Copper-free asymmetric allylic alkylation using grignard reagents on bifunctional allylic bromides

Article abstract of DOI:10.1021/ol300299b

A series of substrates containing a vinylic bromide were employed in a copper-free methodology using bidendate NHC ligands. The desired compounds are generally obtained with good enantioselectivity and good regioselectivity. Importantly the copper-catalyzed system afforded a lower enantioselectivity value. The catalytic products could be transformed into a broad scope of new 1,1-disubstituted olefins in a single step transformation without erosion of the enantioselectivity.

Full text of DOI:10.1021/ol300299b

ORGANIC
LETTERS
2012
Vol. 14, No. 6
1568–1571
Copper-Free Asymmetric Allylic
Alkylation Using Grignard Reagents on
Bifunctional Allylic Bromides
David Grassi and Alexandre Alexakis*
Department of Organic Chemistry, University of Geneva, 30 quai Ernest Ansermet
CH-1211 Geneva 4, Switzerland
alexandre.alexakis@unige.ch
Received February 6, 2012
ABSTRACT
A series of substrates containing a vinylic bromide were employed in a copper-free methodology using bidendate NHC ligands. The desired
compounds are generally obtained with good enantioselectivity and good regioselectivity. Importantly the copper-catalyzed system afforded a
lower enantioselectivity value. The catalytic products could be transformed into a broad scope of new 1,1-disubstituted olefins in a single step
transformation without erosion of the enantioselectivity.
In the context of asymmetric allylic alkylation (AAA),1
the copper-catalyzed AAA turned out to be one of the
most attractive and efficient reactions in the enantioselec-
tive construction of CꢀC bonds. Several ligand families,
such as phosphoramidites, ferrocene-based ligands, pep-
tide-type ligands, or N-heterocyclic carbenes (NHC), were
used to coordinate copper. In addition toa broad substrate
scope, a large variety of organometallic reagents such as
organoaluminum, organozinc, and Grignard reagents
could be used. Among them, the Grignard reagents are
easily available nucleophiles and have been applied to a
wide range of prochiral substrates.2 More recently, they
have been extended to racemic chiral substrates in
DYKAT processes, providing highly enantioenriched
products.3 In most cases, the substrates do not bear other
functionalities, and the only handle for molecular complex-
ity lies on the further transformations of the resulting double
bond. Examples of functionalized substrates are scarce, with
mainly another functional group4 or an additional unsatu-
ration.5 A method that could be able to produce enantio-
merically enriched SN20 products, which could be potent
(4) (a) Mao, B.; Geurts, K.; Faranas-Mastral, M.; van Zijl, A. W.;
Fletcher, S. P.; Minnaard, A. J.; Feringa, B. L. Org. Lett. 2011, 13, 948–
951. (b) Selim, K. B.; Yamada, K.; Tomioka, K. Chem. Commun. 2008,
5140–5142. (c) van Zijl, A. W.; Szymanski, W.; Lopez, F.; Minnaard,
A. J.; Feringa, B. L. J. Org. Chem. 2008, 73, 6994–7002. (d) Gillingham,
D. G.; Hoveyda, A. H. Angew. Chem., Int. Ed. 2007, 46, 3860–3864. (e)
Kacprzynski, M. A.; May, T. L.; Kazame, S. A.; Hoveyda, A. H. Angew.
Chem., Int. Ed. 2007, 46, 4554–4558. (f) van Zijl, A. W.; Lopez, F.;
Minnaard, A. J.; Feringa, B. L. J. Org. Chem. 2007, 72, 2558–2563. (g)
Okamoto, S.; Tominaga, S.; Saino, N.; Kase, K.; Shimoda, K.
J. Organomet. Chem. 2005, 690, 6001–6007. (h) Piarulli, U.; Daubos,
P.; Claverie, C.; Monti, C.; Gennari, C. Eur. J. Org. Chem. 2005, 895–
906. (i) den Hertog, T.; Macia, B.; Minnaard, A.; Feringa, B. L. Adv. Synth.
Catal. 2010, 352, 999–1013. (j) Falciola, C.; Alexakis, A. Angew. Chem., Int.
Ed. 2007, 46, 2619–2676. (k) Falciola, C.; Alexakis, A. Chem.;Eur. J. 2008,
14, 10615–10627. (l) Murphy, K. E.; Hoveyda, A. H. J. Am. Chem. Soc. 2003,
125, 4690–4691.
(1) Transition Metal Catalyzed Enantioselective Allylic Substitution in
Organic Synthesis; Kazmaier, U., Ed.; Springer: Berlin/Heidelberg, 2012;
Topics in Organometallic Chemistry, Vol. 38.
(2) For reviews, see: (a) Alexakis, A.; Malan, C.; Lea, L.; Tissot-
Croset, K.; Polet, D.; Falciola, C. Chimia 2006, 60, 124–130. (b) Falciola,
C. A.; Alexakis, A. Eur. J. Org. Chem. 2008, 3765–3780. (c) Alexakis, A.;
€
ꢀ
ꢁ
Backvall, J. E.; Krause, N.; Pamies, O.; Dieguez, M. Chem. Rev. 2008,
108, 2796–2823. (d) Harutyunyan, S. R.; den Hartog, T.; Geurts, K.;
Minnaard, A. J.; Feringa, B. L. Chem. Rev. 2008, 108, 2824–2852. (e)
Langlois, J.-B.; Alexakis, A., In Transition Metal Catalyzed Enantioselective
Allylic Substitution in Organic Synthesis; Kazmaier, U., Ed.; Springer:
Berlin/Heidelberg, 2012; Topics in Organometallic Chemistry, Vol. 38,
pp 235ꢀ268.
(3) (a) Langlois, J.-B.; Alexakis, A. Chem. Commun. 2009, 3868–
3870. (b) Langlois, J.-B.; Alexakis, A. Angew. Chem., Int. Ed. 2011, 50,
3868–3870. (c) Langlois, J.-B.; Emery, D.; Mareda, J.; Alexakis, A.
Chem. Sci. 201110.1039/C2SC00868H.
(5) (a) Giacomina, F.; Riat, D.; Alexakis, A. Org. Lett. 2010, 12,
1156–1159. (b) Teichert, J. F.; Zhang, S.; van Zijl, A. W.; Slaa, J. W.;
Minnaard, A. J.; Feringa, B. L. Org. Lett. 2010, 12, 4658–4660. (c) Li, H.;
Alexakis, A. Angew. Chem., Int. Ed. 2012, 51, 1055–1058.
r
10.1021/ol300299b
Published on Web 03/06/2012
2012 American Chemical Society

pivotal intermediates toward several new adducts result
ing from an AAA reaction, would be very desirable. This
achievement could save several steps in the synthesis of
substrates and methodology development. We thought that
using a halogen binding to a sp2 carbon may be the candidate
of choice for such desired pivotal intermediates (Scheme 1).
Scheme 1. Concept of Versatile Intermediate
In our laboratory, we have extensively used Grignard
reagents in combination withcopper and phosphoramidite
ligands, and very recently, the use of NHC’s ligands in a
copper-free methodology has proven to be very efficient in
the AAA.6 Having both isomers of the model substrate
in hand, we tested the outcome of the reaction on the
Z-isomer using our NHC’s bidendate ligand (Figure 1).
Different conditions were screened using copper salt and
ligand L1, affording poor regioselectivity and reactivity
(Table 1, entries 1ꢀ2). This disappointing result was not a
surprise, as Z-isomers are known to react sluggishly in the
AAA reaction. By submitting the E-isomer using the same
ligand L1 in the copper-free methodology, we obtained an
encouraging regioselectivity and enantioselectivity (79/21 in
favor of the SN20 adduct, 65% ee) with a high conversion
(Table 1, entry 3). Performing the reaction with copper/NHC
ligand L1 (Table 1, entry 13) or with copper in combination
with our best phosphoramidite L10 (Table 1, entry 14) did
not lead to any valuable improvement of the enantioselec-
tivity, while higher regioselectivities were accessed.
The next logical step was the screening of our NHC’s
library of ligands in a copper-free way in order to try to
improve the existing results. The switch from (methyl/
methyl) L1 to more sterically hindered (ethyl/ethyl moiety)
L2 led to a decrease of regioselectivity, enantioselectivity,
and reactivity (Table 1, entries 3ꢀ4). We then decided to
introduce dissymmetry on the blocking part, and this
choice turned out to be fruitful, showing that the ethyl/
methyl combination L3 was the best (76% conv, 77:23,
71% ee) (Table 1, entry 5 vs entries 3ꢀ4). Thus, the
modulation of the electronic properties of the phenolic
part of the ligand was tested. We tested both electron-
withdrawing and electron-donating groups in the para
position of the hydroxyl group. The introduction of an
electron-donating group, such as a benzyloxy goup L6,
afforded comparable results (Table 1, entry 8), while
Figure 1. Chiral ligands used.
Table 1. Ligand Screening
entry substrate
ligand
conv (%)a γ:R (%)a ee (%)b
1
Z
Z
E
E
E
E
E
E
E
E
E
E
E
E
2% L1
8
100
95
45
76
97
86
85
95
99
99
99
99
99
40:60
43:57
79:21
71:29
77:23
82:18
83:17
77:23
82:18
84:16
82:18
75:25
90:10
92:8
2
1.5% CuTCc, 2% L1
2% L1
3
65
55
71
70
71
68
81
72
85
60
30
65
4
2% L2
5
2% L3
6
2% L4
7
2% L5
8
2% L6
9
2% L7
10
11
12
13
14
2% L8
2% L9
2% L11
1.5% CuTCc, 2% L1
3% CuTCc, 4% L10
a By 1H (NMR). b By chiral SFC. c CuTC = copper thiophene
carboxylate.
introduction of electron-withdrawing group L11 led to
lower results (Table 1, entry 12). Nevertheless, to our
delight, by using the combination of the ideal blocking
part and a para phenyl group L9, we could yield our best
result (99% conv, 82:18, 85% ee) (Table 1, entry 11). We
then screened the generality of our methodology toward
(6) Jackowski, O.; Alexakis, A Angew. Chem., Int. Ed. 2010, 49,
3346–3350. See also: (a) Lee, Y.; Hoveyda, A. H. J. Am. Chem. Soc.
2006, 128, 15604–15605. (b) Lee, Y.; Li, B.; Hoveyda, A. H. J. Am.
Chem. Soc. 2009, 131, 11625–11633.
Org. Lett., Vol. 14, No. 6, 2012
1569

other halogens to see which one was the best compromise
in terms of enantioselectivy, reactivity, selectivity, and
isolated yield (Table 2). With the chloro derivative, the
Cu/phosphoramidite system proved to be completely in-
efficient, while the NHC L9 delivered the desired com-
pound with 82% ee and a 80:20 γ:R selectivity, albeit with a
low isolated yield of 25% (Table 2, entry 3), whereas on
fluorine derivatives, both systems were unreactive (Table 2,
entries 7ꢀ8). Interestingly, the vinylic iodide gavecomplete
conversion with or without copper (Table 2, entries 5, 6).
Obviously, the presence of copper increases the reactivity,
as the reaction could be performed at low temperature
(ꢀ78 °C). However, despite a much better gamma selec-
tivity, the ee remains rather low, particularly when com-
pared to the copper-free process (86% ee).
Table 3. Grignard Reagent Screen
time
(h)
conv
(%)a
yield
(%)b
ee
entry
1
RMgBr
Et
γ:R (%)a
(%)c
2
99
71
82:18
(88:12)d
85
2
3
Me
24
24
99
99
40
55
80:20 (93:7)
73:27
73
78
nBu
(81:19)
4
5
6
7
iAmyl
24
24
24
24
99
99
99
99
60
58
63
45
77:23 (93:7)
84:16 (98:2)
88:12 (98:2)
94:6 (97:3)
81
84
85
85
Butenyl
PhCH2CH2
iBu
a By 1H (NMR). b Yield after purification on SiO2. All reactions were
performed at ꢀ15 °C. c By chiral SFC. d Ratio in parentheses is equal to
regioselectivity after purification on SiO2.
Table 2. Copper-Free AAA to Other Halogens
Table 4. Substrate Scope
temp
conv
(%)a
γ:R
ee
entry R1
ligand
(°C)
(%)a (%)b
1
2
3
4
5
6
7
8
Br 2% L9
ꢀ15
ꢀ78
ꢀ15
ꢀ78
ꢀ15
ꢀ78
ꢀ15
ꢀ78
99 (70)c 82:18
99 92:8
85
65
82
Br 3% CuTCd, 4% L10
Cl 2% L9
99 (25) 80:20
0
Cl 3% CuTCd, 4% L10
conv (%)a
entry
R1
R2
(yield %)b
γ:R (%)a
ee (%)c
I
2% L9
99 (50) 83:17
86
60
I
3% CuTCd, 4% L10
2% L9
99
5
95:5
1
2
3
4
5
6
7
8
oClPh
Et
93 (54)
88 (60)
99 (58)
99 (53)
99 (61)
70 (50)
99 (70)
99 (65)
88:12 (89:11)d
100:0
80
93
86
85
88
93
69
11
F
F
oMePh Et
pMePh Et
pClPh Et
3% CuTCd, 4% L10
5
80:20 (86:14)
85.15 (92:8)
100:0
a By 1H (NMR). b By chiral SFC. c 20% insertion of Grignard into
vinylic iodine. Yield after purification on SiO2. d CuTC = copper
thiophene carboxylate.
Cy
Et
tBu
iPr
Me
PhCH2CH2
PhCH2CH2
PhCH2CH2
100:0
95:5 (100:0)
90:10 (100:0)
Obviously the next step was the evaluation of various
Grignard reagents on our model substrate (Table 3). The
temperature was kept constant to ꢀ15 °C and the reaction
time set to 24 h to ensure maximum conversion.
a By 1H (NMR). b Yield after purification on SiO2. All reactions were
performed at ꢀ15 °C. c By chiral GC or SFC. d Ratio in parentheses is
equal to regioselectivity after purification on SiO2.
Different primary Grignard reagents could be intro-
duced with decent to good regioselectivities on the crude
product (from 73:27 to 94:6 in favor of the γ adduct) and
interesting enantioselective values (73ꢀ85% ee). After
purification on silica gel, isolation of pure SN20 adducts
were possible in most of the cases with decent to good
yields (40ꢀ71%). We should also mention that the challen-
ging MeMgBr could be introduced with fair results
(Table 3, entry 2). In sharp contrast, secondary Grignard
reagents such as isopropyl and cyclohexyl led to messy
reactions. After having determined the limitation of the
reaction in the scope of Grignard reagent, we began to
explore the scope of substrate. A series of aliphatic and
aromatic substrates were synthesized (Table 4).
good to excellent enantioselectivities (80ꢀ93% ee), show-
ing that the steric hindrance is well tolerated (Table 4,
entries 1ꢀ2). Next we probed the influence of electronic
properties by substitution in the para position of the
phenyl ring. We were able to get results (80:20 to 85:15 in
favor of the γ adduct, 85ꢀ86% ee) comparable to the
simple phenyl ring (Table 4, entries 3ꢀ4). We focused next
on the aliphatic substrates. On those substrates, a very
sharp trend of steric hindrance was noticed. Excellent
resultswereobtainedusingstericallydemanding substrates
as cyclohexyl or tert-butyl moieties (100:0 in favor of the
γ adduct and 88ꢀ93% ee) (Table 4, entries 5ꢀ6). By decreas-
ing gradually the steric hindrance to isopropyl or to a
methyl group, this led to a dramatic decrease of enantios-
electivity (69% ee for isopropyl, 11% ee for methyl) while
keeping high levels of regioselectivty (95:5, 90:10 in favor
of the γ adduct) (Table 4, entries 7ꢀ8). On quaternary
The influence of the steric hindrance was studied by
substitution in the ortho position of the phenyl ring next to
the reactive center. We observed perfect to good regios-
electivities (85:15 to 100:0 in favor of the γ adduct) and
1570
Org. Lett., Vol. 14, No. 6, 2012

centers, the reaction proceeded sluggishly, and addition of
copper saltwas beneficialtothe catalytic systemin terms of
conversion, though affording modest enantioselectivity,
showing the limitation of our methodology at the present
time of our research. The further development of a more
active catalyst could afford high enantioselectivity, as a
sharp correlation of temperature/enantioselectivity was
observed (Table 5, entry 1 vs entry 2).
Scheme 2. Application toward New Classes of Substrates
Table 5. Application to Quaternary Centers
temp time
conv
γ:R
ee
entry
ligand
(°C)
(h) (%)a (yield %)b (%)a (%)c
1
2
3
2% L9
2% L9
ꢀ5
14 99 (50)
14 20
82:18 60
90:10 79
90:10 65
ꢀ15
2% CuTC, 3% L9 ꢀ15
14 100 (60)
a By 1H (NMR). b Yield after purification on SiO2. c By chiral SFC.
First we propose herein, as an application of our system,
the extension of the methodology developed by Wood-
ward (Scheme 2).7 Woodward reported the AAA to
BaylisꢀHillman derived substrates using amine catalysts.
The addition of diethyl zinc to such substrate was possible,
with ee values ranging from 76 to 90% ee. However,
because of the synthetic sequence, aliphatic substrates
cannot be accessed, and the system was limited to a single
nuleophile (diethyl zinc). With the present results, we could
extend this methodology to several other carbonyl deriva-
tives (aldehyde, acid) that could be easily obtained, de-
pending on the electrophile used, with a rich scope of
substrates and nucleophiles. Second, wepropose the synth-
esis of 1,1-disubstituted silane derivatives. This intermedi-
ate represents itself a new class of substrates, which has not
been reported yet by AAA. Importantly, this synthon is a
versatile intermediate, as it can be used in the Hiyama
coupling reaction. Lastbut not least isthe use ofourvinylic
bromide as coupling partner in Suzuki and Sonogashira
reactions. An alkyne containing a TMS unit could be
introduced in good yield using the Sonogashira reaction.
We managed also to enter a large panel of aryl or heter-
oaryl moieties using a Suzuki coupling reaction. The pre-
sence of chiral synthons containing aryls or heteroaryls in
vinylic positions have never been reported in the literature.
To conclude, we have disclosed a class of substrates
where a copper-free system using bidendate NHC ligand
and Grignard reagents led to interesting results in terms of
enantioselectivity and regioselectivity. The substrate scope
is large, and more importantly, the products display an
unprecedented potential of further transformations. The
extension to new classes of BaylisꢀHillman derived sub-
strates, a new family of silane derivatives, and a large panel
of 1,1-disubstituted aryls and heteroaryls has been dis-
closed. With a single catalytic product in hand, several
series of nonreported compounds could be synthetized,
in a single step transformation without erosion of the
enantioselectivity.
Acknowledgment. The authors thank the Swiss National
Research Foundation (Grant No. 200020-126663) and
COST action D40 (SER Contract No. C07.0097) for
financial support.
Supporting Information Available. Experimental pro-
cedures, NMR spectra, and chiral separations for all
compounds. This material is available free of charge via
the Internet at http://pubs.acs.org.
€
(7) (a) Borner, C.; Gimeno, J.; Gladiali, S.; Goldsmith, P. J.; Ramazotti,
D.; Woodward, S. Chem. Commun. 2000, 2433–2434. (b) Goldsmith, P. J.;
Teat, S. J.; Woodward, S. Angew. Chem., Int. Ed. 2005, 40, 2235–2237.
The authors declare no competing financial interest.
Org. Lett., Vol. 14, No. 6, 2012
1571