Enantioselective α-benzylation of aldehydes via photoredox organocatalysis

Article abstract of DOI:10.1021/ja106593m

The first enantioselective aldehyde α-benzylation using electron-deficient aryl and heteroaryl substrates has been accomplished. The productive merger of a chiral imidazolidinone organocatalyst and a commercially available iridium photoredox catalyst in the presence of household fluorescent light directly affords the desired homobenzylic stereogenicity in good to excellent yield and enantioselectivity. The utility of this methodology has been demonstrated via rapid access to an enantioenriched drug target for angiogenesis suppression.

Full text of DOI:10.1021/ja106593m

Published on Web 09/10/2010
Enantioselective r-Benzylation of Aldehydes via Photoredox Organocatalysis
Hui-Wen Shih, Mark N. Vander Wal, Rebecca L. Grange, and David W. C. MacMillan*
Merck Center for Catalysis at Princeton UniVersity, Princeton, New Jersey 08544
Received July 25, 2010; E-mail: dmacmill@princeton.edu
Abstract: The first enantioselective aldehyde R-benzylation using
electron-deficient aryl and heteroaryl substrates has been ac-
complished. The productive merger of a chiral imidazolidinone
organocatalyst and a commercially available iridium photoredox
catalyst in the presence of household fluorescent light directly
affords the desired homobenzylic stereogenicity in good to
excellent yield and enantioselectivity. The utility of this methodol-
ogy has been demonstrated via rapid access to an enantioen-
riched drug target for angiogenesis suppression.
The benzylic alkylation of chiral enolates has become a mainstay
transformation in chemical synthesis, mainly due to the seminal
research efforts of Evans, Oppolzer, Seebach, and Myers.1 Surpris-
ingly, however, catalytic enantioselective variants of this venerable
reaction remain confined to a small but valuable group of substrate
types. Phase transfer benzylation of glycine-derived imines,2 chiral
triamine ligation of ketone-derived lithium enolates,3 and the use
of Jacobsen’s Cr(salen) complex with preformed stannous enolates4
are among the most successful examples.5 Direct organocatalytic
methods to form R-benzylated aldehydes have also been reported
wherein the electrophilic partner requires multiple and/or electron-
rich aryl rings (to enable the intermediate formation of stabilized
benzylic carbocations).6 Recently, our laboratory introduced a new
mode of activation termed photoredox organocatalysis,7,8 the
mechanistic foundation of which relies on the propensity of
electrophilic radicals (derived from photocatalytic reduction of alkyl
halides) to combine with facially biased enamines (derived from
aldehydes and a chiral amine catalyst). Inspired by this strategy,
we hypothesized that the enantioselective R-benzylation of alde-
hydes might also be possible via the marriage of these activation
pathways. Herein, we present a new direct and enantioselective
catalytic protocol for the R-alkylation of aldehydes, a transformation
that is successful with a wide array of simple monoaryl9 and
monoheteroaryl methylene halides.
Figure 1. Proposed catalytic cycle for aldehyde R-benzylation.
dine), commonly used as a green triplet emitter in OLEDs, can
readily accept a photon from weak fluorescent light to generate
the strong reductant excited state fac-*Ir(ppy)3 (2) (E1/2 ) -1.73
V vs saturated calomel electrode (SCE) in CH3CN).10 We hoped
that single electron transfer (SET) from fac-*Ir(ppy)3 to a suitable
benzylic bromide might render an aromatic radical anion that would
rapidly undergo σ-bond cleavage to afford the bromide anion and
an electrophilic benzyl radical 3. Within the same time frame,
condensation of an aldehyde substrate with imidazolidinone catalyst
4 should form a highly π-nucleophilic enamine 5 which should
then combine with the electron-deficient radical 3 to enantioselec-
tively forge the crucial homobenzylic center. Rapid oxidation of
the resultant R-amino radical11 6 by fac-Ir(ppy)3+ (7) (E1/2 ) 0.77
V vs SCE in CH3CN)10 would then close the redox cycle while
regenerating the photocatalyst fac-Ir(ppy)3 (1). Moreover, subse-
quent hydrolysis of iminium 8 would reconstitute the organocatalyst
4 while liberating the enantioenriched R-benzyl aldehyde adduct.
Evaluation of this tandem catalysis strategy was first examined
with octanal, imidazolidinone catalyst 9, fac-Ir(ppy)3 (1), a 26 W
compact fluorescent lamp, and 2,4-dinitrobenzyl bromide, benzyl
bromide, or 4-(bromomethyl)pyridine as the alkylating reagent. As
Design Plan. As detailed in Figure 1, it has been established
that commercially available fac-Ir(ppy)3 (1) (ppy ) 2-phenylpyri-
9
13600 J. AM. CHEM. SOC. 2010, 132, 13600–13603
10.1021/ja106593m  2010 American Chemical Society

C O M M U N I C A T I O N S
Table 1. Evaluation of Amine Catalyst for Benzylation
Table 3. Asymmetric Aldehyde R-Benzylation: Bromide Scope
a Isolated yields. b Enantiomeric excess determined by chiral HPLC.
c 4-(Bromomethyl)pyridine added as the hydrobromic acid salt with an ad-
ditional equivalent of 2,6-lutidine.
shown in Table 1, initial experiments revealed that the proposed
benzylation was indeed possible; however, the coupling component
was restricted to the most electron-deficient dinitro-aryl system
(entry 2). Intriguingly, the use of benzyl bromide resulted in the
complete recovery of starting materials12 while 4-(bromomethyl)-
Table 2. Asymmetric Aldehyde R-Benzylation: Aldehyde Scope
a Stereochemistry assigned by chemical correlation or by analogy.
b Enantiomeric excess determined by chiral SFC or HPLC. c 30 mol%
organocatalyst used. d Performed at 15 °C using Ru(bpy)3Cl2 as the photoredox
catalyst; ref 14. e Substrate added as the hydrobromic acid salt with an
additional equivalent of 2,6-lutidine. The free base organocatalyst was used.
f
Yield determined by H NMR. g Ir(dF(CF3)ppy)2(dtbbpy)PF6 was employed
1
as the photoredox catalyst; ref 15. h Isolated yield of the corresponding alcohol.
pyridine rapidly decomposed under photoredox conditions (entries
1, 5). These data suggest that electron-neutral benzylic halides do
not readily undergo the initial reduction step, while the pyridinium
radical is formed but does not couple with the enamine derived
from amine 9. To address this latter limitation, we next examined
amine catalysts of diminished steric demand with the presumption
that a higher concentration and increased spatial exposure of the
enamine intermediate should increase the rate of the critical radical-
olefin addition step. Indeed, the pseudo-C2-symmetric catalyst 10
(wherein the tert-butyl substituent has been replaced by a methyl
group) does enable the pyridinium methylene bromide to become
a viable coupling partner, albeit with low selectivities (entry 6, 81%
yield, 78% ee). Enantiocontrol is regained, however, when the
benzyl, methyl disubstituted amine 4 is employed (entry 7, 90%
ee), a catalyst that likely operates via the preferred E-enamine
geometry with shielding of the π-nucleophilic Re-face by the
benzylic moiety on the catalyst framework (DFT-5).13 The superior
levels of scope, induction, and efficiency exhibited by the combina-
tion of catalysts 4 and fac-Ir(ppy)3 (1) in DMSO at 23 °C prompted
us to select these conditions for further exploration.
a Stereochemistry assigned by chemical correlation or by analogy.
b Enantiomeric excess determined by chiral SFC or HPLC. c Substrate
added as the hydrobromic acid salt with an additional equivalent of
2,6-lutidine. The free base organocatalyst was used. d Isolated yield of
the corresponding alcohol.
We next performed a series of experiments to determine the scope
of the aldehydic component in this asymmetric benzylation protocol.
9
J. AM. CHEM. SOC. VOL. 132, NO. 39, 2010 13601

C O M M U N I C A T I O N S
As revealed in Table 2, these mild redox conditions are compatible
with a wide range of functional groups, including ethers, amines,
imides, carbamates, and aromatic rings (72-91% yield, 87-90%
ee). Moreover, significant variation in the steric demand of the
aldehyde substituent can be accommodated without loss in enan-
tiocontrol (entries 4 and 5, g73% yield, 90% ee).
Figure 2. Fluorescence quenching by reaction substrates.
aryl and heteroaryl methylene bromides display efficient quenching
of the fac-*Ir(ppy)3 excited state, strongly suggesting that the
electron transfer event occurs directly between fac-*Ir(ppy)3 and
the aryl coupling partner. This is in accord with the catalytic cycle
set out in Figure 1.
In conclusion, the first enantioselective aldehyde R-benzylation
using electron-deficient aryl and heteroaryl substrates has been
accomplished. The productive merger of a novel imidazolidinone
organocatalyst and a commercially available iridium photoredox
catalyst directly allows the stereocontrolled formation of ho-
mobenzylic stereogenicity in good to excellent yield. This new
alkylation reaction, which exhibits broad scope and wide
functional group tolerance, has been successfully utilized for
the rapid and enantioselective construction of a previously
developed angiogenesis inhibitor in only three linear chemical
steps.
We have found that a broad range of electron-deficient aryl
and heteroaryl methylene bromides also participate in this
enantioselective benzylation reaction (Table 3). For example,
benzyl systems that incorporate a nitro-substituent with other
electron-withdrawing groups such as nitriles and esters (Table
3, entries 1-3) are well-tolerated. Moreover the 1,2-nitro-fluoro
aryl ring can serve to produce a suitably electrophilic radical
without the intervention of SNAr byproducts (entry 3). Perhaps
more notable with respect to medicinal agent synthesis, a large
range of heteroaryl rings can be successfully employed such as
pyridines, quinolines, benzimidazoles, pyrimidines, and triazines.
As highlighted in entries 4 and 5, bromomethyl pyridines that
have electron-donating substitution (entry 5, 2-methyl, 91% ee)
or electron-withdrawing groups (entry 4, 5-nitro, 90% ee) are
both competent in this enantioselective coupling. Moreover,
fused bicycles such as 4-quinolinyl and 2-benzimidazolyl also
perform well (81-90% yield, 82-88% ee, entries 6, 10). We
postulate that, for substrates with a basic nitrogen (4-pyridinyl,
and entries 5-6, 10),16 substrate protonation facilitates the initial
reduction step17 and thereafter enhances the electrophilicity of
the resulting radical intermediate. However, heterocycles con-
taining two nitrogens, such as 2-pyrazinyl and 4-pyrimidinyl,
and three nitrogens, such as 2-triazinylswhich all lack a basic
nitrogensalso react with good efficiency and excellent enanti-
oselectivity (68-78%, 87-91% ee, entries 7-9).18
Acknowledgment. Financial support was provided by NIHGMS
(R01 01 GM093213-01), and kind gifts from Amgen and Abbott.
Supporting Information Available: Experimental procedures,
structural proofs, and spectral data for all new compounds are provided.
This material is available free of charge via the Internet at http://
pubs.acs.org.
References
We fully expect that the R-heteroarylmethyl aldehyde products
generated in this study will be of value for the generation and testing
of medicinal agents. To highlight this possibility, we have applied
our new catalytic enantioselective benzylation strategy to the
synthesis of the angiogenesis inhibitor 12,19 a drug candidate that
was developed for the treatment of diseases such as diabetic
retinopathy and tumor proliferation. As illustrated above, exposure
of propionaldehyde and 4-(bromomethyl)pyridine to our photoredox
coupling protocol followed by in situ oxime formation yields the
R-methylene pyridyl intermediate in 82% yield and 93% ee.
Subsequent oxime reduction followed by urea coupling with amine
11 (available in two chemical steps) allows the rapid construction
of angiogenesis inhibitor 12 in only three linear steps (34% overall
yield, 93% ee).
In an effort to provide insight into the mechanistic details of
this tandem catalysis coupling, we have performed several fluo-
rescence quenching experiments (Stern-Volmer studies) with our
fac-Ir(ppy)3 photoredox catalyst (Figure 2). Notably, fac-*Ir(ppy)3
oxidation of enamine 5 is not observed and, as such, cannot be the
first step in the photocatalytic cycle (in contrast to previous studies
from our laboratory that centered upon Ru(bpy)3Cl2).7 Instead, both
(1) (a) Evans, D. A.; Ennis, M. D.; Mathre, D. J. J. Am. Chem. Soc. 1982,
104, 1737. (b) Myers, A. G.; Yang, B. H.; Chen, H.; Gleason, J. L. J. Am.
Chem. Soc. 1994, 116, 9361. (c) Seebach, D.; Wasmuth, D. HelV. Chim.
Acta 1980, 63, 197. (d) Oppolzer, W.; Moretti, R.; Thomi, S. Tetrahedron
Lett. 1989, 30, 5603.
(2) (a) Maruoka, K.; Ooi, T. Chem. ReV. 2003, 103, 3013. (b) Ooi, T.; Maruoka,
K. Angew. Chem., Int. Ed. 2007, 46, 4222.
(3) Imai, M.; Hagihara, A.; Kawasaki, H.; Manabe, K.; Koga, K. J. Am. Chem.
Soc. 1994, 116, 8829.
(4) (a) Doyle, A. G.; Jacobsen, E. N. J. Am. Chem. Soc. 2005, 127, 62. (b)
Doyle, A. G.; Jacobsen, E. N. Angew. Chem., Int. Ed. 2007, 46, 3701.
(5) An intramolecular enamine alkylation strategy has been demonstrated; see:
Vignola, N.; List, B. J. Am. Chem. Soc. 2004, 126, 450.
(6) (a) Shaikh, R. R.; Mazzanti, A.; Petrini, M.; Bartoli, G.; Melchiorre, P.
Angew. Chem., Int. Ed. 2008, 47, 8707. (b) Cozzi, P. G.; Benfatti, F.; Zoli,
L. Angew. Chem., Int. Ed. 2009, 48, 1313. (c) Brown, A. R.; Kuo, W.-H.;
Jacobsen, E. N. J. Am. Chem. Soc. 2010, 132, 9286.
(7) Nicewicz, D. A.; MacMillan, D. W. C. Science 2008, 322, 77.
(8) For recent examples of photoredox catalysis, see: (a) Ischay, M. A.;
Anzovino, M. E.; Du, J.; Yoon, T. P. J. Am. Chem. Soc. 2008, 130, 12886.
(b) Koike, T.; Akita, M. Chem. Lett. 2009, 38, 166. (c) Narayanam, J. M. R.;
Tucker, J. W.; Stephenson, C. R. J. J. Am. Chem. Soc. 2009, 131, 8756.
(d) Nagib, D. A.; Scott, M. E.; MacMillan, D. W. C. J. Am. Chem. Soc.
2009, 131, 10875. (e) Du, J.; Yoon, T. P. J. Am. Chem. Soc. 2009, 131,
14604. (f) Condie, A. G.; Gonza´lez-Go´mez, J. C.; Stephenson, C. R. J.
J. Am. Chem. Soc. 2010, 132, 1464. (g) Tucker, J. W.; Narayanam, J. M. R.;
Krabbe, S. W.; Stephenson, C. R. J. Org. Lett. 2010, 12, 368. (h) Tucker,
J. W.; Nguyen, J. D.; Narayanam, J. M. R.; Krabbe, S. W.; Stephenson,
C. R. J. Chem. Commun. 2010, 46, 4985. (i) Furst, L.; Matsuura, B. S.;
Narayanam, J. M. R.; Tucker, J. W.; Stephenson, C. R. J. Org. Lett. 2010,
9
13602 J. AM. CHEM. SOC. VOL. 132, NO. 39, 2010

C O M M U N I C A T I O N S
12, 3104. (j) Ischay, M. A.; Lu, Z.; Yoon, T. P. J. Am. Chem. Soc. 2010,
132, 8572.
(9) The racemic homolytic alkylation of stoichiometric enamines using
p-nitrobenzyl chloride as radical precursor is known; see: Russell, G. A.;
Wang, K. J. Org. Chem. 1991, 56, 3475.
(10) Flamigni, L.; Barbieri, A.; Sabatini, C.; Ventura, B.; Barigelletti, F. Top.
Curr. Chem. 2007, 281, 143.
(11) R-Amino radicals possess oxidation potentials in the range of -0.92
to -1.12 V vs SCE in CH3CN; see: Wayner, D. D. M.; Dannenberg, J. J.;
Griller, D. Chem. Phys. Lett. 1986, 131, 189.
(15) dF(CF3)ppy ) 2-(2,4-difluorophenyl)-5-trifluoromethylpyridine; dtbbpy )
4,4′-di-tert-butyl-2,2′-dipyridyl; see: Lowry, M. S.; Goldsmith, J. I.; Slinker,
J. D.; Rohl, R.; Pascal, R. A., Jr.; Malliaras, G. G.; Bernhard, S. Chem.
Mater. 2005, 17, 5712.
(16) These compounds have pKa values within 2 orders of magnitude of 2,6-
lutidine, and the reduction potentials of the unprotonated species are
too high to be reduced by the photocatalyst. For pKa values in DMSO,
see: Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456. For reduction
potentials, see: Wiberg, K. B.; Lewis, T. P. J. Am. Chem. Soc. 1970, 92,
7154.
(12) We recognized from the outset that this strategy might not be successful
with electron-neutral aryl systems due to the high reduction potential
of systems such as benzyl bromide (E1/2 ) -1.85 V vs SCE in CH3CN)
and the nucleophilic nature of the unsubstituted benzylic radical. For
benzyl bromide reduction potential, see: Koch, D. A.; Henne, B. J.;
Bartak, D. E. J. Electrochem. Soc. 1987, 134, 3062. For benzylic radical
nucleophilicity, see: De Vleeschouwer, F.; Van Speybroeck, V.; Waroquier,
M.; Geerlings, P.; De Proft, F. Org. Lett. 2007, 9, 2721.
(17) The´venot, D.; Buvet, R. J. Electroanal. Chem. 1972, 39, 429.
(18) The parent bromomethyl-heterocycles lacking additional substitution (e.g.,
2-(bromomethyl)pyrazine) are reported to be unstable; for pyrimidines, see:
(a) Brown, D. J.; Waring, P. Aust. J. Chem. 1974, 27, 2251. For pyrazines,
see: (b) Sasaki, K.; Sugou, K.; Miyamoto, K.; Hirai, J.; Tsubouchi, S.;
Miyasaka, H.; Itaya, A.; Kuroda, Y. Org. Biomol. Chem. 2004, 2, 2852.
For triazines, see: (c) Von Angerer, S. Sci. Synth. 2003, 17, 449.
(19) Matsuoka, H.; Nishimura, K.; Seike, H.; Aono, H.; Murai, M. Preparation
of pyridylalkylurea derivatives as angiogenesis inhibitors. U.S. Patent 2008/
0161270 A1, July 3, 2008.
(13) DFT calculations were performed with the use of B3LYP/6-311+G(2d,2p)//
B3LYP/6-31G(d).
(14) Reaction does proceed with Ir(III) catalyst, but with decreased yield.
Ru(bpy)3Cl2 is a competent photoredox catalyst for these transformations
but usually results in lower conversion; for photophysical properties, see:
Kalyanasundaram, K. Coord. Chem. ReV. 1982, 46, 159.
JA106593M
9
J. AM. CHEM. SOC. VOL. 132, NO. 39, 2010 13603