Copper-catalyzed synthesis of enantioenriched tetraarylethanes

Article abstract of DOI:10.1021/ol7027543

Herein we report the asymmetric synthesis of 1,2-dipyridyl-1,2- diarylethanes via an unusual Cu(l)-catalyzed dimerization reaction. Subjection of a variety of enantioenriched substituted 2-pyridyl alcohols to a one-pot protocol generates the desired products in good yields and diastereoselectivities and with ee's up to >99%.

Full text of DOI:10.1021/ol7027543

ORGANIC
LETTERS
2008
Vol. 10, No. 5
741-744
Copper-Catalyzed Synthesis of
Enantioenriched Tetraarylethanes
Wendy S. Jen,* Matthew D. Truppo, Deborah Amos, Paul Devine,
Michael McNevin, Mirlinda Biba, and Kevin R. Campos
Department of Process Research, Merck Research Laboratories, P.O. Box 2000,
Rahway, New Jersey 07065
wendy_jen@merck.com
Received November 14, 2007
ABSTRACT
Herein we report the asymmetric synthesis of 1,2-dipyridyl-1,2-diarylethanes via an unusual Cu(I)-catalyzed dimerization reaction. Subjection
of a variety of enantioenriched substituted 2-pyridyl alcohols to a one-pot protocol generates the desired products in good yields and
diastereoselectivities and with ee’s up to >99%.
Significant advances have been made in the field of metal-
catalyzed couplings of activated and unactivated C-sp3 alkyl
halides in recent years.1 Several notable communications
detailing the metal-mediated cross-couplings of primary alkyl
and benzylic C-sp3 electrophiles in Negishi, Kumada, Stille,
Hiyama, and Suzuki-Miyaura reactions have significantly
advanced the field.1,2 Although efforts have been focused
largely on reactions involving primary C-sp3 electrophiles,
there have also been noteworthy achievements utilizing
secondary alkyl halides. Fu has demonstrated that secondary
C-sp3 electrophiles can be used in nickel-mediated cross-
coupling reactions with alkyl Zn and Si nucleophiles;3 of
particular note is the example of asymmetric Negishi cross-
couplings of secondary R-bromoamides.3a Despite these
seminal contributions, to our knowledge, there are no
examples in the literature of metal-catalyzed couplings of
C-sp3 secondary alkyl groups to generate vicinal C-sp3 chiral
centers in a stereodefined manner (eq 1). Herein we report
the enantioselective synthesis of 1,1,2,2-tetraarylethanes (1
and 2) which employs an unusual copper-catalyzed dimer-
ization of enantioenriched aryl-heteroaromatic secondary
phosphonates. To our knowledge, this is the first reported
method for generating these potentially interesting medicinal
and ligand scaffolds in an enantioselective manner.4
During the course of our studies toward the synthesis of
a medicinal target, we began studying the feasibility of
displacing dibenzylic phosphonate ester 3 with isopropyl
magnesium chloride to generate 4 (eq 2). To our surprise,
subjection of racemic 3 to 5 mol % of CuCN and i-PrMgCl
at room temperature produced none of the desired product;
(1) For a recent review, see: Frisch, A. C.; Beller, M. Angew. Chem.,
Int. Ed. 2005, 44, 674-688.
(2) Benzylic phosphonates/halides: (a) Molander, G. A.; Elia, M. D. J.
Org. Chem. 2006, 71, 9198-9202. (b) Dohle, W.; Lindsay, D. M.; Knochel,
P. Org. Lett. 2001, 3, 2871-2873. (c) McLaughlin, M. Org. Lett. 2005, 7,
4875-4878. (d) Kofink, C. C.; Knochel, P. Org. Lett. 2006, 8, 4121-
4124. (e) Nobre, S.; Monteiro, A. L. Tetrahedron Lett. 2004, 45, 8225-
8228. Allylic phosphonates: (f) Yanagisawa, A.; Nomura, N.; Yamamoto,
H. Synlett 1993, 689-690. (g) Yanagisawa, A.; Nomura, N.; Yamamoto,
H. Tetrahedron 1994, 50, 6017-6028. Alkyl halides: (h) Ohmiya, H.;
Wakabayashi, K.; Yorimitsu, H.; Oshima, K. Tetrahedron 2006, 62, 2207-
2213. (i) Netherton, M. R.; Dai, C.; Neuschutz, K.; Fu, G. C. J. Am. Chem.
Soc. 2001, 123, 10099-10100.
(3) (a) Fischer, C.; Fu, G. C. J. Am. Chem. Soc. 2005, 127, 4594-4595.
(b) Powell, D. A.; Fu, G. C. J. Am. Chem. Soc. 2004, 126, 7788-7789. (c)
Zhou, J.; Fu, G. C. J. Am. Chem. Soc. 2003, 125, 14726-14727. (d)
Bonzalex-Bobes, F.; Fu, G. C. J. Am. Chem. Soc. 2006, 128, 5360-5361.
(e) Strotman, N. A.; Sommer, S.; Fu, G. C. Angew. Chem., Int. Ed. 2007,
46, 3556-3558.
10.1021/ol7027543 CCC: $40.75
© 2008 American Chemical Society
Published on Web 02/02/2008

instead dipyridyl diarylethane 5 was isolated as a mixture
of diastereomers (83:17 5a:5b) in 73% yield. Of particular
note was the unexpectedly high preference for the formation
of the C2-symmetric diastereomer 5a over the meso-5b
isomer.
examined (entries 4-8), although none were as effective as
i-PrMgCl. An improved yield and modest improvement in
dr was observed when MTBE was used as a solvent instead
of THF (entry 10, 84% yield). Alternative copper sources
were examined (entries 11 and 12) and found to be effective
catalysts for the dimerization; however, CuCN consistently
provided superior results. It is interesting to note that both
Cu(I) or Cu(II) salts provided competent catalysts (entry 11);
however, when the reaction was performed in the absence
of copper (entry 13), no product was observed, thus high-
lighting the importance of the presence of catalytic copper
in this reaction.
With optimized conditions in hand, we sought to examine
the effect of the activating group on the dimerization process.
As Table 2 illustrates, substrates containing a variety of
Table 2. Study of Various Leaving Groups
Given the unexpected nature of this product, a variety of
reaction conditions were examined to optimize its yield and
diastereoselectivity (Table 1). The reactions were typically
% yield
dr
entry
X
(C2 + meso)
(C2:meso)
Table 1. Reaction Optimization
1
2
3
4
5
6
OP(O)(OEt)2 (3)
Cl (6)
Br (7)
OMs (8)
OTs (9)
84
83
43
28
0
88:12
88:12
62:38
78:22
OCOCF3 (10)
0
% yield
(C2 + meso) (C2:meso)
dra
leaving groups were prepared and assessed with respect to
their ability to participate in the dimerization. Chloride 6
provided a nearly equivalent result to that observed with
phosphonate 3. Although bromide 7 and mesylate 8 provided
the desired product, the yields were significantly diminished,
largely due to the formation of side products. Tosylate 9 and
trifluoroacetate 10 both failed to generate 5; the former
reaction produced a complex mixture of products, and the
latter hydrolyzed quickly to the alcohol under reaction
conditions. Although both chloride 6 and phosphonate 3 were
viable substrates, the phosphonate ester was chosen for the
study of the asymmetric dimerization due to its ease of
synthesis in optically enriched form.
entry
CuX/solvent
base/temp
1
2
CuCN/THF
CuCN/THF
CuCN/THF
CuCN/THF
CuCN/THF
CuCN/THF
CuCN/THF
CuCN/THF
CuCN/Et2O
CuCN/MTBE
CuI/MTBE
i-PrMgCl, rt
73
74
0
83:17
85:15
i-PrMgCl, 0 °C
i-PrMgCl, -40 °C
PhMgCl, 0 °C
MeMgCl, 0 °C
t-BuMgCl, 0 °C
MeLi or PhLi, 0 °C
Et2Zn, 0 °C
3
4
<5
11
28
0
5
85:15
85:15
6
7
8
0
9
i-PrMgCl, 0 °C
i-PrMgCl, 0 °C
i-PrMgCl, 0 °C
78
84
79
60
0
87:13
88:12
87:13
85:15
10
11
12
13
Cu(OTf)2/MTBE i-PrMgCl, 0 °C
no Cu/MTBE i-PrMgCl, 0 °C
a Diastereomeric ratio was determined by HPLC.
Having developed an optimized method for the synthesis
of C2-symmetric 1,1,2,2-tetraaromatic ethanes from benz-
hydryl phosphonates, we turned our attention to the enan-
tioselective synthesis of 5a to determine whether the integ-
rity of the stereogenic center would survive the dimeriza-
tion process. Chiral alcohol (-)-12 was obtained in >99%
ee via a biocatalytic asymmetric reduction of the corre-
sponding ketone 11 using a commercially available ketore-
ductase (KRED19) and a subsequent recrystallization. Alco-
hol (-)-12 was then subjected to ClP(O)(OEt)2 and i-PrMgCl
to generate phosphonate ester 3 in 96% isolated yield
(Scheme 1). When 3 was subjected to our optimized
run with 2.0 equiv of base; lower amounts would generally
lead to incomplete conversion. Several other bases were
(4) Alternative methods for accessing racemic tetraarylethanes: (a)
Khurana, J. M.; Chauhan, S.; Maikap, G. C. Org. Biomol. Chem. 2003, 1,
1737-1740. (b) Habibi, M. H.; Farhadi, S. Tetrahedron Lett. 1999, 40,
2821-2824. (c) Li, Y.; Izumi, T. Synth. Commun. 2003, 33, 3583-3588.
(d) Schlogl, K.; Weissensteiner, W. Synthesis 1982, 50-53. (e) Ymada,
Y.; Momose, D. Chem. Lett. 1981, 1277-1278. (f) Wakui, H.; Kawasaki,
S.; Satoh, T.; Miura, M.; Nomura, M. J. Am. Chem. Soc. 2004, 126, 8658-
8659. (g) Canty, A. J.; Minchin, N. J. Inorg. Chim. Acta 1985, 100, L13-
L14. (h) Canty, A. J.; Minchin, N. J. Aust. J. Chem. 1986, 39, 1063-1069.
(i) Skattebol, L.; Boulette, B. J. Organomet. Chem. 1970, 24, 547-548. (j)
Newkome, G. R.; Roper, J. M. J. Org. Chem. 1979, 44, 502-505.
742
Org. Lett., Vol. 10, No. 5, 2008

asymmetric biocatalytic reduction of the corresponding
ketones. Each of the corresponding ketones was subjected
to a small ketoreductase (KRED) library screen, and the
optimal enzyme was chosen based on conversion and
enantioselectivity.5 The biocatalytic reductions were then
scaled up, and when possible, the resulting alcohols were
recrystallized to upgrade enantiomeric purity. This process
consistently provided alcohols with excellent ee’s (83 to
>99% ee).
Scheme 1. Asymmetric Two-Pot Dimerization
Each alcohol was then subjected to dimerization conditions
(Table 3). Several enantioenriched 3-bromo-2-pyridyl alco-
hols were studied to analyze the effect of the aryl group on
the dimerization (entries 2-5). Electron-deficient substrates
(entries 2 and 3) retained their enantiopurity well and
produced highly enantioenriched tetraarylethanes; however,
the diastereoselectivities were slightly diminished relative
to 12. Although more electron-rich substrates (entries 4 and
5) generated products with good levels of diastereoselectivity,
there was some erosion of enantioselectivity during the
course of the reaction.
dimerizationconditions, we were delighted to observe a yield
and diastereoselectivity comparable to that obtained with
(()-3 (86% yield, 88:12 dr); however, the C2-symmetric
product 5a was obtained in >99% ee! Since both the
phosphonate formation and dimerization involved the use
i-PrMgCl, this reaction had the potential to be carried out
as a one-pot process. Although phosphonate ester formation
was low yielding in MTBE, this problem could be circum-
vented by carrying out the one-pot reaction in THF/MTBE
to give the desired product in 89% yield, 86:14 dr, and >99%
ee for 5a (eq 3).
As entries 6-8 illustrate, unsubstituted pyridyl alcohols
17-19 were found to be completely unreactive. The impor-
tance of a 3-substituent for dimerization can be attributed to
deactivation of the copper catalyst via pyridine-Cu com-
plexation. In support of this argument, subjection of the more
sterically hindered 3-methyl-2-pyridyl alcohol 20 to dimer-
ization conditions did afford the expected dimer in 51% yield,
72:28 dr, and 50% ee. Additionally, 3-chloro-2-pyridyl
alcohol 21 performed well, providing a nearly equivalent
result to that observed with 12 (entry 10).
To determine if the pyridine component was mechanisti-
cally essential for dimerization, two dibenzylic alcohols were
examined (Scheme 2). In the presence of 100 mol % of
CuCN, benzhydrol could be dimerized to provide tetraphen-
ylethane in 30% yield.6 Although this reaction proceeds with
Having established conditions for the one-pot dimerization
protocol, we next examined the scope of this reaction with
a variety of enantioenriched chiral alcohols. Again, optically
active alcohols 13-16, 20, and 21 were synthesized via an
Table 3. Substrate Scope
% ee of
alcohol
product,b
% yield
dr
% ee of
product
entry
alcohola
R1
R2
(C2:meso)
1
2
3
4
5
6
7
8
9
12
13
14
15
16
17
18
19
20
21
Br
Br
Br
Br
Br
H
H
>99
>99
>99
83
5, 89
22, 86
23, 91
24, 89
25, 80
26, 0
27, 0
28, 0
29, 51
30, 90
86:14
73:27
75:25
>95:5
86:14
>99
>99
>99
44
4-F
4-Cl
3-Me
3-OMe
H
H
H
H
H
87
64
H (3-pyridyl)
H (4-pyridyl)
Me
Cl
>99
83
72:28
85:15
81
83
10
a Except for 13, absolute stereochemistry of alcohols is unknown. b Only relative stereochemistry shown for product.
Org. Lett., Vol. 10, No. 5, 2008
743

inversion, rather than retention, of stereochemistry at the
chiral center.
Scheme 2. One-Pot Dimerizations of Dibenzylic Alcohols
It is difficult to conclusively state the mechanism of this
interesting dimerization reaction;7 however, the results
presented can be collectively used to provide some insight.
The absolute configuration of the substrates and products
reveals that the reaction proceeds with inversion of stereo-
chemistry and, therefore, is likely to proceed through a SN2-
type displacement of the phosphonate with a putative benzyl
metal (e.g., organocopper) species.8 It is also of particular
note that the diastereoselectivities of the racemic reactions
are comparable to those observed in the enantiopure reac-
tions. This result alone implies that the dimerization process
is inherently stereoselective and preferentially leads to the
formation of the C2-symmetric isomer over the meso-
diastereomer. Studies are ongoing to elucidate the exact
nature of this mechanism.
In conclusion, we have described a method for the
asymmetric synthesis of tetraarylethanes via the dimerization
of optically active aryl-heteroaromatic secondary alcohols.
This novel method has been shown to generate a variety of
products in good yields, high diastereoselectivities, and
excellent enantioselectivities. This dimerization proceeds
through an unusual copper-catalyzed pathway that results in
the net inversion of stereochemistry, and studies are ongoing
to elucidate the mechanism of this reaction.
only modest efficiency, this confirms that the pyridine moiety
is not essential for this reaction to proceed. The presence of
the pyridine functional group does, however, significantly
increase the efficiency of the process. A chiral dibenzylic
alcohol was also examined: dimerization of 33 in the
presence of catalytic CuCN generated the corresponding
tetraarylethane in 54% yield. Interestingly, although the
diastereoselectivity ratio was lower, there was little loss in
enantiopurity in the course of this reaction (>99 to 93% ee).
To gain a better mechanistic understanding of this unusual
reaction, the absolute configuration of a substrate and product
was determined through single-crystal X-ray analysis. As
Figure 1 illustrates, crystal structures reveal that dimerization
of alcohol (R)-12 leads to the formation of (1S,2S)-5. These
results suggest that the dimerization process proceeds through
Acknowledgment. The authors thank Robert Reamer,
Adam Beard, David Pollard, and Artis Klapars (Merck
Process Research) for their help and insight.
Supporting Information Available: Experimental details
and spectroscopic data for new compounds are available. This
material is available free of charge via the Internet at
http://pubs.acs.org.
OL7027543
(5) Truppo, M. D.; Pollard, D.; Devine, P. Org. Lett. 2007, 9, 335-338.
(6) The control reaction was run using these conditions on 31 in the
presence of 1 equiv of pyridine to give 32 in 21% yield.
(7) It is well-established in the literature that Cu(I) combined with
Grignard reagents forms organocuprates which can then undergo substitution
reactions; for reviews, see: (a) Normant, J. F. Synthesis 1972, 63-75. (b)
Erdik, E. Tetrahedron 1984, 40, 641-657. Therefore, it is possible that
the observed dimerization occurs through a cuprate intermediate that goes
on to form the observed product. Ullman biaryl coupling reactions are also
well-documented as being catalyzed by Cu(I) via either a radical or a copper
insertion pathway. For recent reviews, see: (c) Hassan, J.; Sevignon, M.;
Gozzi, C.; Lemaire, M. Chem. ReV. 2002, 102, 1359-1467. (d) Nelson, T.
D.; Crouch, R. D. Org. React. 2004, 63, 265-289.
(8) Interestingly, the reaction in which 3 was treated with preformed,
stoichiometric “i-Pr2CuMgCl•MgCN” (formed from 2 i-PrMgCl + CuCN)
produced none of the desired product 5.
Figure 1. Absolute configuration of alcohol 12 and dimer 5.
744
Org. Lett., Vol. 10, No. 5, 2008