48
P. Nandi et al. / Journal of Catalysis 284 (2011) 42–49
enantioselectivity sensitively depends on the calix[4]arene inher-
ent chirality when using -chloroacetophenone and changes from
10% for 12 to 19% for 13 in Table 5. The observed enantioselectivity
with -chloroacetophenone in this experiment is in stark contrast
MPV catalysis rate and ketone and alkoxide juxtapositioning at
the active site is consistent with trends observed in intramolecular
reaction systems [2], as well as comparative studies of mature and
germ-line catalytic antibodies for reactions requiring the bringing
together of two reactants in the transition state [5]. The repercus-
sions of two-point versus one-point ketone binding on enantiose-
lective MPV catalysis are also investigated using inherently chiral
calix[4]arenes. Increasing denticity of ketone reactant binding is
observed to increase the sensitivity of the interplay between inher-
ently direction of inherent chirality in the calixarene ligand and
MPV reaction enantioselectivity.
a
a
to the lack of enantioselectivity for the same ketone in the control
experiments described previously. Altogether, these results
emphasize the importance of cooperativity between normal chiral-
ity arising from the alcohol and inherent chirality arising from the
calixarene scaffold in these experiments. Cooperativity between
normal and inherent chirality was previously shown to be critical
for asymmetric induction by Matt and coworkers, for alkylation
and hydrogenation reactions [39]. MPV enantioselectivity is also
sensitive to calix[4]arene inherent chirality when using ortho-
chloroacetophenone and changes from 4% for 12 to 12% for 13 in
Table 5. The observed several-fold rate increase for the Cl-contain-
ing ketones over F-containing ketone in Table 5 is consistent with
the expected tighter binding of the former, though this cannot be
rigorously deconvoluted from electronic and steric factors, which
also contribute to the rate difference. Results shown in Table 5
demonstrate acetophenone to be an unselective reactant from
the standpoint of enantioselectivity. As such, it is insensitive to
chirality within catalysts 12 and 13 and provides no useful infor-
mation regarding the correlation between rigidity of ketone bind-
ing and sensitivity of enantioselectivity to the particular catalyst
diastereomer used. The results in Table 5 demonstrate that the
sensitivity of the interplay between MPV enantioselectivity and
chirality in the ligand increase upon increasingly rigid reactant
juxtapositioning. They also demonstrate the absolute enantioselec-
tivity need not increase upon increasingly rigid reactant juxtaposi-
4. Experimental procedure
General procedure for conducting MPV reduction: In 5 mL of
anhydrous toluene (freshly distilled over sodium benzophenone
ketyl), 1,3-bis-substituted calix[4]arene (e.g., for 1,3-bispropoxy
calix[4]arene, 73 mg, 0.1 mmol, 10 mol%) is dissolved. To this solu-
tion, Me3Al (40 lL, 0.1 mmol, 10 mol%, 25% w/w solution in hex-
ane) is added under Ar. Gaseous methane bubbles indicate the
formation of methylaluminum calix[4]arene complex, which upon
treatment with alcohol forms the active MPV catalyst and evolves
methane. A 5 mL solution of alcohol (e.g., 300
4 equiv. anhydrous 2-propanol) and substrate ketone (e.g., 155 mg,
1 mmol, 1 equiv. -chloroacetophenone) in toluene is added. The
lL, 240 mg, 4 mmol,
a
reaction is subsequently performed at room temperature and is
monitored by 1H NMR spectroscopy.
Acknowledgments
tioning (i.e., comparison of entries for a-chloroacetophenone and
ortho-chloroacetophenone with ortho-fluorobenzophenone in
Table 5).
The authors are indebted to the Office of Basic Energy Sciences
at the US Department of Energy (DE-FG02-05ER15696) and to the
US National Science Foundation (CHE-0840505) for financial sup-
port. The authors are grateful to Mr. Jarred Ghilarducci for assis-
tance with GC experiments, Dr. Andrew Solovyov for helpful
discussions, and the Berkeley Center for Green Chemistry (BGGC).
3. Conclusion
Catalyst and reactant structural requirements for MPV reduc-
tion have been elucidated in a system involving AlIII–calix[4]arene
complexes as active sites. The calix[4]arene provides an oxo envi-
ronment that covalently binds to the aluminum center in a biden-
tate configuration and acts as an organizational scaffold at the
active site, which permits the design and synthetic tuning of
catalyst structure. The large kinetic isotope effect measured with
calix[4]arene catalysts suggests hydride transfer to be the rate-
limiting step in the reaction. A comparative approach is used for
deconvoluting electronic and steric effects at the active site for
MPV reduction catalysis. Reactant electronic effects are understood
on the basis of increasing the electrophilicity of the carbonyl car-
bon of the ketone, which results in a better hydride acceptor, as
well as increasing hydride donor ability of the bound alkoxide de-
rived from the alcohol via hyperconjugative stabilization. In addi-
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
References
[1] A. Radzicka, R. Wolfenden, Science 267 (1995) 90.
[2] (a) A.J. Kirby, Adv. Phys. Org. Chem. 29 (1994) 87;
(b) D.E. Storm, D.E. Koshland Jr., Proc. Natl. Acad. Sci. USA 66 (1970) 445.
[3] T.C. Bruice, F.C. Lightstone, Acc. Chem. Res. 32 (1999) 127.
[4] D.-H. Lee, R.H. Crabtree, S.-K. Park, Bull. Korean Chem. Soc. 23 (2002) 1157.
[5] (a) F.E. Romesberg, B. Spiller, P.G. Schultz, R.C. Stevens, Science 279 (1998)
1929;
(b) X. Zhang, Q. Deng, S.H. Yoo, K.N. Houk, J. Org. Chem. 67 (2002) 9043;
(c) A. Heine, E.A. Stura, J.T. Yli-Kauhaluoma, C. Gao, Q. Deng, B.R. Beno, K.N.
Houk, K.D. Janda, I.A. Wilson, Science 279 (1998) 1934;
(d) J. Zimmermannn, F.E. Romesberg, C.L. Brooks, I.F. Thorpe, J. Phys. Chem. B
114 (2010) 7359;
(e) A. Piatesi, D. Hilvert, ChemBioChem 5 (2004) 460;
(f) D. Hilvert, K.W. Will, K.D. Nared, M.-T.M. Auditor, J. Am. Chem. Soc. 111
(1989) 9261;
tion, there is
a 40% increase in MPV catalysis rate upon
increasing the Lewis acidity of the metal center via changing ca-
lix[4]arene upper-rim substituents from tert-butyl to less electron
donating H. Steric considerations accounting for the effect of a
cavity formed by calix[4]arene lower-rim substituents result in a
factor of two lower reaction rate when using either sterically
demanding calix[4]arene substituents or bulky reactants. This
result suggests formation of a shape-selective cavity that is defined
by lower-rim substituents.
The most crucial feature that affects MPV reaction rate is the
two-point versus one-point binding of ketone reactant to the Lewis
acid site. This effect alone leads to a 7.7-fold higher rate during
MPV catalysis favoring the more rigid two-point binding ketone.
The denticity of ketone binding is independently investigated
using DFT-based calculations. The above correlation between
(g) L.T. Chong, Y. Duan, L. Wang, I. Massova, P.A. Kollman, Proc. Natl. Acad. Sci.
USA 96 (1999) 14330.
ˇ
[6] S. Marti, J. Andres, V. Moliner, E. Silla, I. Tunón, J. Bertrán, Chem. Eur. J. 14
(2008) 596.
[7] (a) N. de Silva, S.J. Hwang, K.A. Durkin, A. Katz, Chem. Mater. 21 (2009) 1852;
(b) J.M. Notestein, E. Iglesia, A. Katz, J. Am. Chem. Soc. 126 (2004) 16478.
[8] C.R. Graves, H. Zhou, C.L. Stern, S.T. Nguyen, J. Org. Chem. 72 (2007) 9121.
[9] (a) For recent reviews on MPV reduction see: C.F. de Graauw, J.A. Peters, H. van
Bekkum, J. Huskens, Synthesis (1994) 1007;
(b) J.S. Cha, Org. Process Res. Dev. 10 (2006) 1032;
(c) C.R. Graves, J.E. Campbell, S.T. Nguyen, Tetrahedron: Asymmetry 16 (2005)
3460;