Catalysis of the Kemp elimination by natural coals

Article abstract of DOI:10.1021/ol000137u

(matrix presented) Commercially available coals were found to be efficient heterogeneous catalysts of the Kemp elimination reaction in aqueous solutions. A pH-rate profile study suggests that catalysis originates from specific catalytic groups and not sim

Full text of DOI:10.1021/ol000137u

ORGANIC
LETTERS
2
000
Vol. 2, No. 23
747-3750
Catalysis of the Kemp Elimination by
Natural Coals
3
Hagit Shulman and Ehud Keinan*,†,‡,§
†
Department of Chemistry and Institute of Catalysis Science and Technology,
Technion-Israel Institute of Technology, Technion City, Haifa 32000, Israel,
and Department of Molecular Biology and The Skaggs Institute for Chemical Biology,
The Scripps Research Institute, 10550 North Torrey Pines Road,
La Jolla, California 92037
keinan@tx.technion.ac.il
Received May 30, 2000 (Revised Manuscript Received September 27, 2000)
ABSTRACT
Commercially available coals were found to be efficient heterogeneous catalysts of the Kemp elimination reaction in aqueous solutions. A
pH−rate profile study suggests that catalysis originates from specific catalytic groups and not simply from the large graphitic surface area.
The low-quality lignite coals, which exhibit similar catalytic efficiency per weight to that of molecularly imprinted polymers, are better catalysts
for this reaction in comparison with the bituminous coals.
The proton-transfer reaction, which is a key transformation
in chemistry and biology, represents an essential step in
because it is a simple, one-step transformation, which is
highly sensitive to the medium. Efficient catalysis of this
reaction by nonenzymatic systems could shed light on the
importance of several catalytic factors that operate in concert
within the preorganized environment of an enzyme active
1
almost any biocatalyzed process. This step can become rate
determining, especially when the proton is transferred to or
from carbon. A well-studied model for proton transfer from
2
4
a carbon atom is the Kemp elimination reaction. This
site. The recently reported catalytic systems for this reaction
5
exothermic, concerted E
2
elimination of benzisoxazoles to
include host molecules with rigid, preorganized clefts, serum
6
7
8
salicylonitriles (e.g., 1 to 2) is highly sensitive to the base
strength. In the case of carboxylate bases, the reaction rate
is strongly influenced by solvent polarity, primarily through
albumins, polymeric amines, imprinted polymers, surfac-
9
tant aggregates, and specifically designed catalytic antibod-
1
0
ies.
3
desolvation-activation of the catalytic base. This reaction
Naturally occurring coals are versatile heterogeneous
catalysts of several gas-phase reactions as well as of a few
has recently become a prime target for enzyme mimics
1
1
reactions carried out in solution. The reported examples
include various substitution, isomerization, and redox reac-
tions. For example, the coal-catalyzed ligand substitution
†
Technion-Israel Institute of Technology.
The Scripps Research Institute.
Incumbent of the Benno Gitter & Ilana Ben-Ami chair of Biotechnol-
‡
§
12
ogy, Technion.
reactions with cobalt complexes and solvolysis of phosphate
(1) (a) Kuliopulos, A.; Talalay, P.; Mildvan, A. S. Biochemistry 1990,
2
9, 10271. (b) Rose, I. A.; Warms, J. V. B.; Kuo, D. J. Biochemistry 1992,
1, 9993. (c) Gerlt, J. A.; Gassman, P. G. Biochemistry 1993, 32, 11943.
(4) (a) Na, J.; Houk, K. N.; Hilvert, D. J. Am. Chem. Soc. 1996, 118,
6462. (b) Kirby, A. J. Angew Chem., Intl. Ed. Engl. 1996, 35, 707.
(5) (a) Kennan, A. J.; Whitlock, H. W. J. Am. Chem. Soc. 1996, 118,
3027. (b) Hannak, R. B.; Rojas, C. M. Tetrahedron Lett. 1998, 39, 3465.
(6) (a) Kikuchi, K.; Thorn, S. N.; Hilvert, D. J. Am. Chem. Soc. 1996,
118, 8184. (b) Hollfelder, F.; Kirby, A. J.; Tawfik, D. S. Nature 1996,
383, 60. (c) Hollfelder, F.; Kirby, A. J.; Tawfik, D. S.; Kikuchi, K.; Hilvert,
D. J. Am. Chem. Soc. 2000, 122, 1022.
3
(d) Guthrie, J. P.; Kluger, R. J. Am. Chem. Soc. 1993, 115, 11569. (e)
Glickman, M. H.; Wiseman, J. S.; Klinman, J. P. J. Am. Chem. Soc. 1994,
1
16, 793.
(2) Casey, M. L.; Kemp, D. S.; Paul, K. G.; Cox, D. D. J. Org. Chem.
1
973, 38, 2294.
(
3) (a) Kemp, D. S.; Casey, M. L. J. Am. Chem. Soc. 1973, 95, 6670.
b) Kemp, D. S.; Cox.; D. D.; Paul, K. G. J. Am. Chem. Soc. 1975, 97,
312.
(
(7) Hollfelder, F.; Kirby, A. J.; Tawfik, D. S. J. Am. Chem. Soc. 1997,
119, 9578.
7
1
0.1021/ol000137u CCC: $19.00 © 2000 American Chemical Society
Published on Web 10/21/2000

esters13 were attributed to the acidic and basic functional
groups on the carbon surface. Conversely, the racemization
of 1,1′-binaphthyl and its derivatives was found to be
catalyzed by the graphitic surface and not by specific
functional groups.14
In general, the coal catalytic activities stem from three
main features: (a) a graphitic structure that is characterized
by a highly porous hydrophobic surface area with a strong
binding affinity toward aromatic molecules; (b) abundance
of various functional groups, mostly containing oxygen,
nitrogen, and sulfur, e.g., carboxylic acids, phenols, amines,
various sulfur functional groups, and heteroaromatic rings;
and (c) electronic conductivity. We reasoned that these
features would render naturally occurring coals effective
catalysts of the Kemp elimination reaction.
the reciprocal of catalyst concentration. From this relationship
site
one can measure the Michaelis Menten parameters
K . Accordingly, we carried out the reaction with variable
M
kcat and
16
amounts of catalyst and used the linear correlation of a
double reciprocal plot (of kobs vs the catalyst weight) in order
site
to estimate the apparent catalytic constant
The apparent K
kcat (Table 1).
M
values in this heterogeneous system could
not be determined in our case because the relationship
between the catalyst weight and the active site concentration
could not be estimated.
As expected for a heterogeneous catalyst, the catalytic
efficiency was found to increase with increased coal surface
1
7
area (Figure 1). Probably a less expected finding is the
Here we show that commercially available coals indeed
catalyze this reaction when mixed with 1 in a buffered
solution and that catalysis originates from specific catalytic
groups and not simply from the large graphitic surface area.
Three German lignites from three major mining areas in
Germany (coals 1613, 1650, and 1551) and three bituminous
coals (RSA, AUS, COL) were examined as potential catalysts
15
of the Kemp elimination reaction. All reactions were carried
out by shaking a mixture of the coal powder and substrate 1
(0.5-2.0 mM) in phosphate buffer (50 mM, pH 7.4, 0.5 mL)
containing acetonitrile (2% v/v) at 4 °C, and the progress of
the reaction was monitored by HPLC. All coal samples were
shaken in a buffer solution at room temperature for 2 days
to reach equilibrium with solvent before using them as
catalysts.
To compare the catalytic effect of these coal samples, we
carried out a series of reactions using constant amounts of
coal (10 mg) and varying concentrations of substrate 1.
Assuming that under these conditions there was a large
excess of catalytic sites with respect to the substrate, we
expected that the reaction would follow pseudo-first-order
kinetics. Accordingly, we obtained the observed rate con-
stants (kobs) from the linear regression of the experimental
rates as a function of substrate concentration (Table 1).
Figure 1. Dependence of catalytic efficiency on the surface area
of different coals.
apparently linear correlation between the kinetic data and
the carbon content of the catalyst (Figure 2). We found that
coals with decreased abundance of carbon atoms exhibit
(
(
8) Liu, X. C.; Mosbach, K. Macromol. Rapid Commun. 1998, 19, 671.
9) Perez-Juste, J.; Hollfelder, F.; Kirby, A. J.; Engberts, J. B. F. N. Org.
Lett. 2000, 2, 127.
10) (a) Thorn, S. N.; Daniels, R. Q.; Auditor, M. T. M.; Hilvert, D.
(
Nature 1995, 373, 228. (b) Genre-Grandpierre, A.; Tellier, C.; Loriat, M.-
J.; Blanchard, D.; Hodgson, D. R. W.; Hollfelder, F.; Kirby, A. J. Bioorg.
Med. Chem. Lett. 1997, 7, 2497. (c) Sergeeva, M. V.; Tomtova, V.;
Parkinson, A.; Overgaauw, M.; Pomp, P.; Schots, A.; Kirby, A. J.; Hilhorst,
R. Israel J. Chem. 1996, 36, 177.
Table 1. Observed Catalytic Parameters for the Different Coal
Samplesa
coal
kobs (min-1)
sitekcat (min-1
)
(11) Spiro, M. Catal. Today 1990, 7, 167.
(
(
12) Tomita, A.; Tamai, Y. J. Colloid Interface Sci. 1971, 36, 153.
13) Yumiko, Y.; Pincock, R. E. J. Can. J. Chem. 1980, 58, 134.
1
1
1
650
551
613
0.0045
0.0041
0.0033
0.0015
0.0014
0.0009
0.82
0.10
(14) (a) Pincock, R. E.; Johnson, W. M.; Wilson, K. R.; Haywood-
Farmer, J. J. Am. Chem. Soc. 1973, 95, 6477. (b) Pincock, R. E.; Johnson,
W. M.; Haywood-Farmer, J. Can. J. Chem. 1976, 54, 548. (c) Hutchins, L.
G.; Pincock, R. E. J. Org. Chem. 1980, 45, 2414.
AUS
COL
RSA
0.0016
0.0039
(15) Lignites: coal 1613, source Lower Lusatia, East Germany, open-
cast mine, Nochten; coal 1650, source Central Germany (Halle/Leipzig),
open-cast mine, Profen; coal 1551, source Rhine/Ruhr (Rhenish brown coal),
open-cast mine, Hambach. Bituminous coals: RSA, South Africa, Goodhope
mine, Angloamerican’s coal F-6486; AUS, Stewarton-lilyvale mine, BHP-
Utha coal Ltd.; COL, El Cerrejon mine, Carbocol #4903/0985.
a
The reactions were carried out using a shaker, and the progress of the
reaction was monitored by HPLC equipped with a reverse phase, Supelcosil
C18 column. Calculation of kobs was corrected for the buffer-catalyzed
reaction. Calculation of kcat was carried out according to Klotz. The
correlation coefficients (R ) for the pseudo-first-order rate plots were
between 0.96 and 0.99.
site
16
(16) Suh. J.; Scarpa, I. S.; Klotz, I. M.; J. Am. Chem. Soc. 1976, 98,
2
7060.
2
(17) Coal: surface area (m /g), elemental analysis: 1613, 91.8, 62.3 (%
C), 0.64 (% N), 24.8 (% O), 4.7 (% H); 1551, 80.2, 64.2 (% C), 0.78 (%
N), 25.2 (% O), 4.5 (% H); 1650, 161.3, 59.9 (% C), 0.52 (% N), 18.5 (%
O), 4.8 (% H); AUS, 5.8, 75.7 (% C), 4.3 (% H), 0.6 (%S); COL, 10.7,
It has been shown that under conditions of excess catalyst
67.5 (% C), 4. 31 (% H), 0.95 (% S); RSA, 3.15, 76.4 (% C), 4.0 (% H),
a linear relationship exists between the reciprocal of kobs and
1.1 (% S).
3748
Org. Lett., Vol. 2, No. 23, 2000

significant. This observation, together with the sigmoidal
shape of the pH-rate profile for both catalysts, suggests that
catalysis stems from specific catalytic groups and not simply
from the large graphitic surface area. Unlike common
titration curves of homogeneous acids and bases, which are
characterized by steep sigmoid turning points, here we
observe a very mild turning point at a pH range between
5
.5 and 8.5. This observation probably reflects cumulative
and average effects of various catalytic groups, each pos-
sessing a different pK value. A very similar curve that
envelops several pK values (with visible turning points at
a
a
pH 6.2 and 8.3) was observed when alkylated polyethylene
imines were used to catalyze the Kemp elimination reaction.7
By contrast, catalytic antibodies with carboxylate groups in
1
0a
their active site exhibit pK
a
values of 5.5-6.0, and BSA,
whose basic functional group is an amine, exhibits a distinct
6
b
a
pK value of 8.95.
Figure 2. Dependence of catalytic efficiency on the carbon content
of different coals.
We reasoned that adsorption of a hydrophobic salt onto
the coal surface could enhance its catalytic efficiency, in
analogy to the role played by cofactors in enzymatic catalysis.
Thus, coals 1551 and COL (10 mg) were mixed with the
appropriate buffer solution (0.49 mL) containing varying
amounts of methylquinolinium iodide, 3, and the mixture
increased catalytic efficiency. Assuming that the basic
catalytic groups which are relevant to this reaction include
carboxylate anions, phenols, amines, sulfur functionalities,
and heteroaromatic rings, one would expect that decreased
carbon/hetroatom proportions reflect an increased abundance
of such catalytic sites. This may explain why the lignite coals,
which are characterized by a larger surface area and a higher
abundance of heteroatoms in comparison with the bituminous
coals, are better catalysts for this reaction.
was shaken at room temperature for 2 days. The reaction
rates with these pretreated catalysts were measured as
described above. Indeed, the coal that was impregnated with
methylquinolinium iodide was found to be a better catalyst
than the coal itself (Figure 4). This effect can be attributed
The pH-rate profile study of the catalytic Kemp elimina-
tion reaction was carried out at a pH range between 4.5 and
8.5 using the lignite 1551 and the bituminous coal COL
(Figure 3). Although the catalytic superiority of the lignite
over the bituminous coal is preserved across the entire pH
range, at higher pH values this superiority becomes more
Figure 4. Dependence of catalytic efficiency of coals 1551 (b)
and COL (9) on added methylquinolinium iodide.
Figure 3. pH-rate profile of the reactions catalyzed by 1551 (2),
COL (9), and the buffer (b): coals (10 mg) were mixed with the
appropriate buffer solution (0.49 mL) for 2 days at room temper-
ature. An acetonitrile solution (10 uL) of substrate 1 (final
concentration: 2.56 mM) was added and the mixture was shaken
at 4 °C. The progress of the reaction was followed by HPLC.
to the increase of a general base concentration on the surface
of the catalyst and the stabilization of the negatively charged
transition state by the hydrophobic methylquinolinium
Org. Lett., Vol. 2, No. 23, 2000
3749

7
cation. As shown in Figure 4, the additional catalytic effect
reaches saturation at high concentrations of 3 and is more
pronounced with coal 1551 than with COL. While coal 1551
reaches saturation only when the concentration of 3 exceeds
0.05 M, COL is already saturated at 0.007 M. This difference
in the adsorbance capacity correlates well with the difference
2
in the surface area of these two coals (80 and 11 m /g,
respectively). The cocatalytic effect of 3 with coal 1551 is
observed only beyond the concentration of 0.007 M, probably
reflecting partial inhibition of the coal catalytic groups.
It is interesting to compare the efficiency of our coal
catalysts with molecularly imprinted polymers (MIP) because
both are heterogeneous catalysts. Although the work of Liu
8
and Mosbach refers to a the nonsubstituted substrate, 4,
which is less reactive than 1, it appears that both catalysts
are comparable in terms of their catalytic efficiency per
weight (Figure 5).
In conclusion, we have shown here that commercially
available coals are efficient heterogeneous catalysts of the
Kemp elimination reaction in aqueous mixtures. The pH-
rate profile suggests that catalysis originates from specific
Figure 5. Comparison of the catalytic efficiency of coal 1551 (left)
and MIP2 (data taken from ref 8). The reaction was carried out
under the conditions used for MIP2 and substrate 4: coal 1551
a
catalytic groups, with pK values ranging from 5.5 to 8.5,
and not simply from the large graphitic surface area.
Interestingly, the coals of lower commercial value are the
more efficient catalysts. Thus, the lignite coals, which are
characterized by both larger surface area and higher abun-
dance of heteroatoms in comparison with the bituminous
coals, are better catalysts for this reaction. The efficiency of
these catalysts per weight is comparable with that of a
heterogeneous MIP.
(
10 mg) in ethanol-water (1:3, v/v, 8 mL) with substrate 1 (3.075
mM).
it is necessary to use specific catalysts that mimic the
enzymatic precision. Specific positioning of the general base
at the appropriate distance and relative orientation with
respect to the substrate is required for more effective
catalysis. Indeed, high catalytic efficiency (rate acceleration
8
All previously reported catalysts of the Kemp elimination
reaction combine a general base with a hydrophobic environ-
ment. Further rate enhancement can be achieved by stabiliza-
tion of the delocalized, negatively charged transition state.
This work suggests that the various coals possess generic
hydrophobic pockets with basic functional groups, as is the
of 3.4 × 10 ) has been achieved with the appropriately
designed catalytic antibody 34E4, which represents an
encouraging step toward ideal enzyme mimics.1
0a
Acknowledgment is made to Dr. Haim Cohen and Vered
Nehemia of Ben Gurion University for the coal samples and
their physical data. We thank the Israel Science Foundation
and the Skaggs Institute for Chemical Biology for financial
support.
6
7
case with serum albumins, alkylated polyethylene imine,
imprinted polymers,8 and surfactant aggregates. These
9
nonspecific catalysts can achieve remarkable rate enhance-
4
ments of up to 10 . Yet, to reach higher rate accelerations,
OL000137U
3750
Org. Lett., Vol. 2, No. 23, 2000