Copper-catalyzed enantioselective β-boration of acyclic enones

Article abstract of DOI:10.1002/chem.200802150

The enantioselective β-boration of various acyclic enones has been studied by using chiral diphosphinecopper complexes. Good to excellent yields and enantioselectivities (up to 97% ee (enantiomeric excess)) were observed for a range of substrates under op

Full text of DOI:10.1002/chem.200802150

FULL PAPER
DOI: 10.1002/chem.200802150
Copper-Catalyzed Enantioselective ACTHUNTGRNEUNGb-Boration of Acyclic Enones
Hak-Suk Sim, Xinhui Feng, and Jaesook Yun*[a]
Abstract: The enantioselective b-bora-
transformation, the addition of a con-
trolled amount of alcohol, especially
methyl alcohol, is critical to obtain
products in high ee and yield. This
tion of various acyclic enones has been
studied by using chiral diphosphine–
copper complexes. Good to excellent
yields and enantioselectivities (up to
97% ee (enantiomeric excess)) were
methodology accommodates structural
variation of acyclic enones and pro-
vides access to a range of functional-
ized chiral organoboronates in high en-
antiomeric purity.
Keywords: asymmetric catalysis
·
boron · conjugate addition · copper ·
enones
observed for
a range of substrates
under optimized conditions. In this
Introduction
nitriles.[8,9] Boronate esters are versatile intermediates in or-
ganic synthesis but the current routes to chiral boronates
are relatively limited. In this regard, the asymmetric conju-
gate addition of diboron reagents to acyclic enones would
provide an expedient preparation method of chiral b-bory-
lated ketones.
The conjugate addition of organometallic nucleophiles to
electron-deficient olefins constitutes one of the most impor-
tant strategies for bond construction in organic synthesis.[1]
Transition-metal-catalyzed enantioselective conjugate addi-
tions have been extensively studied and remarkable progress
is continually being made. Acyclic enones have proven to be
particularly challenging substrates in many asymmetric con-
jugate additions, although a few notable examples have re-
cently been disclosed.[2]
In recent years, the transition-metal-catalyzed addition of
diboron reagents to a,b-unsaturated carbonyl compounds,
which affords organoboronate products with carbonyl func-
tionalities at the b position, has been a subject of interest.
This transformation has been studied with platinum,[3] rhodi-
um,[4] copper,[5] and nickel[6] catalyst systems by several
groups, including ours. Despite recent developments in
metal-catalyzed conjugate boration methodology, there are
few catalytic enantioselective methods that provide access
to enantiomerically enriched organoboronates by this ap-
proach;[7] only recently have we described the first enantio-
selective b-boration reaction of a,b-unsaturated esters and
The development of a highly enantioselective b-boration
of enones with copper catalysts appeared particularly chal-
lenging, apart from the structural variations of acyclic
enones. Neither Hosomiꢀs[5a] nor Miyauraꢀs[5b] copper catalyt-
ic systems developed for the b-boration of enones can be di-
rectly applied to enantioselective reactions because of their
low reactivity and the type of ligands employed.[10] In our
recent study of the b-boration of a,b-unsaturated esters and
nitriles by using CuCl, NaOtBu, a diphosphine ligand, and
methanol, we observed a dramatic rate acceleration effect
due to the alcohol addiACTHNUTRGNEUNG
tive.[5c] Although our catalyst system
can possibly solve the low reactivity problem of the previous
copper systems, it does not guarantee high enantioselectivity
even with chiral ligands of adequate architecture in the light
of possible nonselective background reactions with the in-
clusion of the alcohol additive. Herein we report the copper-
catalyzed enantioselective addition of bis(pinacolato)dibor-
on (B2pin2, 1) to the challenging class of acyclic enones. A
variety of acyclic enones undergo highly enantioselective
and efficient conjugate addition of the boron reagent cata-
lyzed by a chiral copper–phosphine complex under mild re-
action conditions.
[a] H.-S. Sim, X. Feng, Prof. J. Yun
Department of Chemistry and Institute of Basic Science
Sungkyunkwan University
Suwon 440-746 (Korea)
Fax : (+82)31-290-7075
E-mail: jaesook@skku.edu
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.200802150.
Chem. Eur. J. 2009, 15, 1939 – 1943
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1939

Results and Discussion
ing the amount of methanol to one equivalent (Table 1, en-
tries 5 and 8) or employing a slight excess of iPrOH
(Table 1, entry 11) was effective in giving 3a in good yields
with the expected levels of enantioselectivity.
Next the effect of solvent on enantioselectivity was deter-
mined with 2a and chalcone (2b) by employing one equiva-
lent of MeOH (Table 2). When mandyphos (L2) was used as
the ligand, reactions in toluene resulted in a significant de-
First, to estimate the extent of the “methanol effect” on the
b-boration of acyclic enones, we set up an experiment in
which one equivalent of 4-phenyl-3-buten-2-one (2a) and
one equivalent of trans-ethyl cinnamate competitively react-
ed with one equivalent of diboron (1) in the absence of co-
ordinating ligands (Scheme 1). As expected, the enone re-
crease
in
stereoinduction
(Table 2, entries 2 and 4). On
the other hand, with josiphos
ligand L1 no significant drop in
ee was detected in either THF
or toluene (Table 2, entries 6
and 7). However, reactions in
diethyl ether appeared to be
slightly less selective (Table 2,
entries 5 and 8). Of note is the
Scheme 1. Competitive b-boration reaction of 2a and cinnamate in the absence of ligand.
Table 1. Optimization of the asymmetric boration of enone 2a.
acted faster than the ester, but surprisingly, complete con-
sumption of the enone to the product was detected whereas
less than 1% of the ester react-
ed; the addition of MeOH was
sufficient for complete conver-
sion of the enone! Bearing the
easy progress of this ligandless
reaction in mind, we sought to
Entry Ligand Additive
(equiv)
T
[8C]
Conv.
[%]
Yield[a]
[%]
ee[b]
[%]
explore the asymmetric b-bora-
tion of acyclic enones.
Initially, we chose (R)-(S)-jo-
siphos (L1) and (R)-(S)-mandy-
1
2
3
4
5
6
7
8
L1
L1
L1
L1
L1
L2
L2
L2
L2
L2
L2
–
–
RT
60
RT
RT
RT
60
RT
RT
RT
RT
RT
21
53
100
92
100
56
100
100
100
95
18
48
80
80
93
54
68
95
95
90
92
68 (S)
56 (S)
37 (S)
68 (S)
68 (S)
77 (S)
74 (S)
80 (S)
79 (S)
80 (S)
80 (S)
MeOH (2)
tBuOH (2)
MeOH (1)
–
MeOH (2)
MeOH (1)
tBuOH (2)
iPrOH (1)
iPrOH (1.2)
phos (L2) as starting nonrace-
mic ligands for the investigation
of the b-boration of 2a.[11]
9
10
11
Copper catalysts complexed by
these ligands alone and without
100
an alcohol additive were not ef-
fective for the complete b-bora-
[a] Isolated yield. [b] Determined by chiral HPLC analysis of the corre-
sponding b-hydroxy ketone obtained by oxidation.
tion even at elevated tempera-
tures (Table 1, entries 1, 2, and
6). While some of the ees (en-
Table 2. Effect of solvent on the enantioselectivity of the boration.
antiomeric excesses) obtained under these conditions ap-
peared to reasonably reflect the enantioselectivity of each
ligand (68% ee with L1 at RT and 77% ee with L2 at 608C),
the reaction performed at 608C with ligand L1 resulted in
poor conversion and ee (Table 1, entry 2). Therefore, using
alcohol additives for rate acceleration in this transformation
was unavoidable and a number of conditions with different
alcohols were screened to identify the optimal conditions.
When 2a was subjected to reaction conditions with two
equivalents of methanol, lower levels of enantioselectivity
were obtained (Table 1, entries 3 and 7).[12] Sterically bulky
tBuOH (Table 1, entry 4) and iPrOH (Table 1, entry 10)
were not as effective as MeOH in terms of turnover num-
bers,[13] but afforded product 3a with the desirable levels of
enantioselectivity. We were pleased to find that either reduc-
Entry
Substrate
Ligand
Solvent
ee [%]
1
2
L2
L2
THF
toluene
80
67
3
4
5
L2
L2
L2
THF
toluene
Et2O
70
62
68
6
7
8
L1
L1
L1
THF
toluene
Et2O
96
95
90
1940
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1939 – 1943

b-Boration of Acyclic Enones
FULL PAPER
fact that substrate 2b, which has a phenyl group instead of a
methyl group (2a, 68% ee), reacted with an excellent level
of enantioselectivity (96% ee; Table 2, entry 6) with a
copper–josiphos complex.[14]
Encouraged by the enantioselectivities achieved with the
model substrates, we continued to survey additional acyclic
enone substrates with the following sets of reaction condi-
tions; L1 or L2 was used as the ligand and MeOH (1 equiv)
or iPrOH (1.2 equiv) as the additive with 3 mol% CuCl and
3 mol% NaOtBu, in THF at room temperature (Table 3).[15]
was used, the highest enantioselectivity (93% ee; Table 3,
entry 3) was obtained with the iso-propyl-substituted enone
and lower enantioselectivities were observed with the tert-
butyl- (8% ee; Table 3, entry 5) and phenyl-substituted
enones (70% ee; Table 2).
In the reactions of alkyl enones 2e and 2 f, ligand L1 was
more selective than L2 and afforded both the desired prod-
ucts in 90% ee (entries 6 and 8). The josiphos ligand was
again more efficient for the reaction of other phenyl enones
(2g–j), affording products with excellent enantioselectivity
up to 97% ee.[17] It appears that steric factors at the b-posi-
tion, which the boron addition takes place at, do not play a
major role in determining the enantioselectivity with L1 as
the ligand, as the different phenyl enones were b-borylated
with similar levels of enantioselectivity. It is worth noting
that the extended conjugate substrates 2i[16] and 2j reacted
stereoselectively to give the corresponding 1,4-addition
products in high yield and excellent ee (Table 3, entries 13
and 15). iPrOH was generally as effective as MeOH, but in
some cases it led to poorer results in terms of reactivity. For
example, using iPrOH in the reaction of 2j resulted in a
very slow reaction (30% conversion in 22 h, 70% in 50 h;
Table 3, entry 16). Overall, the combination of Cu–josiphos
(L1) and one equivalent of MeOH in THF allows for com-
plete conversion and good to high ee for a range of acyclic
enones.
Table 3. Enantioselective b-boration of acyclic enones.
Entry Substrate
Ligand Additive Yield[a] [%] ee[b] [%]
1
2
3
2b
L1
L1
L2
iPrOH
MeOH
MeOH
94
79
93
95
89[c]
93[c]
4
5
L1
L2
MeOH
MeOH
89
91
81[c]
8[c]
6
7
L1
L2
MeOH
MeOH
93
86
90
30
A possible catalytic cycle for the conjugate boration of
enones is proposed in Scheme 2. The conjugate additions
8
9
10
L1
L1
L2
MeOH
iPrOH
MeOH
95
90
96
90
88
30
11
12
L1
L1
MeOH
MeOH
97
97
97
94
13
14
L1
L1
MeOH
iPrOH
72[d]
72[d]
92
91
15
16
L1
L1
MeOH
iPrOH
93
–
96
95
[e]
[a] Isolated yield. [b] Determined by chiral HPLC or GC analysis of the
corresponding b-hydroxy ketone obtained by the oxidation of 3. [c] De-
termined by chiral HPLC analysis of the borylated product. [d] Com-
pound 3i was unstable upon silica gel chromatography and was thus
transformed to the corresponding b-hydroxy ketone. Yield over two
steps. [e] ꢀ70% conversion after 50 h.
Scheme 2. Proposed reaction mechanism.
proceed via an intermediate copper enolate that undergoes
protolytic cleavage by an alcohol to form the protonated
product and a copper alkoxide. The copper alkoxide reacts
with diboron to regenerate the active copper boryl complex.
To obtain evidence on the proposed reaction pathway, an
experiment with enone 2c and MeOD as the alcohol addi-
tive was conducted (Scheme 3).[18] The reaction afforded
[D]-3c with deuterium incorporated at the a position in ac-
cordance with the proposed mechanism. This result also sug-
gests that the rate enhancement of the alcohol additive may
In general, with MeOH as the additive all the reactions
reached completion in between 9 and 24 h and good yields
of the desired products were obtained. For b-phenyl-substi-
tuted enones with methyl (2a; Table 1), iso-propyl (2c;
Table 3), and phenyl (2b; Table 2) side groups, the enantio-
selectivities increased from 68 to 96% ee, and then de-
creased to 81% ee for tert-butyl enone 2d (Table 3) when
L1 was employed as the ligand. On the other hand, when L2
Chem. Eur. J. 2009, 15, 1939 – 1943
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1941

J. Yun et al.
Conclusion
In summary, we have developed a catalytic enantioselective
conjugate boration method of acyclic enones that provides
ready access to chiral organoboronates that have a boronate
group at the stereocenter b to the carbonyl. It was found
that the addition of alcohol additives was crucial for higher
yields of the desired products and that controlling the
amount of alcohol depending on its size was essential to
obtain high and reproducible ee. Notably, a copper–josiphos
complex generally gave excellent levels of enantioselectivity,
up to 97% ee, with various acyclic enones.
Scheme 3. Deuterium labeling study.
be due to more rapid quenching of the copper enolate by al-
cohol than by the diboron.
Recently, Marder, Lin, and coworkers have shown by
using DFT studies on the diboration of aldehydes that s-
À
À
bond metathesis between a Cu O bond and a B B bond is
almost barrierless, while the metathesis process between a
Cu C bond and a B B bond has a higher barrier.[19] We rea-
soned that the slow reaction rates observed by Hosomi et al.
and Miyaura et al. and in some of our results without alco-
hol should result from slow s-bond metathesis between the
copper enolate and diboron reagent. Therefore, the sluggish-
ness of the enolate boration in comparison to the boration
of copper alkoxides might be the result of a preference for a
C-bound copper enolate over an O-bound copper enolate.
Overall, the rate acceleration effect of alcohol additives can
Experimental Section
À
À
General procedure for the asymmetric b-boration of acyclic enones:
CuCl (0.015 mmol, 1.5 mg), NaOtBu (0.015 mmol, 1.4 mg), and (R)-(S)-
josiphos ligand (0.015 mmol, 9.7 mg) were placed in a resealable Schlenk
tube and THF (0.40 mL) was added under nitrogen. The reaction mixture
was stirred for 30 min at RT, after which time bis(pinacolato)diboron
(0.55 mmol, 140 mg) in THF (0.30 mL) was added. The reaction mixture
was stirred for 10 min, then the enone substrate (0.5 mmol) and MeOH
(0.5 mmol, 0.02 mL) were successively added. The reaction tube was
washed with THF (0.30 mL), sealed, and stirred until no starting material
was detected (monitored by TLC). The reaction mixture was filtered
through a Celite pad and concentrated, then the product was purified by
silica gel chromatography. See the Supporting Information for details of
the synthesis and characterization of individual compounds.
À
be explained by the facile formation of Cu OR bonds from
the copper enolate, followed by the barrierless boration of
the copper alkoxide with B2pin2 to regenerate the active
À
Cu B catalyst.
The experiment conducted in Scheme 1 shows that the
rate of the nonselective pathway can also be significantly en-
hanced by the addition of alcohol additives, a mechanistic
representation of which is also shown in Scheme 2. The fact
that a highly enantioselective b-boration of enones has been
realized in the presence of alcohol additives indicates that
the nonselective pathway is effectively suppressed under our
optimal conditions. Because keeping the concentration of
Acknowledgements
This work was supported by a Korea Research Foundation grant (KRF-
2007-531-C00034) and by a Korea Science and Engineering Foundation
grant (R01-2008-000-20332-0) funded by the Korean Government. We
thank Solvias for supplying the ligands used in this study.
À
À
the free Cu B species, Cu Bpin, as low as possible is essen-
tial for high enantioselectivity, the coordination ability of
the chiral ligand (L*) to the metal as well as its inherent
facial selectivity should be an important factor to consider
in this asymmetric transformation.[11] We also observed that
the level of enantioselectivity is dependent on the amount
and size of the alcohol additive and the nonselective reac-
tion takes place to a great extent with excess methanol.
With bulkier alcohols (iPrOH and tBuOH), slower reaction
rates were observed but the enantioselectivity was preserved
in most cases. This might be attributed either to the rate of
the background reaction being slowed down to a greater
extent than the rate of the enantioselective route or to less
perturbation of the ligand–metal association by an increase
in the size of the alcohol. The effect of solvent on ee can be
explained by the degree of the nonselective pathway al-
lowed in different solvents.
[1] For representative examples, see: a) K. Fagnou, M. Lautens, Chem.
Rev. 2003, 103, 169–196; b) B. L. Feringa, R. Naasz, R. Imbos, L. A.
Arnold in Modern Organocopper Chemistry (Ed.: N. Krause),
Wiley-VCH, Weinheim, 2002, pp. 224–258; c) P. Perlmutter, Conju-
gate Addition Reactions in Organic Synthesis, Tetrahedron Organic
Chemistry Series 9, Pergamon, Oxford, 1992.
[2] a) H. Mizutani, S. J. Degrado, A. H. Hoveyda, J. Am. Chem. Soc.
2002, 124, 779–781; b) B. H. Lipshutz, J. M. Servesko, Angew.
Chem. 2003, 115, 4937–4940; Angew. Chem. Int. Ed. 2003, 42, 4789–
4792; c) F. Lꢂpez, S. R. Harutyunyan, A. J. Minnaard, B. L. Feringa,
J. Am. Chem. Soc. 2004, 126, 12784–12785.
[3] a) Y. G. Lawson, M. J. G. Lesley, T. B. Marder, N. C. Norman, C. R.
Rice, Chem. Commun. 1997, 2051–2052; b) H. Abu Ali, I. Gold-
berg, M. Srebnik, Organometallics 2001, 20, 3962–3965; c) N. J. Bell,
A. J. Cox, N. R. Cameron, J. S. O. Evans, T. B. Marder, M. A. Duin,
C. J. Elsevier, X. Baucherel, A. A. D. Tulloch, R. P. Tooze, Chem.
Commun. 2004, 1854–1855.
[4] G. W. Kabalka, B. C. Das, S. Das, Tetrahedron Lett. 2002, 43, 2323–
2325.
[5] a) H. Ito, H. Yamanaka, J. Tateiwa, A. Hosomi, Tetrahedron Lett.
2000, 41, 6821–6825; b) K. Takahashi, T. Ishiyama, N. Miyaura, J.
Organomet. Chem. 2001, 625, 47–53; c) S. Mun, J.-E. Lee, J. Yun,
Org. Lett. 2006, 8, 4887–4889; d) J.-E. Lee, J. Kwon, J. Yun, Chem.
Commun. 2008, 733–734.
[6] K. Hirano, H. Yorimitsu, K. Oshima, Org. Lett. 2007, 9, 5031–5033.
1942
www.chemeurj.org
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2009, 15, 1939 – 1943

b-Boration of Acyclic Enones
FULL PAPER
[7] E. Burks, H. J. P. Morken, Chem. Commun. 2007, 4717–4725.
[8] J.-E. Lee, J. Yun, Angew. Chem. 2008, 120, 151–153; Angew. Chem.
Int. Ed. 2008, 47, 145–147.
[9] For a recent preparation of b-borylated amides by the catalytic
asymmetric hydroboration of b,g-unsaturated amides, see: S. M.
Smith, N. C. Thacker, J. M. Takacs, J. Am. Chem. Soc. 2008, 130,
3734–3735.
[10] Miyauraꢀs conditions also indicate the possible interference of non-
selective reactions by nonligated copper species at elevated temper-
atures.
[11] We have screened other types of ligands, such as salen-type ligands,
oxazolines, and Pfaltzꢀs-type ligands, but all led to an almost racemic
product. See the Supporting Information for details.
[12] When a 1:2 ratio of copper–L1 was employed with two equivalents
of MeOH as the additive, we observed that the enantioselectivity in-
creased to 70% ee. However, with L2, both 1:1 and 1:2 combination
of copper–ligand in the presence of two equivalents of MeOH gave
variably lower levels of enantioselectivity than the expected ee. In
this study, we pursued the development of a simple, general protocol
in which excess ligand is not employed, for future screening of vari-
ous ligands for their effectiveness.
[14] The (di-tert-butyl)phosphine analogue of josiphos gave
enantioselectivity (90% ee).
a lower
[15] All of the products (3) could be isolated by silica gel chromatogra-
phy except for 3i,[16] which partially decomposed into the corre-
sponding protodeboronated product upon chromatography. Assign-
ments of the absolute configuration of 3a,b,g,h, and i were made by
comparison of the optical rotations of their hydroxy derivatives with
those in the literature. The other unknown products were assumed
to have the same configuration as the known cases.
[16] Filtration of the crude reaction mixture through a Celite pad afford-
ed a fairly clean 1H NMR of 3i that was slightly contaminated with
boron impurities. See the Supporting Information for the spectra.
[17] Mandyphos ligand gave poor to moderate ees for these substrates
(25–64% ee).
[18] DPEphos (bis(2-diphenylphosphinophenyl)ether) was used as the
ligand, see: P. C. J. Kamer, P. W. N. M. Van Leeuwen, J. N. H. Reek,
Acc. Chem. Res. 2001, 34, 895–904.
[19] H. Zhao, L. Dang, T. B. Marder, Z. Lin, J. Am. Chem. Soc. 2008,
130, 5586–5594.
Received: October 17, 2008
[13] See also entry 16 in Table 3.
Published online: January 8, 2009
Chem. Eur. J. 2009, 15, 1939 – 1943
ꢁ 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
1943