Copper-Catalyzed Asymmetric 1,2-Addition of Grignard Reagents to 3-Acyl 2 H -chromenes

Article abstract of DOI:10.1055/s-0036-1588532

Enones in which the carbon-carbon double bond is part of the pharmacologically important 2 H -chromene (2 H -1-benzopyran) nucleus undergo asymmetric copper-catalyzed 1,2-addition of Grignard reagents. High yields and enantiomeric excesses up to 84% are o

Full text of DOI:10.1055/s-0036-1588532

SYNLETT
0
9
3
6
-
5
2
1
4
1
4
3
7
-
2
0
9
6
© Georg Thieme Verlag Stuttgart · New York
2017, 28, A–E
letter
A
en
B. C. Calvo, A. J. Minnaard
Letter
Synlett
Copper-Catalyzed Asymmetric 1,2-Addition of Grignard Reagents
to 3-Acyl 2H-chromenes
Beatriz C. Calvo
MgBr
HO
O
Adriaan J. Minnaard*
0
0
0
0
0-
0
0
2
5-
9
6
6
1-
3
0
0
Ph2P
Cy2P
cat. CuBr•SMe2 /L
t-BuOMe, –78 °C
Fe
Stratingh Institute for Chemistry,
Nijenborgh 7, 9747 AG Groningen,
The Netherlands
O
O
L
70% yield, 80% ee
a.j.minnaard@rug.nl
Received: 09.06.2017
Accepted after revision: 10.07.2017
Published online: 17.08.2017
addition, the method was applied in natural product syn-
thesis.16 The originally used enone substrates, however, had
a rather fixed substitution pattern with a methyl, phenyl, or
bromide substituent at the α-position of the double bond
(Scheme 1). In order to get more insight into the substrate
DOI: 10.1055/s-0036-1588532; Art ID: st-2017-b0450-l
Abstract Enones in which the carbon–carbon double bond is part of
the pharmacologically important 2H-chromene (2H-1-benzopyran) nu-
cleus undergo asymmetric copper-catalyzed 1,2-addition of Grignard
reagents. High yields and enantiomeric excesses up to 84% are obtained
and access to these novel enantio-enriched tertiary alcohols is provided.
General structure of the 1,2-addition substrates
O
R1
R3
Key words asymmetric catalysis, Grignard reagents, copper catalysis,
1,2-addition, 2H-chromenes
R2
R1 = Ph, Me, p-BrC6H4, Cy, p-CF3C6H4
R2 = Me, Ph, Br
Complementary to the well-known asymmetric copper-
catalyzed conjugate addition (1,4-addition) of Grignard re-
agents1 we reported in 2011 on a copper-diphosphine cata-
lyst that performs enantioselective 1,2-addition reactions
of Grignard reagents to enones.2–4 The alkylcopper species
apparently outcompetes the Grignard reagent in the addi-
tion to the carbonyl group, and for several types of enones
the yields and ee obtained are very high.5 Until then, just a
few examples of enantioselective 1,2-addition reactions
employing Grignard reagents had been described. Seebach
reported on the use of stoichiometric TADDOL ligand com-
bined with alkyl Grignard reagents and organolithium re-
agents for the enantioselective 1,2-addition to ketones.6 A
catalytic enantioselective 1,2-addition of Grignard reagents
to ketones had been reported in 2006 by Hatano et al.7 In
this system, the addition of catalytic ZnCl2to the alkyl
Grignard reagents is required. More recent examples, also
using aryl Grignard reagents, were reported by the groups
of Yus8 and Gilheany.9
R
3 = Me, Ph, Et, i-Pr, CH2CH(CH3)2
O
HO
CuBr•SMe2 (5 mol%)
L1 (6 mol%)
+
t-BuOMe, –78 °C, 2 h
MgBr
95% yield, 94% ee
O
HO
CuBr•SMe2 (5 mol%)
L1 (6 mol%)
Br
+
t-BuOMe, –78 °C, 2 h
MgBr
94% yield, 96% ee
Ph2P
Cy2P
Fe
L1
(S,RFe)-Rev. Josiphos
It turned out that the substrate scope of the alkyl
Grignard reagent/copper-diphosphine catalyst system could
be considerably expanded with arylalkyl ketones,10 aryl he-
teroaryl ketones,11 acylsilanes,12 silyl ketimines,13 alkenyl-
substituted aromatic N-heterocycles,14 and ketimines.15 In
Scheme 1 Substituted enones previously used in the enantioselective
copper-catalyzed 1,2-addition of Grignard reagents
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–E

B
B. C. Calvo, A. J. Minnaard
Letter
Synlett
requirements for this reaction, and to expand the substrate
scope, we desired to vary the substituent at the α-position
and study the effect on the selectivity and the enantioselec-
tivity of the copper-catalyzed Grignard addition reaction.
A hydroxy-methylene unit at the α-position, as in 1 (Fig-
ure 1), was chosen as a suitable expansion of the substrate
scope.
MgBr
(3 equiv)
O
HO
CuBr•SMe2 (5 mol%)
L1 (6 mol%)
OH
t-BuOMe, –78 °C, 2 h
OH
1
7, 6% yield, no ee
Scheme 3 Copper-catalyzed 1,2-addition of isobutylmagnesium bro-
mide to enone 1
O
(preferred) s-cis conformation to the s-trans conformation
(Scheme 4), the last one being apparently ineffective in the
reaction. Alternatively, the copper catalyst is inhibited by
bidentate coordination to this alkoxy ketone. As depicted in
Scheme 4, the reaction mechanism for the 1,2-and 1,4-ad-
dition reactions probably both run via copper–Grignard
complex I. It forms a π-complex with the double bond of
the s-cis enone and, upon alkyl transfer, the 1,2-addition
product is formed.
The reaction was subsequently studied using the
TBDMS-protected derivative of 1, but the result was the
same; a low yield and absence of enantioselectivity. We as-
sume that as long as the oxygen can participate in bidentate
coordination of either the magnesium or the copper, this
will prohibit (enantioselective) addition.
To challenge this hypothesis, it was decided to ‘lock’ the
oxygen in a cyclic ether, so that the s-trans conformation
could no longer be stabilized by coordination of the alkoxy
group and the ketone to the copper/magnesium. The 3-
acyl-substituted 2H-chromenes 8–13 were synthesized by
reacting a series of substituted salicyl aldehydes with meth-
yl vinyl ketone under basic conditions (Scheme 5).19 The
deprotonated hydroxyl group of the salicyl aldehyde adds in
a Michael-type addition reaction to the enone. Subsequent
aldol condensation gives the corresponding 2H-chromenes.
These substrates are interesting as the 2H-chromene
core is pharmacologically important and is found in natural
products.20–22 Nevertheless, except for 9, the synthesis of
which gave in our hands a somewhat lower yield than re-
ported, the other compounds had not been described be-
fore. 3-Acyl-substituted 2H-chromenes have not been used
in 1,4- or 1,2-addition reactions.
The copper-catalyzed asymmetric 1,2-addition to 9 was
carried out employing isobutylmagnesium bromide. Prod-
uct 14 was obtained in good yield and a high ee of 80% (Ta-
ble 1, entry 1). This result supports our hypothesis that pre-
venting the alkoxide or ether from coordinating to the cata-
lyst or magnesium leads to (a larger amount of) the cis
enone and in turn to an enantioselective addition reaction.
The yields and the enantioselectivities are slightly lower
compared to the previous studied substrates, the α-bromo-
and α-methyl-substituted enones. With the same substrate
we performed the reaction with the linear Grignard reagent
OH
1
Figure 1 Target substrate for the 1,2-addition reaction
The synthesis of 1 proceeded according to literature
(Scheme 2).17 Baylis–Hillman reaction of benzaldehyde and
methyl vinyl ketone afforded 2, although due to concomi-
tant polymerization of the methyl vinyl ketone the yield
was with 50% just moderate. As the starting materials are
low cost, this was compensated for by performing the reac-
tion at a large scale. Subsequent acetylation, rearrange-
ment, and acetate hydrolysis,18 gave 1 in 50% yield over two
steps.
O
OH
O
O
DABCO
H
+
DMF, r.t., 7 d
2
3
4
50%
O
OH
O
O
O
H2SO4
+
O
CH2Cl2, r.t., 2 h
OAc
6
5
2
O
K2CO3
MeOH/H2O, r.t., 20 h
OH
1
50%
Scheme 2 Synthesis of enone 1
With 1 in hands, the 1,2-addition reaction was studied
at low temperature employing isobutylmagnesium bro-
mide, one of the Grignard reagents that gave the highest ee
in this reaction with the previous substrates employed. Ter-
tiary alcohol 7 was obtained in a very low yield, however,
and turned out to be racemic (Scheme 3). Mainly unreacted
starting material was observed, even with an excess of
Grignard reagent.
From previous studies we know that enolization of the
substrate takes over if the addition reaction is slow. We
speculate that the lack of reactivity is due to coordination of
the magnesium alkoxide, formed upon deprotonation by
the Grignard reagent, to the ketone. This coordination re-
sults in a change of conformation of the enone from the
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–E

C
B. C. Calvo, A. J. Minnaard
Letter
Synlett
Proposed mechanish for the 1,4- and 1,2-addition reactions
P
P
P
P
Br
Br
Cu
+
2 RMgBR
Cu
O
MgBr
MgBr
O
P
P
Br
R
Br
R
Cu
P
P
O
Cu
X
P
R
Cu
MgBr
Br
P
X
X
complex I
π−complex
α-complex
P
= L1
P
X = Me, Br, Ph
X = H
R
R
OMgBr
OMgBr
X
X
1,2-addition
1,4-addition
Proposed ineffective s-trans enone conformations:
O
O
P
Cu
MgBr
O
P
O
Scheme 4 The proposed mechanism for the 1,2- and 1,4-addition reaction
ethylmagnesium bromide, and product 15 was obtained in
a lower yield and a considerably lower ee, in line with our
previous reports.4
Subsequently, we decided to study the effect of substit-
uents on the aromatic ring of the substrate on the yield and
enantioselectivity of the addition reaction. Substrates 10,
11, and 12 were reacted with isobutylmagnesium bromide,
and in all cases the products were obtained in good yields
and high ee. Varying the substituent on the ring apparently
does not have a large influence on the ee, a similar trend as
Table 1 Copper-Catalyzed 1,2-Addition of Grignard Reagents to 3-Acyl 2H-chromenes23
O
R4
R3
HO
CuBr•SMe2 (5 mol%)
R1
R1
R3
L1 (6 mol%)
+
R4MgBr
R2
O
R2
t-BuOMe, –78 °C, 2 h
O
14–22
Entry
Starting material
R1
R2
R3
Me
R4
Product
Yield (%)a
ee (%)b
1
2
3
4
5
6
7
8
9
9
9
H
H
i-Bu
Et
14
15
16
17
18
19
20
21
22
70
50
91
97
80
78
70
70
62
80
40
80
84
80
58
18
80
70
H
H
Me
Me
Me
Me
Me
Me
Et
11
12
10
9
Cl
F
H
i-Bu
i-Bu
i-Bu
CH2Cy
i-Pr
H
Me
H
H
H
9
H
H
8
H
H
i-Bu
i-Bu
13c
H
Me
Me
a Isolated yield.
b Enantiomeric excess determined by chiral HPLC.
c de = 68%.
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–E

D
B. C. Calvo, A. J. Minnaard
Letter
Synlett
tion. Upon locking the oxygen substituent in a 2H-
chromene core, thereby preventing chelation, high ee and
good yields are obtained. The regioselectivity for the reac-
tion was remarkable; no conjugate addition product was
observed. The observation that the 2H-chromene core can
be equipped with an enantio-enriched tertiary hydroxyl
group might be interesting for medicinal chemistry appli-
cations, all the more so because methods to prepare enan-
tio-enriched tertiary alcohols are still scarce.
O
O
O
R1
R1
K2CO3
R3
R3
+
dioxane
reflux, 2 d
R2
R2
(1.2 equiv)
OH
O
8; R1 = H, R2 = H, R3 = Et, 36%
9; R1 = H, R2 = H, R3 = Me, 60%
10; R1 = Me, R2 = H, R3 = Me, 27%
11; R1 = Cl, R2 = H, R3 = Me, 45%
12; R1 = F, R2 = H, R3 = Me, 62%
13; R1 = H, R2 = Me, R3 = Me, 33%
Funding Information
Financial support from The Netherlands Organization for Scientific
Research (NWO-CW) is acknowledged.
)(
Scheme 5 Synthesis of the desired substrates for the asymmetric 1,2-
addition reaction
Acknowledgment
observed for asymmetric 1,2-addition reactions to methyl
arylketones.10
T. D. Tiemersma (Stratingh Institute for Chemistry) is acknowledged
for high-resolution mass spectrometry support.
Subsequently, we used different branched Grignard re-
agents in combination with 9. Cyclohexylmethylmagne-
sium bromide was employed, giving the expected product
with a lower ee compared to isobutylmagnesium bromide.
When isopropylmagnesium bromide was used, the product
was obtained with very low ee (Table 1, entry 7). Isopropyl-
magnesium bromide is an α-branched Grignard reagent,
that consistently gives low enantioselectivities in asymmet-
ric 1,2-addition reactions.5 Apparently the highest ee are
obtained with β-branched Grignard reagents.
Next, the steric effect of a larger substituent on the car-
bonyl function was studied. The combination of 8, in which
the 2H-chromene has a propionyl instead of an acyl substit-
uent, with isobutylmagnesium bromide afforded the prod-
uct with the same yield and ee as for 9. Increasing the ster-
ics at that position apparently does not affect the ee. Finally,
steric effects at the 2-position were studied. When the re-
action was performed with rac-13, the product was ob-
tained with a de of 68% and an ee of 70%. Apparently, a sub-
stituent at the 3-position does have an effect on the asym-
metric 1,2-addition and leads to some extend to a kinetic
resolution.
Unfortunately, the absolute configuration of the prod-
ucts could not be determined despite attempts to obtain
crystals suitable for X-ray diffraction or chemical correla-
tion to a compound of known configuration. The absolute
configuration is therefore conferred from that of the sub-
strates studied previously (Scheme 1).
In this study, the substrate scope of the enantioselective
copper/diphosphine-catalyzed 1,2-addition of Grignard re-
agents has been enlarged. Employing substrates with a
CH2OH or CH2OTBDMS group at the α-position of the dou-
ble bond led to very poor conversions and no enantioselec-
tivity. We hypothesize that this is due to coordination of
this oxygen to copper or magnesium, leading to a substrate
conformation that is not suitable for asymmetric 1,2-addi-
Supporting Information
Supporting information for this article is available online at
https://doi.org/10.1055/s-0036-1588532.
S
u
p
p
o
nrtogI
f
rmoaitn
S
u
p
p
ortiInfogrmoaitn
References and Notes
(1) Harutyunyan, S. R.; den Hartog, T.; Geurts, K.; Minnaard, A. J.;
Feringa, B. L. Chem. Rev. 2008, 108, 2824.
(2) Collados, J. F.; Solà, R.; Harutyunyan, S. R.; Maciá, B. ACS Catal.
2016, 6, 1952.
(3) Madduri, A. V. R.; Harutyunyan, S. R.; Minnaard, A. J. Chem.
Commun. 2012, 48, 1478.
(4) Madduri, A. V. R.; Minnaard, A. J.; Harutyunyan, S. R. Org.
Biomol. Chem. 2012, 10, 2878.
(5) Madduri, A. V. R.; Harutyunyan, S. R.; Minnaard, A. J. Drug
Discov. Today 2013, 10, 21.
(6) Weber, B.; Seebach, D. Angew. Chem., Int. Ed. Engl. 1992, 31, 84.
(7) Hanato, M.; Suzuki, S.; Ishihara, K. J. Am. Chem. Soc. 2006, 9998.
(8) Fernández-Mateos, E.; Maciá, B.; Yus, M. Eur. J. Org. Chem. 2014,
6519.
(9) Bieszczad, B.; Gilheany, D. G. Angew. Chem. Int. Ed. 2017, 56, 1.
(10) Madduri, A. V. R.; Harutyunyan, S. R.; Minnaard, A. J. Angew.
Chem. Int. Ed. 2012, 51, 3164.
(11) Ortiz, P.; del Hoyo, A. M.; Harutyunyan, S. R. Eur. J. Org. Chem.
2015, 72.
(12) Rong, J.; Oost, R.; Desmarchelier, A.; Minnaard, A. J.;
Harutyunyan, S. R. Angew. Chem. Int. Ed. 2015, 54, 3038.
(13) Rong, J.; Collados, J. F.; Ortiz, P.; Jumde, R. P.; Otten, E.;
Harutyunyan, S. R. Nat. Commun. 2016, 7, 13780.
(14) Jumde, R. P.; Lanza, F.; Veenstra, M. J.; Harutyunyan, S. R. Science
2016, 352, 433.
(15) Ortiz, P.; Collados, J. F.; Jumde, R. P.; Otten, E.; Harutyunyan, S.
R. Angew. Chem. Int. Ed. 2017, 56, 3041.
(16) Wu, Z.; Harutyunyan, S. R.; Minnaard, A. J. Chem. Eur. J. 2014, 20,
14250.
(17) De Paula, B. R. S.; Zampieri, D. S.; Rodrigues, J. A. R.; Moran, O. J.
S. Tetrahedron: Asymmetry 2013, 24, 973.
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–E

E
B. C. Calvo, A. J. Minnaard
Letter
Synlett
(18) Lee, K. Y.; GowriSankar, S.; Kim, J. M. Bull. Korean Chem. Soc.
2004, 413.
(19) Brenna, E.; Gatti, F. G.; Malpezzi, L.; Monti, D.; Parmeggiani, F.;
Sacchetti, A. J. Org. Chem. 2013, 78, 4811.
(20) Sharif, S. A. I.; Calder, E. D. D.; Harkiss, A. H.; Maduro, M.;
Sutherland, A. J. Org. Chem. 2016, 81, 9810.
(21) Calmus, L.; Corbu, A.; Cossy, J. Adv. Synth. Catal. 2015, 357, 1381.
(22) Song, L.; Huang, F.; Guo, L.; Ouyanga, M.-A.; Tong, R. Chem.
Commun. 2017, 53, 6021.
(23) General Procedure for the Enantioselective 1,2-Addition
1-[2-(2H-Chromen-3-yl)]-4-methylpentan-2-ol (14)
To a flame-dried Schlenk tube containing a magnetic stirring
bar, CuBr·SMe2 (15 μmol, 3.1 mg), L1 (18 μmol, 10.7 mg), and 3
mL of dry t-BuOMe were added. The mixture was left to stir for
10 min. Subsequently, 9 (0.3 mmol, 52 mg) was added to the
solution. The mixture was left to stir for 30 min at –78 °C. Isobu-
tylmagnesium bromide (2 M in Et2O, 1.7 equiv, 0.25 mL) was
then added dropwise over 15 min, and the reaction was left to
stir for 2 h at –78 °C. The reaction was quenched with H2O (2 mL),
allowed to warm up to r.t., and diluted with Et2O. NH4Claq was
added, and the layers were separated. The aqueous layer was
extracted with Et2O, and the combined organic layers were dried
over MgSO4, filtered, and concentrated at reduced pressure to
afford 14 in 70% yield as a yellowish oil after flash chromato-
graphy (SiO2, n-pentane/Et2O (90:10)); 80% ee. Retention times
on chiral HPLC: tR = 28.2 min and 31.8 min. 1H NMR (400 MHz,
CDCl3): δ = 7.11 (td, J = 7.5, 1.3 Hz, 1 H), 7.03 (dd, J = 7.4, 1.5 Hz,
1 H), 6.92–6.86 (m, 1 H), 6.82 (d, J = 8.0 Hz, 1 H), 6.46 (s, 1 H),
4.73 (d, J = 0.7 Hz, 2 H), 1.75 (m, 1 H), 1.57 (dd, J = 5.9, 2.2 Hz, 3
H), 1.40 (s, 3 H), 0.96 (t, J = 7.0 Hz, 6 H). 13C NMR (101 MHz,
CDCl3): δ = 153.2, 134.0, 128.7, 126.8, 123.0, 121.5, 117.7, 115.4,
20
74.3, 65.6, 48.8, 28.0, 24.5, 24.5, 24.4. [α]D –7.8 (c 1.0, CHCl3).
HRMS (ESI–): m/z calcd for [C15H20O2 – H]–: 231.139; found:
231.138.
Analytical Data for Compound 21
80% ee. Retention times on chiral HPLC: tR = 13.2 min and 14.4
min.
1H NMR (400 MHz, CDCl3): δ = 7.11 (td, J = 7.7, 1.7 Hz, 1 H), 7.04
(dd, J = 7.5, 1.6 Hz, 1 H), 6.90 (td, J = 7.4, 1.1 Hz, 1 H), 6.82 (d, J =
8.0 Hz, 1 H), 6.48 (s, 1 H), 4.63 (s, 2 H), 1.86–1.74 (m, 1 H), 1.64–
1.46 (m, 5 H), 0.99 (d, J = 6.6 Hz, 3 H), 0.93 (d, J = 6.7 Hz, 3 H),
0.87 (t, J = 7.4 Hz, 3 H). 13C NMR (101 MHz, CDCl3): δ = 153.1,
138.1, 128.6, 126.7, 123.0, 121.5, 119.0, 115.4, 65.8, 47.9, 32.9,
24.6, 24.4, 24.2, 7.5. [α]D20 +11.0 (c 1.0, CHCl3). HRMS (ESI+): m/z
calcd for [C16H22O2 + H]+: 247.169; found: 247.169.
© Georg Thieme Verlag Stuttgart · New York — Synlett 2017, 28, A–E