Chiral pyrrolidine thioethers: Effective nitrogen-sulfur donating ligands in palladium-catalyzed asymmetric allylic alkylations

Article abstract of DOI:10.1016/j.tetasy.2004.03.015

Enantiomeric cyclic, five-membered alcohols and vic-diols were converted into the corresponding mono- and bis(arylsulfanyl) derivatives with complete inversion of configuration at one or both stereogenic centers. The thus obtained chiral thioethers were tested in the Pd-catalyzed allylic alkylation of dimethyl malonate with rac-1,3-diphenyl-2-propenyl acetate. The purely S,S-donating C₂-symmetric chiral ligands showed poor to moderate enantioselectivity (up to 42%), while 1-alkyl-3,4-bis(arylthio)pyrrolidines afforded much higher results, 81-89% ee. When chiral pyrrolidine mono-thioethers were applied the observed enantioselectivity improved further to 86-90% ee. These results suggest that the pyrrolidine thioethers served as the N(sp ³),S-donating chiral ligands. Examination of molecular models showed that the sense of stereoinduction, namely (R)-product from with (3R)-1-benzyl-3-(arylthio)pyrrolidine is in agreement with the nucleophilic attack being directed at the allylic carbon located trans to the sulfur atom in the intermediate η³-allylpalladium complex.

Full text of DOI:10.1016/j.tetasy.2004.03.015

Tetrahedron:
Asymmetry
Tetrahedron: Asymmetry 15 (2004) 1437–1444
Chiral pyrrolidine thioethers: effective nitrogen–sulfur donating
ligands in palladium-catalyzed asymmetric allylic alkylations
*
_
ꢀ
Renata Siedlecka, Elzbieta Wojaczynska and Jacek Skarzewski
_
Institute of Organic Chemistry, Biochemistry and Biotechnology, Wrocław University of Technology, 50-370 Wrocław, Poland
Received 21 January 2004; accepted 11 March 2004
Abstract—Enantiomeric cyclic, five-membered alcohols and vic-diols were converted into the corresponding mono- and bis(aryl-
sulfanyl) derivatives with complete inversion of configuration at one or both stereogenic centers. The thus obtained chiral thioethers
were tested in the Pd-catalyzed allylic alkylation of dimethyl malonate with rac-1,3-diphenyl-2-propenyl acetate. The purely S,S-
donating C2-symmetric chiral ligands showed poor to moderate enantioselectivity (up to 42%), while 1-alkyl-3,4-bis(aryl-
thio)pyrrolidines afforded much higher results, 81–89% ee. When chiral pyrrolidine mono-thioethers were applied the observed
enantioselectivity improved further to 86–90% ee. These results suggest that the pyrrolidine thioethers served as the N(sp3),S-
donating chiral ligands. Examination of molecular models showed that the sense of stereoinduction, namely (R)-product from with
(3R)-1-benzyl-3-(arylthio)pyrrolidine is in agreement with the nucleophilic attack being directed at the allylic carbon located trans to
the sulfur atom in the intermediate g3-allylpalladium complex.
Ó 2004 Elsevier Ltd. All rights reserved.
1. Introduction
highly demanding requirement, the reaction remains a
useful platform for testing new ligands for transition-
metal catalyzed processes.
Asymmetric catalysis of organic reactions to provide
enantiomerically enriched products is of central impor-
tance to modern synthetic chemistry.1 Within this field
palladium-catalyzed allylic nucleophilic substitution is
one of the most important tools for the enantioselective
formation of carbon–carbon and carbon–heteroatom
bonds.2 The reaction mechanism is well understood and
the process has already found numerous applications in
the synthesis of natural products.3 The reaction enantio-
selectivity is determined by the steric and stereoelec-
tronic properties of the chiral ligand coordinated to the
palladium atom. The ligand must provide an effective
chiral environment at the reacting face of the p-allyl
unit, the face opposite to the metal. Because of this
In the rational design of chiral ligands two general
concepts have led to the most effective structures.4 The
first one relies on the reduction of the number of pos-
sible diastereomeric transition states, which are attained
using homodonating ligands with C2-symmetry.5 In this
case, stereodifferentiation comes from purely steric
interactions that raise the energy of some structures,
while at the same time making the others prevalent. This
idea was exploited fruitfully in the title reaction by the
use of P,P- and N,N-donating ligands.6 The second
strategy takes advantage of the stereoelectronic differ-
entiation exerted by the application of heterodonating
D
Nu
Nu
D
D
D
R
R
Nu
Pd
R
Pd
R
Pd
Pd
Nu
A
A
2
A
A
R
R
R
R
3
2
W-shaped η -complex
η -complex
3
η -complex
M-shaped η -complex
Scheme 1.
* Corresponding author. Fax: +71-328-4064; e-mail: jacek.skarzewski@pwr.wroc.pl
0957-4166/$ - see front matter Ó 2004 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2004.03.015

1438
R. Siedlecka et al. / Tetrahedron: Asymmetry 15 (2004) 1437–1444
ligands. Thus, due to the trans-effect,7 an attacking
nucleophile should approach the allylic system from the
site opposite to the more p-accepting ligand center.
(Scheme 1).
2. Results and discussion
2.1. Ligands synthesis
Homochiral 1,2-bis(phenylsulfanyl)cyclopentane and its
heterocyclic analogues 2a–f and 3 were synthesized
directly from the readily available enantiomeric diols 1
via double substitution occurring with the inversion of
configuration at both stereogenic centers. To achieve
this, the hydroxy groups were activated as the oxo-
phosphonium salts generated from tributylphosphine
oxidized with diphenyl disulfide. Under the reaction
conditions these intermediates underwent SN2 reactions
with the generated thiophenolate anion, thus bringing
about the desired conversion of diols into bis-thiophenyl
ethers (the Hata reaction).12 We applied the Hata reac-
tion in a sealed tube according to our published proto-
col13 and that simple procedure gave the required
products in good to reasonable yields (Scheme 2).
Accordingly P,N-, P,S-, and N,S-bidentate ligands have
been applied successfully in the Pd-catalyzed allylic
substitutions and their electronic effects were considered
as the main cause of enantioselectivity.8 Nonetheless,
even for these ligands the stereochemical outcome of the
reaction was often influenced by purely steric effects.
Recently various C2-symmetric chiral S,S-donating
ligands were derived from tartaric acid and examined in
the Pd-catalyzed allylic alkylation.9 Dithioethers formed
from 2,3-isopropylidenethreitol (the sulfur analogues of
DIOP) showed low activities and poor enantioselectivi-
ties (up to 42%). However, it was discovered that
(3R,4R)-1-benzyl-3,4-bis(phenylsulfanyl)pyrrolidine effi-
ciently catalyzed the alkylation and, in spite of the for-
mation of four diastereomeric complex species, gave
81% ee.9 This unexpectedly high enantioselectivity was
ascribed to the chiral backbone rigidity of the S,S-
donating ligand.9;10b Since we have recently developed
easy enantioselective syntheses of the corresponding
dithioethers,11 we could investigate a collection of sim-
ilar ligands. Herein we present our results that provide
an explanation of the sense of the enantioselectivity and
the ligand complexation mode different from the current
interpretation offered.9;10b We also report new highly
selective chiral N,S-ligands, readily available in both
enantiomeric forms.
The antipodes of 1 were also converted to the respective
ent-2 and ent-3. The complete stereospecificity of the
reaction on homochiral substrates was confirmed by the
absence (NMR, GC) of any meso-disubstituted product.
Moreover, under Hata reaction conditions commercially
available (S)- and (R)-1-benzyl-3-hydroxypyrrolidine 4
were easily converted into (R)- and (S)-1-benzyl-3-
phenylsulfanylpyrrolidine 5, respectively and also (S)-1-
benzyl-2-pyrrolidinemethanol 7 was smoothly changed
into the thioether 8. As described in the literature
hydrogenolysis of the N-benzyl group in 1c14 allowed
HO
2-NaphS
PhS
(2-NaphS)2
(PhS)2, Bu3P
X
(S,S)-1c
X
NCH2Ph
(R,R)-3
C6H6, 80°C
Bu3P, C 6H6, 80°C
HO
PhS
2-NaphS
(S,S)-1
(R,R)-2
X: CH 2 (a); O (b); NCH2Ph (c); NC12H25 (d); NCOPh (e); NBoc (f)
(S,S)-1e
PhCOCl
PhS
PhS
HO
H2, 10% Pd/C
CF3COOH
CH2Cl2
(R,R)-2f
NH
(R,R)-2g
(S,S)-1c
NH
Boc2O
C2H5OH, AcOH
HO
(S,S)-1f
HO
RS
(RS)2, Bu3P
C6H6, 80°C
(PhS)2, Bu3P
C6H6, 80°C
NCH2Ph
(R)-4
NCH2Ph
CH2OH
NCH2Ph
CH2SPh
NCH2Ph
R: Ph
R: 2-Naph (S)-6
( S)-5
(S)-8
(S)-7
PhS
PhS
Ph
PhS
Ph
Ph 3CCl
Et 3N
(R,R)-2g
NCPh3
SPh
(R,R)-2h
(R,R)-11
O
PhS
PhS
PhS
PhS
30% H2O2
*
*
NCH2Ph
NCH2Ph
NCH2Ph
L*, VO(acac)2
*
PhS
PhS
(R R ,S,S)-10
S,
(R ,S,S)-9
S
(S,S)-2c
S
O
O
Scheme 2.

R. Siedlecka et al. / Tetrahedron: Asymmetry 15 (2004) 1437–1444
1439
subsequent N-acylation to give the diols 1e and 1f,
which, in turn, were submitted to the Hata reaction. A
standard deprotection of the N-Boc function15 in 2f
furnished (3R,4R)-3,4-bis(phenylsulfanyl)pyrrolidine 2g.
This compound was smoothly transformed into the
corresponding N-trityl derivative 2h using the recently
developed procedure.16 Finally, (S,S)-2c was oxidized
2-propenyl acetate, using N,O-bis(trimethylsilyl)acet-
amide (BSA)–potassium acetate (3 mol %) as the base,
[Pd(g3-C3H5)Cl]2 (2.5 mol %) as the palladium pre-cat-
alyst and the respective sulfide as the chiral ligand
(10 mol %). The reaction was carried out in methylene
dichloride at 25 °C for 3–10 days. In the absence of the
ligand no product could be detected after the same time
(Scheme 3).
with 30% H2O2 in the presence of catalytic
L-N-(3-
phenyl-5-nitrosalicylidene)valinol and VO(acac)2.17 The
sulfoxidation gave the corresponding mono- (RS,S,S)-9
(45%, 80% de) and bis-sulfoxide (RS,RS,S,S)-10 (40%,
64% de) isolated as the pure diastereomeric products.
The absolute configuration of the introduced phenyl-
sulfinyl group was ascribed according to the sign of the
Cotton effects (CE) observed in their CD spectra. It is
well established that for the (R)-configuration of this
inherently chiral chromophore the primary UV band
(235–255 nm, log e ca. 3.6) demonstrates a positive CE of
De ca. 20 and the shorter wavelength band (ca. 220 nm)
shows a negative CE of similar strength.18
Firstly, the ligands, which we consider being purely S,S-
donating were tested and the results obtained are pre-
sented in Table 1. It can be clearly seen that these
derivatives show relatively low to moderate activity and
selectivity and the results are similar to those previously
reported for the other C2-symmetric bis-thioethers.9;10b
Table 1. Pd-catalyzed alkylation of dimethyl malonate with 1,3-
diphenyl-2-propenyl acetate in the presence of C2-symmetric S,S-
donating ligands
Entry
Ligand
Reaction
time (d)
Yield (%) Ee (%) Configu-
ration of
product
The CD spectra obtained for 9 and 10 fitted to this
pattern strictly (Fig. 1) Moreover, the C2-symmetric
bis(sulfide) derivative (R,R)-11 was available from our
earlier work.11
1
2
3
4
5
(R,R)-2a
(R,R)-2b
(R,R)-2e
(R,R)-2f
(R,R)-11
3
8
68
42
71
40
82
42
20
15
18
36
(ꢁ)-(S)
(ꢁ)-(S)
(ꢁ)-(S)
(ꢁ)-(S)
(þ)-(R)
3
3
10
Then, we turned to 1-alkyl-3,4-bis(arylthio)pyrrolidines
and the results are shown in Table 2. Generally, the
enantioselectivities are much higher than the previous
ones. Thus, for (R,R)-2c9 the reported high yield and
enantioselectivity were reproduced and its N-dodecyl 2d
Table 2. Pd-catalyzed alkylation of dimethyl malonate with 1,3-
diphenyl-2-propenyl acetate in the presence of C2-symmetric 1-alkyl-
3,4-bis(phenylthio)pyrrolidines
Entry
Ligand
Reaction
time (d)
Yield (%) Ee (%) Configu-
ration of
product
1
2
3
4
(R,R)-2c
(S,S)-2c
(S,S)-2c
40% Ee
(S,S)-2c
40% Ee
(R,R)-2c
(R,R)-2d
(S,S)-2d
(R,R)-2g
(R,R)-2h
(R,R)-3
(S,S)-3
3
3
8
3
88
46
77
47
819
85
86
30
(ꢁ)-(S)
(þ)-(R)
(þ)-(R)
(þ)-(R)
Figure 1. CD spectra of (S,S)-2c (ꢀ ꢀ ꢀ), (RS,S,S)-9 (––), and
(RS,RS,S,S)-10 (- - -) in acetonitrile solution in 10ꢁ4 M concentration
range.
5
3
58
33
(ꢁ)-(S)
2.2. Asymmetric allylic alkylation
6
3
3
3
8
3
3
75
64
70
61
82
81
87
89
49
39
82
90
(ꢁ)-(S)
(þ)-(R)
(ꢁ)-(S)
(ꢁ)-(S)
(ꢁ)-(S)
(þ)-(R)
7
8
With the above mentioned set of chiral sulfides in hand
we examined their effectiveness in the Pd-catalyzed
allylic alkylation. As a model reaction we studied the
alkylation of dimethyl malonate with rac-1,3-diphenyl-
9
10
11
Ph
Ph
H2C(CO2Me)2, BSA(3 eq.), AcOK (3 mol%)
Ph
Ph
CO2Me
Pd2(C3H5)2Cl2 (2.5 mol%), Ligand* (10 mol%)
CH2Cl2, r.t.
OAc
MeO2C
Scheme 3.

1440
R. Siedlecka et al. / Tetrahedron: Asymmetry 15 (2004) 1437–1444
Table 3. Pd-catalyzed alkylation of dimethyl malonate with 1,3-
diphenyl-2-propenyl acetate in the presence of N,S-donating ligands
tate complex of the bis-sulfoxide with palladium is
known and moderately active in asymmetric allylic
alkylation,19 we also tested the catalytic performance of
the respective bis-sulfoxide (RS,RS,S,S)-10 and mono-
sulfoxide (RS,S,S)-9. The first derivative was almost
completely inactive, while the second one gave the (R)-
alkylation product in 88% ee. Remarkably the last result
lies between 86% ee for (S,S)-2c and 90% ee for (R)-5.
All these facts made us believe that the examined mono-
and bis-thioethers of pyrrolidine operate mainly as the
sp3-nitrogen and sulfur-donating ligands. This conclu-
sion can be additionally supported by the result
obtained using the other mono-thioether of pyrrolidine
8 (Table 3, entry 8). Although the previous works on
the N,S-heterodonating ligands in the Pd-catalyzed
allylic alkylation concerned mainly those of N(sp2),
S-type, a recent precedent involved the use of chiral
N(sp3),S-donating ligands derived from ephedrine
and pseudoephedrine.20 The stereochemical outcome
of that reaction was plausibly accounted for by both
trans to sulfur and trans to nitrogen nucleophilic
attack.20b
Entry Ligand
Reaction Yield Ee (%) Configu-
(%)
time (d)
ration of
product
1
2
3
4
5
6
7
8
(S)-5
(R)-5
(R)-5a
(S)-6
(R)-6
3
3
8
8
8
8
8
3
75
73
78
78
79
71
<5
78
90
90
90
87
86
88
(ꢁ)-(S)
(þ)-(R)
(þ)-(R)
(ꢁ)-(S)
(þ)-(R)
(þ)-(R)
(RS,S,S)-9
(RS,RS,S,S)-10
(S)-8
79
(þ)-(R)
a Half of the usual catalyst amount was used.
and bis(2-naphthylsulfanyl) 3 analogues were found to
perform even better. Two derivatives with a very large
(Tr) and the smallest (H) substituent at the pyrrolidine
nitrogen (2g and 2h) were the only exceptions from this
tendency. Moreover, the ligands of (R,R)-configuration
gave the allylic alkylation product of (S)-configuration
and their enantiomers afforded the (R)-product. Inter-
estingly, the (R,R)-ligands seemed to be more active with
only slightly less enantioselectivity than their enantiomers
(cf. Table 2, entries: 1 vs 2, 6 vs 7, 10 vs 11). Nonetheless,
no meaningful nonlinear effects were detected (entries 4
and 5) and the cause for the small differences observed
remains obscure. Additionally, the substrate recovered
after the reaction did not show any enantioenrichment.
2.3. Stereochemical models
The generally accepted reaction mechanism2 involves
the formation of the intermediate Pdþ(chiral ligand)(g3-
trans-1,3-diphenylpropenyl). The symmetrical p-allyl
system can be complexed in an M or W-configuration.
Each of the complexes, after nucleophilic attack rear-
ranges into two possible Pd0(chiral ligand)(g2-olefin)
complexes leading to the enantiomeric substituted
products (Scheme 1). If A denotes a p-accepting,
r-donating ligand center and D is only a r-donating
center, the nucleophilic attack can be expected to occur
from the trans to A direction. On the other hand, the
favorable site of this nucleophilic addition should be
that which corresponds to the release of more steric
strain when coming from g3- to g2-complex. Generally,
thioether ligands are considered rather poor r-donors
and weakly p-accepting ligands and their trans influence
within the coordination sphere is regarded to be higher
than that of amines.21 With all these features in mind,
we examined the molecular models representing all the
possible intermediates formed in the reaction catalyzed
by (S,S)-2c and (R)-5. It seems that for the S,S-donating
(S,S)-2c both M and W-g3-complexes, for example (Fig.
2, A), are of similar energy and in both cases the
nucleophilic attack at any of the prochiral centers
releases only little and similar steric strain when coming
to the corresponding g2-complexes. Thus, this model
barely explains the high enantioselectivity observed for
2c. On the other hand, this reaction mode is perhaps
operating for the purely S,S-donating ligands, which
give poor to moderate enantioselectivities.
Simple inspection of a molecular model for C2-sym-
metric 1-benzyl-3,4-bis(phenylsulfanyl)pyrrolidine 2c
revealed that essentially this derivative can operate as
both an S,S- and/or an S,N-donating ligand. The liter-
ature data did not prove its solely S,S-donating char-
acter,9 so, in order to examine the second possibility we
used the corresponding ligands having within their
structure only one arylsulfanyl group. Thus, both
enantiomers of 1-benzyl-3-(arylsulfanyl)pyrrolidine (5
and 6) were used in the same allylic alkylation and the
respective results are presented in Table 3 (entries 1–5).
We obtained higher activity and enantioselectivity than
with the corresponding bis-thioethers. Even half of the
normal catalyst/ligand loading did not decrease the ee
value. Here, the (S)-ligand gave the alkylation product
of (S)-configuration and the application of its enantio-
mer resulted in the formation of the (R)-product.
The stereochemical outcomes of the reaction with both
types of ligands, mono- and bis-thioethers are in full
agreement. Mono-thioether, for example, (R)-5 can be
regarded as a catalytically active part of the structure of
bis-thioether (S,S)-2c and their absolute configuration
descriptor change results from the CIP rules.
PhS
PhS
PhS
NCH2Ph
(R ,S,S)-9 (R)-5
NCH2Ph
Contrary to this, for the N,S-donating (S,S)-2c the
M-shaped allylic system is clearly favored (B) and the
nucleophilic addition should preferentially occur trans
to S leading to the R-configuration in the product.
Similarly, for (R)-5 the M-shaped g3-allylic complex (C)
seems to be privileged over its W-shaped isomer (E),
*
PhS
(S,S)-2c (R)-5
S
O
Since the phenylsulfinyl functionality can act as S- or O-
donating group and the chiral C2-symmetric S,S-biden-

R. Siedlecka et al. / Tetrahedron: Asymmetry 15 (2004) 1437–1444
1441
3. Conclusions
In conclusion, chiral pyrrolidine arylthioethers are
effective catalysts in the Pd-catalyzed asymmetric
alkylation operating as the N(sp3),S-donating chiral
ligands. The models show that the sense of stereoin-
duction is in agreement with the nucleophilic attack
directed at the allylic carbon located trans to the sulfur
atom in the intermediate complex.
4. Experimental
4.1. General
Melting points were determined using a Boetius hot-
stage apparatus and are uncorrected. IR spectra were
recorded on a Perkin Elmer 1600 FTIR spectropho-
tometer. 1H NMR and 13C NMR spectra were measured
on a Bruker CPX (1H, 300 MHz) or a Bruker Avance
(1H, 500 MHz) spectrometer using TMS as an internal
standard. UV and CD spectra were recorded for
CH3CN solutions using a Hewlett-Packard 8452 diode
array spectrophotometer and a JASCO J 600 spectro-
polarimeter, respectively. Observed rotations at 589 nm
were measured using an Optical Activity Ltd. Model
AA-5 automatic polarimeter. GC–MS analyses were
determined on a Hewlett-Packard 5890 II gas chro-
matograph (25 m capillary column) with a Hewlett-
Packard mass spectrometer 5971 A operating on the
electron impact mode (70 eV). Separations of products
by chromatography were performed on silica gel 60
(230–400 mesh) purchased from Merck. Thin layer
chromatography analyses were performed using silica
gel 60 precoated plates (Merck).
Figure 2. Selected molecular models of the intermediate complexes
(see, the text).
which also differs in configuration at the sulfur atom.
Moreover, again the nucleophilic attack on the M-iso-
mer leading to the (R)-product releases the steric strains
(D). The respective attack on the W-form produces the
overcrowded g2-complex (F) and this route can be
excluded on this basis. Summarizingly, the attack pref-
erentially takes place at the carbon located trans to the
sulfur atom in the M-shaped complex (B, C), thus giving
the (R)-enantiomer of product. This suggests that both
factors, steric and stereoelectronic are responsible for
the enantioselectivity. The role of the second effect can
be additionally supported by the fact that replacement
of phenylsulfanyl groups in 2c by isopropylsulfanyl
functions,9 with the more r-donor and less p-accepting
character of the sulfur-donating center, resulted in a
substantial decrease of enantioselectivity (from 81% to
30% ee).9 Interestingly, it is an opposite effect to that
observed for the chiral N(sp3),S-donating ligands de-
rived from ephedrine and pseudoephedrine. There the
highest ee was observed for the tert-butylthioether (89%
ee), and arylthioethers were much inferior (15% ee) and
consequently both types of nucleophilic attack (trans to
S and trans to N) were postulated.20 Finally, it is note-
worthy that the diminished enantioselectivity of 2c ver-
sus 5 can be explained by the parallel participation of
the S,S-along with the S,N-complexing species in the
former case. This interpretation is in line with the
observation of four different Pdþ(2c)(allyl) complexes in
4.2. Ligand synthesis
4.2.1. General procedure for the preparation of mono- and
bis(sulfides). A solution of diol (0.8 mmol), ArSSAr
(4.8 mmol) and Bu3P (1.59 mL, 6.4 mmol) in dry benzene
(5 mL) was placed under argon in a reaction ampoule.
For mono-alcohols amounts of the reagents were
halved. The sealed tube was heated at 80 °C for 3 days.
Thereafter the cooled mixture was diluted with ether
(20 mL), washed with 2 M NaOH, brine, and dried over
anhyd Na2SO4. After evaporation, the crude product
was purified by chromatography on silica gel. Selected
data for the obtained products 2a–d were reported
before.11
4.2.2. (+)-(3R,4R)-1-Benzoyl-3,4-bis(phenylsulfanyl)pyr-
20
D
rolidine 2e. Yield 67%. Mp ¼ 131–133 °C. ½aꢂ ¼ þ84:2
1
(c 0.76, CH2Cl2, >95% ee). H NMR (CDCl3): 3.45 (d,
1H, J ¼ 11:7 Hz), 3.56 (d, 1H, J ¼ 4:5 Hz), 3.67–3.77
(m, 2H), 4.12–4.15 (m, 1H), 4.29–4.33 (m, 1H), 7.14–
7.23 (m, 10H, ArH), 7.37–7.43 (m, 3H, ArH), 7.49–7.51
(m, 2 H, ArH); 13C NMR (CDCl3): 49.7, 50.4, 51.5, 53.0,
127.6, 128.2, 128.3, 128.8, 129.7, 130.6, 132.6, 132.9,
136.7, 170.3; IR (KBr): 3047, 2948, 1625, 1427, 1250,
741, 690. Anal. Calcd for C23H21NOS2 (M ¼ 391.551):
1
the respective H NMR spectrum.9

1442
R. Siedlecka et al. / Tetrahedron: Asymmetry 15 (2004) 1437–1444
C, 70.55; H, 5.41; N, 3.58; S, 16.38. Found: C, 70.78; H,
5.58; N, 3.36; S, 16.35.
(m, 10H, ArH); 13C NMR (CDCl3): 32.8, 43.5, 53.6,
60.5, 61.2, 126.5, 127.4, 128.7, 129.2, 129.3, 130.1, 137.1,
139.2; IR (film): 3060, 3027, 2959, 2792, 1584, 1480,
1439, 1145, 738, 699. GC–MS (EI, 70 eV): 269 (2% Mþ),
159 (47% MþꢁPhSH), 91 (100%, PhCH2þ). Rf ¼ 0:27 (n-
hexane–ethyl acetate 9:1). Anal. Calcd for C17H19NS
(M ¼ 269.322): C, 75.80; H, 7.11; N, 5.19; S, 11.88.
Found: C, 75.50; H, 7.34; N, 4.82; S, 12.14.
4.2.3. (+)-(3R,4R)-1-tert-Butoxycarbonyl-3,4-bis(phen-
ylsulfanyl)pyrrolidine 2f. Yield 79%. Mp ¼ 97–99 °C.
20
D
½aꢂ ¼ þ50:0 (c 0.90, CH2Cl2, >95% ee). 1H NMR
(CDCl3): 1.48 (s, 9H, OC(CH3)3), 3.41–3.52 (m, 2H),
3.58–3.60 (m, 2H), 3.92–4.02 (m, 2H), 7.17–7.35 (m,
10H, ArH); 13C NMR (CDCl3): 28.9, 49.9, 51.0, 80.3,
128.1, 129.6, 132.6, 133.8, 154.7; IR (KBr): 3067, 2977,
2876, 1698, 1402, 1154, 1117, 739, 690. Rf ¼ 0:44 (n-
hexane–ethyl acetate 90:10). Anal. Calcd for
C21H25NO2S2 (M ¼ 387.561): C, 65.08; H, 6.50; N, 3.61;
S, 16.55. Found: C, 65.38; H, 6.33; N, 3.70; S 16.72.
4.2.8. ())-(S)-1-Benzyl-3-phenylsulfanylpyrrolidine 5.
20
Yield 90%. ½aꢂ ¼ ꢁ20:0 (c 1.26, CH2Cl2).
D
4.2.9. (+)-(R)-1-Benzyl-3-(2-naphthylsulfanyl)pyrrolidine
20
D
CH2Cl2). H NMR (CDCl3): 1.87–1.92 (m, 1H, H4),
6. Yield 90%. Mp ¼ 37–38 °C. ½aꢂ ¼ þ36:1 (c 1.08,
1
The BOC protecting group was removed according to
the literature procedure.15
2.38–2.42 (m, 1H, H4), 2.54 (dd, 1H, J1 ¼ 9:8 Hz,
J2 ¼ 6:2 Hz, H2), 2.63–2.67 (m, 1H, H5), 2.67–2.74 (m,
1H, H5), 3.13 (dd, 1H, J1 ¼ 9:5 Hz, J2 ¼ 7:7 Hz, H2),
3.62 and 3.69 (ABq, 2H, J ¼ 12:9 Hz, CH2Ph), 3.89–3.92
(m, 1H, H3), 7.17–7.78 (m, 12H, ArH); 13C NMR
(CDCl3): 32.8, 43.4, 53.6, 60.5, 61.2, 126.1, 126.9, 127.5,
127.6, 127.9, 128.1, 128.2, 128.7, 128.8, 129.2, 132.2,
134.2, 134.7, 139.1; IR (film): 3054, 2959, 2792, 1625,
1589, 1500, 1453, 1133, 812, 743. Rf ¼ 0:35 (n-hexane–
ethyl acetate 3:1). Anal. Calcd for C21H21NS
(M ¼ 319.378): C, 78.96; H, 6.62; N, 4.38; S, 10.01.
Found: C, 78.70; H, 6.68; N, 4.22; S, 9.70.
4.2.4. (+)-(3R,4R)-3,4-Bis(phenylsulfanyl)pyrrolidine 2g.
20
D
Yield 90%. ½aꢂ ¼ þ40:0 (c 0.68, CH2Cl2, >95% ee). 1H
NMR (CDCl3): 2.96 (dd, 2H, J1 ¼ 12:7 Hz,
J2 ¼ 3:2 Hz), 3.49–3.52 (m, 2H), 3.56–3.58 (m, 2H), 3.69
(br s, 1H, NH), 7.22–7.27 (m, 10H, ArH); 13C NMR
(CDCl3): 52.4, 52.5, 127.8, 129.6, 132.1, 134.4; IR (film):
3058, 2876, 1727, 1683, 1481, 1210, 1142, 741. GC–MS
(EI, 70 eV): 274 (53% (PhS)2C3H4NHþ2 ), 161 (100%,
PhSNC3Hþ2 ), 109 (54%, PhSþ).
4.2.5. Preparation of N-tritylpyrrolidine. Triethylamine
(100 mg, 1 mmol) was added in one portion to a solution
of trityl chloride (95 mg, 0.34 mmol) and (þ)-(3R,4R)-
3,4-bis(phenylsulfanyl)pyrrolidine (90 mg, 0.31 mmol) in
CH2Cl2 (3 mL). The mixture was stirred for about 20 h
at room temperature and the reaction was monitored by
TLC. Dichloromethane (10 mL) was then added and the
solution was washed with water (10 mL). The organic
layer was separated, dried over Na2SO4, filtered, and
evaporated. The crude product was purified by chro-
matography on silica gel.
4.2.10. ())-(S)-1-Benzyl-3-(2-naphthylsulfanyl)pyrrolid-
20
ine 6. Yield 90%. ½aꢂ ¼ ꢁ34:9 (c 1.06, CH2Cl2).
D
4.2.11.
(+)-(3R,4R)-1-Benzyl-3,4-bis(2-naphthylsulfa-
nyl)pyrrolidine 3. Yield 88%. Mp ¼ 98–100 °C
20
(CH2Cl2–hexane). ½aꢂ ¼ þ108 (c 0.78, CH2Cl2, >95%
D
ee). 1H NMR (CDCl3): 2.73 (dd, 2H, J1 ¼ 10:3 Hz,
J2 ¼ 3:9 Hz), 3.21–3.31 (m, 2H), 3.65 and 3.79 (ABq, 2H,
J ¼ 13:7 Hz), 3.79–3.83 (m, 2H), 6.98–7.74 (m, 19H,
ArH); 13C NMR (CDCl3): 52.6, 59.7, 60.0, 126.4, 126.9,
127.5, 127.7, 128.0, 128.7, 128.8, 128.9, 129.1, 130.3,
132.5, 132.9, 134.0, 138.8; IR (KBr): 3050, 2961, 1727,
1271, 1130, 819, 740. Rf ¼ 0:45 (n-hexane–ethyl acetate
9:1). Anal. Calcd for C31H27NS2 (M ¼ 477.685): C,
77.95; H, 5.70; S, 13.43. Found: C, 78.30; H, 5.71; S,
13.50.
4.2.6. (+)-(3R,4R)-1-Trityl-3,4-bis-(phenylsulfanyl)pyr-
rolidine 2h. Yield 84%. Mp ¼ 98–100 °C (CH2Cl2–hex-
20
D
ane). ½aꢂ ¼ þ41:6 (c 0.48, CH2Cl2, >95% ee). 1H NMR
(CDCl3): 2.53 (dd, 2H, J1 ¼ 10:4 Hz, J2 ¼ 3:9 Hz), 2.93
(dd, 2H, J1 ¼ 9:9 Hz, J2 ¼ 5:9 Hz), 3.55–3.60 (m, 2H),
7.19–7.34 (m, 19H, ArH), 7.55–7.59 (m, 6H, ArH); 13C
NMR (CDCl3): 51.3, 52.4, 73.9, 126.3, 126.8, 127.5,
129.0, 129.3, 131.1, 135.6, 142.1; IR (KBr): 3446, 2929,
1728, 1583, 1481, 1438, 1023, 740. Rf ¼ 0:73 (n-hexane–
ethyl acetate 9:1).
4.2.12. ())-(S)-1-Benzyl-2-(phenylsulfanylmethyl)pyrroli-
20
D
1
dine 8. Yield 81%. ½aꢂ ¼ ꢁ67:5 (c 1.20, CH2Cl2). H
NMR (CDCl3): 1.66–1.68 (m, 1H, H3), 1.71–1.77 (m,
2H, H3, H4), 1.99–2.02 (m, 1H, H4), 2.15–2.23 (m, 1H,
H5), 2.75–2.76 (m, 1H, H2), 2.89–2.94 (m, 2H, H5,
CH2S), 3.14–3.17 (m, 1H, CH2S) 3.33 and 3.99 (ABq,
2H, JAB ¼ 13:0 Hz, CH2Ph), 7.10–7.30 (m, 10H, ArH);
13C NMR (CDCl3): 23.0, 30.9, 39.0, 54.9, 59.3, 63.2,
126.0, 127.3, 128.6, 129.2, 129.3, 129.3, 137.7, 139.9; IR
(film): 3060, 2964, 2791, 1584, 1480, 1026, 738, 698. GC–
MS (EI, 70 eV): 283 (0.03%, Mþ), 160 (86%
MþꢁPhSCH3), 91 (100%, PhCH2þ). Rf ¼ 0:38 (n-hex-
ane–ethyl acetate 85:15). Anal. Calcd for C18H21NS
4.2.7. (+)-(R)-1-Benzyl-3-phenylsulfanylpyrrolidine 5.
20
D
1
Yield 84%. ½aꢂ ¼ þ21:1 (c 1.04, CH2Cl2). H NMR
(CDCl3): 1.74–1.78 (m, 1H, H4), 2.25–2.29 (m, 1H, H4),
2.39 (dd, 1H, J1 ¼ 9:9 Hz, J2 ¼ 6:2 Hz, H2), 2.51–2.56
(m, 1H, H5), 2.61–2.65 (m, 1H, H5), 3.01 (dd, 1H,
J1 ¼ 9:9 Hz, J2 ¼ 7:3 Hz, H2), 3.53 and 3.60 (ABq, 2H,
J ¼ 12:9 Hz, CH2Ph), 3.67–3.73 (m, 1H, H3), 7.10–7.81

R. Siedlecka et al. / Tetrahedron: Asymmetry 15 (2004) 1437–1444
1443
(M ¼ 283.432): C, 76.28; H, 7.47; N, 4.94; S, 11.31.
4.2.17. ())-(3S,4S)-1-Benzoyl-3,4-dihydroxypyrrolidine
20
Found: C, 76.50; H, 7.33; N, 4.95; S, 11.53.
1e. Yield 86%. Mp ¼ 131–133 °C. ½aꢂ ¼ ꢁ86:0 (c 0.78,
D
CH3OH, >95% ee). H NMR (CDCl3): 2.15 (br s, 1H,
1
Diastereoselective sulfoxidation of (þ)-(3S,4S)-1-benzyl-
3,4-bis(phenylsulfanyl)pyrrolidine 2c was performed
according to the earlier described procedure.17 The
products were isolated by chromatography and purified
by recrystallization from the n-hexane–dichloromethane
mixture.
OH), 2.29 (br s, 1H, OH), 3.32 (d, 1H, J ¼ 11:7 Hz),
3.58 (d, 1H, J ¼ 13:6 Hz), 3.74–3.76 (m, 1H), 3.88–3.91
(m, 1H), 4.10 (s, 1H), 4.20 (s, 1H), 7.29–7.48 (m, 5H,
ArH); 13C NMR (CDCl3–CD3OD 2:1): 52.2, 54.9, 73.8,
75.0, 127.1, 128.5, 130.4, 136.2, 171.3; IR (KBr): 3395,
2921, 1611, 1564, 1454, 1077, 988, 695.
4.2.13.
(+)-(3S,4S,RS)-1-Benzyl-3-(phenylsulfinyl)-4-
4.3. Catalytic reaction procedure
(phenylsulfanyl)pyrrolidine 9. Yield 40%. Mp ¼ 106–
20
107 °C (CH2Cl2–hexane). ½aꢂ ¼ þ34:6 (c 0.52, CH2Cl2,
A solution of the allylpalladium chloride dimer (4 mg,
0.01 mmol) and ligand (0.04 mmol) in dry dichloro-
methane (1.0 mL) was stirred under argon atmosphere
at room temperature for 15 min before a solution of rac-
1,3-diphenyl-2-propenyl acetate (100 mg, 0.4 mmol) in
dichloromethane (1.5 mL) was added. The resulting
yellow solution was treated successively with dimethyl
malonate (0.137 mL, 1.2 mmol), N,O-bis(trimethyl-
silyl)acetamide (0.296 mL, 1.2 mmol) and anhydrous
potassium acetate (1.2 mg, 0.012 mmol). The reaction
was carried out at room temperature for 3–10 days
(monitored by TLC). The reaction mixture was diluted
with ether (10 mL) and quenched with the satd NH4Cl
solution. The separated organic layer was washed with
brine, dried (MgSO4), and evaporated. The product was
purified by column chromatography (hexane–ethyl ace-
tate 3:1). The enantiomeric excess was determined by 1H
NMR using Eu(hfc)3 (0.5 equiv) as a chiral shift reagent.
Under these conditions the chemical shift difference
observed for one of the diastereotopic ester methyl
groups was ca. 11 Hz [downfield for the dextrorotatory
(R)-enantiomer]. The assignment of absolute configu-
ration was based on the specific rotation according to
the literature data.22
D
>95% ee). H NMR (CDCl3): 2.64–2.71 (m, 2H), 3.04–
1
3.24 (m, 3H), 3.60 and 3.70 (ABq, 2H, J ¼ 13:1 Hz),
3.76–3.82 (m, 1H), 7.23–7.32 (m, 10H, ArH), 7.44–7.48
(m, 3H, ArH), 7.52–7.56 (m, 2H, ArH); 13C NMR
(CDCl3): 45.8, 50.8, 59.2, 60.8, 69.1, 124.6, 127.2, 127.7,
128.4, 128.6, 129.2, 129.3, 131.4, 132.2, 133.9, 138.0,
142.2; IR (KBr): 3436, 2794, 1438, 1033, 748, 690; UV
(CH3CN): k ffi 210 nm (sh), k ¼ 250 nm (e ¼ 10,200);
Rf ¼ 0:56 (n-hexane–ethyl acetate 1:1).
4.2.14.
(+)-(3S,4S,RS,RS)-1-Benzyl-3,4-bis(phenylsulf-
inyl)pyrrolidine 10. Yield 40%. Mp ¼ 156–158 °C
20
(CH2Cl2–hexane). ½aꢂ ¼ þ56:0 (c 0.36, CH2Cl2, >95%
D
ee). 1H NMR (CDCl3): 2.59 (dd, 2H, J1 ¼ 10:3 Hz,
J2 ¼ 6:7 Hz), 3.12 (dd, 2H, J1 ¼ 10:4 Hz, J2 ¼ 4:7 Hz),
3.35–3.41 (m, 2H), 3.69 (s, 2H, CH2Ph), 7.21–7.31 (m,
5H, ArH), 7.45–7.54 (m, 10H, ArH); 13C NMR
(CDCl3): 51.3, 58.9, 63.2, 124.7, 127.3, 128.4, 128.5,
129.5, 131.8, 137.8, 141.6; IR (KBr): 3434, 2797, 1493,
1442, 1085, 1036, 746; UV (CH3CN): k ffi 210 nm (sh),
k ¼ 254 nm (e ¼ 10,700); Rf ¼ 0:22 (n-hexane–ethyl ace-
tate 1:1).
Acknowledgement
4.2.15. Preparation of 1-acyl-3,4-dihydroxypyrrolidines.
3,4-Dihydroxypyrrolidine acetate14 (1.0 g, 6.1 mmol)
and 4.6 g of anhydrous K2CO3 were placed into 7 mL of
the chloroform–methanol mixture (1:1 v/v). Benzoyl
chloride (0.85 mL, 7.3 mmol, 1.2 equiv) was added
dropwise to the stirred, cooled solution. The resultant
mixture was stirred overnight and then evaporated to
dryness. The solid residue was extracted continuously
with diethyl ether and after removal of solvent the crude
product was recrystallized from ethyl acetate. 1-tert-
Butoxycarbonyl-3,4-dihydroxypyrrolidine was obtained
in an analogous manner using di(tert-butyl) dicarbonate
instead of benzoyl chloride.
The authors thank the State Committee for Scientific
Research for financial support (KBN Grant 7 T09A 109
21).
References and notes
1. Catalysis in Asymmetric Synthesis; Ojima, I., Ed.; 2nd ed.;
Wiley-VCH: New York, 2000.
2. For reviews, see: (a) Pfaltz, A.; Lautens, M. In Compre-
hensive Asymmetric Catalysis; Jacobsen, E. N., Pfaltz, A.,
Yamamoto, H., Eds.; Springer: Berlin, 1999; Vol. II, pp
833–884; (b) Trost, B. M.; Van Vranken, D. L. Chem. Rev.
1996, 96, 395–422.
3. Trost, B. M.; Crawley, M. L. Chem. Rev. 2003, 103, 2921–
2944.
4.2.16. ())-(3S,4S)-1-tert-Butoxycarbonyl-3,4-dihydro-
xypyrrolidine 1f. Yield 79%. Mp ¼ 163–164 °C.
20
D
4. Handy, S. T. Curr. Org. Chem. 2000, 4, 363–395.
5. Whitesell, J. K. Chem. Rev. 1989, 89, 1581–1590.
6. (a) Trost, B. M. Acc. Chem. Res. 1996, 29, 355–364; (b)
Pfaltz, A. Acc. Chem. Res. 1993, 26, 339–345.
7. (a) Pfaltz, A. Synlett 1999, 835–842; (b) Szabo, K. J.
Organometallics 1996, 15, 1128–1133.
½aꢂ ¼ ꢁ19:1 (c 1.02, CH3OH, >95% ee). 1H NMR
(CDCl3): 1.38 (s, 9H, OC(CH3)3), 1.80 (br s, 2H, OH),
3.24 (d, 1H, J ¼ 11:4 Hz), 3.31 (d, 1H, 11.4 Hz), 3.50–
3.58 (m, 2H), 4.25 (br s, 2H); 13C NMR (CDCl3–
CD3OD 2:1): 28.2, 51.2, 51.7, 74.3, 74.9, 79.9, 155.5; IR
(KBr): 3407, 2917, 1621, 1471, 1077. Rf ¼ 0:44 (n-hex-
ane–ethyl acetate 90:10).
8. For recent examples of the P,N-ligands, see: (a) Helmchen,
G.; Pfaltz, A. Acc. Chem. Res. 2000, 33, 336–345; (b)

1444
R. Siedlecka et al. / Tetrahedron: Asymmetry 15 (2004) 1437–1444
Constantine, R. N.; Kim, N.; Bunt, R. C. Org. Lett. 2003,
15. Cran, G. A.; Gibson, C. L.; Handa, S. Tetrahedron:
Asymmetry 1995, 6, 1553–1556.
16. Canle, L. M.; Clegg, W.; Demirtas, I.; Elsegood, M. R. J.;
Maskill, H. J. Chem. Soc., Perkin Trans. 2 2000, 85–92.
5, 2279–2282; (c) P,S-ligands: Nakano, H.; Okuyama, Y.;
Yamagida, M.; Hongo, H. J. Org. Chem. 2001, 66, 620–
625; (d) N,S-ligands: Adams, H.; Anderson, J. C.; Cub-
bon, R.; James, D. S.; Mathias, J. P. J. Org. Chem. 1999,
64, 8256–8262.
_
17. Skarzewski, J.; Ostrycharz, E.; Siedlecka, R. Tetrahedron:
Asymmetry 1999, 10, 3457–3461.
ꢀ
ꢀ
9. Jansat, S.; Gomez, M.; Muller, G.; Dieguez, M.; Aghmiz,
A.; Claver, C.; Masdeu-Bulto, A. M.; Flores-Santos, L.;
18. (a) Mislow, K.; Green, M. M.; Laur, P.; Melillo, J. T.;
Simmons, T.; Ternay, A. L. J. Am. Chem. Soc. 1965, 87,
1958–1976; (b) Rosini, C.; Donnoli, M. I.; Superchi, S.
Chem. Eur. J. 2001, 7, 72–79.
19. Tokunoh, R.; Sodeoka, M.; Aoe, K.; Shibasaki, M.
Tetrahedron Lett. 1995, 36, 8035–8038.
20. (a) Rassias, G. A.; Page, P. C. B.; Reignier, S.; Christie, S.
D. R. Synlett 2000, 379–381; (b) Page, P. C. B.; Heaney,
H.; Reignier, S.; Rassias, G. A. Synlett 2003, 22–28.
21. (a) Hutton, A. T. In Comprehensive Coordination Chem-
istry; Wilkinson, G., Ed.; Pergamon: Oxford, 1987; Vol. 5,
ꢀ
Martin, E.; Maestro, M. A.; Mahia, J. Tetrahedron:
Asymmetry 2001, 12, 1469–1474.
10. For reviews on the catalytical use of sulfur containing
ꢀ
ligands, see: (a) Bayon, J. C.; Claver, C.; Masdeu-Bulto,
A. M. Coord. Chem. Rev. 1999, 193–195, 73–145; (b)
ꢀ
Masdeu-Bulto, A. M.; Dieguez, M.; Martin, E.; Gomez,
M. Coord. Chem. Rev. 2003, 242, 159–201; For the most
ꢀ
recent example, see: (c) Khiar, N.; Araujo, C. S.; Alvarez,
E.; Fernandez, I. Tetrahedron Lett. 2003, 44, 3401–3404.
ꢀ
ꢀ
ꢀ
ꢀ
_
ꢀ
€
11. Skarzewski, J.; Gupta, A.; Wojaczynska, E.; Siedlecka, R.
Synlett 2003, 1615–1619.
pp 1131–1155; (b) Muller, A.; Diemann, E. In Compre-
hensive Coordination Chemistry; Wilkinson, G., Ed.;
Pergamon: Oxford, 1987; Vol. 2, pp 551–558.
12. (a) Hata, T.; Sekine, M. Chem. Lett. 1974, 837–838; For a
review, see: (b) Valentine, D. H.; Hillhouse, J. H. Synthesis
2003, 317–334.
_
13. Siedlecka, R.; Skarzewski, J. Synlett 1996, 757–758.
14. Cardona, F.; Goti, A.; Picasso, S.; Vogel, P.; Brandi, A.
22. (a) Sprintz, J.; Helmchen, G. Tetrahedron Lett. 1993, 34,
1769; (b) Leutenegger, U.; Umbricht, G.; Fahrni, C.; Von
Matt, P.; Pfaltz, A. Tetrahedron 1992, 48, 2143–2156; (c)
Hayashi, T.; Yamamoto, A.; Hagihara, T.; Ito, Y.
Tetrahedron Lett. 1986, 27, 191–194.
J. Carbohydr. Chem. 2000, 19, 585–601.