Exploiting π shielding interactions in a η6 arene-complexed chiral auxiliary

Article abstract of DOI:10.1016/S0040-4039(00)00713-9

The origin of asymmetric induction in a chiral auxiliary derived from isomannide was probed using a series of η⁶ arene chromium carbonyl complexes. Diastereoselectivity as high as 99% was attainable, and subsequent spectroscopic analysis shed light on the role of π-π interactions. (C) 2000 Elsevier Science Ltd.

Full text of DOI:10.1016/S0040-4039(00)00713-9

Tetrahedron Letters 41 (2000) 4695±4699
6
Exploiting p shielding interactions in a Z arene-complexed
chiral auxiliary
Graham B. Jones * and Mustafa Guzel
Bioorganic and Medicinal Chemistry Laboratories, Department of Chemistry, Northeastern University,
Boston, MA 02115, USA
Received 12 April 2000; revised 1 May 2000; accepted 2 May 2000
Abstract
The origin of asymmetric induction in a chiral auxiliary derived from isomannide was probed using a
6
series of Z arene chromium carbonyl complexes. Diastereoselectivity as high as 99% was attainable, and
subsequent spectroscopic analysis shed light on the role of p±p interactions. # 2000 Elsevier Science Ltd.
All rights reserved.
Arene chromium(0) carbonyl complexes are versatile components of the synthetic arsenal, a
1
consequence of the broad range of chemical and stereochemical properties they possess. On
complexation to an arene ring, signi®cant changes to its chemistry occur, including increased
2
3
susceptibility to nucleophilic attack, increased acidity of ring protons, enhanced solvolytic
4
5
properties, in addition to introduction of the stereodirective capacity of the metal carbonyl tripod.
Due to the ability of arene chromium carbonyl complexes to in¯uence the electronic properties of
both ring substituents and the p orbitals of the arene, we have become interested in developing
chiral controller systems which harness these e€ects, as depicted in 1.
We previously demonstrated that mixed ligand complexes of 8-phenylmenthol acrylates 2 exert
electronic in¯uence on the diastereoselectivity of cycloadditions to the acrylate, presumably via
modulation of aryl-vinyl p±p stacking interactions, and that the process can be tuned by appropriate
6
selection of ligand X. To investigate the scope of such modulation, we sought to demonstrate the
*
Corresponding author. Fax: 617-373-8795; e-mail: grjones@lynx.neu.edu
0040-4039/00/$ - see front matter # 2000 Elsevier Science Ltd. All rights reserved.
PII: S0040-4039(00)00713-9

4
696
e€ects in a more readily accessible system, and selected isomannide acrylates 3 for our study.
Loupy has reported that cycloaddition to this acrylate may be in¯uenced by p-stacking interactions,
helping account for the good levels of asymmetric induction observed. The basic criteria for two
7
entities to engage in a p stack are satis®ed when p clouds interact through space at a distance of
Ê
Ê
8
between 3±3.5 A, with optimal overlap at approximately 3.4 A. To investigate this system further,
and attempt modulation of the p±p interactions, ligand 4 was transformed into a monobenzylated
common precursor, which could be converted to 3 or control substrate 6 as desired (Scheme 1).
6
Alternatively, Z complexation followed by ligand substitution of TMS ether 5 gave ligand
precursors 7, which were deprotected and converted to give a family of acrylates 8. Surprisingly,
9
although acrylates 8, [X=CO, P(OMe) , P(OEt) ] were stable, the bulkier triphenylphosphite
3
3
analog [X=P(OPh) ] was not isolable, resulting in decomplexation even under a variety of con-
3
ditions. Cycloaddition of the acrylates 3 and 8 was conducted using cyclopentadiene, and various
Lewis acid combinations (Scheme 2). Following decomplexation, the de of the cycloadduct 9 was
assessed using chiral HPLC (OD column), yielding some important ®ndings (Table 1). As had
7
been reported, appreciable de is obtained with 3, which could be improved further by an appropriate
choice of Lewis acid (entries 1 and 5). In all cases examined the endo-R enantiomeric cycloaddition
product predominated. Since p±p interactions were reasoned to contribute to diastereocontrol in
7
substrate 3, it was surprising to ®nd that in the case of the p de®cient 8, X=CO, diastereocontrol
is enhanced, contrary to that observed in the analogous 8-phenylmenthol acrylate 2, X=CO.6
The more electron-rich mixed ligand systems performed even better, the trend seemingly re¯ecting
steric rather than electronic factors (Table 1).
6
Scheme 1. Preparation of mixed ligand Z arene isomannide acrylates. Reactions and conditions: (a) BnCl, LiH, DMF,
6
DMF 99%; (e) Cr(CO)
8%; (b) CH
3
CH
2
COOH, HOBt, EDAC, THF, 94%; (c) ClCOCHˆCH
2
, 2,6-lutidine 30±40%; (d) TMSCl, imidazole,
CO , MeOH 90±95%
3
ꢀ
6
, nBu
2
O, THF, 125 C, 16 h, 95%; (f) X, hꢀ, PhH 75±90%; (g) K
2
In an attempt to rationalize these ®ndings, a number of spectroscopic studies were undertaken
using ligands 3, 6 and 8. Assuming s-trans enoate geometry is adopted in the Lewis acid catalyzed
cycloaddition, a p±p stacked model (A) does not account for the observed endo-R cycloadduct,
which would require addition to the (p-shielded) upper face, and which also imposes unfavorable

4
697
Scheme 2. Diastereocycloaddition of isommanide acrylates
Table 1
2 2
Cycloaddition of acrylates 3 and 8 with cyclopentadiene. All cycloadditions conducted in CH Cl .
Cycloadducts were decomplexed in Et O, then analyzed by chiral HPLC (Chiracel OD, 10% 2-
2
propanol:90% hexanes as eluent, 1 mL/min ¯ow rate) using a racemic endo-9 prepared by alternate
R
methods as both an internal and external control (t endo-S=18.9 min endo-R=20.2 min)
geometry on the ester function.10 Fluorescence-quenching studies indicated a concentration-
dependent interaction between the acrylate and aryl moieties of 3 in the absence of Lewis acid.11
No quenching was observed with the analogous arene complexes 8 (or control element 6), either
in the presence or absence of Lewis acid, which may indicate the metal carbonyl tripod either
imparts a conformational bias to the benzyl group, or electronically insulates the arene. Chemical
ꢀ
shift analysis of the acrylate protons in 3 and 8 (CDCl , 25 C) did not reveal a trend for up®eld
3
8
shift consistent with p stacking with the phenyl ring, instead giving near uniform values for the
entire series, even in the presence of Lewis acid (Table 2). More revealing were NOE studies,
which failed to show any NOE between the vinyl group protons and either the aryl moiety or
benzylic hydrogens, in both 3 and 8. However, in the presence of EtAlCl , NOEs were observed
2
between the phenyl H and the acrylate H (5%) and H (2%), suggesting that a conformational
4
b
c
12
change ensues when Lewis acid coordinates to the enoate. Taken together, the data is suggestive
of a conformation B, where approach of the diene is directed to the underside of the enoate,
leading to the observed endo-R cycloadduct.

4
698
Table 2
H NMR analysis of isomannide acrylates (ꢁ)
1
The correlation of steric bulk with de in the case of derivatives 8, can be satis®ed with structure
C, the metal carbonyl group serving to amplify p-shielding of the upper face of the enoate. Other
possibilities include interaction of (the metal-complexed) arene with the acrylate by some form of
13
charge transfer process that is not detected by ¯uorescence quenching, or the participation of
the metal carbonyl tripod itself in binding to the Lewis acid.
14
15
Given the commercial availability of the ligand 4 and the very high levels of asymmetric
induction attainable (Table 1, entry 12), immediate application of this auxiliary can be anticipated,
as can the design of catalysts which bene®t from such stereodirective elements.16 The results
herein suggest that in systems where arene-vinyl p-shielding plays an intergral role in stereo-
control, metal carbonyl-complexed aryl derivatives may serve to amplify selectivity.
Acknowledgements
We thank the PRF (Administered by the American Chemical Society) for ®nancial support of
this work (33920-AC1).
References
1
. For some examples, see: Merlic, C. A.; Walsh, J. C. Tetrahedron Lett. 1998, 39, 2083; Taniguchi, N.; Uemura, M.
Tetrahedron Lett. 1997, 38, 7199; Arin, A.; Blake, A. J.; Li, W.-S.; Simpkins, N. S. Synlett 1997, 1453; Cowton,
E. L. M.; Gibson, S. E.; Schneider, M. J.; Smith, M. H. J. Chem. Soc., Chem. Commun. 1996, 839; Hayashi, Y.;
Sakai, H.; Kaneta, N.; Uemura, M. J. Organometal. Chem. 1995, 503, 143; Davies, S. G.; Goodfellow, C. L.;
Sutton, K. H. Tetrahedron: Asymmetry 1992, 3, 1303.
2
. Card, R. J.; Trahanovsky, W. S. Tetrahedron Lett. 1973, 3823; Semmelhack, M. F.; Hall, H. T.; Yoshifuji, M.;
Clark, G. J. Am. Chem. Soc. 1975, 97, 1247; Semmelhack, M. F.; Wul€, W.; Garcia, J. L. J. Organometal. Chem.
1
982, 240; C5; Kozikowski, A. P.; Isobe, R. J. Chem. Soc., Chem. Commun. 1978, 1076.
3
. Card, R. J.; Trahanovsky, W. S. J. Org. Chem. 1980, 45, 2556; Semmelhack, M. F.; Bisaha, J.; Czarny, M. J. Am.
Chem. Soc. 1979, 101, 768; Jaouen, G.; Meyer, A.; Simmonneaux, G. J. Chem. Soc. Chem. Commun., 1975, 813;
Masters, N. F.; Mathews, N.; Nechvatal, G.; Widdowson, D. Tetrahedron 1989, 45, 5955.
4
. Top, S.; Jaouen, G. J. Org. Chem. 1981, 46, 78.
5
. Bolm, C.; Muniz, K. Chem. Soc. Rev. 1999, 28, 51; Son, S. U.; Jang, H.-Y.; Han, J. W.; Lee, I. S.; Chung, Y. K.
Tetrahedron: Asymmetry 1999, 10, 347; Nelson, S. G.; Hil®ker, M. A. Org. Lett. 1999, 9, 1379; Pasquier, C.; Naili,
S.; Pelinski, L.; Brocard, J.; Mortreux, A.; Agbossou, F. Tetrahedron: Asymmetry 1998, 9, 193; Merlic, C. A.;
Walsh, J. C. Tetrahedron Lett. 1998, 39, 2083; Malfait, S.; Pelinski, L.; Brocard, J. Tetrahedron: Asymmetry 1996,
7, 653; Uemura, M.; Miyake, R.; Nishimura, H.; Matsumoto, Y.; Hayashi, T. Tetrahedron: Asymmetry 1992, 3,
2
13, and references cited therein.
6
7
. Jones, G. B.; Chapman, B. J.; Mathews, J. E. J. Org. Chem. 1998, 63, 2928.
. Loupy, A.; Monteux, D. Tetrahedron Lett. 1996, 37, 7023.

4
699
8
9
. Jones, G. B.; Chapman, B. J. Synthesis 1995, 475.
. Satisfactory spectroscopic and analytical data were obtained for all new compounds.
1
0. For discussion of geometries of Lewis acid coordinated enoates and enals, see: Seyden-Penne, J. Chiral Auxiliaries
and Ligands in Asymmetric Synthesis; John Wiley & Sons: New York, 1995.
1. Analysis conducted at 10- through 10-^7 M substrate in CH Cl (3; l abs 240 nm, l emm 280 nm). A 100-fold
^5
1
2
2
dilution resulted in 20% reduction in ¯uorescence quenching. For protocols, see: Whitesell, J. K.; Lawrence,
R. M.; Chen, H.-H. J. Org. Chem. 1986, 51, 4779; Whitesell, J. K. Chem. Revs. 1992, 92, 953.
1
1
2. Interestingly, these NOEs became negative when aggregation ensued (at higher concentration).
3. For a complex in which an arene chromium tricarbonyl group serves to stabilize an electron de®cient arene has
been reported, see: Kaneta, N.; Mitamura, F.; Uemura, M.; Murata, Y.; Komatsu, K. Tetrahedron Lett. 1996, 37,
5
385.
14. Lokshin, B. V.; Rusach, E. B.; Kolobova, N. E.; Makarov, Y. V.; Ustynyuk, N. A.; Zdanovich, V. I.; Zhakaeva,
A. Z.; Setkina, V. N. J. Organometal. Chem. 1976, 108, 353.
1
1
5. Isomannide: Aldrich cat. No. 32,926-6, 25 g=$34.80.
6. For recent examples, see: Jones, G. B.; Guzel, M. Tetrahedron: Asymmetry 1998, 9, 2023; Jones, G. B.; Guzel, M.;
Chapman, B. J. Tetrahedron: Asymmetry 1998, 9, 901; Jones, G. B.; Huber, R. S.; Chapman, B. J. Tetrahedron:
Asymmetry 1997, 8, 1797; Jones, G. B.; Heaton, S. B.; Guzel, M.; Chapman, B. J. Tetrahedron: Asymmetry 1997,
8, 3625; Jones, G. B.; Chapman, B. J. Synlett 1997, 439.