Nonlinear effects as 'indicators' in the tuning of asymmetric catalysts

Article abstract of DOI:10.1016/S0957-4166(98)00168-2

Nonlinear effects (NLEs) have been used as a new tool to characterize the various modifications of a chiral titanium complex used in stoichiometric or catalytic mode in asymmetric sulfoxidation by cumyl hydroperoxide. Spectacular changes from (+)- to (-)-NLEs were observed for minor modifications of the experimental conditions.

Full text of DOI:10.1016/S0957-4166(98)00168-2

TETRAHEDRON:
ASYMMETRY
Pergamon
Tetrahedron: Asymmetry 9 (1998) 1941–1946
Nonlinear effects as ‘indicators’ in the tuning of asymmetric
catalysts
Jean-Michel Brunel, Timo O. Luukas and Henri B. Kagan∗
Laboratoire de Synthèse Asymétrique, URA CNRS 1497, Institut de Chimie Moléculaire d’Orsay, Université Paris-Sud, 91405
Orsay, France
Received 7 April 1998; accepted 27 April 1998
Abstract
Nonlinear effects (NLEs) have been used as a new tool to characterize the various modifications of a chiral
titanium complex used in stoichiometric or catalytic mode in asymmetric sulfoxidation by cumyl hydroperoxide.
Spectacular changes from (+)- to (−)-NLEs were observed for minor modifications of the experimental conditions.
© 1998 Elsevier Science Ltd. All rights reserved.
1. Introduction
Nonlinear effects (NLEs) have been discovered for various classes of reactions,1–8 both in catalytic and
stoichiometric reactions (with respect to the chiral auxiliary). It is a good indication that diastereomeric
species are involved directly in the catalytic cycle or in the reaction scheme, or remain as dormant and
unreactive species. The presence of a (+)-NLE (amplification) may have some practical consequences.
A nonlinear effect may be a useful piece of information to discuss mechanistic details of a reaction,
as exemplified in the case of the catalyzed asymmetric addition of diethylzinc to benzaldehyde.9 A
change of experimental conditions may also perturb the size and sign of a nonlinear effect. This has
been recently shown by Feringa et al. who found that addition of anthraquinone reversed the NLE in
the catalytic enantioselective allylic oxidation of cyclohexene using copper proline complexes and t-
butyl hydroperoxide.10 Kobayashi et al. also described important perturbations of NLE in Diels–Alder
reactions catalyzed by lanthanide triflates after introduction of achiral additives or after change of the
lanthanide ion.11
We wish to report here the use of the nonlinear effect as an ‘indicator’ for the comparison of closely
related chiral catalysts or chiral mediators. This approach will be illustrated in the specific case of
asymmetric sulfoxidation.
∗
Corresponding author. E-mail: kagan@icmo.u-psud.fr
0957-4166/98/$19.00 © 1998 Elsevier Science Ltd. All rights reserved.
PII: S0957-4166(98)00168-2

1942
J.-M. Brunel et al. / Tetrahedron: Asymmetry 9 (1998) 1941–1946
Fig. 1.
2. Results and discussion
We have developed the enantioselective oxidation of sulfides by hydroperoxides in the presence of
titanium complexes derived from the Sharpless reagent. We initially prepared the combination Ti(O-
iPr)4:(R,R)-diethyl tartrate (DET):H2O=1:2:1 which gave enantiomeric excesses of up to 90% in p-tolyl
methyl sulfide 1 oxidation with t-butyl hydroperoxide as oxidant.12 Modena et al. independently prepared
a reagent of stoichiometry Ti(O-iPr)4:(R,R)-DET=1:4 providing sulfoxide 2 in 88% ee.13 Careful control
of the generation of our titanium complex allowed >99% ee to be reached with cumene hydroperoxide
(CHP) as oxidant.14 We also studied a combination of Ti(O-iPr)4:(R,R)-DET:iPrOH=1:4:4 giving 90%
ee in (R)-2 (CHP as oxidant). This latter combination in the presence of 4 Å molecular sieves recently
allowed a catalyst (10% mol equiv.) to be set up giving up to 95% ee in (R)-2.15,16 A wide diversity
of titanium species17 with different behaviors are obviously generated by minor modifications of the
experimental conditions. We investigated some of the above sulfoxidations using DET of various ees
(Fig. 1).
We call A, B, C the following cases (oxidation of 1 into sulfoxide 2 by CHP):
A: 1 mol equiv. Ti(O-iPr)4:(R,R)-DET=1:4
B: Ti(O-iPr)4:(R,R)-DET:iPrOH=1:4:4
C: Ti(O-iPr)4:(R,R)-DET:iPrOH:MS 4 Å=1:4:4:1
Titanium complexes A (Modena reagent),13 B and C were prepared at 16°C in CH2Cl2 and the
reactions were performed at −25°C in the presence of 2 equiv. of CHP. These three systems were studied
at the 2 mmol scale and were compared as closely as possible (same batches of Ti(O-iPr)4, DET, CHP
and MS 4 Å and much care in the oxidation procedure).18 In the above experiments (R,R)-DET (100%
ee) gave 85–90% ee for stoichiometric reactions A and B and for the catalytic reaction C. However some
dramatic differences between the three systems could be seen by investigating the reaction on the full
range of ees of diethyl tartrate (Scheme 1, data in Tables 1 and 2).
Scheme 1.
Reagent A gave rise to a negative NLE quite similar to the NLE produced by our initial water-modified
reagent.3 Reagent B (1 mol equiv., B1) is characterized by a positive NLE, while its use in the catalytic

J.-M. Brunel et al. / Tetrahedron: Asymmetry 9 (1998) 1941–1946
1943
Table 1
Oxidation of sulfide 1 in the absence of molecular sieves (Scheme 1)
Table 2
Oxidation of sulfide 1 in the presence of molecular sieves (Scheme 1)
mode (10% mol equiv., B2) leads to a disappearance of NLE (Scheme 1). The catalyst B2 in the presence
of molecular sieves gave rise to a (−)-NLE (C1, Scheme 1). A second set of experiments in the presence
of molecular sieves was performed in which the amount of the titanium combination was modified from
10% mol equiv. to 1 mol equiv. (at a fixed concentration, 7×10−2 M, of sulfide 1). For a Ti/1 ratio of
between 1 and 10 the ee of sulfoxide 2 remains close to 90%, while there is a strong influence on the
NLE (Scheme 1). A 70% mol equiv. of combination C, case C2, gave a weak (−)-NLE while with 90%
and 100% mol equiv., cases C3 and C4, the NLE curve crosses the straight line. Here, DET of low ee
provides a (−)-NLE while DET of ee >40% gives a (+)-NLE. Such two-shape curves are quite rare.19
This surprising inversion of the sign of the nonlinear effect with the amount of the chiral titanium complex
may have several origins. The most likely is the influence of the concentration of the complex.
In order to gain some information on the mechanism of the reaction, the catalytic sulfoxidation C1
(DET: 100% ee) was followed as a function of the reaction time (Scheme 2, data in Table 3). Surprisingly
the ee of sulfoxide 2 increases during the course of the reaction (from 68% ee at 13% conversion) to 90%
ee at 90% conversion. It is not a kinetic resolution since the amount of the corresponding sulfone remains
very low (5% at 90% conversion), furthermore we checked that racemic sulfoxide 2 gives no kinetic
resolution in the presence of catalyst C1. The most likely explanation is an evolution of the catalyst

1944
J.-M. Brunel et al. / Tetrahedron: Asymmetry 9 (1998) 1941–1946
structure during the first half of the reaction.20,21 Since the ee of the product is conversion-dependent
it is necessary to check that the NLE curves of Scheme 1 involve similar conversions in order to make
comparisons. Hopefully all the reactions provided almost the full conversion of the sulfide 1, whatever
the ee of DET.
Scheme 2. Catalytic sulfoxidation of 1 (C1, DET 100% ee)
3. Conclusion
In conclusion, we established that the nonlinear effects may provide an extremely sensitive tool to
trace the modifications of a chiral catalyst, acting as a ‘finger print’, while the variation of the ee of the
product for an enantiopure chiral auxiliary may be less informative.22 This approach should also apply
to other systems giving NLE.23
Table 3
Oxidation of sulfide 1 (Scheme 2) as a function of reaction time

J.-M. Brunel et al. / Tetrahedron: Asymmetry 9 (1998) 1941–1946
1945
4. Experimental
Measurements of ees were made by HPLC analysis on a Spectroseries P100 pump module with a
Spectroseries UV100 detector and a Daicel Chiralcel OD-H column. Determinations for kinetics were
made on a Fisons CG8000 series gas chromatograph with a DB1 30 m column.
4.1. General procedure for catalytic oxidation of sulfide 1
All the operations were performed under argon. The Schlenk and stirrer bar were dried in the oven and
then capped with a septum and cooled under vacuum. The cold Schlenk was flushed with argon. CH2Cl2
was distilled from calcium hydride and stored over activated molecular sieves under argon. Molecular
sieves (pellets 4 Å) were purchased from Prolabo and activated at 200°C under reduced pressure (0.01
mmHg) for 48 hours. The diethyl tartrate as well as titanium tetraisopropoxide was distilled over a
Vigreux column before use. The commercially available cumene hydroperoxide (80% in cumene alcohol)
was purchased from Aldrich and used without purification. The active free peroxide was determined by
iodometric titration.
A solution of 50 µl of (R,R)-DET (0.3 mmol) in 1 ml of CH2Cl2 at 16°C was added to a Schlenk
containing 5% weight equivalent (with respect to the sulfide) of 4 Å molecular sieves. This solution
was stirred for 2.5 min and 22 µl of Ti(O-iPr)4 (0.07 mmol) was added slowly. The resulting mixture
was stirred for 10 min at 16°C and 22 µl of 2-propanol (0.3 mmol) was added slowly. Stirring was
maintained for 10 min followed by cooling in a freezer (−22°C) without stirring for an additional 20
min. The reaction was allowed to take place after rapid addition of 100 mg of sulfide 1 (0.7 mmol) in 9
ml of CH2Cl2 and pre-cooled (−22°C) CHP (1.4 mmol), with storage of the flask in the fridge (−22°C)
without stirring. After 16 h the mixture was poured into a solution of 1 g of ferrous sulfate heptahydrate
(3.6 mmol) and 300 mg of citric acid (1.4 mmol) in 15 ml of water, 1 ml of 1,4-dioxan, 20 ml of ether, and
was stirred for 15 min. The aqueous phase was extracted with ether (3×15 ml). The combined organic
phases were stirred vigorously with 15 ml of 2 M aqueous NaOH for 1 h. The aqueous solution was
then extracted with ether (3×15 ml). The combined organic solutions were dried over MgSO4, filtered
and evaporated under reduced pressure. Flash chromatography on silica gel (eluent: ethyl acetate) of the
crude product first afforded unreacted sulfide, sulfone, cumyl alcohol then pure sulfoxide. The fractions
of sulfoxide 2 were mixed before ee measurement (chiral HPLC) to avoid ee enrichment.18
Acknowledgements
We acknowledge Rhone-Poulenc Co for a fellowship to J.M.B. and for its financial support, and the
Academy of Finland for a fellowship to T.O.L. We thank Dr. P. Pitchen for useful discussions.
References
1. For recent reviews on NLE see Ref. 2. Initial reports on NLE between 1986 and 1990: Refs. 3–6. Simplified models for
discussing NLE: Refs. 7 and 8.
2. (a) Kagan, H. B.; Girard, C.; Guillaneux, D.; Rainford, D.; Samuel, O.; Zhang, S. Y.; Zhao, S. H. Acta Chem. Scand.,
1996, 50, 345; Avalos, M.; Babiano, R.; Cintas, P.; Jiménez, J. L.; Palacios, J. C. Tetrahedron: Asymmetry, 1997, 8, 2997;
(c) Bolm, C. In Advanced Asymmetric Catalysis, Ed. Stephenson, G. R., Blackie Academic and Professional, New York,
1996, pp. 9–26.
3. Puchot, C.; Samuel, O.; Dunach, E.; Zhao, S.; Agami, C.; Kagan, H. B. J. Am. Chem. Soc., 1986, 108, 2353.

1946
J.-M. Brunel et al. / Tetrahedron: Asymmetry 9 (1998) 1941–1946
4. Oguni, N.; Matsuda, Y.; Kaneko, T. J. Am. Chem. Soc., 1988, 110, 7877.
5. Kitamura, M.; Okada, S.; Suga, S.; Noyori, R. J. Am. Chem. Soc., 1989, 111, 4028.
6. Terada, M.; Mikami, K.; Nakai, T. J. Chem. Soc., Chem. Commun., 1990, 1623.
7. Guillaneux, D.; Zhao, S. H.; Rainford, D.; Samuel, O.; Kagan, H. B. J. Am. Chem. Soc., 1994, 116, 9430.
8. Blackmond, D. G. J. Am. Chem. Soc., 1997, 119, 12934.
9. Kitamura, M.; Suga, S.; Niwa, M.; Noyori, R. J. Am. Chem. Soc., 1995, 117, 4832.
10. Zondervan, C.; Feringa, B. L. Tetrahedron: Asymmetry, 1996, 7, 1895.
11. Kobayashi, S.; Ishitani, H.; Araki, M.; Hachiya, I. Tetrahedron Lett., 1994, 35, 6325.
12. Pitchen, P.; Dunach, E.; Deshmukh, M. N.; Kagan, H. B. J. Am. Chem. Soc., 1984, 106, 8188.
13. Di Furia, F.; Modena, G.; Seraglia, R. Synthesis, 1984, 325.
14. Brunel, J. M.; Diter, P.; Duetsch, M.; Kagan, H. B. J. Org. Chem., 1995, 60, 8036.
15. Brunel, J. M.; Kagan, H. B. Synlett, 1996, 404.
16. Brunel, J. M.; Kagan, H. B. Bull. Soc. Chim. Fr., 1996, 133, 1109.
17. For a discussion on the various titanium species which may be generated from mixing Ti(O-iPr)4 and tartrates in several
relative amounts see, Berrisford, D. J.; Bolm, C.; Sharpless, K. B. Angew. Chem. Int. Ed. Engl., 1995, 34, 1059.
18. Diter, P.; Taudien, S.; Samuel, O.; Kagan, H. B. J. Org. Chem., 1994, 59, 370.
19. For a prediction of multi-shape NLE curves based on a simple kinetic model see Ref. 7. For a case of the NLE curve
crossing the straight line see, Tanaka, K.; Mitsui, J.; Kawabata, H.; Watanabe, A. J. Chem. Soc., Chem. Comm., 1991,
1632.
20. Sulfoxide 2 has 81% ee at 50% conversion and 90% ee at 90% conversion. A simple calculation indicates that the catalyst
must produce essentially enantiomerically pure sulfoxide beyond 50% conversion.
21. Asymmetric autoinduction (modification of the catalyst by the product) may be operating here (see, Alberts, A. H.;
Wynberg, H. J. Am. Chem. Soc., 1989, 111, 7265; Bolm, C.; Bienewald, H.; Seger, A. Angew. Chem. Int. Ed. Engl.,
1996, 35, 1657). However, addition of either enantiomer of 2 at the beginning of the reaction did not improve the
enantioselectivity.
22. We are currently investigating the various equilibria between monomeric and dimeric titanium complexes involved in the
asymmetric oxidation. The concentration effect on NLEs may be quantitatively correlated with the distribution of titanium
complexes. Ti(O-iPr)4 has been found to be devoid of catalytic activity in sulfoxidation and cannot be responsible for a
(−)-NLE.
23. We also studied NLEs in Ti catalyzed TMSCN addition on benzaldehyde (see, Hayashi, M.; Matsuda, T.; Oguni, N.
J. Chem. Soc., Perkin Trans 1, 1992, 3135) or in Sharpless epoxidation of PhCH2CH_CH–CH2OH catalyzed by Ti(O-
iPr)4/DET. In both cases an inversion of the sign of NLE with catalytic amount has been observed, as in sulfoxidation
(unpublished results).