Asymmetric chemo- and regiospecific addition of organozinc reagents to Baylis-Hillman derived allylic electrophiles

Article abstract of DOI:10.1039/b006943o

The copper-catalysed S(N)2' addition of ZnR₂ to allylic (Z)-ArCH=C(CH₂X)(CO₂Et) (X = Br, Cl, OSO₂Me) fashions only ArCH(R)C(=CH₂)(CO₂Et); use of a chiral ligand gives up to 64% ee for this

Full text of DOI:10.1039/b006943o

Asymmetric chemo- and regiospecific addition of organozinc reagents to
Baylis–Hillman derived allylic electrophiles
Christoph Börner,a Josep Gimeno,b Serafino Gladiali,c Paul J. Goldsmith,a Daniela Ramazzottic and Simon
Woodward*a
a School of Chemistry, Nottingham University, University Park, Nottingham, UK NG7 2RD.
E-mail: simon.woodward@nottingham.ac.uk; Fax: (+44) 115-9513564; Phone: (+44) 115-9513564
b Departament de Química, Universitat Autònoma de Barcelona, Bellaterra, 08193 Barcelona, Spain
c
Dipartimento di Chimica, Università di Sassari, Via Vienna 2, 07100 Sassari, Italy
Received (in Liverpool, UK) 21st August 2000, Accepted 23rd October 2000
First published as an Advance Article on the web 22nd November 2000
The copper-catalysed SN2A addition of ZnR2 to allylic (Z)-
ArCHNC(CH2X)(CO2Et) (X = Br, Cl, OSO2Me) fashions
only ArCH(R)C(NCH2)(CO2Et); use of a chiral ligand gives
up to 64% ee for this demanding reaction.
Hillman chemistry (Scheme 1).11 Additionally, bromination of
1 proceeds with very high (Z)-selectivity12 to afford solid
products which may be crystallised to high chemical and
stereochemical purity. Simple mixture of THF solutions of 2
13
and ZnEt2 in the presence of [Cu(MeCN)4]BF4 (3 mol%) at
The reactions of organometallic nucleophiles, MR, with allylic
halide (or pseudohalide) substrates R1CHNCR2CH2X often lead
to a mixture of SN2 (attack a to the leaving group X) and SN2A
(attack g to X) products.1 This feature causes problems in metal-
catalysed asymmetric g-additions to unsymmetrical allylic
electrophiles where the catalyst must control both the regio- and
enantioselectivity. Catalysts showing superlative selectivity for
both these aspects are rare.2–6 Pointers in the stoichiometric
literature suggested to us that the regiochemical problem might
be overcome with substrates where R2 is an ester function7 and
if an organozinc reagent8 is used as the terminal organometallic
in the catalytic cycle. In particular, we were keen to apply these
ideas to enantioselective copper-catalysed SN2A additions of MR
to allylic halides as very few successes have been reported for
this otherwise intrinsically useful transformation.9,10 Aside
from controlling the regiochemistry the ester could play two
additional roles. Firstly, a more rigorously bound asymmetric
transition state might be attained, via carbonyl co-ordination.
Secondly, the products from such carbonyl containing sub-
strates are present in many compounds of biological interest.
The allylic bromides 2 (Scheme 1) were selected for initial
trials with commercial ZnEt2 to demonstrate the viability of this
approach. Compounds 2 are attractive as they are available in
just two, chromatography-free, steps via known Baylis–
220 °C leads to rapid formation of 4 (R = Et) as apparently the
Scheme 1 Reagents and conditions: i, conc. HX/H2SO4, 16–24 h, rt; ii,
ZnR2, THF, 220 °C, cat. [Cu(MeCN)4]BF4; iii, ZnEt2, THF, 220 °C, cat.
[Cu(MeCN)4]BF4/(Sa)-7; iv, for 5 NEt3–HCO2H reflux; v, for 6 treatment
of 5 with aq. HCl in MeOH followed by mesylation (MeSO2Cl–NEt3) of the
derived alcohol.
Table 1 Reaction of allylic electrophiles 2–6 with organometallics (RM) in the presence of [Cu(MeCN)4]BF4†
Ligand/
mol%
Halide
RM
CuI/mol%
Temp./°C
Time/min
Yield/%a
ee/%b
2a
2a
2b
2c
2d
2e
2a
2a
2a
3a
3a
3a
3b
3b
3d
3e
5
ZnEt2
ZnEt2
ZnEt2
ZnEt2
ZnEt2
ZnEt2
Zn(CH2TMS)2
ZnEt2
ZnEt2
ZnEt2
ZnEt2
AlEt3
ZnEt2
ZnEt2
ZnEt2
ZnEt2
ZnEt2
ZnEt2
3
0.5
3
3
3
None
None
None
None
None
None
None
7 20
7 20
7 20
7 20
7 20
7 20
7 20
7 20
7 20
7 20
7 20
220
220
220
220
220
220
0
220
240
220
240
0
220
240
220
220
220
220
40
40
40
40
40
40
180
20
40
40
40
40
40
40
40
40
20
20
100 (78)
100 (73)
100 (63)
100 (80)
86
—
—
—
—
—
—
—
3 (+)
13 (+)
36 (2)
34 (2)
8 (2)
60 (2)
64 (2)
30 (2)
22 (2)
—
3
7
84
c
47 (25)
100 (80)
(44)
56 (35)
(19)
12 ( < 5)
100 (76)
64
25 (19)
27 (14)
< 5
10
10
10
10
10
10
10
10
10
10
10
6
100 (90)
2 (+)
a Based on NMR conversion, isolated yields in parentheses. b Determined by HPLC on a Chiracel OD column, predominant stereoisomer in parentheses.
c Used in the presence of MgCl2.
DOI: 10.1039/b006943o
Chem. Commun., 2000, 2433–2434
This journal is © The Royal Society of Chemistry 2000
2433

sole product (Scheme 1). Extremely high chemo- and regio-
selectivity, technical simplicity, and high rate ( ~ 50 TON h21
Notes and references
)
†
Representative procedures and compound data. An argon-protected
characterise these reactions (Table 1).† The reactions can be run
at even lower copper loadings, for example, 2a gives a near
quantitative yield of 4a (R = Et) at 0.5 mol% CuI ( > 290
TON h21). In the absence of CuI polar co-solvents are necessary
to provide high chemical yields7 (in the absence of any copper,
yields of < 12% were attained in our control reactions). The
new catalytic protocol is superior to use of classical Gilman
cuprates. For example, reaction of 2a with LiCuMe2·LiI in THF
affords only a 8+1 SN2A+SN2 mixture. Compounds 4 (R = Et)
are the intimate precursors of a number of biologically active
molecules showing, for example, inhibition of angiotensin-
converting and epithelial neutral endopetidase enzymes14 or
those showing molluscicidal activity against endoparasite
carrying Biomphalaria glabrata.15
solution of [Cu(MeCN)4]BF4 (6.3 mg, 0.02 mmol, 3 mol%) in THF (0.7
cm3) at 220 °C was treated sequentially with solid 2b (211 mg, 0.67 mmol)
and ZnEt2 (1.0 cm3 of 1.0 M hexane solution, 1.0 mmol). After 40 min the
pale yellow solution was quenched with aqueous HCl (2 M, 3 cm3), the
product was extracted with diethyl ether, and the organic fraction dried
(MgSO4). The solvent was removed and the product assayed directly by 1H
NMR spectroscopy. In all cases the spectra were consistent with a > 20+1
SN2A+SN2 selectivity.
For asymmetric runs, ligand 7 (38 mg, 0.10 mmol, 20 mol%) and
[Cu(MeCN)4]BF4 (15.7 mg, 0.05 mmol, 10 mol%) in dry THF (1 ml) were
stirred at 220 °C in the presence of ZnEt2 (0.10 mmol). Solutions of the
allylic chloride 3b (140.2 mg, 0.52 mmol) in THF (0.55 cm3) and ZnEt2
(0.77 ml of a 1.0 M hexane solution, 0.77 mmol) were added by syringe
pump over 20 min and the reaction stirred for a further 20 min at 220 °C.
The reaction was worked up as above. In cases (Table 1) where the reaction
was not complete unreacted allylic chloride was removed by flash
chromatography or treatment with DABCO.
Preliminary approaches to rendering these reactions viable as
catalytic asymmetric syntheses are also reported in Table 1.†
Relatively high copper and ligand loadings were used to
maximise the asymmetric induction obtained as this is noted to
be a problematic area.9,10 Ligand 7 was selected as an initial
(2)-Ethyl 2-methylene-3-(4-nitrophenyl)pentanoate 4b (R = Et) Yield
21
76% (60% ee); [a]546 284 (c = 0.33, in CHCl3); dH (400 MHz, CDCl3)
0.88 (3 H, t, J = 7.3, CHCH2Me), 1.21 (3 H, t, J = 7.1, OCH2Me), 1.79 (1
H, ddq, J = 13.4, 8.8, 7.3, CHCH2aMe), 1.78 (1 H, ddq, J = 13.4, 6.3, 7.3,
CHCH2bMe), 3.84 (1 H, dd, J = 8.8, 6.3, CHCH2), 4.11 (2 H, m, OCH2Me),
5.76 (1 H, s, NCH2a), 6.41 (1 H, s, NCH2b), 7.39 (2 H, d, J = 8.7, C6H4), 8.14
(2 H, d, J = 8.7, C6H4); dC (67.8 MHz, CDCl3) 12.0 (CHCH2Me), 13.8
(OCH2Me), 26.9 (CHCH2), 47.8 (CH), 60.6 (OCH2), 123.3 (Ar-H), 124.5
(NCH2), 128.8 (Ar-H), 142.4 (Ar-i), 146.3 (NCCO), 150.7 (Ar-i), 166.2
(CO); nmax(thin film)/cm21 2967m, 2936m, 2876 (3 3 C-H), 1714s (CNO),
1520s, 1347s (2 3 NNO), 1254m, 1152m, 851m; m/z (FAB) 264 ([M + H]+,
13%), 221 (14), 207 (16), 147 (38), 77 (13), 73 (100). [Found (HRMS,
FAB): [M + H]+, 264.1244. C14H18NO4 requires M + H, 264.1236].
1 R. M. Magrid, Tetrahedron, 1980, 36, 1901.
2 T. Hayashi, M. Kawatasura and Y. Kozumi, Chem. Commun., 1997,
561.
3 J. P. Janssen and G. Helmchen, Tetrahedron Lett., 1997, 38, 8025.
4 R. Pretôt and A. Pfaltz, Angew. Chem., Int. Ed., 1998, 37, 323.
5 R. Pretôt, G. C. Lloyd-Jones and A. Pfaltz, Pure Appl. Chem., 1998, 70,
1035.
6 B. M. Trost and I. Hachiya, J. Am. Chem. Soc., 1998, 120, 1104.
7 L.-H. Xu and E. P. Kündig, Helv. Chim. Acta, 1994, 77, 1480.
8 F. Lambert, B. Kirschleger and J. Villieras, J. Organomet. Chem., 1991,
406, 71.
9 J. E. Bäckvall, S. E. Karlström, M. van Klaveren, G. van Koten, G. J.
Meuzelaar, E. S. M. Persson and A. del Villar, Tetrahedron, 2000, 56,
2895 [18–53% ee values attained].
10 F. Dübner and P. Knochel, Angew. Chem., Int. Ed. Engl., 1999, 38, 379;
P. Knochel and F. Duebner, (Avecia Limited, UK), PCT Int. Appl., 9
March, 2000, [Chem. Abs., 2000, 132, 222063; 2–87% ee values
attained].
11 D. Basavaiah, P. D. Rao and R. S. Hyma, Tetrahedron, 1996, 52,
8001.
12 F. Ameer, S. E. Drewes, N. D. Emslie, P. T. Kaye and R. L. Mann,
J. Chem. Soc., Perkin Trans. 1, 1983, 2293.
13 G. Kubas, Inorg. Synth., 1990, 28, 68.
candidate based on its efficacy in asymmetric conjugate
addition.16 Varying the leaving group indicated chloride to be
the best leaving group with respect to enantioselectivity,
although the chemical yield suffered somewhat in this case.
Very high chemical yields were realised with the mesylate 6, but
the product 4a is essentially racemic, while the formate 5 does
not participate in the reaction. Ligand 7 was confirmed as the
optimal structure by screening a small library of compounds
against a test reaction of 3a with ZnEt2; none of the other
structures lead to very active catalysts. Similarly, changing to a
terminal AlEt3 organometallic source is not tolerated.
In conclusion, a new type of efficient catalytic SN2A chemistry
has been developed. The degree of stereocontrol realised in
these reactions appears to be due more to electronic than steric
factors, however, more experiments are required before the
details of the asymmetric transition state become clear. These
studies together with applications of the compounds 4 to the
synthesis of biologically active compounds are underway in our
laboratories.
14 C. A. Almansa, A. F. de Arriba, E. Carceller, F. L. Cavalcanti, J. Forn,
J. Garcia-Rafanell, L. A. Gomez and R. Rodriguez, J. Med. Chem.,
1996, 39, 2197.
15 G. Ruecker, K. Hostettmann, W. Gajewski, M. Loebbert and P. Boeken,
Arch. Pharm. (Weinheim, Ger.), 1993, 326, 941.
16 S. M. W. Bennett, S. M. Brown, A. Cunningham, M. R. Dennis, J. P.
Muxworthy, M. A. Oakley and S. Woodward, Tetrahedron, 2000, 56,
2847.
We thank the EPSRC for support of this project through
grants GR/M75341, GR/M84909, GR/N37339 and for access to
their Mass Spectrometry Service (University of Swansea). S. G.
and S. W. are grateful to the EU for support through COST
(working groups D12/0009/98 and D12/0022/99) and SOC-
RATES. J. G. acknowledges the support of the Generalitat de
Catalunya.
2434
Chem. Commun., 2000, 2433–2434