Experimental transition state for the B-chlorodiisopinocampheylborane (DIP-Cl) reduction

Article abstract of DOI:10.1016/j.tetlet.2009.03.216

Asymmetric reductions of prochiral ketones are important transformations in the syntheses of natural products, pharmaceuticals, and fine chemicals. B-Chlorodiisopinocampheylborane (DIP-Cl), a stoichiometric reagent that is capable of reducing prochiral aralkyl ketones with high selectivity. Here, we utilize a recently developed ¹³C kinetic isotope effect (KIE) methodology to probe the symmetry breaking process inherent to this asymmetric reduction. Experimental KIEs and computed transition structures indicate significant roles for non-bonding interaction, specific directed orbital interactions, and hydrogen tunneling in this reaction.

Full text of DOI:10.1016/j.tetlet.2009.03.216

Tetrahedron Letters 50 (2009) 3027–3030
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Experimental transition state for the B-chlorodiisopinocampheylborane
(
DIP-Cl) reduction
Sean E. Stafford, Matthew P. Meyer *
School of Natural Sciences, University of California, Merced, PO Box 2039, Merced, CA 95344, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Asymmetric reductions of prochiral ketones are important transformations in the syntheses of natural
products, pharmaceuticals, and fine chemicals. B-Chlorodiisopinocampheylborane (DIP-Cl), a stoichiom-
etric reagent that is capable of reducing prochiral aralkyl ketones with high selectivity. Here, we utilize a
Received 10 February 2009
Revised 29 March 2009
Accepted 31 March 2009
Available online 5 April 2009
13
recently developed C kinetic isotope effect (KIE) methodology to probe the symmetry breaking process
inherent to this asymmetric reduction. Experimental KIEs and computed transition structures indicate
significant roles for non-bonding interaction, specific directed orbital interactions, and hydrogen tunnel-
ing in this reaction.
Ó 2009 Elsevier Ltd. All rights reserved.
Stereoselective reductions of prochiral ketones are of
prime importance in the syntheses of natural products, pharma-
ceuticals, and fine chemicals. However, it is often difficult to
choose an appropriate reduction system, a priori. The B-chloro-
diisopinocampheylborane (DIP-Cl) reagent is an exemplary case,
where subtle changes in substrate can have profound effects upon
selectivity and reactivity.1 While isobutyrophenone is reduced in
reaction pathways. Qualitative differences for the two modes of
attack are evident in Figure 1.
Primary among the factors that determine stereoselection are
non-bonding interactions. Early seminal work illustrated a signifi-
6
cant role for steric interactions in directing stereoselection. From
this work came numerous qualitative models in which steric inter-
actions play key roles in determining transition structures. In spite
of the obvious importance of non-bonding interactions to stereose-
lection, few efforts have been made to develop quantitative models
of how these interactions influence the outcomes of stereoselective
reactions. While notable work on kinetic isotope effects (KIEs) that
0
high yield with good selectivity (90% ee), 4 -methylphenyl isopro-
pyl ketone is reduced in high yield with very high selectivity (>99%
2
0
0
ee). More surprisingly, in our hands, 2 ,5 -dimethylphenyl isopro-
pyl ketone is inert to reduction by DIP-Cl at 25 °C.
7
Numerous processes can erode selectivity in asymmetric reac-
tions: (1) racemization occurring after the selective step, (2) com-
petition from non-selective pathways, and (3) close competition
between diastereomeric transition states. Much like the chiral
reductant, B-3-Pinanyl-9-borabicyclo[3.3.1]nonane (Alpine-Bor-
derive from steric interactions has been performed, the systems in
which substantial and unequivocal steric KIEs have been reliably
measured are constrained to inversions involving aromatic com-
8
–10
pounds.
In fact, after a dearth of activity in this area, a recent,
elegant contribution has re-asserted the importance of non-bond-
1
1
ane), DIP-Cl utilizes
a-pinene as a chiral auxiliary. While low selec-
ing interactions in chemical reactions.
tivity in the Alpine–Borane reduction of prochiral ketones has been
shown to arise from a competing formation of the dehydroboration
3
,4
product, 9-BBN, it is likely that poorly selective DIP-Cl reductions
are the result of close competition between diastereomeric transi-
tion states. This can be surmised by the negligible reaction pro-
A
B
5
Cl
Cl
B
gress observed over a 24-h period when attempts at reducing
B
Ipc
O
Ipc
O
H
0
0
2
,5 -dimethylphenyl isopropyl ketone were made. To our knowl-
H
H
edge, no reports of side reactions have been made by other re-
search groups investigating DIP-Cl reductions. As a means of
understanding the interplay of structure and energetics in the
competition between diastereomeric transition states, we have
computed transition structures for both the favored and unfavored
H
*
Corresponding author. Tel.: +1 209 228 2982.
Figure 1. Qualitative models for (A) favored (Re) and (B) unfavored (Si) approach in
E-mail address: mmeyer@ucmerced.edu (M.P. Meyer).
the DIP-Cl reduction.
0
040-4039/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2009.03.216

3
028
S. E. Stafford, M. P. Meyer / Tetrahedron Letters 50 (2009) 3027–3030
The purpose of this Letter is to explore the factors that lead to a
significant preference for the favored diastereomeric transition
structure (Fig. 1A) over the unfavored diastereomer shown in
Figure 1B. The simplest explanation is that steric occlusion be-
tween the proximal methyl group on the reactive isopinocampheyl
group and the aromatic ring preclude Re attack. In large part, this
simple view is borne out in recent 2H KIE experiments designed
1
2
to probe the role of non-bonding interactions in stereoselection.
Here, we attempt to understand the combined roles of orbital
interaction and non-bonding interactions as directing forces in
the transition state for the DIP-Cl reduction.
Figure 3. Computed transition structures [B3LYP/6-31G(d)] for the (A) favored Re
attack and (B) disfavored Si attack of (ꢁ)-DIP-Cl. One isopinocampheyl group has
been eliminated for clarity.
Following the same guiding principle as in earlier work, we
have endeavored to utilize 13C KIEs to gain insight into the origins
1
3
of both selectivity and reactivity in the DIP-Cl reduction. The
method employed here is capable of resolving individual KIEs upon
both of the enantiotopic methyl groups on the isopropyl substitu-
ent (Fig. 2). This approach is designed to gain information regard-
ing the fundamental symmetry breaking process that occurs when
a reactant possessing one or more symmetry elements is exposed
to an asymmetric reactant, catalyst, or reactant–catalyst complex.
As a means of interpreting these experimental measurements, we
have computed transition structures for both the favored (Re
attack) and disfavored (Si attack) reaction channels (Fig. 3).
The KIE results shown in Figure 2 were acquired in a method
similar to that utilized in a recent study of the Corey–Bakshi–Shi-
ceeds cleanly and with very high selectivity (>99% ee). Likewise,
stock ketone was also desymmetrized using the CBS reduction.
Reduction using (S)-Me–CBS and BH –THF was used for desym-
3
metrization in preference to the DIP-Cl reduction because it is
more easily worked up and resulting products are more easily puri-
fied. The resonances of the pro-R and pro-S methyl groups were as-
1
4
signed using NMR predictions using the CSGT methodology and
1
5
the IGAIM variation upon a fully optimized B3LYP/6-31G(d)
model of the anticipated R enantiomer of the benzylic alcohol
product.
bata (CBS) reaction. Fractionation of naturally occurring 13C in 4 -
1
3
0
Among the results shown in Figure 2, the KIEs at the carbonyl
and prochiral methyl groups lend insight into the determinants
of reactivity and selectivity in the DIP-Cl reduction. First, the KIE
upon the carbonyl carbon will be discussed. The experimentally
determined KIE measurements at the carbonyl have an average va-
lue of 1.030. This measurement is in accordance with what might
be expected in a reaction where the carbonyl carbon experiences
a net reduction in bond order. The net reduction of bond order at
the transition state is reflective of an early transition state where
the bond order of the nascent C–H bond does not compensate
for the reduction of bond order in the carbonyl.
methylphenyl isopropyl ketone is measured by comparing quanti-
1
3
tative C NMR spectra of remaining reactant taken from a high
conversion (ꢀ85–95%) reaction and stock ketone reactant. Before
1
3
quantitative C NMR analysis, however, we desymmetrize the
re-isolated reactant and stock ketone using a highly stereoselective
reduction. This process converts the enantiotopic methyl groups to
diastereotopic groups, each having a distinct 1 C resonance. In the
3
0
experiments performed here, four (ꢁ)-DIP-Cl reductions of 4 -
methylphenyl isopropyl ketone were taken to 82.5%, 90.8%,
9
1.9%, and 93.2% conversion. The reactions were quenched using
NaOH/H to ensure conversion of all products to the free alcohol
form. The unreacted ketone was re-isolated using flash chromatog-
raphy and desymmetrized using the CBS reduction, which pro-
2
O
2
We utilized two commonly used barrier shapes to compute a
one-dimensional tunnel correction: the truncated parabolic bar-
1
6
17
rier and the Eckart barrier. Estimations of the free energy of
reaction were accomplished using empirical estimates of the rate
constant for this reaction (see Supplementary data). This value
was used to compute the barrier heights for the truncated parabola
and Eckart barriers. The concave downward curvature of the barri-
ers was computed using the reduced mass and frequency for the
imaginary frequency associated with the unstable mode in the
computed transition structure. Typically, inclusion of the effects
of tunneling upon the computed KIE results in greater agreement
between measured and computed values. Here, the opposite is ob-
1
1
1
1
.002(3) 1.000(4)
.002(3) 1.000(4)
.001(3) 1.000(3)
.005(2) 1.000(4)
A
1.000(4)
1.000(2)
1
.002(3)
.000(3)
O
1
1
.000
H
(
assumed)
1.005(5)
1
.005(3)
.010(3)
pro-R
pro-S
1
18
served. While tunneling can reasonably be expected to contrib-
0
.998(5)
.998(3)
.999(3)
.001(3)
1.012(3)
0
ute to the KIE in this reaction, given the substantial imaginary
1
1
1
.036(4)
0
1
1.002(4)
ꢁ1
.024(3)
.029(2)
frequency of 776i cm , the overestimation of tunneling suggests
1.001(2)
1
1
.001(2)
.003(4)
two possible reasons for disagreements between experimental
and computed estimates of the KIE at the carbonyl position. One
possibility is that the basis set employed does not provide enough
coverage upon and near the transferred hydride, which could arti-
ficially inflate the imaginary frequency. Artificial inflation of the
imaginary frequency would result in spurious overestimations of
the effect of tunneling upon the KIE. Another possibility is that
the first order saddle point upon the potential energy surface is
not an appropriate representation of the transition state. The
inherently quantum nature of hydride transfer may distort the po-
sition and orientation of the dividing surface that separates the
1
1
1
1
.001(4) 1.031(3)
.001(2)
.000(2)
.000(3)
B
1.028 (w/o tunneling)
1.043 (1-D Eckart)
O
1
.040 (1-D Bell)
1.000
H
0
.999
0.999
pro-R 0.997
pro-S 0.996
0
.999
0.997
0.999
19
reactant and product portions of the potential energy surface.
Figure 2. (A) Experimental 13C KIEs (k12C/k13C) for the (ꢁ)-DIP-Cl reduction of 4 -
0
This effect has been noted recently in various gas phase reactions
that involve simultaneous and substantial hydrogenic and heavy
atom motion.
methylphenyl isopropyl ketone. (B) Computed KIEs based on the [B3LYP/6-31G(d)]
transition structure for Re attack.
2
0

S. E. Stafford, M. P. Meyer / Tetrahedron Letters 50 (2009) 3027–3030
3029
Measurements of the KIEs at the prochiral methyl groups that
Table 1
Key lengths and angles in the TS for Re and Si attack
reside upon the reactant ketone suggest that non-bonding interac-
tions in the transition states of asymmetric reactions are more
complex than simple steric repulsion. The pro-R methyl group on
the isopropyl moiety displays an isotope effect that is greater than
unity. It is puzzling that a KIE that might reasonably be expected to
arise from steric interaction would be greater than unity. In fact,
non-bonding interactions, in general, are more complicated. Both
attractive and repulsive forces are present in non-bonding interac-
Structural element
Re attack (preferred)
Si attack (disfavored)
C(@O)ꢂ ꢂ ꢂH
C@O
1.3412 Å
1.3245 Å
1.3663 Å
1.4782 Å
1.7719 Å
111.89°
776i
1.2979 Å
1.3375 Å
1.3969 Å
1.4672 Å
1.7736 Å
116.92°
659i
C
B–O
Ipcꢂ ꢂ ꢂH
C
Ipcꢂ ꢂ ꢂB
\
S L
R –C(@O)–R
–
m
13
tions. It is somewhat reassuring that the C KIEs measured here
2
correspond with H KIEs measured upon the enantiotopic methyl
2
groups. In an earlier report, we measured an average H KIE upon
2
the pro-S group as 0.972. The average of H KIE values measured
isopropyl ketone, key structural elements (Table 1) suggest the ori-
gins of stereoselection in this system. These quantitative estimates
agree in essence with the qualitative transition structures shown in
Figure 1. Not surprisingly, the transition structure for Si attack is
later, with greater transfer of the hydride to the carbonyl. A sur-
prising structural characteristic is the advanced distortion of the
for the pro-R group was 0.990. As in the 2H KIEs, the smaller
12
1
3
C KIE is observed at the pro-S position. This similar trend sug-
gests that repulsive forces are stronger for the pro-S position. How-
ever, it appears from the measured 13C KIEs presented here, that
more than conventional steric repulsion is operative.
2
Secondary KIEs, especially those which are essentially decou-
pled from the reaction coordinate, are largely indicators in fre-
quency shifts in going from the reactant to the transition state.
bond angle at the formerly sp carbonyl carbon. The transition
structure for favored Re attack shows significant distortion of this
angle toward what might be expected in the product alcohol. The
modest disagreement between computed and measured KIE values
at the carbonyl may be, in part, due to the exaggerated deforma-
tion of the carbonyl. Another interesting, although expected obser-
vation is the attenuated magnitude of the imaginary frequency for
disfavored Si attack. The closer approach between the acceptor
(C@O) and donor (CIpc) carbons implies a more adiabatic transfer,
leading to an attenuated negative force constant. In fact, the com-
puted transition structures show an attenuated negative force con-
stant for Si attack relative to that computed for favored Re attack.
Ultimately, the simplest explanation for this difference in transi-
tion structures can be attributed to the greater steric repulsion be-
tween the aromatic group on the ketone and the proximal methyl
upon the isopinocampheyl group from which the hydride transfers.
A greater degree of bond formation is necessary to compensate for
the increased steric repulsion.
2
1,22
Several studies have been performed upon the IR
Raman
and
2
3,24
frequency shifts that occur with applied pressure in
ground state molecules. Nearly universally, C–H bonds become
red-shifted as pressure is applied until sufficient pressure develops
such that repulsive forces override attractive intermolecular forces.
Although examples of frequency shifts in C–C single bond stretches
are fewer in number, the associated frequency is blue-shifted upon
the application of pressure.2
3,24
Contrarily, heavy atom bonds with
, the C@C bond in
CN are red-shifted over a signif-
significant dipoles, such as the C–F bond in CFCl
CH CCl , and the C„N bond in CH
icant portion of the pressure range explored.
some triple- and double-bonded C–C stretches experience red-
3
2
2
3
2
1,22
Furthermore,
shifts upon the application of pressure.2
1,22
Infrared and Raman
spectra of ground state molecules under pressure are perhaps not
perfectly suited to comparisons with transition states experiencing
steric occlusion. However, the same inter- and intramolecular
forces are present and can be expected to play a role in determin-
ing changes in vibrational frequencies. The data mentioned above
suggest that dipolar and polarizable bonds can experience reduc-
tions in frequency upon close contact with a polarizable solvent.
Perhaps a similar situation prevails at the transition state. An in-
crease in polarizability is expected to be a general feature of tran-
sition states. In reactions like the DIP-Cl reduction of ketones, the
creation of partial charges and resulting dipoles can be expected.
1
3
In conclusion, we have measured C KIEs for the (ꢁ)-DIP-Cl
0
reduction of 4 -methylphenyl isopropyl ketone and computed a
model transition structure that yields insights into the process of
stereoselection in this reaction. We have also presented an isotope
1
3
effect methodology by which C KIEs may be measured for each
individual enantiotopic group in reactions where symmetry break-
ing makes the groups inequivalent. Finally, the results of our
experimental and computational work have suggested that, in
addition to steric repulsion, other non-bonding interactions such
as dispersion forces and inductive interactions may be important
in mediating stereoselection.
2
The difficulty arises in explaining the simultaneous inverse H KIEs
1
3
that occur at both prochiral methyl groups and the normal
C
KIE that occurs at the pro-R methyl group. At least three scenarios
are possible: First, dispersion forces attenuate as the inverse sixth
power of the separation distance, while repulsive forces are well
modeled by inverse twelfth power interactions. It is possible that
dispersion forces act upon the entire methyl group but the repul-
sive interactions only extend to the hydrogen atoms on the methyl.
However, repulsive forces upon the hydrogen atoms should trans-
late into inverse C KIEs, if the predominant effect of steric inter-
action is upon the C–H(D) stretch. A second possibility is that steric
interactions affect the vibrational manifolds of methyl groups lar-
gely via blue shifts in asymmetric bending and stretching motions
that include little motion of the carbon atom. Finally, it may be that
Acknowledgments
We thank the University of California for funding and Mike
Colvin (UC Merced) for the use of his Linux cluster.
Supplementary data
1
3
Supplementary data (experimental procedures, NMR integra-
tion results, computational procedures, and energies and geome-
tries of all calculated structures) associated with this article can
be found, in the online version, at doi:10.1016/j.tetlet.2009.03.216.
3
the C–CH bond is red-shifted as the result of inductive effects aris-
ing from charge and dipolar localization at the reaction center. This
scenario might be likely to affect the carbon atom more due to the
relative proximity of the methyl carbon to the carbonyl undergoing
reduction.
References and notes
1. Chandrasekharan, J.; Ramachandran, P. V.; Brown, H. C. J. Org. Chem. 1985, 50,
5
445–5450.
Brown, H. C.; Chandrasekharan, J.; Ramachandran, P. V. J. Am. Chem. Soc. 1988,
10, 1539–1546.
3. Midland, M. M.; Petre, J. E.; Zderic, S. A. J. Organomet. Chem. 1979, 182, C53–C56.
2.
In computing transition structures for both the favored Re at-
1
0
tack and disfavored Si attack of (ꢁ)-DIP-Cl upon 4 -methylphenyl

3
030
S. E. Stafford, M. P. Meyer / Tetrahedron Letters 50 (2009) 3027–3030
4
.
.
.
Midland, M. M.; Petre, J. E.; Zderic, S. A.; Kazubski, A. J. Am. Chem. Soc. 1982, 104,
28–531.
Ramachandran, P. V.; Gong, B.; Brown, H. C.; Francisco, J. S. Tetrahedron Lett.
004, 45, 2603–2605.
(a) Cornforth, J. W.; Cornforth, R. H.; Mathew, K. K. J. Chem. Soc. 1959, 112–127;
b) Cram, D. J.; Elhafez, F. A. A. J. Am. Chem. Soc. 1952, 74, 5828–5835; (c)
13. Saavedra, J.; Stafford, S. E.; Meyer, M. P. Tetrahedron Lett. 2009, 50, 1324–1327.
5
14. Keith, T. A.; Bader, R. F. W. Chem. Phys. Lett. 1992, 194, 1–8.
15. Keith, T. A.; Bader, R. F. W. Chem. Phys. Lett. 1993, 210, 223–231.
16. Bell, R. P. The Tunnel Effect in Chemistry; Chapman and Hall: New York,
1980.
5
6
2
(
17. Eckart, C. Phys. Rev. 1930, 35, 1303–1309.
Karabatsos, G. J. J. Am. Chem. Soc. 1967, 89, 1367–1371; (d) Cherest, M.; Felkin,
H.; Prudent, N. Tetrahedron Lett. 1968, 9, 2199–2204; (e) Brown, H. C.; Dickason,
W. C. J. Am. Chem. Soc. 1970, 92, 709–710; (f) Zimmerman, H. E.; Traxler, M. D. J.
Am. Chem. Soc. 1957, 79, 1920–1923.
18. Meyer, M. P.; Delmonte, A. J.; Singleton, D. A. J. Am. Chem. Soc. 1999, 121,
10865–10874.
19. González-Lafont, A.; Villà, J.; Lluch, J. M.; Bertrán, J.; Steckler, R.; Truhlar, D. G. J.
Phys. Chem. A 1998, 102, 3420–3428.
7.
8.
9.
Carter, R. E.; Melander, L. Adv. Phys. Org. Chem. 1973, 10, 1–27.
Melander, L.; Carter, R. E. J. Am. Chem. Soc. 1964, 86, 295–296.
Mislow, K.; Graeve, R.; Gordon, A. J.; Wahl, G. H., Jr. J. Am. Chem. Soc. 1964, 86,
20. (a) Marinkovic, M.; Gruber-Stadler, M.; Nicovich, J. M.; Soller, R.; Mülhäuser,
M.; Wine, P. H.; Bache-Andreassen, L.; Nielsen, C. J. J. Phys. Chem. A 2008, 112,
12416–12429; (b) Gruber-Stadler, M.; Mülhäuser, M.; Sellevåg, S. R.; Nielsen, C.
J. J. Phys. Chem. A 2008, 112, 9–22.
1
733–1741.
1
1
0. Carter, R. E.; Dahlgren, L. Acta Chem. Scand. 1969, 23, 504–514.
1. Hayama, T.; Baldridge, K. K.; Wu, Y.-T.; Linden, A.; Siegel, J. S. J. Am. Chem. Soc.
21. Benson, A. M., Jr.; Drickamer, H. G. J. Chem. Phys. 1957, 27, 1164–1174.
22. Wiederkehr, R. R.; Drickamer, H. G. J. Chem. Phys. 1958, 28, 311–316.
23. Schindler, W.; Jonas, J. J. Chem. Phys. 1980, 73, 3547–3552.
24. Lee, M.-R.; Ben-Amotz, D. J. Chem. Phys. 1993, 99, 10074–10077.
2
008, 130, 1583–1591.
1
2. West, J. D.; Stafford, S. E.; Meyer, M. P. J. Am. Chem. Soc. 2008, 130, 7817–7818.