C-Phosphanylated sulfoximines: synthesis and applications in asymmetric allylic substitution reactions

Article abstract of DOI:10.1016/j.tetasy.2006.01.008

Starting from cyclic sulfonimidates, a number of C-phosphanylated enantiomerically pure sulfoximines have been prepared either as such or in protected form. Two of them, 5a and epi-5a, have been used as ligands in palladium complex catalyzed allylic subst

Full text of DOI:10.1016/j.tetasy.2006.01.008

Tetrahedron: Asymmetry 17 (2006) 500–503
C-Phosphanylated sulfoximines: synthesis and applications
in asymmetric allylic substitution reactions
Volker Spohr, Jan Philipp Kaiser and Michael Reggelin*
Clemens-Scho¨pf Institut fu¨r Organische Chemie und Biochemie, Technische Universita¨t Darmstadt,
Petersenstraße 22, 64287 Darmstadt, Germany
Received 6 December 2005; accepted 5 January 2006
Available online 20 February 2006
Abstract—Starting from cyclic sulfonimidates, a number of C-phosphanylated enantiomerically pure sulfoximines have been pre-
pared either as such or in protected form. Two of them, 5a and epi-5a, have been used as ligands in palladium complex catalyzed
allylic substitution reactions delivering the substitution product with up to 95% ee.
Ó 2006 Elsevier Ltd. All rights reserved.
1. Introduction
tion. Amongst others, these and structurally related sys-
tems have been applied to the asymmetric allylic
alkylation (AAA) reaction13,14 and, for the PHOX li-
gands, their mode of action has been studied thoroughly
by NMR spectroscopy.15–19 In the course of our ongo-
ing studies to elaborate upon enantiomerically pure sul-
foximines as powerful reagents for stoichiometric
asymmetric transformations,20–24 we became struck by
the idea of using these versatile compounds as ligands
for asymmetric catalysis (Scheme 1).25
Chiral bidentate P,N-ligands comprise an important
class of ligands for asymmetric metal catalyzed transfor-
mations.1 This is true, especially for systems displaying
imino nitrogens and carbon-bound phosphanes as
donor atoms, which in turn can be traced back to the
favorable balance between the r-donor ability of the
former and the p-acceptor character of the latter binding
site. Since the pioneering work of Faller on the elec-
tronic differentiation of allylic termini2,3 in chiral molyb-
denum complexes, a large number of very successful
ligands based on that principle have been described
(Fig. 1).
O
S
O
S
O
O
X
N
N
O
S
Y
5
N
R1
O
O
S
R1
Z
R2
N
4
epi-4
N
Ph2P
N
PPh2
R
R2
PPh2
Y = (CH2)nPPh2; X = H; Z = OH
X = H; Y = CH3; Z = PPh2 X = PPh2; Y = tBu; Z = OTBS
1
2
3
Figure 1. Important C-phosphanylated imino P,N-ligands.
Due to their structural resemblence to the ligands to be
reported herein, the imino phosphanes 1,4–9 sulfinyl-
imino phosphanes 2 (described by Schenkel and Ell-
man10,11) and phosphinooxazoline (PHOX) ligands 3
developed by Helmchen and Pfaltz12 merit special atten-
O
N
O
S
S
N
N
S
PPh2
n
PPh2
PPh2
O
H3C
5a (Type I)
OH
OH
n = 1: 5b (Type II)
n = 2: 5c (Type II)
5d (Type III)
*
Corresponding author. Tel.: +49 6151 165540; fax: +49 6151
165531; e-mail: re@punk.oc.chemie.tu-darmstadt.de
Scheme 1. C-Phosphanylated sulfoximines from cyclic sulfonimidates.
0957-4166/$ - see front matter Ó 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetasy.2006.01.008

V. Spohr et al. / Tetrahedron: Asymmetry 17 (2006) 500–503
501
A closer inspection of the generic structure 5 summariz-
ing a whole class of novel C-phosphanylated sulfoxi-
mines reveals not only interesting similarities between
ligands 1–3, but also differences related to the interplay
of the C- and S-stereogenic centers. When comparing 1
to 5a and 5b, it is obvious that all compounds should
form five-membered ring chelates after metal complexa-
tion. However contrary to 1, the stereogenic sulfur
atoms appear as additional stereochemical control
elements. Moreover, in 5a and 5b the position of both
stereogenic centers is inverted. This raises interesting
questions concerning their relative influence on the
stereochemical outcome of reactions and to a possible
mutual reinforcement of the asymmetric induction
provoked by them. Ligands 2 and 3 are similar to 5c
and 5d with respect to chelation (six-membered ring),
but here again a concerted action of both stereogenic
centers is to be expected. For these reasons and due to
the fact that all four types of phosphino sulfoximines
can be easily obtained by the commercially available sul-
fonimidates 4 and epi-4, we started a research program
to explore the suitability of these ligand types for asym-
metric catalysis.
Figure 2. Crystal structures of the S-epimeric borane complexes 8 (A)
and epi-8 (B). Black: C; red: O; blue: N; yellow: S; magenta: P; cyan: B.
It is interesting to note that their solid state structures
are quite different. Especially the NCCP torsional angle
amounting 76° for 8 and 175° for epi-8, which indicates
that the former might be much better suited for chela-
tion of a metal in comparison to the latter. Due to the
fact that the free phosphanes 5a and epi-5a are rather
sensitive to oxidation, their liberation from the boranes
by DABCO should be part of the complexation
procedure.
Herein we report the syntheses of 5a, epi-5a, 5b (as
oxide), and 5c (as borane complex) as well as applica-
tions of the first two ligands in the AAA-reaction.
2. Results and discussion
For the synthesis of a protected form of ligand 5b we
first tried to react the lithiated borane complexed methyl
phosphane 9a with the sulfonimidate 4 (Scheme 3).
The ring opening of 4 and epi-4 by MeLi delivers the
methyl sulfoximines 6 and epi-6 in nearly quantitative
yield (Scheme 2).22,24,26,27 The subsequent mesylation
required the application of the anhydride. With mesyl
chloride, variable amounts of chlorinated products were
obtained, presumably via aziridinium intermediates.25
The reaction to the mesylates 7 and epi-7 was complete
after 2 h at 0 °C and therefore, could be used without
further purification in the phosphanylation with the
potassium diphenylphosphide–borane complex. This
latter compound was prepared by deprotonation of the
corresponding diphenylphosphane complex using potas-
sium tbutanolate (KOtBu) as a base. The borane pro-
tected phosphanylated sulfoximines 8 and epi-8 were
isolated in 90% and 73% yield from the alcohols.
O
1) nBuLi, 0°C, THF
S
P
2) 4
N
PPh2
X
X
CH3
OH
X = BH3 : 10a
10b
X = BH3 : 9a
9b
X = O :
X = O :
Scheme 3. Type II ligands.
To our surprise no product could be isolated and we
therefore next used oxide 9b in our second attempt. This
time the desired phosphanylated sulfoximine 10b was
obtained in 31% yield, after 12 h of reaction time within
Both compounds are crystalline solids, which were ame-
nable to X-ray structural analysis (Fig. 2).32
N
O
O
N
S
OH
N
S
OMs
N
S
PPh2
BH3
N
S
PPh2
a)
b)
c)
d)
S
96%
90% (b,c)
quant.
O
O
O
O
H3C
H3C
H3C
H3C
4
6
7
8
5a
a)
b)
c)
d)
epi-4
epi-6
epi-7
epi-8
epi-5a
98%
73% (b,c)
quant.
Scheme 2. Synthesis of 5a and epi-5a. Reagents and conditions: (a) MeLi, THF, À78 °C; (b) methanesulfonyl anhydride, NEt3, CH2Cl2, 0 °C; (c)
KPPh2(BH3), THF, rt; (d) DABCO, toluene, 65 °C.

502
V. Spohr et al. / Tetrahedron: Asymmetry 17 (2006) 500–503
which the temperature was slowly raised from À50 °C to
rt. Finally we tried to gain access to phosphino sulfoxi-
mines incorporating a PCCSN bond path, as found in
structures 5c and 5d. In fact, we were successful in pre-
paring the latter compound via ortho-lithiation of the
corresponding phosphane free precursor. However the
product turned out to be very sensitive to oxidation
and especially to acidic conditions.28 Therefore, we syn-
thesized 13 as an alternative (Scheme 4).
To the best of our knowledge, this is the first application
of C-phosphanylated sulfoximines in asymmetric allylic
substitutions. In fact there is only one other application
of this type of compounds in asymmetric catalysis. In a
very recent paper, Bolm et al. described Ir-catalyzed
hydrogenation reactions of imines using diphenylphos-
phanyl sulfoximines as ligands.29 As substrates we used
1,3-diphenylallyl acetate rac-14a, (E)-3-penten-2-ol ace-
tate rac-14b and cyclohexenyl acetate rac-14c. The C-
nucleophile was prepared from dimethylmalonate using
the BSA (N,O-bis(trimethylsilyl)acetamide) method in
the presence of a catalytic amount of potassium ace-
tate.30 The reactions were carried out in either CH2Cl2,
THF, or toluene at room temperature for reaction times
as indicated in Table 1 at a substrate concentration of
0.15 M. The catalysts were generated by first deboranat-
ing 8 and epi-8 with DABCO in toluene for 4 h at 65 °C
followed by addition of [Pd(allyl)Cl]2 as the palladium
source. The first three entries in Table 1 clearly indicate
that CH2Cl2 is the best solvent for the process, deliver-
ing (R)-15a in an almost quantitative yield with 93%
ee in only 10 min.
O
N
O
N
O
N
b), c)
d)
S
S
S
R = CH2OH
R
PPh2
BH3
OTBS
OTBS
OTBS
R = H: 11a
R = CH2OH: 11b
12
13 80% (b-d)
a)
Scheme 4. Reagents and conditions: (a) nBuLi, THF, À78 °C then
CH2O; (b) nBuLi, THF, À78 to 0 °C then ClCO2Me; (c) KOtBu, THF,
0 °C rt; (d) HPPh2(BH3), KOtBu, THF, rt then 12.
The other solvents, especially THF, led to a drastic
reduction of the reaction rate, although the ee remained
almost constant. A 10-fold reduction of the catalyst
load in the privileged solvent was possible (90% yield,
91% ee) but entailed unacceptably long reaction times
(entry 4). To gain insight into the interplay of the differ-
ent sources of chirality in the ligand, we inverted the
absolute configuration at the sulfur without changing
the one at carbon. Unsurprisingly this change had no
effect on the absolute configuration of the product (entry
5). Moreover, only a minor (if any) effect on the stereo-
selectivity was observed (95% ee), which is a strong
hint that the asymmetric induction is mainly governed
by the C-stereogenic center as part of the assumed
five-membered ring chelate being formed after complex-
ation. Similar results were obtained with monodentate
BINOL derived N-phosphino sulfoximines.31 Although
a little bit more pronounced than in our case, the abso-
lute configuration at sulfur plays only a minor role in
these systems too. The only recognizable significant
change was on reactivity. With epi-5a as ligand, the reac-
tion proceeded at least 6 times slower than with 5a,
which might be a consequence of the unfavorable con-
formational preferences of the former prior to complex-
ation (Fig. 2). Unfortunately, aliphatic substrates 14b
and 14c gave only unsatisfying stereoselectivites (entries
6 and 7). If one assumes that the stereochemical out-
come of the reaction is governed by similar factors as
found with the PHOX-ligands, this result is not very
surprising.12
Hydroxyalkylation of the lithiated methyl sulfoximine
11a24–26 with gaseous formaldehyde (generated from
paraformaldehyde by heating) followed by acylation
with methyl chloroformate and KOtBu induced elimina-
tion furnished vinyl sulfoximine 12. Conjugate addition
of KPPh2(BH3) delivered the borane complexed phos-
phino sulfoximine 13, which was fully characterized by
2D NMR and high resolution mass spectroscopy.
Whereas the three-step sequence from alcohol 11b to
the target compound worked with an 80% overall yield
without purification of intermediates, the reaction of
metalated 11a with formaldehyde turned out to be
rather unsatisfactory (30%). To our surprise it was not
possible to prepare 12 directly, via ring opening of 4
with vinyl lithium.
2.1. Pd-complex catalyzed allylic substitutions
Next we turned our attention to the application of 5a
and epi-5a in the palladium complex catalyzed allylic
substitution reaction (Scheme 5).
O
O
OAc
R
MeO
R
OMe
[Pd(allyl)Cl]2 / ligand
R
CH2(CO2Me)2, BSA, KOAc, t
R
R = Ph: rac-14a
R = Me: rac-14b
R = Ph: rac-15a
R = Me: rac-15b
3. Conclusions
O
O
OAc
MeO
OMe
C-phosphanylated, enantiomerically pure sulfoximines
of types I and II (Scheme 1) have been prepared for
the first time by applying the commercially available sul-
fonimidates 4 and epi-4 as common precursors. After
deboranation of 8 and epi-8 the resulting sulfoximino
phosphanes can be used as ligands in the palladium
[Pd(allyl)Cl]2 / ligand
CH2(CO2Me)2, BSA, KOAc, t
rac-14c
rac-15c
Scheme 5. Reagents and conditions used for the AAA reactions.

V. Spohr et al. / Tetrahedron: Asymmetry 17 (2006) 500–503
503
Table 1. Palladium complex catalyzed allylic substitution reactions as illustrated in Scheme 5a
7. Suzuki, Y.; Ogata, Y.; Hiroi, K. Tetrahedron: Asymmetry
1999, 10, 1219.
8. Saitoh, A.; Achiwa, K.; Morimoto, T. Tetrahedron:
Asymmetry 1998, 9, 741.
9. Saitoh, A.; Morimoto, T.; Achiwa, K. Tetrahedron:
Asymmetry 1997, 8, 3567.
10. Schenkel, L. B.; Ellman, J. A. J. Org. Chem. 2004, 69,
1800.
11. Schenkel, L. B.; Ellman, J. A. Org. Lett. 2003, 5, 545.
12. Helmchen, G.; Pfaltz, A. Acc. Chem. Res. 2000, 33, 336.
13. Trost, B. M.; Vranken, D. L. v. Chem. Rev. 1996, 96, 395.
14. Trost, B. M.; Breit, B.; Organ, M. G. Tetrahedron Lett.
1994, 35, 5817.
complex catalyzed allylic substitution reaction. Both
S-epimeric complexes proved to be very active catalysts
delivering the substitution product (R)-15 with nearly
quantitative yield on a minutes time scale (with 5a)
and P93% ee. These results clearly indicate that the
major source of enantioselectivity is the valine derived
C-stereogenic center, whereas the electron rich imino
nitrogen of the sulfoximine serves as a rate enhancer very
much like the amidine moiety in the VALAP-ligand.9
The major difference between the two epimers is the
much higher reactivity of 5a, as compared to epi-5a.
The ortho-phosphanylated sulfoximine 5d turned out
to be too acid sensitive to be of any practical value,
whereas the type II ligands were not only stable com-
pounds, but also appeared very promising from a stereo-
chemical point of view. Here the S-stereogenic center
would be part of the metal chelate, thus a pronounced
effect on the stereochemical outcome of the reaction in
combination with the rate enhancing effect of the sulfox-
imine nitrogen can be expected. Work along these lines
is currently in progress as well as solution NMR studies
on the structure of the complexes.
15. Kollmar, M.; Goldfuss, B.; Reggelin, M.; Rominger, F.;
Helmchen, G. Chem. Eur. J. 2001, 7, 4913.
16. Junker, J.; Reif, B.; Steinhagen, H.; Junker, B.; Felli, I. C.;
Reggelin, M.; Griesinger, C. Chem. Eur. J. 2000, 6, 3281.
17. Reif, B.; Steinhagen, H.; Junker, B.; Reggelin, M.;
Griesinger, C. Angew. Chem., Int. Ed. 1998, 37, 1903.
18. Steinhagen, H.; Reggelin, M.; Helmchen, G. Angew.
Chem., Int. Ed. 1997, 36, 2108.
19. Sprinz, J.; Kiefer, M.; Helmchen, G.; Reggelin, M.
Tetrahedron Lett. 1994, 35, 1523.
20. Reggelin, M.; Junker, B.; Heinrich, T.; Slavik, S.; Buhle,
¨
P. J. Am. Chem. Soc., submitted for publication.
21. Reggelin, M.; Junker, B. Chem. Eur. J. 2001, 7, 1232.
22. Reggelin, M.; Gerlach, M.; Vogt, M. Eur. J. Org. Chem.
1999, 1011.
23. Reggelin, M.; Heinrich, T. Angew. Chem., Int. Ed. 1998,
37, 2883.
24. Reggelin, M.; Weinberger, H.; Gerlach, M.; Welcker, R. J.
Am. Chem. Soc. 1996, 118, 4765.
25. Reggelin, M.; Weinberger, H.; Spohr, V. Adv. Synth.
Catal. 2004, 346, 1295.
Acknowledgements
We thank Ms. S. Foro (TU Darmstadt) for her careful
work on the X-ray structures and Professor H. Fueß
(Department of Material Science, TU-Darmstadt) for
measuring time on his diffractometer. Continuous sup-
port from the Solvay Pharmaceuticals (Hannover, Ger-
many) is gratefully acknowledged.
26. Reggelin, M.; Weinberger, H. Tetrahedron Lett. 1992, 33,
6959.
27. Reggelin, M.; Weinberger, H. Angew. Chem., Int. Ed.
1994, 33, 444.
28. Dupas, G.; Levacher, V. Tetrahedron 2005, 61, 8138.
29. Moessner, C.; Bolm, C. Angew. Chem., Int. Ed. 2005, 44,
7564.
References
1. Guiry, P. J.; Saunders, C. P. Adv. Synth. Catal. 2004, 346,
497.
30. Trost, B. M.; Lautens, M. J. Am. Chem. Soc. 1987, 109,
1469.
2. Faller, J. W.; Chao, K.-H.; Murray, H. H. Organometal-
lics 1984, 3, 1231.
31. Reetz, M. T.; Bondarev, O. G.; Gais, H. J.; Bolm, C.
Tetrahedron Lett. 2005, 46, 5643.
3. Adams, R. D.; Chodosh, D. F.; Faller, J. W.; Rosan, A.
M. J. Am. Chem. Soc. 1979, 101, 2570.
4. Hiroi, K.; Watanabe, K. Tetrahedron: Asymmetry 2001,
12, 3067.
5. Morimoto, T.; Yamaguchi, Y.; Suzuki, M.; Saitoh, A.
Tetrahedron Lett. 2000, 41, 10025.
32. Crystallographic data (excluding structure factors) for the
structures in this paper have been deposited with the
Cambridge Crystallographic Data Centre as supplemen-
tary publication numbers CCDC 289283 and 289284.
Copies of the data can be obtained, free of charge, on
application to CCDC, 12 Union Road, Cambridge CB2
1EZ, UK [fax: +44(0) 1223 336033 or e-mail: deposit@
ccdc.cam.ac.uk].
6. Saitoh, A.; Achiwa, K.; Tanaka, K.; Morimoto, T. J. Org.
Chem. 2000, 65, 4227.