Novel ferrocene modified P,N-ligands for enantioselective palladium-catalyzed allylic substitution reactions

Article abstract of DOI:10.1016/S0022-328X(01)00911-1

Novel ferrocene modified P,N-ligands (R,R,Sp)-8 and (S,S,Sp)-9 were synthesized conveniently from enantiopure cyclohexanediamine and (Sp)-2-phosphinoferrocenyl carboxylic acid. They were used as chiral ligands for palladium-catalyzed allylic substitutions of rac-1, 3-diphenylprop-2-enyl acetate with dimethylmalonate and benzylamine. (R,R,Sp)-8 was found to be a better ligand containing matched chiralities (central chirality and planar chirality) and showed a higher enantioselectivity.

Full text of DOI:10.1016/S0022-328X(01)00911-1

Journal of Organometallic Chemistry 637–639 (2001) 762–766
www.elsevier.com/locate/jorganchem
Note
Novel ferrocene modified P,N-ligands for enantioselective
palladium-catalyzed allylic substitution reactions
Shu-Li You, Xue-Long Hou *, Li-Xin Dai
Laboratory of Organometallic Chemistry, Shanghai Institute of Organic Chemistry, Chinese Academy of Sciences, 354 Fenglin Lu,
Shanghai 200032, People’s Republic of China
Received 2 January 2001; received in revised form 29 January 2001; accepted 29 January 2001
Abstract
Novel ferrocene modified P,N-ligands (R,R,Sp)-8 and (S,S,Sp)-9 were synthesized conveniently from enantiopure cyclohexane-
diamine and (Sp)-2-phosphinoferrocenyl carboxylic acid. They were used as chiral ligands for palladium-catalyzed allylic
substitutions of rac-1, 3-diphenylprop-2-enyl acetate with dimethylmalonate and benzylamine. (R,R,Sp)-8 was found to be a better
ligand containing matched chiralities (central chirality and planar chirality) and showed a higher enantioselectivity. © 2001
Elsevier Science B.V. All rights reserved.
Keywords: Alkylation; Amination; Ferrocenyl-derived ligands; Planar chirality
ity and 82% ee. However, poor regioselectivities and zero
enantioselectivity were obtained with the corresponding
1. Introduction
phenyl derived ligands. Bolm [5] studied the role of
planar chirality on addition reaction of diethylzinc to
aldehydes with the hydroxyl-ferrocenyl-oxazoline ligand
2. At the same time, we investigated the role of planar
chirality in palladium-catalyzed allylic alkylation reac-
tion with thioether ferrocenyl oxazoline 3 [6]. Both Bolm
and us compared the results with that from the corre-
sponding phenyl derived ligands and concluded that
introduction of a matched chirality on the ferrocene
system often increases the enantioselectivity. In addition
to the increase of enantioselectivity the ferrocenyl-
derived ligands frequently showed higher reactivity in
some metal-catalyzed reactions than that of phenyl
derived ligands [6a].
Ligand 4 has been proved very effective in asymmetric
palladium-catalyzed allylic substitutions [7,8]. Recently,
ligands 6 and 7, the ferrocene analogues of ligand 4,
containing two kinds of chiralities were synthesized by
Zhang [9c] and us [9a,b]. Among them, the chiralities in
ligand 6 were found to be matched and better results
were provided by comparison with ligand 4 in some
cases [9a,b]. We have demonstrated that the presence of
an additional proper planar chirality would improve the
Chirotechnology is one of the most interesting and
challenging technologies of recent years [1]. The recogni-
tion of the importance of chirality in biological systems
has made the production of enantiomerically pure chiral
drugs to be an important requisite in drug design and
development for the pharmaceutical industry. As a
consequence, great efforts have been made towards the
asymmetric catalytic processes by chemists. Therein
design and synthesis of chiral ligands are always the key
point. Among them, ferrocene containing ligands are
one kind of the most interesting ligands because of the
easily introduction of planar chirality and the inherent
special electronic properties of ferrocene itself [2]. In
comparison with the phenyl analogues, the ferrocenyl
derived ligands with planar chirality usually showed
higher selectivity and reactivity [3]. Sammakia [4] found
that ferrocene derived ligand 1 was efficient in copper
catalyzed conjugate addition of Grignard reagents to
enones. The product was formed with high regioselectiv-
* Corresponding author. Tel.: +86-21-6416-3300; fax: +86-21-
6416-6128.
E-mail address: xlhou@pub.sioc.ac.cn (X.-L. Hou).
0022-328X/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0022-328X(01)00911-1

S.-L. You et al. / Journal of Organometallic Chemistry 637–639 (2001) 762–766
763
efficiency of asymmetric induction [10,11]. Ligand 5 was
also found to be effective in palladium-catalyzed alkyla-
tion recently [12]. It was applied into enantioselective
cyclization to form 9-azabicyclo[4,2,1] nonane skeleton.
The later was the key intermediate for synthesis of
(−)-anatoxin-a [12]. The former success encouraged our
further efforts to introduce the planar chiral ferrocene
moiety into this ligand. In this paper, we report the
synthesis of ferrocene modified P,N-ligands (R,R,Sp)-8
and (S,S,Sp)-9 and their applications as chiral ligands in
palladium-catalyzed asymmetric allylic substitutions.
(R,R,Sp)-11 with DCC and 2-Pyridine carboxylic acid
gave rise to the desired chiral ligand (R,R,Sp)-8 in high
yield. Following the same methods, (S,S,Sp)-9 was
synthesized from (S,S)-cyclohexanediamine in moderate
yield.
To examine the catalytic efficiency of 8 and 9 as chiral
ligands in palladium-catalyzed allylic substitution, the
usual model reaction, reaction of 1,3-diphenylprop-2-
enyl acetate with the nucleophile derived from dimethyl-
malonate was chosen (Scheme 2) and the results are
summarized in Table 1.
From the results listed in Table 1, we found that the
reaction ran smoothly with ligand (R,R,Sp)-8 in THF or
CH2Cl2 to give the product 13 in high yield and moder-
ate enantioselectivity (entries 1 and 2 in Table 1). The
enantioselectivity was raised to 85% ee when Cs2CO3
was used as the base in CH2Cl2 (entry 3 in Table 1).
Change of solvent from CH2Cl2 to CH3CN or toluene
resulted in the decrease of both yield and enantioselectiv-
ity (entries 4 and 5 in Table 1). It is interesting that
change of configuration of central chirality from (R,R)
in ligand 8 to (S,S) in 9 does not invert the configuration
2. Results and discussion
The synthesis of ligands 8 and 9 was straightforward
and is shown in Scheme 1. (Sp)-2-Phosphinoferrocenyl
carboxylic acid 10 is a key intermediate, which was
synthesized from phosphinoferrocenyl oxazoline easily
as reported previously [9a]. Condensation of (Sp)-10 and
(R,R)-cyclohexanediamine in the presence of DCC af-
forded (R,R,Sp)-11 in moderate yield. Treatment of
Scheme 1. Reagent and conditions: (a) (R,R)-cyclohexanediamine, DCC, DMAP, DCM, 45%; (b) 2-pyridinyl carboxylic acid, DCC, DMAP,
DCM, 84%; (c) (R,R)-cyclohexanediamine, DCC, DMAP, DCM and procedure b, 35% (two steps).

764
S.-L. You et al. / Journal of Organometallic Chemistry 637–639 (2001) 762–766
Scheme 2.
of product but decreases the enantioselectivity (entries 3
and 6). These results demonstrate that the planar chiral-
ity is responsible mainly for dominating both the
configuration and enantioselectivity of the product. It is
also indicated that (R,R,Sp)-8 possessed matched
chiralities.
In order to reveal the efficiency of the ligands and also
the role of the planar chirality further, asymmetric allylic
amination reaction was carried out by using BnNH2 as
nucleophile (Scheme 3). The results were listed in Table
2. Both ligand 8 and 9 gave moderate yield and enan-
tioselectivity even the reaction was performed at 40 °C.
Ligand 8 gave better yield and enantioselectivity still
because the presence of matched chiralities.
Although fair enantioselectivities can be achieved with
(R,R,Sp)-8 in palladium-catalyzed alkylation, there is no
significant increase on enantioselectivity compared with
phenyl-derived ligand 5 (82% yield and 86% ee). We
believe that better results may well be achieved in other
type of asymmetric syntheses.
standard method. The commercially available reagents
were used as received without further purification.
1
Melting points are uncorrected. H-NMR spectra were
recorded on a Bruker AMX-400 (400 MHz) spectrome-
ter in CDCl3 at room temperature (r.t.). 31P-NMR
spectra were recorded on a Bruker AMX-400 (162
MHz) spectrometer and the chemical shifts were refer-
enced to external 85% H3PO4. Chemical shifts are given
in parts per million relative to tetramethylsilane as an
internal standard. Optical rotations were measured us-
ing a Perkin–Elmer 241 MC polarimeter with a ther-
mally jacketed 10 cm cell at 25 °C (concentration c
given as g 100 ml−1). IR spectra were measured in
cm−1, using a Shimadzu IR-440 infrared spectrophoto-
meter. Mass spectra and high-resolution mass spectra
were taken using HP5989A and Finnigan MAT mass
spectrometers, respectively. Elemental analyses were
performed on a Foss-Heraeus Vario EL instrument. Ee
values were determined by chiral HPLC on Chiralcel
OD or OJ column.
4.2. (R,R,Sp)-N-[2-(Diphenylphosphino)ferrocenyl-
carbonyl]-diaminocyclohexane (11)
3. Conclusions
Novel ferrocene modified P,N-ligands (R,R,Sp)-8
and (S,S,Sp)-9 were synthesized, and their catalytic
efficiency for palladium-catalyzed allylic alkylation and
amination of rac-1,3-diphenyl-2-propenyl acetate were
examined. In both reactions, (R,R,Sp)-8 was found to
be the better one of the two, perhaps because it con-
tains matched chiralities. Moderate to good enantiose-
lectivity was achieved with this ligand in palladium
catalyzed allylic amination and alkylation, respectively.
Unfortunately, no significant increase of the enantiose-
lectivity or activity was observed with two ferrocenyl-
derived ligands than that of the phenyl-derived ligand.
Further study with these ligands in other metal-cata-
lyzed asymmetric reactions is in progress.
To a solution of (Sp)-2-phosphinoferrocenyl car-
boxylic acid (10) (414 mg, 1 mmol), (R,R)-cyclohexane-
diamine (23 mg, 2 mmol) and DMAP (5.5 mg, 0.05
mmol) in dry dichloromethane (5 ml) were added DCC
(226 mg, 1 mmol) at r.t. After the completion of the
material, the reaction mixture was diluted with 15 ml
Table 1
Asymmetric palladium-catalyzed allylic alkylation with rac-1,3-
diphenyl-2-propenyl acetate and dimethylmalonate a
Entry
Ligand
Solvent
Base
Yield (%) b
ee (%) c
1
2
3
4
5
6
8
8
8
8
8
9
THF
NaH
BSA
Cs2CO3 84
Cs2CO3 53
Cs2CO3 23
Cs2CO3 86
95
98
72 (R)
74 (R)
85 (R)
79 (R)
80 (R)
69 (R)
CH2Cl2
CH2Cl2
CH3CN
Toluene
CH2Cl2
4. Experimental
a Molecular ratio: [Pd(h3-C3H5)Cl]2/ligand/13/CH2(CO2Me)2/
Base=2/6/100/300/300.
4.1. General methods
b Isolated yield based on 1,3-diphenyl-2-propenyl acetate.
c Determined by HPLC (chiracel OD column) and the absolute
configuration of product was assigned through comparison of the
sign of specific rotations with the literature data [13].
All reactions were performed under an atmosphere of
either argon or nitrogen using oven-dried glassware.
Solvents were treated prior to use according to the

S.-L. You et al. / Journal of Organometallic Chemistry 637–639 (2001) 762–766
765
Scheme 3.
dichloromethane and quenched with saturated aqueous
4.4. (S,S,Sp)-N-[2-(Diphenylphosphino)
ferrocenylcarbonyl]-N-[2-pyridinylcarbonyl]-
diaminocyclohexane (9)
NaHCO3 solution. The organic layer was extracted
twice with dichloromethane. The combined organic
layer was dried over Na2SO4 and concentrated in vacuo
to give the crude product, which was subsequently
purified by column chromatography (EtOAc:Et3N=
1:1) to give the main compound 11 (230 mg, 45%) as an
A similar procedure as for (R,R,Sp)-8 but from
(S,S)-cyclohexanediamine gave (S,S,Sp)-9 in 35% yield
1
(two steps) as an orange solid: H-NMR l 8.59–8.61
1
orange solid: H-NMR l 7.42–7.55 (m, 5H), 7.18–7.19
(m, 1H), 8.40 (d, 1H, J=8.3 Hz), 8.23 (d, 1H, J=7.8
Hz), 7.81 (td, 1H, J1=7.7 Hz; J2=1.7 Hz), 7.51 (td,
2H, J1=7.8 Hz; J2=1.6 Hz), 7.37–7.47 (m, 5H), 7.12–
7.22 (m, 5H), 5.02 (m, 1H), 4.35 (m, 1H), 3.99–4.08 (m,
1H), 3.92(s, 5H), 3.70 (m, 1H), 2.20–2.22 (m, 1H),
1.90–1.93 (m, 1H), 1.67–1.79 (m, 2H), 1.17–1.44 (m,
5H); 31P-NMR l −20.02; MS (EI) m/z (rel intensity)
615 ([M+], 3), 550 (100), 412 (8); IR (KBr, cm−1) 3302
(m), 3053 (w), 2930 (m), 1645 (s), 1589 (s), 1523 (s),
1434 (m); [h]2D0= −108° (c 0.45, CHCl3); m.p. 94–
96 °C; Anal Calc. for C35H34FeN3O2P: C, 68.30; H,
5.57; N, 6.82. Found: C, 68.10; H, 5.62; N, 6.62%.
(m, 6H), 5.18 (t, 1H, J=1.2 Hz), 4.45 (t, 1H, J=2.5
Hz), 4.15( s, 5H), 3.71 (m, 1H), 3.55–3.62 (m, 1H), 2.28
(td, 1H, J1=10.8 Hz; J2=4.0 Hz), 2.13–2.16 (m, 1H),
1.88–1.92 (m, 1H), 1.74–1.77 (m, 2H), 1.01–1.45 (m,
6H); 31P-NMR l –19.92; MS (EI) m/z (rel intensity)
510 ([M+], 9), 445 (100), 411 (32), 346 (8), 309 (13), 201
(9), 121 (5); IR (KBr, cm−1) 3297 (m), 2926 (m), 1737
(w), 1634 (s), 1434 (m), 1262 (m), 818 (s); [h]2D0=
−223° (c 0.55, CHCl3); m.p. 70–73 °C; HRMS Calc.
for C29H31FeN2OP: 510.15530. Found: 510.15693.
4.3. (R,R,Sp)-N-[2-(Diphenylphosphino)-
ferrocenylcarbonyl]-N-[2-pyridinyl carbonyl]-diamino
cyclohexane (8)
4.5. General procedure for the palladium-catalyzed
allylic alkylation of rac-1,3-diphenyl-2-propenyl acetate
To a solution of (R,R,Sp)-11 (204 mg, 0.4 mmol),
2-pyridine carboxylic acid (98 mg, 0.8 mmol) and
DMAP (2.2 mg, 0.02 mmol) in dry dichloromethane (5
ml) were added DCC (181 mg, 0.8 mmol) at r.t. After
the completion of the material, the reaction mixture
was diluted with 10 ml dichloromethane and quenched
with saturated aqueous NaHCO3 solution. The organic
layer was extracted twice with dichloromethane. The
combined organic layer was dried over Na2SO4 and
concentrated in vacuo to give the crude product, which
was subsequently purified by column chromatography
(EtOAc:Et3N=1:1) to give the main compound 8 (207
[Pd(h3-C3H5)Cl]2 (3.7 mg, 0.01 mmol) and the ligand
(18.5 mg, 0.03 mmol) were dissolved in dry CH2Cl2 (2
ml), and then stirred for 30 min at r.t. under an
atmosphere of argon. To this solution rac-1,3-diphenyl-
2-propenyl acetate (126 mg, 0.5 mmol), dimethyl-
malonate (0.17 ml, 1.5 mmol), and proper base (1.5
mmol) were added successively. The reaction mixture
was stirred at r.t. and monitored by TLC. After com-
pletion, the reaction mixture was diluted with CH2Cl2
(20 ml) and washed twice with ice-cold saturated
aqueous ammonium chloride. The organic phase was
1
mg, 84%) as an orange solid: H-NMR l 8.49–8.51 (m,
Table 2
1H), 8.18 (d, 1H, J=7.2 Hz), 8.10 (d, 1H, J=8.7 Hz),
7.77 (td, 1H, J1=8.7 Hz; J2=1.6 Hz), 7.46 (td, 2H,
J1=7.6 Hz; J2=1.6 Hz), 7.31–7.37 (m, 4H), 7.01–7.07
(m, 6H), 4.89 (m, 1H), 4.33 (m, 1H), 3.92 (s, 5H), 3.68
(m, 1H), 2.15–2.24 (m, 2H), 1.24–1.79 (m, 8H); 31P-
NMR l −18.88; MS (EI) m/z (rel intensity) 615
([M+], 3), 550 (100), 412 (8); IR (KBr, cm−1) 3336 (m),
3261 (w), 3067 (w), 2929 (m), 1659 (s), 1639 (s), 1531
(s), 1434 (m); [h]D20= −112° (c 0.38, CHCl3); m.p.
180–185 °C (dec); Anal Calc. for C35H34FeN3O2P: C,
68.30; H, 5.57; N, 6.82. Found: C, 68.11; H, 5.68; N,
6.56%.
Asymmetric palladium-catalyzed allylic amination with rac-1,3-
diphenyl-2-propenyl acetate and benzylamine a
b
Entry
Ligand
Yield (%)
ee (%) c
1
2
8
9
52
37
79 (S)
39 (S)
a Reaction was performed in THF at 40 °C. Molecular ratio:
[Pd(h3-C3H5)Cl]2/ligand/13/BnNH2=2/6/100/200.
b Isolated yield based on 1,3-diphenyl-2-propenyl acetate.
c Determined by HPLC (chiracel OJ column) and the absolute
configuration of product was assigned through comparison of the
sign of specific rotations with the literature data [14].

766
S.-L. You et al. / Journal of Organometallic Chemistry 637–639 (2001) 762–766
(d) J. Seyden-Penne, in: Chiral Auxiliaries and Ligands in Asym-
metric Synthesis, Wiley, New York, 1995.
[2] Reviews: (a) T. Hayashi, A. Togni (Eds.), Ferrocene, VCH,
Weinheim, Germany, 1995;
(b) C.J. Richards, A.J. Locke, Tetrahedron: Asymmetry 9 (1998)
2377.
[3] (a) T. Hayashi, M. Tajika, K. Tamao, M. Kumada, J. Am.
Chem. Soc. 98 (1976) 3718;
dried over anhydrous MgSO4 and then concentrated
under reduced pressure. The residue was purified by
preparative TLC (EtOAc:petroleum ether=1:15) to
give the pure product. The enantiomeric purities were
determined by HPLC analysis (Chiralcel OD column,
hexane:isopropanol (80:20); flow rate=0.7 ml min−1
tR=18.7 min, tS=20.4 min).
;
(b) T. Hayashi, M. Konoshi, M. Fukushima, T. Mise, M.
Kagotani, M. Tajika, M. Kumada, J. Am. Chem. Soc. 104
(1982) 180;
(c) C.J. Richards, D.E. Hibbs, M.B. Hurthouse, Tetrahedron
Lett. 36 (1995) 3745;
4.6. General procedure for the palladium-catalyzed
allylic amination of rac-1,3-diphenyl-2-propenyl acetate
(d) S.D. Pastor, A. Togni, J. Am. Chem. Soc. 111 (1989) 2333;
(e) A. Togni, S.D. Pastor, J. Org. Chem. 55 (1990) 1649.
[4] E.L. Stangeland, T. Sammakia, Tetrahedron 48 (1997) 16503.
[5] C. Bolm, K. Muniz-Fernandez, A. Seger, G. Raabe, K. Gunther,
J. Org. Chem. 63 (1998) 7860.
[Pd(h3-C3H5)Cl]2 (3.7 mg, 0.01 mmol) and the ligand
(18.5 mg, 0.03 mmol) were dissolved in dry THF (2 ml),
and then stirred for 30 min at r.t. under an atmosphere
of argon. To this solution rac-1,3-diphenyl-2-propenyl
acetate (126 mg, 0.5 mmol), BnNH2 (0.11 ml, 1.0
mmol) were successively added. The reaction mixture
was stirred at 40 °C and monitored by TLC. After
completion, the reaction mixture was diluted with Et2O
(20 ml) and washed twice with ice-cold saturated
aqueous ammonium chloride. The organic phase was
dried over anhydrous MgSO4 and then concentrated
under reduced pressure. The residue was purified by
preparative TLC (EtOAc:petroleum ether=1:10) to
give the pure product. The enantiomeric purities were
determined by HPLC analysis (Chiralcel OJ column,
[6] (a) S.-L. You, Y.-G. Zhou, X.-L. Hou, L.-X. Dai, Chem.
Commun. (1998) 2765;
(b) L.-X. Dai, X.-L. Hou, W.-P. Deng, S.-L. You, Y.-G. Zhou,
Pure Appl. Chem. 71 (1999) 1401.
[7] Reviews and recent examples for palladium-catalyzed allylic
substitution: (a) G. Consiglio, R.M. Waymouth, Chem. Rev. 89
(1989) 257;
(b) C.G. Frost, J. Howarth, J.M.J. Williams, Tetrahedron:
Asymmetry 3 (1992) 1089;
(c) B.M. Trost, D.L. Van Vranken, Chem. Rev. 96 (1996) 395;
(d) M.T. Reetz, S. Sostmann, J. Organomet. Chem. 603 (2000)
105;
(e) N.F.K. Kaiser, U. Bremberg, M. Larhed, C. Moberg, A.
Hallberg, J. Organomet. Chem. 603 (2000) 2.
[8] (a) B.M. Trost, D.L. Van Vranken, C. Bingel, J. Am. Chem.
Soc. 114 (1992) 9327;
hexane:isopropanol (87:13); flow rate=0.6 ml min−1
tS=20.3 min, tR=24.3 min).
;
(b) B.M. Trost, Acc. Chem. Res. 29 (1996) 355 and references
cited therein.
[9] (a) S.-L. You, X.-L. Hou, L.-X. Dai, B.-X. Cao, J. Sun Chem.
Commun. (2000) 1933;
Acknowledgements
(b) S.-L. You, X.-L. Hou, L.-X. Dai, X.-Z. Zhu, Org. Lett. 3
(2001) 149;
(c) J.M. Longmire, B. Wang, X. Zhang, Tetrahedron Lett. 41
(2000) 5435.
[10] (a) W.-P. Deng, X.-L. Hou, L.-X. Dai, Y.-H. Yu, W. Xia Chem.
Commun. (2000) 285;
This research was financially supported by the Major
Basic Research Development Program (grant no.
G2000077506), National Natural Science Foundation
of China, National Outstanding Youth Fund, Chinese
Academy of Sciences, and Shanghai Committee of Sci-
ence and Technology.
(b) S.-L. You, X.-L. Hou, L.-X. Dai, Tetrahedron: Asymmetry
11 (2000) 1495;
(c) X.-T. Zhou, Y.-R. Lin, L.-X. Dai, J. Org. Chem. 64 (1999)
1331;
(d) W.-P. Deng, X.-L. Hou, L.-X. Dai, X.-W. Dong Chem.
Commun. (2000) 1483;
(e) X.-W. Wu, X.-L. Hou, L.-X. Dai, B.-X. Cao, J. Sun Chem.
Commun. (2000) 1195.
References
[11] (a) C. Bolm, K. Muniz-Fernandez, J.P. Hildebrand, Org. Lett. 1
(1999) 991;
(b) K. Muniz, C. Bolm, Chem. Eur. J. 6 (2000) 2309.
[12] B.M. Trost, J.D. Oslob, J. Am. Chem. Soc. 121 (1999) 3057.
[13] P. Wimmer, M. Widhalm, Tetrahedron: Asymmetry 6 (1995)
657.
[1] Some reviews: (a) H. Brunner, in: E. Eliel, S.H. Wilen (Eds.),
Topics in Stereochemistry, vol. 18, Wiley, New York, 1988, p.
129;
(b) R. Noyori, M. Kitamamura, in: R. Scheffold (Ed.), Modern
Synthetic Methods, Springer, Berlin, 1989, p. 115;
(c) I. Ojima, in: Catalytic Asymmetric Synthesis, VCH, New
York, 1993;
[14] P.V. Matt, O. Loiseleur, G. Koch, A. Pfaltz, C. Lefeber, T.
Feucht, G. Helmchen, Tetrahedron: Asymmetry 5 (1994) 573.