species produced following
an initial rearrangement of the protonated reactant). Ideally, the
model reaction system should have the potential to undergo a
variety of competing reaction pathways, allowing selectivity to
be probed. Despite the wide-ranging studies of catalytic appli-
cations of zeolitic materials, comparatively few studies of
reactions under ambient conditions have been reported, with
representative examples given in refs 12-20.
In this paper, we report a reversible acid catalyzed transfor-
mation that occurs at ambient temperature within a zeolitic solid
and proceeds with substantially greater selectivity than the
corresponding transformation under comparable conditions in
the liquid state. We also discuss in more general terms the
approaches and strategies that should be adopted in addressing
relevant issues in this area.
The model reaction identified in this study is the transforma-
tion of acetaldehyde (Figure 1) within the zeolite host material
ferrierite (Figure 2). The ferrierite structure21 contains two types
of tunnels (with cross sections 4.2 Å × 5.4 and 3.5 Å × 4.8
Å), both of which are accessible (as verified by molecular
* To whom correspondence should be addressed. E-mail: K.D.M.H.,
K.D.M.Harris@bham.ac.uk; J.M.T., jmt@ri.ac.uk.
10.1021/jp012440y CCC: $22.00 © 2002 American Chemical Society
Published on Web 01/18/2002

Acid-Catalyzed Trimerization of Acetaldehyde
J. Phys. Chem. B, Vol. 106, No. 6, 2002 1323
Figure 4. Molecular structure of the cyclic trimer P.
Figure 2. Computer graphic picture showing the structure of ferrierite
viewed along the direction of the main tunnel. Three molecules of
acetaldehyde and one molecule of the cyclic trimer P are shown, with
molecular surfaces represented.
Figure 5. Proposed mechanism for the Brønsted acid-catalyzed
cyclotrimerization of acetaldehyde to produce P.
lines are observed for both the methyl carbon and the ring carbon
of P suggests that only one configurational isomer of P is
produced, as elaborated in ref 24. As discussed below, this
conclusion is further supported by results from solution state
NMR studies.
From integration of the signals due to P and acetaldehyde in
the solid-state 13C NMR spectrum, the percentage conversion
from acetaldehyde to P in ferrierite at ambient temperature is
estimated to be 93%, corresponding to a molar ratio of
P:acetaldehyde of 4.43 (see ref 25). There was no change in
the relative amounts of acetaldehyde and P during successive
measurements of the solid-state 13C NMR spectrum, indicating
that the reaction had already reached final conversion by the
time of starting the solid-state NMR experiments. Thus, with
the experimental approach used here, kinetic aspects of the
reaction could not be investigated.
Figure 3. In situ high-resolution solid-state 13C NMR spectrum
recorded at ambient temperature for a sealed ampule of acetaldehyde/
ferrierite. Peaks due to acetaldehyde (denoted A) and the product P
are marked.
modeling investigations carried out as part of this study) by
molecules of the size of acetaldehyde. At ambient temperature,
pure acetaldehyde is a stable liquid; in acidic environments, it
is protonated on the oxygen atom of the carbonyl group, and a
variety of competing processes may then ensue, as discussed
below.
To further characterize the reaction product, the quartz ampule
used in the solid-state 13C NMR experiments was broken under
a nitrogen atmosphere, and the organic guest component inside
the ferrierite was washed out with DMSO and studied by
solution-state 1H and 13C NMR spectroscopy and GC-MS
analysis. All analyses carried out indicated that acetaldehyde
and P were the only species extracted from the ferrierite (see
Figure 6a). From integration of the signals due to P and
acetaldehyde in the solution-state 1H NMR spectrum, the
percentage conversion from acetaldehyde to P was estimated
to be 87%, in close agreement with the results from our solid-
state NMR studies. In further agreement with the solid-state
NMR results, our solution-state 1H and 13C NMR spectra support
our assignment that only one configurational isomer of P is
produced from the reaction of acetaldehyde in ferrierite (see
ref 24).
2. Results and Discussion
Acetaldehyde was included within the H+ form of ferrierite
using the procedure described in section 4.2, followed by in
situ characterization at ambient temperature using high-resolu-
tion solid-state 13C NMR spectroscopy (for the sample of
acetaldehyde/ferrierite contained in a sealed quartz ampule).
From the solid-state 13C NMR spectrum (Figure 3), it is clear
that a reaction occurs to generate a single product (denoted P),
together with some unreacted acetaldehyde. The product P is
identified as the cyclic trimer shown in Figure 4. As discussed
in section 4.4, this assignment was further confirmed from
studies involving an authentic sample of pure P obtained
commercially. A proposed mechanism for the Brønsted acid-
catalyzed conversion of acetaldehyde to P is given in Figure 5.
We note that all lines in the solid-state 13C NMR spectrum
of acetaldehyde/ferrierite (and those of the other samples
discussed below) are very narrow. The fact that single narrow
In principle, linear oligomers [-OCH(CH3)-]n (with n > 3)
could also arise if the intermediate B in Figure 5 were to react
with another monomer of acetaldehyde rather than undergoing

1324 J. Phys. Chem. B, Vol. 106, No. 6, 2002
Lee et al.
5. Clearly the linear trimer would contain more than one
distinguishable CH3 carbon environment and more than one
distinguishable CH carbon environment, which should be
resolved in both the solid-state and solution-state NMR spectra.
Finally, a solid-state 13C NMR spectrum was recorded for
the ferrierite sample after the extraction procedure with DMSO.
The spectrum showed no evidence for any carbon-containing
species (e.g. polymer) remaining behind in the ferrierite,
confirming that all reaction products had been extracted.
To provide a comparison between the reaction of acetaldehyde
in the constrained environment of the ferrierite host structure
and the corresponding reaction of acetaldehyde in the liquid
state, a number of control experiments were carried out. First,
a small amount of p-toluenesulfonic acid was added to pure
liquid acetaldehyde at ambient temperature (see section 4.5),
and the mixture was studied subsequently by solution-state 1H
and 13C NMR spectroscopy and GC-MS analysis. These
analyses indicate that acetaldehyde reacts under these conditions
in the liquid state to generate a substantial number of products,
1
including the cyclic trimer P. In the solution-state H and 13C
NMR spectra, the main peaks are in the region of the peak
positions characteristic of the cyclic trimer P but are significantly
1
broader than the peaks in the solution-state H and 13C NMR
spectra for the products extracted from the reaction of acetalde-
hyde in ferrierite, showing that the liquid-state reaction generates
a distribution of products with rather similar chemical environ-
1
ments. In this regard, we recall that the set of H and 13C
chemical shifts for linear oligomers [-OCH(CH3)-]n of dif-
ferent chain length n and for cyclization products (such as P)
should be close to each other due to the similarity of the local
environments of a given type of functional group in each of
these molecules. In addition, the solution-state 1H and 13C NMR
spectra contain several peaks of comparatively low intensity at
other positions, representing other minor products and/or end-
group environments. In Figure 6, the gas chromatogram for the
products from the liquid-state reaction (Figure 6b) is compared
with that for the products extracted following the reaction of
acetaldehyde in ferrierite (Figure 6a).
Another liquid-state reaction was carried out by adding a
small amount of hydrochloric acid to liquid acetaldehyde at
ambient temperature. Analysis by GC-MS (Figure 6c) and
1
solution-state H and 13C NMR spectroscopy led to similar
conclusions to those discussed above for the liquid-state reaction
involving acetaldehyde and p-toluenesulfonic acid. In summary,
it is clear that the reaction of acetaldehyde under conditions of
Brønsted acidity in the liquid state generates a wide range of
products, in contrast to the clear selectivity observed for the
reaction of acetaldehyde in ferrierite.
Two additional investigations were carried out to understand
further the nature of the reaction of acetaldehyde to produce P
within the ferrierite host structure. First, to confirm that Brønsted
acid catalysis is an essential feature of the reaction, we repeated
our experiment within a sample of sodium-exchanged ferrierite,
which is devoid of Brønsted acid sites (see section 4.6). The
solid-state 13C NMR spectrum (Figure 7) recorded for this
sample reveals that only unreacted acetaldehyde is present, with
no detectable amounts of P or any other reaction product.
Figure 6. Comparison of gas chromatograms for the products from
(a) the reaction of acetaldehyde in ferrierite (A ) acetaldehyde; P )
product), (b) the liquid-state reaction of acetaldehyde with p-toluene-
sulfonic acid, and (c) the liquid-state reaction of acetaldehyde with
hydrochloric acid.
cyclization to produce P, but our GC-MS analysis of the guest
species extracted from the ferrierite rules out the presence of
any reaction product of mass higher than P. Furthermore, while
the 1H and 13C chemical shifts in P may be rather close to those
in linear oligomers, the signals due to the end groups in any
oligomeric species would be observable in both the solid-state
and solution-state NMR spectra (provided the number of
monomer units n is not sufficiently high).
Second, after P was identified as the only product of the
reaction in ferrierite, a sample of pure P (commercial name:
paraldehyde) was obtained. To assess whether the transformation
of acetaldehyde to P within ferrierite is genuinely reversible,
the sample of pure P was adsorbed within ferrierite (see section
4.4), using the same procedure as that for the adsorption of
acetaldehyde inside ferrierite. The solid-state 13C NMR spectrum
On a similar basis, the cyclic trimer shown in Figure 4 can
be distinguished from the linear trimer, which would be
produced by chain termination of the intermediate B in Figure

Acid-Catalyzed Trimerization of Acetaldehyde
J. Phys. Chem. B, Vol. 106, No. 6, 2002 1325
catalysts, as demonstrated22,23 in the regioselective oxidation
of alkanes, to further impose geometric specificity over organic
transformations occurring within them.
4. Experimental Section
4.1. Instrumentation. All solid-state 13C and 1H NMR spectra
were recorded using a Chemagnetics CMX Infinity 300
spectrometer, at 75.49 and 300.18 MHz respectively.
Solution-state 13C NMR spectra were recorded at 75.49 MHz
on a Bruker AC300 spectrometer using DMSO-d6 as solvent
and a PENDANT pulse sequence. Solution-state 1H NMR
spectra were recorded at 300.18 MHz on a Bruker AC300
spectrometer using DMSO-d6 as solvent.
Figure 7. In situ high-resolution solid state 13C NMR spectrum
recorded at ambient temperature for a sealed ampule containing the
sample of acetaldehyde adsorbed in sodium-exchanged ferrierite.
Thermogravimetric analysis was carried out using a Perkin-
Elmer TGA6 instrument under a flow of nitrogen gas, with the
sample heated from 30 to 600 °C at a rate of 10 °C min-1
.
GC-MS analysis was carried out using a VG Prospec mass
spectrometer equipped with a Fisons 8000 series gas chromato-
graph. The samples were injected into a 25 m DB5 capillary
column and separated using a temperature program from 60 to
250 °C at 10 °C min-1
.
4.2. Preparation of Acetaldehyde/Ferrierite. A sample of
ferrierite (idealized formula Na2Mg2[Al6Si30O72]‚18H2O) with
an actual Si/Al ratio of 40 was calcined in air at 550 °C for 12
h and dehydrated under vacuum at 400 °C for 1 h. Using
standard vacuum line techniques, acetaldehyde was adsorbed
from the gas phase into the dehydrated ferrierite, which was
contained in a 9.5 mm diameter quartz tube. The temperature
of the acetaldehyde/ferrierite sample was not raised above
ambient temperature during this procedure. After adsorption,
the quartz tube was immersed in liquid nitrogen and sealed to
form an ampule of appropriate length for solid-state NMR
experiments. In a separate series of thermogravimetric analysis
experiments, the loading of acetaldehyde in the ferrierite was
estimated to represent approximately 5.3 molecules of acetalde-
hyde per unit cell of ferrierite.
4.3. Analysis of Acetaldehyde/Ferrierite. Solid-state 13C
NMR spectra were recorded at ambient temperature (25 °C) on
the sealed quartz ampule of acetaldehyde/ferrierite, which was
wrapped in PTFE tape to ensure a tight fit within the magic-
angle spinning NMR rotor (9.5 mm diameter). The spectra were
recorded using a direct polarization spin-echo sequence (to
suppress the signal from the PTFE tape), under conditions of
magic-angle spinning and 1H decoupling. The magic angle
spinning frequency was 1.0 kHz (stable spinning at higher
frequencies was difficult to attain for the sealed ampules,
although the comparatively low spinning frequency used has
no detrimental effect on the quality of the spectra recorded).
1H decoupling was carried out using the WALTZ sequence and
was applied during acquisition and during the spin-echo delays
but not during the recycle delay. We note that, for the samples
studied here, high-power (16 kHz) CW decoupling was not able
to completely remove the 13C-1H J-coupling. The recycle delay
was 30 s.
Figure 8. In situ high-resolution solid-state 13C NMR spectrum
recorded at ambient temperature for a sealed ampule containing the
sample of P adsorbed in ferrierite. Peaks due to acetaldehyde (denoted
A) and P are marked.
(Figure 8) recorded at ambient temperature for a sealed quartz
ampule containing this sample indicates that a reaction occurs
to produce acetaldehyde and that acetaldehyde and P are the
only organic species present. Importantly, the solid-state 13C
NMR spectrum (Figure 8) is essentially identical to that (Figure
3) recorded for the sample prepared by adsorption of acetalde-
hyde inside ferrierite, and the relative proportions of acetalde-
hyde and P determined from these spectra are the same within
experimental errors. The relative amounts of acetaldehyde and
P (estimated by integration of the solid-state 13C NMR
spectrum25) represent a percentage conversion (from acetalde-
hyde to P) of 92%, corresponding to a P:acetaldehyde molar
ratio of 3.83. Thus, as expected from the principle of micro-
scopic reversibility, the same equilibrium distribution of acet-
aldehyde and P is produced whether starting from pure
acetaldehyde or starting from pure P. In this regard, we recall
that each step in the proposed mechanism (Figure 5) for the
acid-catalyzed interconversion between acetaldehyde and P is
a reversible one.
3. Concluding Remarks
In this paper, we have identified a reaction that proceeds
reversibly and with very high selectivity within a zeolitic host
structure under ambient conditions. The high product specificity
(in contrast to the corresponding reaction in the liquid state) is
clearly a consequence of the spatial constraints imposed by the
well-defined microenvironment within the zeolitic host structure.
In view of the wide range of acid catalyzed rearrangements
of organic molecules known to occur in the solution state around
ambient temperature, there are interesting opportunities to
explore the effects of imposing spatial restrictions on these
reactions by carrying them out within microporous Brønsted
acid catalysts containing microenvironments of differing ge-
ometries. Furthermore, there is also scope to “tune” the
positioning of active sites within these inorganic microporous
Following the solid-state NMR experiments, the ampule was
broken in an atmosphere of dry nitrogen (glovebox) and the
organic species present within the ferrierite were extracted in
DMSO. These extracts were analyzed by solution-state 1H and
13C NMR spectroscopy and GC-MS analysis. The results of
these analyses confirm the assignment of the cyclic trimer P as
the only reaction product (in addition to unreacted acetaldehyde).
4.4. Preparation and Analysis of P/Ferrierite. After
identification of P as the only product from the reaction of
acetaldehyde in ferrierite, a sample of pure P (paraldehyde) was

1326 J. Phys. Chem. B, Vol. 106, No. 6, 2002
Lee et al.
1
obtained from Sigma-Aldrich. First, solution-state H and 13C
NMR spectra were recorded for the sample of pure P and
confirmed that the product from the reaction of acetaldehyde
in ferrierite had been correctly assigned as P.
(5) Cheetham, A. K.; Ferey, G.; Loiseau, T. Angew. Chem., Int. Ed.
Engl. 1999, 38, 3268.
(6) Haag, W. O.; Lago, R. M.; Weisz, P. B. Nature 1984, 309, 589.
(7) Wright, P. A.; Thomas, J. M.; Cheetham, A. K.; Nowak, A. K.
Nature 1985, 318, 611.
Second, the sample of pure P was adsorbed inside ferrierite
using the procedure described (for acetaldehyde) in section 4.2.
The solid-state 13C NMR spectrum of a sealed quartz ampule
containing the sample of P/ferrierite was recorded at ambient
temperature using the procedure described in section 4.3.
Following the solid-state NMR experiment, the ampule was
broken in an atmosphere of dry nitrogen (glovebox) and the
organic species present within the ferrierite were extracted in
DMSO. These extracts were analyzed by solution-state 1H and
13C NMR spectroscopy and GC-MS analysis, as described in
section 4.3.
(8) Derouane, E. G. J. Catal. 1986, 100, 541.
(9) Derouane, E. G.; Vanderveken, D. J. Appl. Catal. 1988, 45, L15.
(10) Raja, R.; Ratnasamy, P. J. Mol. Catal. 1995, 100, 93.
(11) van Santen, R. A. Catal. Technol. 1998, 2, 161.
(12) Jacobs, P. A.; Declerck, D. J.; Vandamme, L. J.; Uytterhoeven, J.
B. J. Chem. Soc., Faraday Trans. 1 1975, 71, 1545.
(13) Kwok, T. J.; Jayasuriya, K.; Damavarapu, R.; Brodman, B. W. J.
Org. Chem. 1994, 59, 4939.
(14) Bhaumik, A.; Tatsumi, T. Chem. Commun. 1998, 463.
(15) Wennerberg, J.; Ek, F.; Hansson, A.; Frejd, T. J. Org. Chem. 1999,
64, 54.
(16) Hartmann, M.; Kevan, L. Chem. ReV. 1999, 99, 635.
(17) Hirano, T.; Li, W.; Abrams, L.; Krusic, P. J.; Ottaviani, M. F.;
Turro, N. J. J. Org. Chem. 2000, 65, 1319.
(18) Turro, N. J.; Lei, X. G.; Li, W.; Liu, Z. Q.; McDermott, A.;
Ottaviani, M. F.; Abrams, L. J. Am. Chem. Soc. 2000, 122, 11649.
(19) Turro, N. J. Acc. Chem. Res. 2000, 33, 637.
(20) Lee, S.-O.; Sankar, G.; Kitchin, S. J.; Dugal, M.; Thomas, J. M.;
Harris, K. D. M. Catal. Lett. 2001, 73, 91.
4.5. Reaction of Acetaldehyde in the Liquid State. To
investigate the behavior of acetaldehyde in the presence of
Brønsted acids in the liquid state, small amounts of p-
toluenesulfonic acid and hydrochloric acid were added (in
separate experiments) to acetaldehyde (10 mL) and left at
ambient temperature for 10 h. Each mixture was then studied
(21) Meier, W. M., Olson, D. H., Eds. Atlas of Zeolite Structure Types;
Butterworths: Markham, Canada, 1987; p 64.
(22) Thomas, J. M.; Raja, R.; Sankar, G.; Bell, R. G. Nature 1999, 398,
227.
1
by solution-state 13C and H NMR spectroscopy and GC-MS
analysis.
(23) Dugal, M.; Sankar, G.; Raja, R.; Thomas, J. M. Angew. Chem.,
Int. Ed. 2000, 39, 2310.
4.6. Experiments on Sodium-Exchanged Ferrierite. To
confirm the role of Brønsted acid catalysis in the reaction
observed for acetaldehyde in ferrierite, an experiment was
carried out involving adsorption of acetaldehyde within a sample
of sodium-exchanged ferrierite. To prepare sodium-exchanged
ferrierite, a sample of the H+ form of ferrierite (6 g) was added
to 1.5 L of aqueous sodium acetate (0.1 M), and the solution
was stirred overnight. The ferrierite was then filtered off, washed
with water, and dried in an oven at 100 °C for 3 h. The
procedures described in section 4.2 for calcination and dehydra-
tion of ferrierite were carried out. A solid-state 1H NMR
spectrum recorded for the sample of sodium-exchanged ferrierite
contained in a sealed quartz ampule showed that no detectable
amounts of Brønsted acid sites were present [this spectrum was
recorded using a single H pulse sequence with magic-angle
spinning (spinning frequency 2.5 kHz; rotor diameter 9.5 mm;
recycle delay 30 s)].
Acetaldehyde was adsorbed within the sample of sodium-
exchanged ferrierite using the procedure described in section
4.2. Analysis of the resulting material by solid-state 13C NMR
spectroscopy (using the procedure described in section 4.3)
confirmed that no reaction had occurred (i.e. acetaldehyde was
detected as the only organic species present within the ferrierite).
(24) Molecule P can exist as different configurational isomers, depending
on whether the methyl groups lie on the same face or opposite faces of the
six-membered ring. In particular, there are two configurational isomers:
[A] with all three methyl groups on the same face of the six-membered
ring and [B] with two methyl groups on one face and one methyl group on
the other face. In the case of [A], and assuming a chair conformation of
the six-membered ring, there are two molecular conformations (triequatorial
and triaxial), which may be interconverted by ring inversion. Similarly in
the case of [B], there are two molecular conformations (equatorial-axial-
axial and axial-equatorial-equatorial), which may be interconverted by
ring inversion. It is clear that (either for an individual conformation or on
averaging over the ring-inversion processes discussed above), in the case
of [A], all methyl groups are equivalent, whereas, in the case of [B], there
are two different methyl group environments with populations in a 2:1 ratio.
The solid-state 13C NMR spectrum recorded for acetaldehyde/ferrierite
(Figure 3) and the solution-state 1H and 13C NMR spectra recorded for the
organic species extracted from the ferrierite indicate that all methyl groups
in the molecules of P have the same average environment, implying that
only isomer [A] is produced. For this isomer, the conformation (triequa-
torial) with all methyl groups in equatorial positions should be favored
energetically, at least for the isolated molecule, and furthermore, this
conformation of [A] should be lower in energy than any conformation of
isomer [B]. We note that, as each step in the proposed reaction mechanism
(Figure 5) from acetaldehyde to P can be regarded as an equilibrium, it is
reasonable to find a situation in which only the thermodynamically most
stable isomer of P is obtained (provided, of course, that the production of
this isomer is compatible with the spatial constraints imposed on the reaction
by the ferrierite host structure).
1
(25) If we denote the numbers of moles of acetaldehyde and P present
at any stage during the reaction as nA and nP, respectively, the integrated
intensity of the peaks due to P in the solid-state 13C NMR spectrum,
expressed as a fraction of the total integrated intensity of peaks in the
spectrum (acetaldehyde plus P), is: R ) (3nP)/(3nP + nA). In deriving this
equation, we recognize that, for a molecule of acetaldehyde, each peak in
the spectrum represents a single 13C nucleus, whereas, for a molecule of P,
each peak in the spectrum represents three equivalent 13C nuclei. Similar
arguments hold for integration of the solution-state 1H NMR spectrum.
Given the stoichiometry of the reaction (acetaldehyde f 1/3P), the number
of moles of acetaldehyde that have been consumed at any stage during the
reaction is equal to three times the number of moles of P that have been
produced. Thus, the percentage conversion is the following: C ) 100
(number of moles of acetaldehyde consumed)/(number of moles of
acetaldehyde present initially) ) 100(3nP)/(3nP + nA) ) 100R. Rearrange-
ment of this equation gives the molar ratio: nP/nA ) R/(3 - 3R).
Acknowledgment. Financial support from EPSRC (to
J.M.T.), HEFCE and EPSRC (to K.D.M.H.), and the University
of Birmingham and CVCP (award of a studentship to S.O.L.)
is gratefully acknowledged.
References and Notes
(1) John, C. S.; Clark, D. M.; Maxwell, I. E. In PerspectiVes in
Catalysis; Thomas, J. M., Zamaraev, K. I., Eds.; Blackwell: Oxford, U.K.,
1992; p 387.
(2) Van Bekkum, H., Flanigen, E. M., Jansen, J. C., Eds. Introduction
to Zeolite Science and Practice; Elsevier: Amsterdam, 1991.
(3) Thomas, J. M. Angew. Chem., Int. Ed. Engl. 1994, 33, 913.
(4) Thomas, J. M. Angew. Chem., Int. Ed. Engl. 1999, 38, 3588.