A catalytic and enantioselective desymmetrization of meso cyclic allylic bisdiethylphosphates with organozinc reagents

Article abstract of DOI:10.1002/anie.200390088

The chosen one: The desymmetrization of meso-2 is possible with the use of diethylzinc in the presence of the Cu' complex of 1 (10 mol%; Tf = CF₃SO₂). Ligand 1 was optimized from a screen of a library of 125 compounds. Other ligands

Full text of DOI:10.1002/anie.200390088

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Communications
Desymmetrization of meso Compounds
R1
O
O
NHR2
S
N
A Catalytic and Enantioselective
Desymmetrization of meso Cyclic Allylic
Bisdiethylphosphates with Organozinc
Reagents**
OH
R3
1afk–ejo
a: R1 = Me
b: R1 = iPr
c: R1 = iBu
k: R3 = H
f: R2 = CH2Ph
l: R3 = 3,5-tBu2
m: R3 = 3,5-Cl2
n: R3 = 5,6-(CH)4-
o: R3 = 3-Ph
g: R2 = (R)-CH(Me)Cy
h: R2 = (S)-CH(Me)Cy
i: R2 = iPr
Umberto Piarulli,* Philippe Daubos,
Christelle Claverie, Maryline Roux, and
Cesare Gennari*
d: R1 = CH2Ph
e: R1 = tBu
j: R2 = CHPh2
Catalytic enantioselective desymmetrization of meso com-
pounds is a powerful tool for the construction of enantiomeri-
cally enriched functionalized products.[1] meso Cyclic allylic
diol derivatives are challenging substrates for the asymmetric
allylic substitution reaction,[2] owing to the potential compe-
tition of several reaction pathways. In particular, SN2’ and SN2
substitutions can occur, and both with either retention or
inversion of stereochemistry. In the case of SN2 substitution, in
which an allylic alcohol derivative is obtained, a second allylic
substitution might occur through the SN2’ or SN2 mechanism,
with either retention or inversion of stereochemistry. Based
on this complex scenario, up to 15 isomers (seven pairs of
enantiomers and one meso compound) could, in principle, be
obtained.
Herein we present a new highly regio-, diastereo-, and
enantioselective desymmetrization of meso, cyclic allylic
bisdiethylphosphates with organozinc reagents[3] catalyzed
by copper(i) complexes of chiral Schiff base ligands 1.[4] cis-4-
Cyclopentene-1,3-diol was transformed into the correspond-
ing bisdiethylphosphate 2 by deprotonation with nBuLi and
reaction with diethylchlorophosphate in THF/TMEDA
(4:1).[5] Reaction of meso-4-cyclopentene-1,3-bisdiethylphos-
phate (2) with diethylzinc in the presence of (CuOTf)2¥C6H6
(Tf ¼ CF3SO2) (10 mol%) and chiral ligand 1cjl in toluene/
THF (95:5 v/v) at À788C afforded only the product arising
from the SN2’ mechanism with inversion of stereochemistry,
with an enantiomeric ratio 3/4 of 87:13 in favor of the S,S
enantiomer (Scheme 1).[6 Other copper sources (CuCN,
Cu(OTf)2) and other solvents (pure toluene, pure THF,
CH2Cl2, n-hexane) gave lower yields and poorer selectivities.
A library of 125 ligands 1[4c] was screened: CuI complexes
were preformed in situ by stirring a solution of ligand 1
(10 mol%) with (CuOTf)2¥C6H6 (10 mol%) in toluene/THF
(95:5) at room temperature. Diethylzinc (solution in toluene)
and 4-cyclopentene-1,3-bisdiethylphosphate (2) were then
added to the mixture at À788C, and the reaction mixture was
stirred for 15 h before quenching. The most interesting results
are shown in Table 1: The best enantiomeric ratio (94:6) in
favor of the S,S enantiomer 3 was observed in the presence of
ligands 1cjo and 1cjm.[7,8] We found that an increase in the
temperature to À608C did not have a detrimental effect on
the enantioselectivity. Instead, complete conversion and
almost quantitative yield were observed (for example, ligand
1cjo: > 98% yield, 88% ee; ligand 1cjm: > 98% yield,
88% ee; ligand 1cjl: 80% yield, 74% ee). Interestingly,
ligands with different steric hindrance but with the same
absolute configuration at the stereogenic center bearing R1
may lead to opposite enantiomeric ratios (!) (Table 1, entries
7 9). An enantiomeric ratio of up to 76:24 in favor of (R,R)-4
was obtained in the presence of ligand 1egk. As a rule of
thumb, substituted salicylaldehydes (R3 ¼ 3,5-Cl2; 3-Ph; 3,5-
tBu; 5,6-(CH)4-), bulky amines (R2 ¼ CHPh2), and relatively
small substituents at the stereogenic center (R1 ¼ iBu, Me)
favor the formation of enantiomer (S,S)-3, whereas unsub-
stituted salicylaldehydes (R3 ¼ H) and relatively small amines
(e.g. R2 ¼ CH2Ph) tend to favor the formation of enantiomer
(R,R)-4.
8]
[*] Dr. U. Piarulli
Dipartimento di Scienze Chimiche, Fisiche e Matematiche
Universit‡ dell©Insubria
via Valleggio 11, 22100 Como (Italy)
Fax: (þ39)031-238-6449
To investigate the scope of this new reaction, different
organozinc reagents were tested with bisdiethylphosphate 2 in
the presence of the ligands that gave the best results in the
previous screening (1cjo, 1cjm) (Scheme 2). In the case of
dimethylzinc, the reaction gave exclusively the product
E-mail: umberto.piarulli@uninsubria.it
Prof. Dr. C. Gennari, Dr. P. Daubos, Dr. C. Claverie, Dr. M. Roux
Dipartimento di Chimica Organica e Industriale
Centro di Eccellenza C.I.S.I., Universit‡ di Milano
Istituto di Scienze e Tecnologie Molecolari (ISTM) del CNR
via G. Venezian 21, 20133 Milano (Italy)
Fax: (þ39)02-5031-4072
O
O
O
O
E-mail: cesare.gennari@unimi.it
OP(OEt)2
(EtO)2PO
(EtO)2PO
OP(OEt)2
a
[**] We thank the European Commission (IHP Network grant ™Combi-
Cat∫ HPRN-CT-2000-00014) for financial support and postdoctoral
fellowships to M. Roux (HPRN-CT-2000-00014), C. Claverie (HPRN-
CT-2000-00014), and P. Daubos (™Marie Curie∫ HPMF-CT-2001-
01318). We also thank ™Merck Research Laboratories∫ (Merck©s
Academic Development Program Award to C. Gennari, 2001) for
financial support. U. Piarulli thanks the Dipartimento di Chimica
Organica e Industriale (Milan University) for their hospitality.
+
meso-2
3
4
Scheme 1. Enantioselective allylic alkylation of 2 with Et2Zn, catalyzed
by (CuOTf)2¥C6H6/1. Screening of the library of ligands 1.
a) 1) (CuOTf)2¥C6H6 (10 mol%), 1 (10 mol%), toluene/THF (95:5),
room temperature, 45 min; 2) Et2Zn (1.1m in toluene), À788C, 15 h.
234
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4202-0234 $ 20.00+.50/0
Angew. Chem. Int. Ed. 2003, 42, No. 2

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Angewandte
Chemie
Table 1: Selected results from the high-throughput screening of the library of ligands 1.[a]
from the SN2’ substitution with
inversion of stereochemistry (10 þ
11) are formed, with moderate
enantiomeric excess (e.g. 56% ee
with ligand 1cjk, À608C).[10]
Entry
1
R1
R2
R3
3/4
Yield [%]
1
2
3
4
5
6
7
8
9
cjo
cjm
cjl
ajm
cjk
cjn
afk
bfk
egk
iBu
iBu
iBu
Me
iBu
iBu
Me
iPr
CHPh2
CHPh2
CHPh2
CHPh2
CHPh2
CHPh2
CH2Ph
CH2Ph
3-Ph
3,5-Cl2
3,5-tBu2
3,5-Cl2
H
5,6-(CH)4-
H
H
94:6
94:6
54
47
42
62
54
49
13
12
26
87:13
86:14
84:16
83:17
31:69
30:70
24:76
In conclusion, we have disclosed
a new highly regio-, diastereo-, and
enantioselective desymmetrization
of meso cyclic allylic bisdiethyl-
phosphates with organozinc re-
agents catalyzed by copper(i) com-
plexes of chiral Schiff base ligands
1. Further investigations into the
scope and limitations of this reac-
tion are currently underway.
tBu
(R)-CH(Me)Cy
H
[a] [CuOTf]2¥C6H6 (0.1 equiv), 1 (0.1 equiv), Et2Zn (2.0 equiv), 2 (1.0 equiv), toluene/THF (95:5), À788C,
15 h.
O
O
O
O
OP(OEt)2
(EtO)2PO
OP(OEt)2
(EtO)2PO
a
+
R
R
Experimental Section
5 (R = Me)
7 (R = Ph)
6 (R = Me)
8 (R = Ph)
meso-2
General Procedure: Ligand 1 (0.017 mmol) was dissolved in dry
toluene/THF (95:5 v/v; 1.5 mL) in a flame-dried flask under argon.
(CuOTf)2¥C6H6 (4.7 mg, 0.017 mmol) was subsequently added, and
the resulting greenish solution was stirred at room temperature for
45 min. The reaction mixture was cooled to À788C and treated with
Et2Zn (1.1m solution in toluene; 0.310 mL, 0.340 mmol). After
10 min, meso-2 (60 mg, 0.170 mmol) was added. The reaction mixture
was stirred at À788C for 15 h, then quenched with a saturated
aqueous solution of NH4Cl (1 mL), and diluted with ethyl acetate
(1 mL). The organic phase was separated and filtered through celite.
n-Decane (0.033 mL, 0.170 mmol) was added, and a sample of the
crude reaction mixture (1 mL) was then injected into a GC instrument
equipped with a chiral capillary column for determination of yields
and enantiomeric ratios (3/4). Column: MEGADEX DMEPEb, OV
1701, 25 m, film 0.25 mm; carrier: H2 (70 kPa); injector 2508C;
detector 2508C; oven temperature 1108C, 0.88CminÀ1 to 1408C; tR:
0.91 min (n-decane), 17.7 min ((1R,2R)-4), 18.0 min ((1S,2S)-3), and
39.8 min (meso-2).
Scheme 2. Enantioselective allylic alkylation of 2 with R2Zn, catalyzed
by (CuOTf)2¥C6H6/1cjo or (CuOTf)2¥C6H6/1cjm. a) 1) (CuOTf)2¥C6H6
(10 mol%), 1cjo or 1cjm (10 mol%), toluene/THF (95:5), room tem-
perature, 45 min; 2) R2Zn, À608C, 15 h.
O
O
n
n
OP(OEt)2
OP(OEt)2
Et
Et
O
O
10
11
a
n
(EtO)2PO
OP(OEt)2
O
O
9
n
n
OP(OEt)2
OP(OEt)2
Et
Et
12
13
Scheme 3. Allylic alkylation of 9 with Et2Zn, catalyzed by
Received: October 10, 2002 [Z50333]
(CuOTf)2¥C6H6/1. a) 1) (CuOTf)2¥C6H6 (10 mol%), 1 (10 mol%), room
temperature, 45 min; 2) Et2Zn, À78 or À608C, 15 h.
[1] For a recent review, see: M. C. Willis, J.Chem.Soc.Perkin
Trans.1 1999, 1765 1784.
[2] For leading examples of desymmetrization of meso cyclic allylic
diol derivatives (esters, carbonates, and the closely related meso
oxabicyclic alkenes), see: a) B. M. Trost, J. Dudash, Jr., E. J.
Hembre, Chem.Eur.J. 2001, 7, 1619 1629; b) B. M. Trost, D. E.
arising from the SN2’ substitution with inversion of stereo-
chemistry (ligand 1cjm, À608C), in moderate yield (40%)
and excellent enantiomeric ratio (5/6 97:3) in favor of the S,S
enantiomer (5, R ¼ Me).[7,8] Allylic phenylation was possible
in the reaction of bisdiethylphosphate 2 with a mixture of
diphenylzinc and dimethylzinc (2:1).[9] The phenyl group was
preferably transferred (Ph transfer vs. Me transfer¼ 48:1),
giving the product of SN2’ substitution with inversion of
stereochemistry in moderate yield (60%) and fair enantio-
meric ratio (7/8 84:16 with ligand 1cjo, À608C) in favor of the
S,R enantiomer (7, R ¼ Ph).[7,10]
Patterson, E. J. Hembre, Chem.Eur.J.
2001, 7, 3768 3775;
c) B. M. Trost, J. Dudash, Jr., O. Dirat, Chem.Eur.J. 2002, 8,
259 268; d) M. Lautens, J.-L. Renaud, S. Hiebert, J.Am.Chem.
Soc. 2000, 122, 1804 1805; e) M. Lautens, S. Hiebert, J.-L.
Renaud, Org.Lett. 2000, 2, 1971 1973; f) M. Lautens, S.
Hiebert, J.-L. Renaud, J.Am.Chem.Soc.
2001, 123, 6834
6839; g) M. Lautens, C. Dockendorff, K. Fagnou, A. Malicki,
Org.Lett. 2002, 4, 1311 1314, and references therein; h) F.
Bertozzi, M. Pineschi, F. Macchia, L. A. Arnold, A. J. Minnaard,
B. L. Feringa, Org.Lett. 2002, 4, 2703 2705.
Reaction of diethylzinc with cis-2-cyclohexene-1,4-bisdi-
ethylphosphate[11] (9, n ¼ 1) (Scheme 3) gave the SN2’ prod-
ucts originating from either inversion (10 and 11) or retention
of stereochemistry (12 and 13) with good diastereoselectivity
(81:19 4:96), depending on the solvent and the ligand used.
However, racemic mixtures were invariably produced.[12]
Preliminary studies with cis-2-cycloheptene-1,4-bisdiethyl-
phosphate[11] (9, n ¼ 2) indicate that only the products arising
[3] For copper-catalyzed allylic alkylations in the presence of
dialkylzinc reagents, see: a) F. D¸bner, P. Knochel, Angew.
Chem. 1999, 111, 391 393; Angew.Chem.Int.Ed. 1999, 38, 379
381; b) C. A. Luchaco-Cullis, H. Mizutani, K. E. Murphy, A. H.
Hoveyda, Angew.Chem. 2001, 113, 1504 1508; Angew.Chem.
Int.Ed. 2001, 40, 1456 1460; c) H. Malda, A. W. van Zijl, L. A.
Arnold, B. L. Feringa, Org.Lett. 2001, 3, 1169 1171; d) S.
Angew. Chem. Int. Ed. 2003, 42, No. 2
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4202-0235 $ 20.00+.50/0
235

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Communications
Ongeri, U. Piarulli, M. Roux, C. Monti, C. Gennari, Helv.Chim.
Acta 2002, 3388 3399.
ee values (column: MEGADEX DMEPEb, OV 1701, 25 m, film
0.25 mm, carrier: H2 (70 kPa)). For the determination of 3/4 ratio,
see Experimental Section. Determination of 5/6 (R ¼ Me) ratio:
injector 2508C, detector 2508C, oven temperature 908C,
0.88CminÀ1 to 1308C, tR: 26.2 min ((1R,2R)-6) and 26.7 min
((1S,2S)-5).
[4] a) I. Chataigner, C. Gennari, U. Piarulli, S. Ceccarelli, Angew.
Chem. 2000, 112, 953 956; Angew.Chem.Int.Ed. 2000, 39, 916
918; b) S. Ongeri, U. Piarulli, R. F. W. Jackson, C. Gennari, Eur.
J.Org.Chem. 2001, 803 807; c) I. Chataigner, C. Gennari, S.
Ongeri, U. Piarulli, S. Ceccarelli, Chem.Eur.J. 2001, 7, 2628
2634.
[9] a) C. Bolm, N. Hermanns, J. P. Hildebrand, K. MuÊiz, Angew.
Chem. 2000, 112, 3607 3609; Angew.Chem.Int.Ed. 2000, 39,
3465 3467; b) M. Schinnerl, M. Seitz, A. Kaiser, O. Reiser, Org.
Lett. 2001, 3, 4259 4262.
[5] K. L. Yu, K. Y. Ko, B. Fraser-Reid, Synth.Commun. 1988, 465
468.
[6] The chiral copper complex appears to be essential to obtain the
reaction product. In fact, the addition of Et2Zn in the presence of
(CuOTf)2¥C6H6 and in the absence of the chiral ligand gave only
starting material and minute amounts of 3/4 (8%).
[10] The enantiomeric ratios (7/8 (R ¼ Ph), 10:11, n ¼ 2) was
determined after reduction of the phosphate ester with LiAlH4
in diethyl ether (quant. yield), and derivatization of the resulting
alcohol with (R)-(þ)-a-methoxy-a-(trifluoromethyl)phenylace-
tic acid, dicyclohexyl carbodiimide, 4-dimethylaminopyridine in
dichloromethane (quant. yield). The ratio of the two diastereo-
meric Mosher esters was assessed by 1H NMR spectroscopy; see:
[7] CuCN (3.5 equiv) mediated allylic alkylation of (1R,3S)-(þ)-cis-
4-cyclopentene-1,3-diol 1-acetate with EtMgCl (3.0 equiv) in
THF at À18!08C gave (1S,2S)-trans-2-ethyl-3-cyclopenten-1-ol
selectively, through SN2’ substitution of the acetate with inver-
sion. This compound was then reacted with (EtO)2POCl
(pyridine, DMAP, CH2Cl2) to give enantiomerically pure 3
[a]D ¼ þ 84.08 (CHCl3, c ¼ 1.2). The above synthetic sequence
was also performed with MeMgCl and PhMgCl instead of
EtMgCl, yielding enantiomerically pure (1S,2S)-5 (R ¼ Me) and
(1S,2R)-7 (R ¼ Ph). See: a) M. Ito, M. G. Murugesh, Y. Kobaya-
shi, Tetrahedron Lett. 2001, 42, 423 427; b) M. Ito, M. Matsuu-
a) J. A. Dale, H. S. Mosher, J.Am.Chem.Soc.
1973, 95, 512
519; b) G. R. Sullivan, J. A. Dale, H. S. Mosher, J.Org.Chem.
1973, 38, 2143 2147.
[11] J. E. B‰ckvall, S. E. Bystrˆm, R. E. Nordberg, J.Org.Chem.
1984, 49, 4619 4631.
[12] At present we do not have a convincing rationale for this
unexpected result. However, a possible reason for the different
behavior of the cycloalkene derivatives 2 and 9 (n ¼ 1, n ¼ 2)
might be related to their symmetry. Indeed, whereas the
cyclopentene derivative 2 and the cycloheptene derivative 9
mi, M. G. Murugesh, Y. Kobayashi, J.Org.Chem.
2001, 66,
5881 5889; c) Y. Kobayashi, M. Ito, J. Igarashi, Tetrahedron
Lett. 2002, 43, 4829 4832.
(n ¼ 2) can exist and react in
a CS conformation (plane
symmetric), the cyclohexene derivative 9 (n ¼ 1) is better
[8] The crude reaction mixture was injected into a GC instrument
equipped with a chiral capillary column for the determination of
described as a mixture of readily interconverting enantiomers.
236
¹ 2003 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1433-7851/03/4202-0236 $ 20.00+.50/0
Angew. Chem. Int. Ed. 2003, 42, No. 2