A mechanistic probe for asymmetric reactions: Deuterium isotope effects at enantiotopic groups

Article abstract of DOI:10.1021/ja8026899

The development of a mechanistic probe that is especially suitable for the study of asymmetric reactions is presented. Chemically innocuous enantiotopic methyl groups are utilized as probes for the distinct environments that develop at the transition state for the (-)-B-chlorodiisopinocampheylborane reduction of 4′-methylisobutyrophenone. ²H kinetic isotope effects (KIEs) are determined for both enantiotopic methyl groups using two types of competition reactions. One competition is that between the d3-methyl enantiomeric isotopomers. The other competition reaction is that between the d6-dimethyl and perprotiated isotopologues. The rate constant ratios can be converted into kinetic isotope effects upon each of the individual enantiotopic methyl groups by invoking the rule of the geometric mean. The resulting isotope effect measurements yield highly precise values and contribute further understanding to the transition structure for this stereoselective reduction. The results are discussed in the context of steric isotope effects and the origins of these effects, which arise from the impact of steric crowding upon the anharmonicity of C-H bonds in the transition structure relative to the reactant state. Copyright

Full text of DOI:10.1021/ja8026899

Published on Web 06/03/2008
A Mechanistic Probe for Asymmetric Reactions: Deuterium Isotope Effects at
Enantiotopic Groups
Jason D. West, Sean E. Stafford, and Matthew P. Meyer*
UniVersity of California, Merced, School of Natural Sciences, P.O. Box 2039, Merced, California 95344
Received April 11, 2008; E-mail: matt.meyer@gmail.com
The rational development of stereoselective reagents and catalysts
remains an elusive goal. While significant progress has been made
using largely empirical approaches, quantitative measurements are
necessary to develop predictive transition structure models. Two
primary directing forces are often credited for stereoselectivity: (1)
orbital geometry and (2) steric occlusion. The underlying physical
phenomena responsible for these “forces” are well-understood.
However, experiments capable of quantitatively assessing sym-
metry-breaking interactions at the transition state have yet to be
developed as generally applicable tools. This paper describes a
methodology that uses 2H kinetic isotope effect (KIE) measurements
upon “chemically innocuous” enantiotopic groups as a probe for
symmetry breaking in the transition structure. In asymmetric
reactions, enantiotopic groups become diastereotopic in the transi-
tion state, resulting in distinct average environments for each group.
KIEs result from changes in structure and bonding that occur as
the molecule of interest progresses along the reaction coordinate
Figure 1. The two competition reactions by which 2H KIEs upon
from the reactant to the transition state of the rate-limiting step.
The effects of steric occlusion1 and dispersion interactions2 upon
vibrational force constants are well-known. As a result, the
measurements described here are the first steps in understanding
the forces responsible for symmetry breaking in the transition state.
Stereoselective reductions of asymmetric ketones are of immense
importance, especially in pharmaceutical synthesis.3–8 While all
commonly used reduction systems exhibit exemplary reactivity and
selectivity with select substrates, all reduction systems have a class
or classes of ketones that prove intractable.9,10 Reductions using
B-chlorodiisopinocampheyl-borane (DIP-Cl) have shown outstand-
ing selectivity, especially in the reductions of aralkyl ketones and
R-tertiary dialkyl ketones. However, subtle changes in substrate
structure can seriously impact the effectiveness of DIP-Cl as a
reductant. In fact, 2′,5′-dimethylisobutyrophenone is inert to DIP-
Cl at room temperature; whereas, 4′-methylisobutyrophenone (1)
is cleanly reduced to the S-enantiomer of the reduction product
with >99% ee by (–)-DIP-Cl under the same conditions. These
results are in contrast to the (–)-DIP-Cl reduction of isobutyrophe-
none, which predominately yields the S-enantiomer with lower
selectivity (90% ee).11
Kinetic isotope effect (KIE) studies are exquisitely sensitive
probes of the transition structure.12,13 Empirical determinations of
KIEs offer a means of evaluating computational or heuristic models
of the transition structure. Enantiotopic groups are equivalent by
symmetry in the reactant. Asymmetric reactions convert these
positions into diastereotopic groups in the product-determining step.
As a consequence, the measurement of KIEs at enantiotopic groups
in asymmetric reactions is a natural choice for investigating the
critical symmetry breaking event that occurs during the product-
determining step. The enantiotopic groups utilized in our ongoing
investigations of stereoselective reductions are chemically
innocuoussmeaning that changes in isotopic substitution at these
positions have little effect upon the imaginary frequency associated
enantiotopic methyl groups are determined.
with the reaction coordinate at the transition state. While this aspect
of experimental design could be relaxed to investigate neighboring
group phenomena such as hydrogen bond formation or Lewis acid/
Lewis base interactions, the primary motivation of the current
research is to ascertain the structural and energetic importance of
steric occlusion and dispersion forces. Furthermore, ensuring that
the isotopic substitution on the enantiotopic groups does not affect
the imaginary frequency, makes the assumption of the rule of the
geometric mean more credible (see eq. 2).14
The method for determining 2H KIEs on the enantiotopic methyl
groups in 4′-methylisobutyrophenone relies on determining the
relative rates in two distinct competition reactions. The two
competition reactions performed in these experiments are shown
in Figure 1 panels A and B. Figure 1A illustrates a competition
between isotopomers in the racemically d3-labeled substrates, where
R is the ratio of S-3 to R-3 in reisolated ketone. While this
competition reaction yields the relative isotope effects at each
enantiotopic methyl group (eq 1), the competition reaction shown
in Figure 1B yields the product of the isotope effects at the
enantiotopic methyl groups, assuming the rule of the geometric
mean15 (eq 2). In eq 2, R is the ratio of 2 to 1 in ketone reisolated
from high conversion reactions, and R0 is the ratio of 2 to 1 in the
stock ketone used as starting material in the high conversion
reactions. In eqs 1 and 2, F is the overall fractional conversion.
Using eqs 3 and 4, the absolute KIEs at both the pro-R and pro-S
methyl group can be computed from the numerical values of the
relative rates computed from the competition experiments in Figure
1A,B.
kS-D3
kH6 kH6
ln[2(1 - F) ⁄ (1 + 1 ⁄ R)]
ln[2(1 - F) ⁄ (1 + R)]
)
)
(1)
kR-D3
k
R-D3 kS-D3
9
7816 J. AM. CHEM. SOC. 2008, 130, 7816–7817
10.1021/ja8026899 CCC: $40.75
2008 American Chemical Society

C O M M U N I C A T I O N S
kH6
kH6
kH6
ln[(1 + 1 ⁄ R0)(1 - F) ⁄ (1 + 1 ⁄ R)]
ln[(1 + R0)(1 - F) ⁄ (1 + R)]
the pro-R group in the transition state. In the context of the
qualitative transition structure adapted from Brown,11 it is possible
that antiperiplanar donation of the methine C-H bond into the
nascent C-H bond formed by the attacking hydride orients the
prochiral groups such that the pro-S methyl group comes into strong
incidence with the nearby methyl on the isopinocampheyl group.
While such an explanation is satisfying in the context of the familiar
principles of molecular orbital theory and the semiclassical origins
of steric isotope effects, dispersion forces could contribute com-
pensatory normal contributions to the KIE. While this effect is
probably small and may seem counterintuitive in the context of a
condensed phase transition structure, alkanes under moderate
pressure exhibit red-shifts in C-H bond stretches.18 Likewise, the
transfer of cyclohexane from the gas phase to neat liquid lowers
the frequency of both the axial and equatorial C-H stretches by
)
×
)
(2)
(3)
(4)
kD6 kS-D3 kR-D3
kH6
kS-D3 kH6
)
×
kR-D3
k
kD6
ꢀ
R-D3
kH6
kH6 kS-D3
)
(
)(
)
kS-D3
k
kR-D3
ꢀ
D6
The competition reactions between enantiomeric isotopomers
shown in Figure 1A were taken to 78.1, 84.8, and 80.3% conversion.
The unreacted ketone was reisolated using flash chromatography
and desymmetrized using the CBS reduction, which proceeds
cleanly and with very high selectivity (>99% ee).10 This desym-
metrization was chosen in lieu of using DIP-Cl itself because the
CBS reaction is more easily worked up. The desymmetrization
allows for quantification of the relative amounts of S-3 and R-3 in
more than 10 cm-1 19
In fact, isotope-dependent dispersion forces
.
are thought to be responsible for reverse-phase HPLC separations
of the perdeuterated and perprotiated isotopologues of several
molecules.20 Obviously, dispersion forces can affect force constants
and result in a normal (>1) contribution to observed KIEs. What
remains to be quantitatively determined is the relative importance
of steric interactions and dispersion forces.
We have illustrated a simple and powerful new method for
probing symmetry breaking in the transition states of stereoselective
reactions. Computational work is underway to explain the origins
of the isotope effects observed here and in other systems. A new
method is also under development to estimate the relative impor-
tance of dispersion forces and steric occlusion in these measurements.
1
the reisolated starting material by H NMR. The doublets of each
methyl group are easily resolved in the resulting spectrum performed
in CD2Cl2. While these groups relax quickly, a T1 calibration is
performed to ensure that the resonances are fully relaxed between
transients. The assignment of the resonance to each individual
methyl group is performed using chemical shifts computed by the
CSGT16 methodology and the IGAIM17 variation upon a fully
optimized B3LYP/6-31+G(d,p) model of the anticipated11
R
enantiomer of the benzylic alcohol product. NMR chemical shift
predictions showed that the pro-S position is downfield of the pro-R
position in the 1H NMR spectrum, while it is upfield of the pro-R
position in the 13C NMR spectrum. We acquired a HMQC spectrum
upon the product to determine that the diastereotopic positions did,
in fact, switch relative positions in the 1H and 13C spectra as a
partial check on the accuracy of the NMR predictions.
Acknowledgment. We thank the University of California for
funding, Steve Minter for help with Matlab, and Mike Colvin for
the use of his Linux cluster.
The competition reactions between 1 and 2 shown in Figure 1B
were taken to 87.9, 88.0, and 90.6% conversion. The unreacted
ketone was reisolated as above. Reisolated reactant from these
Supporting Information Available: Detailed experimental proce-
dures, derivation of eqs 1 and 2, and tables of integrations from
quantitative NMR measurements.This material is available free of
charge via the Internet at http://pubs.acs.org.
1
reactions was analyzed using H NMR to determine the relative
amounts of 1 and 2. The methine resonance was used to account
for total ketone present, since its resonance at 3.50 ppm is well-
resolved and unlikely to overlap with any contaminants. The methyl
resonance at 1.18 ppm was used to account for 1. These measure-
ments allowed for an accurate estimate of the ratio of 1 to 2 in the
reisolated ketone, thus yielding R. The same measurement in stock
ketone yields R0.
References
(1) Carter, R. E.; Melander, L. AdV. Phys. Org. Chem. 1973, 10, 1–27.
(2) Schindler, W.; Jonas, J. J. Chem. Phys. 1980, 73, 3547–52.
(3) Farina, V.; Reeves, J. T.; Senanayake, C. H.; Song, J. J. Chem. ReV. 2006,
106, 2734–93.
(4) van Bergen, M.; Gais, H.-J. J. Am. Chem. Soc. 2002, 124, 4321–8.
(5) King, A. O.; Corley, E. G.; Anderson, R. K.; Larsen, R. D.; Verhoeven,
T. R.; Reider, P. J. J. Org. Chem. 1993, 58, 3731–5.
(6) Ramachandran, P. V.; Gong, B.; Brown, H. C. Tetrahedron: Asymmetry
1993, 4, 2399–400.
(7) Thompson, A. S.; Tschaen, D.m.; Simpson, P.; McSwine, D. J.; Reamer,
R. A.; Verhoeven, T. R.; Shinkai, I. J. Org. Chem. 1992, 57, 7044–52.
(8) DeNinno, M. P.; Schoenleber, R.; Asin, K. E.; MacKenzie, R.; Kebabian,
J. W. J. Med. Chem. 1990, 33, 2950–52.
(9) Brown, H. C.; Park, W. S.; Cho, B. T.; Ramachandran, P. V. J. Org. Chem.
1987, 52, 5406–12.
(10) Corey, E. J.; Helal, C. J. Angew. Chem., Int. Ed. 1998, 37, 1986–2012.
(11) Brown, H. C.; Chandrasekharan, J.; Ramachandran, P. V. J. Am. Chem.
Soc. 1988, 110, 1539–46.
The results from the competition reactions are shown in Figure
2. The competition reactions directly yield kS-D3/kR-D3 and kH6/kD6
.
The KIEs upon individual prochiral groups are obtained using eqs
3 and 4. Repulsive steric interactions cause an increase in force
constant, thus further splitting the zero-point energy differences
between C-D and C-H in the transition state, ultimately leading
to an inverse (<1) KIE.1 This appears to explain the inverse nature
of the overall KIE (kH6/kD6 < 1). Similar reasoning suggests that
the pro-S group experiences greater repulsive steric interaction than
(12) Thomas, A. A.; Singleton, D. A. J. Am. Chem. Soc. 1995, 117, 9357.
(13) Li, L.; Lo, M.; Ghanem, M.; Taylor, E. A.; Schramm, V. L. Biochemistry
2008, 47, 2577.
(14) Amin, M.; Price, R. C.; Saunders, W. H., Jr J. Am. Chem. Soc. 1988, 110,
4085.
(15) Bigeleisen, J. J. Chem. Phys. 1955, 23, 2264–67.
(16) Keith, T. A.; Bader, R. F. W. Chem. Phys. Lett. 1992, 194, 1–8.
(17) Keith, T. A.; Bader, R. F. W. Chem. Phys. Lett. 1993, 210, 223–31.
(18) Lee, M.-R.; Ben-Amotz, D. J. Chem. Phys. 1993, 99, 10074.
(19) Remar, G. J.; MacPhail, R. A. J. Chem. Phys. 1995, 103, 4381.
(20) Turowski, M.; Yamakawa, N.; Meller, J.; Kimata, K.; Ikegami, T.; Hosoya,
K.; Tanaka, N.; Thornton, E. R. J. Am. Chem. Soc. 2003, 125, 13836–49.
Figure 2. Relative rate constants computed from competition reactions
and resulting KIEs on enantiotopic methyl groups.
JA8026899
9
J. AM. CHEM. SOC. VOL. 130, NO. 25, 2008 7817