Stereochemical memory effects in alkene radical cation/anion contact ion pairs: Effect of substituents, and models for diastereoselectivity

Article abstract of DOI:10.1021/ja051657t

A series of 12 stereochemically defined 2,m-dimethyl- and 2,m,n-trimethyl-6-benzylamino-2-nitro-3-(diphenylphosphatoxy)hexanes have been synthesized and their cyclization reactions leading to di- and trisubstituted N-benzyl pyrrolidines examined in the presence of tributyltin hydride and azoisobutyronitrile in benzene at reflux. The cyclizations are interpreted in terms of generation of an alkyl radical by abstraction of the nitro group with a stannyl radical. The phosphate leaving group is then expelled in a heterolytic cleavage to give a contact alkene radical cation/phosphate anion pair. For the majority of the examples studied, the cyclizations are best understood in terms of nucleophilic attack by the amine on the opposite face of the alkene radical cation to the one shielded by the leaving group, within the confines of the initial contact ion pair, resulting in overall cyclization with inversion of configuration. Dependent on the relative stereochemistry of the substituents, the cyclization is envisaged as taking place through either chair-like or twist-boat-like transition states with the maximum number of substituents pseudo-equatorial. The model breaks down when cyclization on the initial contact ion pair would engender significant destabilizing steric interactions, especially ^1,3A strain in the alkene radical cation. In these cases a fully equilibrated Beckwith-Houk-type transition state provides a satisfactory model. Interesting examples of matching and mismatching in the Corey-type oxazaborolidine-mediated reduction of alkyl (methyl-1-nitroethyl) ketones by a β-methyl group in the alkyl chain are reported, and the mismatching is attributed to a developing syn-pentane interaction in the transition state.

Full text of DOI:10.1021/ja051657t

Published on Web 06/15/2005
Stereochemical Memory Effects in Alkene Radical Cation/
Anion Contact Ion Pairs: Effect of Substituents, and Models
for Diastereoselectivity
David Crich* and Krishnakumar Ranganathan
Contribution from the Department of Chemistry, UniVersity of Illinois at Chicago,
845 West Taylor Street, Chicago, Illinois 60607-7061
Received March 15, 2005; E-mail: Dcrich@uic.edu
Abstract: A series of 12 stereochemically defined 2,m-dimethyl- and 2,m,n-trimethyl-6-benzylamino-2-
nitro-3-(diphenylphosphatoxy)hexanes have been synthesized and their cyclization reactions leading to
di- and trisubstituted N-benzyl pyrrolidines examined in the presence of tributyltin hydride and azoiso-
butyronitrile in benzene at reflux. The cyclizations are interpreted in terms of generation of an alkyl radical
by abstraction of the nitro group with a stannyl radical. The phosphate leaving group is then expelled in a
heterolytic cleavage to give a contact alkene radical cation/phosphate anion pair. For the majority of the
examples studied, the cyclizations are best understood in terms of nucleophilic attack by the amine on the
opposite face of the alkene radical cation to the one shielded by the leaving group, within the confines of
the initial contact ion pair, resulting in overall cyclization with inversion of configuration. Dependent on the
relative stereochemistry of the substituents, the cyclization is envisaged as taking place through either
chair-like or twist-boat-like transition states with the maximum number of substituents pseudo-equatorial.
The model breaks down when cyclization on the initial contact ion pair would engender significant
destabilizing steric interactions, especially 1,3A strain in the alkene radical cation. In these cases a fully
equilibrated Beckwith-Houk-type transition state provides a satisfactory model. Interesting examples of
matching and mismatching in the Corey-type oxazaborolidine-mediated reduction of alkyl (methyl-1-nitroethyl)
ketones by a â-methyl group in the alkyl chain are reported, and the mismatching is attributed to a developing
syn-pentane interaction in the transition state.
Introduction
sibility of such a phenomenon in alkene radical cation chemistry.
The demonstration that â-(phosphatoxy)alkyl radicals6 undergo
The existence of stereochemical memory effects arising from
nucleophilic attack on carbocations within the confines of
contact ion pairs in classical solvolysis reactions is widely
recognized and appreciated.1,2 Similarly, the existence of
memory effects in free radical reactions has been amply
demonstrated in recent years.3,4 The classical generation of
alkene radical cations by one-electron oxidation of the corre-
sponding alkenes,5 however, has hitherto precluded the pos-
diastereoselective rearrangement in nonpolar solvents by a
heterolytic fragmentation to an alkene radical cation/anion
contact ion pair, followed by the open-shell equivalent of internal
return,7 raised the possibility that stereochemical memory effects
in the nucleophilic substitution of these fascinating reactive
intermediates might finally be realized. We describe an extensive
series of investigations designed to test this hypothesis and
derive a stereochemical model accounting for the effect of single
and multiple substituents in the cyclization of γ-amino alkene
radical cations within the confines of a contact radical ion pair.8
The model has predictive value and should find application in
synthetic schemes employing these intermediates.
(1) (a) Winstein, S.; Clippinger, E.; Fainberg, A. H.; Heck, R.; Robinson, G.
C. J. Am. Chem. Soc. 1956, 78, 328. (b) Raber, D. J.; Harris, J. M.; Schleyer,
P. v. R. In Ions and Ion Pairs in Organic Reactions; Szwarc, M., Ed.;
Wiley: New York, 1974; Vol. 2, p 247. (c) Richard, J. P.; Amyes, T. L.;
Toteva, M. M.; Tsuji, Y. AdV. Phys. Org. Chem. 2004, 39, 1.
(2) For a recent example from our own laboratory, see: Crich, D.; Chan-
drasekera, N. S. Angew. Chem., Int. Ed. 2004, 43, 5386.
(3) For recent examples, see: (a) Buckmelter, A. J.; Powers, J. P.; Rychnovsky,
S. D. J. Am. Chem. Soc. 1998, 120, 5589. (b) Buckmelter, A. J.; Kim, A.
I.; Rychnovsky, S. D. J. Am. Chem. Soc. 2000, 122, 9386. (c) Dalgard, J.
E.; Rychnovsky, S. D. Org. Lett. 2004, 6, 2713. (d) Rychnovsky, S. D.;
Hata, T.; Kim, A. I.; Buckmelter, A. J. Org. Lett. 2001, 3, 807. (e) Sauer,
S.; Schumacher, A.; Barbosa, F. G. B. Tetrahedron Lett. 1998, 39, 685. (f)
Giese, B.; Wettstein, P.; Stahelin, C.; Barbosa, F.; Neuberger, M.; Zehnder,
M.; Wessig, P. Angew. Chem., Int. Ed. 1999, 38, 2586.
(6) (a) Beckwith, A. L. J.; Crich, D.; Duggan, P. J.; Yao, Q. Chem. ReV. 1997,
97, 3273. (b) Crich, D. In Radicals in Organic Synthesis; Renaud, P., Sibi,
M., Eds.; Wiley-VCH: Weinheim, 2001; Vol. 2, p 188.
(7) (a) Newcomb, M.; Horner, J. H.; Whitted, P. O.; Crich, D.; Huang, X.;
Yao, Q.; Zipse, H. J. Am. Chem. Soc. 1999, 121, 10685. (b) Whitted, P.
O.; Horner, J. H.; Newcomb, M.; Huang, X.; Crich, D. Org. Lett. 1999, 1,
153. (c) Bales, B. C.; Horner, J. H.; Huang, X.; Newcomb, M.; Crich, D.;
Greenberg, M. M. J. Am. Chem. Soc. 2001, 123, 3623. (d) Newcomb, M.;
Miranda, N.; Huang, X.; Crich, D. J. Am. Chem. Soc. 2000, 122, 6128. (e)
Bagnol, L.; Horner, J. H.; Newcomb, M. Org. Lett. 2003, 5, 5055. (f)
Horner, J. H.; Bagnol, L.; Newcomb, M. J. Am. Chem. Soc. 2004, 126,
14979. (g) Wang, Y.; Grimme, S.; Zipse, H. J. Phys. Chem. A 2004, 108,
2324.
(4) For a recent review on stereochemical memory effects in synthesis, see:
Zhao, H.; Hsu, D. C.; Carlier, P. R. Synthesis 2005, 1.
(5) For reviews, see: (a) Schmittel, M.; Burghart, A. Angew. Chem., Int. Ed.
Engl. 1997, 36, 2550. (b) Hintz, S.; Heidbreder, A.; Mattay, J. In Topics
in Current Chemistry; Mattay, J., Ed.; Springer: Berlin, 1996; Vol. 177, p
77. (c) Bauld, N. L.; Bellville, D. J.; Harirchian, B.; Lorenz, K. T.; Pabon,
R. A.; Reynolds, D. W.; Wirth, D. D.; Chiou, H.-S.; Marsh, B. K. Acc.
Chem. Res. 1987, 20, 371.
(8) For a preliminary communication, see: Crich, D.; Ranganathan, K. J. Am.
Chem. Soc. 2002, 124, 12422.
9
9924
J. AM. CHEM. SOC. 2005, 127, 9924-9929
10.1021/ja051657t CCC: $30.25 © 2005 American Chemical Society

Stereochemical Memory Effects in Contact Ion Pairs
A R T I C L E S
Scheme 1. Transition State Model A: All Pseudo-equatorial
Results and Discussion
Early stereochemical and isotopic labeling experiments from
our laboratory on the â-(phosphatoxy)alkyl radical rearrange-
ment,9 when interpreted in terms of the ion pair mechanism for
this rearrangement,7 coupled with well-known nucleophilic
attack10 on alkene radical cations, lead directly to the conclusion
that nucleophilic attack on fragmentation-derived alkene radical
cations has the potential to exhibit stereochemical memory
effects. In order for memory effects to be observed, it is
necessary that the rate of nucleophilic attack on the contact ion
pair be more rapid than the rate of memory-destroying solvation
of the contact ion pair. The rate of attack of butylamine on the
p-methoxystyrene radical cation has been determined to be 2.5
× 109 M-1 s-1 in acetonitrile at 25 °C,10e,f whereas Farid and
co-workers find that radical cation/radical anion pairs solvate
with estimated rate constants of 108-109 s-1 in dichloroethane
at room temperature.7f,11 These rate constants, and other
considerations of a more practical nature, prompted us to focus
on cyclization reactions with a view to increasing the effective
molarity of the nucleophile and the rate of the trapping reaction.
Radical Ionic Cyclization and Assignment of Product
Stereochemistry. A series of 12 substrates were prepared as
described in full in the Supporting Information.12 The radical
ionic cascade reactions were conducted by heating a mixture
of substrate (0.02 M), tributyltin hydride (1.5 molar equivalents),
and azobisisobutyronitrile (AIBN, 0.3 molar equivalents) to
reflux in benzene under nitrogen for 40 h with the periodic
addition of further AIBN. The use of syringe pump techniques
and lower reaction temperatures was investigated, but neither
was found to support the radical ionic chain process. The basic
products were isolated from the reaction mixture by extraction
into dilute hydrochloric acid and re-extraction into ether
following basification, with product ratios being determined by
integration of the 1H NMR spectra of the extracts before
chromatographic separation.13 The results obtained are reported
in Table 1. As expected from previous studies on simpler
systems, no evidence was found for the alternative formation
of piperidines by 6-endo ring closure.14,15
Stereochemical Models for Cyclization. Of the 12 cycliza-
tions examined, 9 take place with moderate to high degrees of
diastereoselectivity favoring the product with formal inversion
of configuration (Table 1, entries 1, 3-7, 9, 11, and 12). These
reactions are best understood in terms of fragmentation of the
initial radical to a contact alkene radical cation/phosphate anion
pair with the phosphate shielding the face of the molecule from
which it has just departed. This ion pair undergoes rapid
intramolecular nucleophilic attack by the amine in an ipso-5-
exo manner on the face of the alkene radical opposite to the
one shielded by the phosphate, before any scrambling of the
ion pair. Among these nine examples, five may be accounted
for by a chair-like transition state with all substituents pseudo-
equatorial (Scheme 1), even if the selectivity is reduced, as in
the case of syn-3 (Table 1, entry 5).
The four remaining examples of inversion (Table 1, entries
4, 6, 11, and 12) all proceed with predominant inversion of
configuration but through transition states that necessarily
include pseudo-axial substituents if a chair-like conformation
is invoked. This leads to the hypothesis of competing twist-
boat-like transition states in these cases (Scheme 2).17
The other extreme, retention of configuration, is represented
by entries 2 and 8, and to a lesser extent by entry 10, in Table
1. These systems all share the anti configuration between the
departing phosphate and the methyl group at the vicinal
stereogenic center and cyclize with predominant retention of
configuration. We view these reactions in terms of the transition
state for cyclization of the initial contact ion pair, with attack
(9) (a) Crich, D.; Yao, Q. Tetrahedron Lett. 1993, 34, 5677. (b) Crich, D.;
Yao, Q. J. Am. Chem. Soc. 1994, 116, 2631. (c) Crich, D.; Yao, Q.; Filzen,
G. F. J. Am. Chem. Soc. 1995, 117, 11455. (d) Crich, D.; Sartillo-Piscil,
F.; Quintero-Cortes, L.; Wink, D. J. J. Org. Chem. 2002, 67, 3360. (e)
Crich, D.; Huang, W. J. Am. Chem. Soc. 2001, 123, 9239.
(10) (a) Mihelcic, J.; Moeller, K. D. J. Am. Chem. Soc. 2004, 126, 9106. (b)
Duan, S.; Moeller, K. D. Org. Lett. 2001, 3, 2685. (c) Giese, B.; Beyrich-
Graf, X.; Burger, J.; Kesselheim, C.; Senn, M.; Schafer, T. Angew. Chem.,
Int. Ed. Engl. 1993, 32, 1742. (d) Crich, D.; Huang, X.; Newcomb, M. J.
Org. Chem. 2000, 65, 523. (e) Johnston, L. J.; Schepp, N. P. J. Am. Chem.
Soc. 1993, 115, 6564. (f) Johnston, L. J.; Schepp, N. P. AdV. Electron-
Transfer Chem. 1996, 5, 41.
(11) Arnold, B. R.; Noukakis, D.; Farid, S.; Goodman, J. L.; Gould, I. R. J.
Am. Chem. Soc. 1995, 117, 4399.
(12) In this paper, nitrophosphates and their precursors are numbered as indicated.
The N-benzyl group is considered part of the extended zigzag conformation
backbone, and all methyl groups are considered substituents. For purposes
of consistency throughout the paper, this system is applied even to
compounds 3, 5, and 6, which would be more correctly shown with the
methyl group as part of the backbone and the amine as a substituent.
(15) The isolated yields for these cyclizations, as reported in Table 1, are typical
for reactions of this type.14 As the focus of this study was the stereo-
selectivity of the ring closure, no attempt was made to isolate and
characterize products arising from other reaction pathways. In previous
work, we have identified a number of competing decomposition pathways
for the alkene radical cation, including overall reduction to the correspond-
ing alkene7 and CC bond cleavage.14
(16) In the trisubstituted pyrrolidines 11-13, the first stereochemical descriptor
refers to the relationship between the isopropyl group on the 2-position
and the next substituent around the ring. The second stereochemical
descriptor applies to the relationship between the two methyl groups; thus,
for example, cis,trans-11 refers to 2,3-cis-3,4-trans-1-benzyl-2-isopropyl-
3,4-dimethylpyrrolidine.
(17) It should be noted, however, that 1,3-diaxial interactions are less important
in chair-like transition states with their longer partial bonds than in actual
chair form rings, which leaves open the possibility that these cyclizations
may occur through the chair-like transition states.
(18) In this scheme, anti,anti-6 and trans,cis-13, and the associated transition
states, are represented as their enantiomers for ease of comparison.
(13) No significant difference was observed on the product ratios, whether
determined on the crude reaction mixtures or following the extractive
workup.
(14) Crich, D.; Ranganathan, K.; Neelamkavil, S.; Huang, X. J. Am. Chem. Soc.
2003, 125, 7942.
9
J. AM. CHEM. SOC. VOL. 127, NO. 27, 2005 9925

A R T I C L E S
Crich and Ranganathan
Table 1. Radical-Ionic Cascade Reactions16
9
9926 J. AM. CHEM. SOC. VOL. 127, NO. 27, 2005

Stereochemical Memory Effects in Contact Ion Pairs
A R T I C L E S
Table 1. (Continued)
a Inv and ret refer to overall inversion and retention of configuration, respectively, at the site of substitution.
Scheme 2. Transition State Model B: Twist-Boats18
of the phosphate between the faces of the alkene radical cation
and conformational inversion of the skeleton lead to a second
chair-like transition state for ring closure, resulting in overall
retention of configuration (Scheme 3).
Comparison of entries 5 and 12 in Table 1 (syn-3 and
anti,anti-6), which differ only in the presence of the C-5 methyl
group in anti,anti-6, is informative as both proceed with similar,
reduced diastereoselectivity. Both also afford a minor hexahy-
dropyrroloisoquinoline from the minor cyclization mode as
discussed below. The implication is that both reactions proceed
through similar transition states, with the role of the C-5 methyl
group subordinate to that of the C-6 methyl group in the
cyclization of anti,syn-6. In the above discussion, the cyclization
of syn-3 is assigned to model A, whereas that of anti,anti-6 is
placed under model B. Either, as discussed above, the effect of
the axial methyl group in the chair-like transition state arising
from anti,anti-6 is minimal, or both cyclizations proceed through
twist-boat-like transition states (Scheme 2) resulting from a
minimization of steric interactions between the C-5 methyl and
N-benzyl groups. Comparison between entries 6 and 11 of Table
1 (anti-3 and syn,anti-6), again differing only in the presence
of a methyl group at C-5, is also instructive as again the C-5
methyl group has only a minimal effect on the outcome of the
cyclization. This lack of influence of the substituent at C-5 is
again seen in a comparison of entries 3 and 4 in Table 1 (syn-
and anti-2), and overall, we conclude that models A and B can
reasonably be refined into a single model, most likely that of a
chair-like transition state in which pseudo-axial substituents at
C-5 are only minimally destabilizing. A chair-like transition state
is ultimately favored over a common twist-boat-like transition
state because of the problems arising in the initial transition
state for the cyclization of anti-1, anti,syn-4, and anti,syn-5
(Scheme 3, model C), which ultimately results in predominant
cyclization with retention of configuration. As indicated, the
problem here is one of 1,3A strain in the alkene radical cation,
which would be reduced in any twist-boat-like transition state.
of the amine opposite to the face shielded by the phosphate,
suffering from 1,3A strain in the alkene radical cation moiety.
This retards the initial cyclization to the extent that the migration
9
J. AM. CHEM. SOC. VOL. 127, NO. 27, 2005 9927

A R T I C L E S
Crich and Ranganathan
Scheme 3. Transition State Model C: Retention of Configuration
12 cyclizations (Table 1, entries 5, 6, 11, and 12). In the case
of anti-3 (Table 1, entry 6), the pyrroloisoquinoline 10 arises
from an oxidative cyclization of the major ring-closed radical,21
whereas with syn-3 the same pyrroloisoquinoline 10 is formed
from the minor cyclization mode. In other words, while two
stereoisomeric pyrroloisoquinolines are possible from syn- and
anti-3, only one (10) is formed. The same scenario exists for
the two isomers of susbstrate 6 when pyrroloisoquinoline 14 is
formed from the major cyclization pathway in the cyclization
of the syn,anti-isomer and from the minor pathway in the
cyclization of the anti,anti-isomer. All four cases belong to
radicals arising from model B (Scheme 2), twice as the major
mode and twice as the minor mode. Obviously, in this system,
with the cis-6-methyl group and 2-isopropyl radical, the
pyrrolidine ring adopts a conformation in which the methyl
group buttresses the N-benzyl moiety, forcing it closer to the
isopropyl radical, thereby promoting the second cyclization.
Presumably, in the corresponding trans series, the pyrrolidine
ring adopts a conformation that allows for less compression
between the methyl and benzyl groups.22 The question arises
as to why such a pyrroloisoquinoline was not formed in
measurable amount in the cyclization of either syn,syn- or
anti,syn-5, both of which have the requisite methyl group on
the 5-position of the ring-closed radical. The answer must lie
with the methyl group on the 3-position of the cyclized system,
which may predispose the pyrrolidine ring to a conformation
that minimizes the buttressing between the 5-methyl and
N-benzyl groups. Alternatively, it is conceivable that the
3-methyl group impinges on the conformation of the vicinal
isopropyl radical and prevents it from achieving the correct
orbital alignment with the aromatic ring for cyclization.
Conclusion and Predictions
Alkene radical cations generated by heterolytic fragmentation
of â-(phosphatoxy)alkyl radicals can undergo highly diastereo-
selective cyclization reactions with suitably positioned amine
groups. These cyclizations take place at the level of the initial
contact ion pair with attack by the amine on the opposite face
of the alkene radical cation to the one shielded by the just-
departed phosphate group. Onto this fundamental model for
overall inversion of stereochemistry are layered the conforma-
tional and torsional effects of the chain linking the alkene radical
cation to the nucleophile. The more selective cyclizations are
best described by chair-like transition states, although the
possibility of twist-boat-like analogues cannot be excluded. A
1,2-anti relationship between the departing phosphate and a
vicinal substituent is highly detrimental to this model, owing
to 1,3A strain in the alkene radical cation, and results in a more
Beckwith-Houk-like equilibrated model. Substituents having
a 1,3-relationship with the departing phosphate group have little
influence on diastereoselectivity. The influence of substituents
in 1,4-relationship with the departing phosphate and geminal
Model C (Scheme 3) is akin to the Beckwith-Houk model19
for the cyclization of substituted 5-hexenyl radicals with the
main chain undergoing conformational equilibration before
cyclization through the more stable transition state with the
maximum number of substituents pseudo-equatorial. 5-exo-Alkyl
radical cyclizations typically20 have rate constants of 105-106
s-1, but the nucleophilic cyclizations of amines onto alkene
radical cations examined here are expected to be significantly
faster. Intermolecular additions of butylamine to substituted
styrene radical cations10e,f have rate constants of 109 M-1 s-1
,
and it is reasonable to expect the present cyclizations to proceed
with rate constants of at least 109 s-1. It is therefore reasonable
that cyclizations following models A and B (Schemes 1 and 2)
take place before full scrambling of the geometry of the initial
contact ion pair and, thus, before the system can sample the
full range of conformations available to it. The fully equilibrated
Beckwith-Houk-type model comes into play only when cy-
clization onto the initial contact ion pair is disfavored.
Formation of Pyrrolo[1,2-b]isoquinolines 10 and 14.
Hexahydropyrrolo[1,2-b]isoquinolines are formed in 4 of the
(21) Current thinking suggests that the cyclohexadienyl radical resulting from
the second cyclization is oxidized by the initiator AIBN: (a) Beckwith, A.
L. J.; Bowry, V. W.; Bowman, W. R.; Mann, E.; Parr, J.; Storey, J. M. D.
Angew. Chem., Int. Ed. 2004, 43, 95. (b) Engel, P. S.; Wu, W.-X. J. Am.
Chem. Soc. 1989, 111, 1830. (c) Curran, D. P.; Yu, H.; Liu, H. Tetrahedron
1994, 50, 7343. (d) Curran, D. P.; Liu, H. J. Chem. Soc., Perkin Trans. 1
1994, 1377.
(22) Presumably, these oxidative cyclizations take place at the level of the
pyrrolidinium ion arising from the first cyclization, suggesting that the
configuration at nitrogen and possibly even the bulk of the associated
phosphate anion must be taken into account in any model.
(19) (a) Beckwith, A. L. J. Tetrahedron 1981, 37, 3073. (b) Beckwith, A. L. J.;
Schiesser, C. H. Tetrahedron 1985, 41, 3925. (c) Spellmeyer, D. C.; Houk,
K. N. J. Org. Chem. 1987, 52, 959. (d) Curran, D. P.; Porter, N. A.; Giese,
B. Stereochemistry of Radical Reactions; VCH: Weinheim, 1996.
(20) (a) Newcomb, M. Tetrahedron 1993, 49, 1151. (b) Newcomb, M. In
Radicals in Organic Synthesis; Renaud, P., Sibi, M. P., Eds.; Wiley-VCH:
Weinheim, 2001; Vol. 1, p 317. (c) Ingold, K. U.; Griller, D. Acc. Chem.
Res. 1980, 13, 317.
9
9928 J. AM. CHEM. SOC. VOL. 127, NO. 27, 2005

Stereochemical Memory Effects in Contact Ion Pairs
A R T I C L E S
to the amine depends on configuration. When the configuration
is 1,4-anti the influence is minimized, but in the 1,4-syn-system
diastereoselectivity is considerably reduced. In systems with a
substituent geminal to the amine, a second oxidative radical
cyclization follows the initial nucleophilic attack on the alkene
radical cation, leading to the formation of hexahydropyrrolo-
[1,2-b]isoquinolines. Beckwith-Houk-type cyclic transition
states for the cyclizations of free alkene radical cations generated
by oxidation of alkenes have been discussed previously by
Moeller;10a,b however, the memory effects that are the main
focus of this paper have not been previously observed owing
to the mode of generation of alkene radical cations from alkenes
that has been most commonly employed in other laboratories.5
The reactions described in this paper are conducted in refluxing
benzene in order to ensure that the nitroalkane, a relatively poor
radical precursor, participates in a sustainable radical chain
reaction. Nevertheless, for the most part, the diastereoselectivi-
ties observed are significant. Accordingly, we anticipate that
when more efficient radical precursors, compatible with the
presence of the leaving group and of the nucleophilic amine,
are located, reactions run at lower temperatures will give even
higher selectivities.
Acknowledgment. We thank the National Science Foundation
(CHE 9986200) for support of this work.
Supporting Information Available: Complete description of
substrate preparation, full experimental details, and copies of
spectra of all new compounds. This material is available free
of charge via the Internet at http://pubs.acs.org.
JA051657T
9
J. AM. CHEM. SOC. VOL. 127, NO. 27, 2005 9929