Use of a conformational radical clock for evaluating alkyllithium-mediated cyclization reactions

Article abstract of DOI:10.1021/ol006866r

matrix presented The reductive lithiation of nitrile 9 led to the cyclic product 11 as a single diastereomer in 42% ee. The intermediate radical is a conformational radical clock. The radical lifetime can be determined from the optical purity of the produ

Full text of DOI:10.1021/ol006866r

ORGANIC
LETTERS
2001
Vol. 3, No. 6
807-810
Use of a Conformational Radical Clock
for Evaluating Alkyllithium-Mediated
Cyclization Reactions
Scott D. Rychnovsky,* Takeshi Hata, Angie I. Kim, and Alexandre J. Buckmelter
Department of Chemistry, 516 Rowland Hall, UniVersity of California,
IrVine, California 92697-2025
srychnoV@uci.edu
Received November 13, 2000 (Revised Manuscript Received February 12, 2001)
ABSTRACT
The reductive lithiation of nitrile 9 led to the cyclic product 11 as a single diastereomer in 42% ee. The intermediate radical is a conformational
radical clock. The radical lifetime can be determined from the optical purity of the product 11. We show that the lifetime of the intermediate
radical is too brief to allow a radical cyclization, and thus the cyclization proceeds through an alkyllithium intermediate.
Radical clock reactions are powerful tools for evaluating
reaction mechanisms.1 We recently reported a radical clock
reaction based on the ring inversion of a 2-tetrahydropyranyl
radical.2 Described herein is the study of a cyclization
reaction using a conformational radical clock.
Alkyllithium cyclizations with alkenes provide an interest-
ing approach to five-membered ring synthesis. Work by
Bailey and others has established that reactive alkyllithium
reagents are required, and unless a leaving group is present,3
the alkene must be terminal for efficient cyclization.4 The
mechanism is thought to be a metal-ene type cyclization,4
and a transition state for the cyclization has been identified
using ab initio methods.5 A related cyclization of alkenyl
halides catalyzed by an alkyllithium reagent has been
studied.6 In this case the cyclization was determined to
proceed through a radical intermediate rather than the
expected alkyllithium intermediate.7 Unsaturated alkylmag-
nesium reagents have also been reported to cyclize via the
corresponding radical.8 Evaluating the mechanism of 5-hex-
enyl benzene sulfide reductive cyclizations3 is particularly
problematic, because the radical is an obligatory intermediate
in the generation of the alkyllithium reagent and both radical
and alkyllithium cyclization would produce the observed
product. We report a reductive cyclization in which the
alkyllithium rather than the radical is the active intermediate.
This study is predicated on determining the lifetime of an
intermediate radical using a conformational radical clock
reaction. An optically pure 2-tetrahydropyranyl radical can
be used as a radical clock as shown in Scheme 1.2 The
optically pure radical 2, generated by reduction of cyano-
hydrin 1, racemizes by ring inversion, and we have shown
that the rate of racemization is about 5.7 × 108 s-1 at 22 °C
(1) (a) Newcomb, M. Tetrahedron 1993, 49, 1151-1176. (b) Griller,
D.; Ingold, K. U. Acc. Chem. Res. 1980, 13, 317-323.
(2) Buckmelter, A. J.; Kim, A. I.; Rychnovsky, S. D. J. Am. Chem. Soc.
2000, 122, 9386-9390.
(3) (a) Broka, C. A.; Lee, W. J.; Shen, T. J. Org. Chem. 1988, 53, 1336-
1338. (b) Broka, C. A.; Shen, T. J. Am. Chem. Soc. 1989, 111, 2981-
2984.
(4) (a) Bailey, W. F.; Patricia, J. J.; DelGobbo, V. C.; Jarret, R. M.;
Okarma, P. J. J. Org. Chem. 1985, 50, 1999-2000. (b) Bailey, W. F.; Nurmi,
T. T.; Patricia, J. J.; Wang, W. J. Am. Chem. Soc. 1987, 109, 2442-8. (c)
Chamberlin, A. R.; Bloom, S. H.; Cervini, L. A.; Fotsch, C. H. J. Am.
Chem. Soc. 1988, 110, 4788-4796.
(5) Bailey, W. F.; Khanolkar, A. D.; Gavaskar, K.; Ovaska, T. V.; Rossi,
K.; Thiel, Y.; Wiberg, K. B. J. Am. Chem. Soc. 1991, 113, 5720-5727.
(6) Bailey, W. F.; Carson, M. W. J. Org. Chem. 1998, 63, 9960-9967.
(7) Bailey, W. F.; Carson, M. W. Tetrahedron Lett. 1999, 40, 5433-
5437.
(8) (a) Inoue, A.; Shinokubo, H.; Oshima, K. Org. Lett. 2000, 2, 651-
653. (b) Bodineau, N.; Mattalia, J.-M.; Thimokhin, V.; Handoo, K.; Negrel,
J.-C.; Chanon, M. Org. Lett. 2000, 2, 2303-2306.
10.1021/ol006866r CCC: $20.00 © 2001 American Chemical Society
Published on Web 02/20/2001

Scheme 1. Tetrahydropyranyl Radicals as Radical Clocks
Scheme 2. Synthesis of Nitrile 9
commercially available tetrahydropyran-2-methanol, fol-
lowed by esterification, gave the expected ester. Deproto-
nation and alkylation with 5-iodo-1-pentene gave the unsat-
urated ester 6 in good overall yield. After several unsuccessful
attempts to resolve the acid of 6, we investigated the
resolution of diastereomeric amides. Ester 6 was converted
to amides 7 and 7′ by treatment with (R)-R-methylbenzyl-
amine and isopropylmagnesium chloride.13 The two diaster-
eomers were easily separated by silica gel chromatography.14
Attempts to hydrolyze the secondary amides 7 and 7′ were
unsuccessful, but the secondary amide was converted to
primary amide 8 by dissolving metal reduction with sodium
in ammonia. Dehydration of amide 8 with thionyl chloride
gave the target nitrile 9 as a single enantiomer.15
or 3.9 × 106 s-1 at -78 °C.2 Note that the barrier to radical
inversion is estimated to be less than 0.5 kcal/mol,9 but the
barrier to ring inVersion of 2 is closer to 7 kcal/mol.
Reduction of the radical 2 and its enantiomer 2′ lead to the
configurationally stable alkyllithium reagents 3 and 3′.10 Most
carbonyl electrophiles and protons react with R-alkoxy
lithium reagents such as 3 with retention of configuration.11
Reductive lithiation of nitrile 1 with lithium di-tert-butyl-
biphenylide (LiDBB) at -78 °C leads to a 69.5:29.5 mixture
of 4 and 4′ (39% ee) on trapping with methanol. The optical
purity of the product is directly related to the lifetime of the
radical. The half-life of the radical under the reaction
conditions is given by eq 1.12 For the reductive lithiation of
optically pure cyanohydrin 1, the lifetime of the intermediate
radical 2 under the reaction conditions was found to be 2.8
× 10-7 s using eq 1. Conformational radical clocks are useful
tools for measuring the lifetimes of radical intermediates.
With nitrile 9 in hand, its reductive lithiation and cycliza-
tion was investigated, and the results are illustrated in
Schemes 3 and 4. Treatment with LiDBB in THF for 10
Scheme 3. Lithiation and Cyclization of Nitrile 9
t1/2)ln2/kR×(1-ee)/ee
(1)
The preparation of the substrate for the reductive cycliza-
tion, nitrile 9, is outlined in Scheme 2. Oxidation of
(9) Griller, D.; Ingold, K. U.; Krusic, P. J.; Fischer, H. J. Am. Chem.
Soc. 1978, 100, 6750-6752. See ref 17.
(10) Rychnovsky, S. D.; Buckmelter, A. J.; Dahanukar, V. H.; Skalitzky,
D. J. J. Org. Chem. 1999, 64, 6849-6860.
(11) Sawyer, J. S.; Kucerovy, A.; Macdonald, T. L.; McGarvey, G. J. J.
Am. Chem. Soc. 1988, 110, 842-853.
min at -78 °C, followed by addition of CO2 gas, aqueous
workup, and diazomethane treatment, gave the spirocyclic
ester 11 in 65% yield and 42% ee.15 The relative configu-
ration of 11 was determined by NMR analysis and subse-
quently confirmed by an indirect correlation.16 When the
(12) Here, ee is the fractional enantiomeric excess and kR is the rate of
racemization of the radical at the reaction temperature. Alternatively, the
half-life of the radical can be calculated directly from the concentrations
of the two enantiomers of the product: t1/2 ) 2 ln 2/kR × [4′]/([4] - [4′]).
(13) Williams, J. M.; Jobson, R. B.; Yasuda, N.; Marchesini, G.; Dolling,
U. H.; Grabowski, E. J. J. Tetrahedron Lett. 1995, 36, 5461-5464.
(14) The Rf values for 7 and 7′ in 20% EtOAc/hexanes are 0.34 and
0.41, respectively. The absolute configurations at the quaternary centers in
7 and 7′ were not assigned.
(15) The optical purity was directly evaluated by GC analysis on a
CHIRALDEX G-TA or CHIRALDEX B-PH column.
808
Org. Lett., Vol. 3, No. 6, 2001

reduction of nitrile 9 was run for 30 min at -78 °C prior to
trapping, compound 11 in was isolated in similar yields and
optical purities. We conclude that the cyclization was
essentially complete after 10 min at -78 °C.
Reduction of 9 with lithium in ammonia gave very
different results, Scheme 4. The Li/NH3 reduction of racemic
Scheme 5. In both pathways, nitrile 9 is reduced to radical
15; radical 15 is the key branch point in each pathway. In
the alkyllithium pathway, the racemization of 15 will
compete with reduction to the alkyllithium reagent 16.
Alkyllithium reagents 16 and 16′ are configurationally stable
at -78 °C in THF,19 and the cyclizations of R-alkoxy
alkyllithium reagents have been found to take place with
retention of configuration.20 Thus, in the alkyllithium cy-
clization the ratio of 16:16′ will determine the final ratio of
11:11′.
Scheme 4. Reduction and Reductive Cyclization of Nitrile 9
An alternative view of the cyclization is that it proceeds
via a radical cyclization. On the face of it, the fact that the
cyclization of 9 is complete in 10 min at -78 °C might
suggest a radical cyclization, many of which are known to
be very rapid, over the alkyllithium cyclization. The radical
cyclization pathway is outlined in the lower half of Scheme
5. The intermediate radical 15 is again the key branch point
with cyclization and racemization reactions competing. The
cyclized radicals 19 and 19′ that are produced would then
be reduced to give alkyllithium reagents 17 and 17′. Reaction
with CO2 and esterification with diazomethane generates a
mixture of 11 and 11′. In this case the ratio 11:11′ would be
determined by competition between racemization of 15 and
cyclization of 15.
A radical clock reaction would help distinguish between
the alkyllithium and the radical cyclization pathways. Radical
15 is structurally similar to radical 2, and it should have a
comparable racemization rate at -78 °C. Radical 15 is the
only point in either mechanism where racemization would
be likely, and so the enantiomeric excess of product 11 is a
direct measure of the lifetime of the radical under the reaction
conditions. Using the optical purity of 11, 42% ee, and the
measured rate of racemization of radical 2 in eq 1 gives an
estimated lifetime for radical 15 of 2.4 × 10-7 s.21 How does
this lifetime compare with the expected rate of radical
cyclization? Extrapolating from the published Arrhenius data
for a 5-hexenyl cyclization leads to a predicted cyclization
rate of 4 × 102 at -78 °C.22 Comparing the lifetime of radical
15 and the predicted rate of cyclization suggests that the rate
of cyclization is approximately 5 orders of magnitude too
small to allow cyclization in the lifetime of the radical.
9 at -78 °C gave the uncyclized tetrahydropyran 12 in 76%
yield with none of the corresponding cyclized product 13.
Lithium ammonia reductions of nitriles proceed through
stepwise electron transfers and generally produce results
similar to those of LiDBB reductions with the same
substrates.17 Reduction of 9 with LiDBB, followed by
methanol quenching, led to 12 and 13 in a 1:10 ratio.
Oxidation of 13 with RuO4 gave the known lactone 14.18
Thus, nitrile 9 cyclizes rapidly and efficiently when reduced
with LiDBB, but it did not cyclize at all when reduced with
lithium in liquid ammonia.
One concern that might be raised with the foregoing
analysis is that the rate of cyclization was calculated from a
5-hexenyl radical but the cyclizing radical 15 is highly
substituted. Newcomb, however, has evaluated the cycliza-
tion rates of the 6,6-diphenyl-5-hexenyl radical23 and the
1-methoxy-6,6-diphenyl-5-hexenyl radical.24 The rate of
The cyclization of 9 en route to 11 can be rationalized
either as a radical cyclization or as an alkyllithium cycliza-
tion. These two mechanistic possibilities are illustrated in
(16) Ester 11 was isolated as a single diastereomer, but its relative
configuration was not determined easily. The 13C chemical shifts for both
possible diastereomers were predicted using the computational method of
Forsyth and Sebag, and the 13C data for 11 was more consistent with the
structure shown in Scheme 3. Oxidation of 13 to lactone 14 and correlation
(see ref 18) further support the assignment. Forsyth, D. A.; Sebag, A. B. J.
Am. Chem. Soc. 1997, 119, 9485-9494.
(19) Cohen, T.; Lin, M. T. J. Am. Chem. Soc. 1984, 106, 1130-1131.
(20) (a) Tomooka, K.; Komine, N.; Nakai, T. Tetrahedron Lett 1997,
38, 8939-8942. (b) Woltering, M. J.; Frohlich, R.; Hoppe, D. Angew.
Chem., Int. Ed. Engl. 1997, 36, 1764-1766.
(21) The reviewer noted that the rate of reduction of a radical with LiDBB
can be estimated, in principle, from electrochemical data. Unfortunately,
neither the self-exchange rate nor the solvent reorganization energy for DBB
reduction has been reported. For an insightful discussion, see: (a) Andrieux,
C. P.; Gallardo, I.; Save´ant, J.-M. J. Am. Chem. Soc. 1989, 111, 1620-
1634. (b) Gonzalez, J.; Hapiot, P.; Konovalov, V.; Save´ant, J.-M. J. Am.
Chem. Soc. 1998, 120, 10171-10179.
(22) Griller, D.; Ingold, K. U. Acc. Chem. Res. 1980, 13, 317-323.
(23) Ha, C.; Horner, J. H.; Newcomb, M.; Varick, T. R.; Arnold, B. R.;
Lusztyk, J. J. Org. Chem. 1993, 58, 1194-8.
(17) (a) Rychnovsky, S. D.; Powers, J. P.; Lepage, T. J. J. Am. Chem.
Soc. 1992, 114, 8375-8384. (b) Buckmelter, A. J.; Powers, J. P.;
Rychnovsky, S. D. J. Am. Chem. Soc. 1998, 120, 5589-5590.
(18) Both the cis and trans isomers of lactone 14 were reported, and
NMR data for our product clearly matched the diagnostic methyl shifts for
1
the cis isomer: trans isomer H δ 0.91 ppm, 13C δ 16.06 ppm; cis isomer
1H δ 1.00 ppm, 13C δ 13.00 ppm; lactone 14 1H δ 0.99 (d, J ) 6.6 Hz)
ppm, 13C δ 12.7 ppm. However, several of the other 13C peaks reported for
the cis isomer did not match our data for compound 14, and we cannot
explain the discrepancy. Canonne, P.; Boulanger, R.; Bernatchez, M.
Tetrahedron 1989, 45, 2525-40.
Org. Lett., Vol. 3, No. 6, 2001
809

Scheme 5. Radical and Alkyllithium Cyclization Pathways for the Formation of Esters 11 and 11′
cyclization for the R-alkoxy radical does differ from that of
Conformational radical clocks are useful for determining
the lifetime of radical intermediates. For the reductive
lithiation of nitrile 9, the optical purity of the product 12,
42% ee, implies a lifetime for radical 15 of 2.4 × 10-7 s.
This lifetime is too brief for a radical cyclization to take
place. Thus, the reduction of 9 to 11 proceeds via cyclization
of the alkyllithium 16.
the primary radical, but only by about a factor of 2 over the
extrapolated range from -70 to 80 °C. Variations of this
magnitude do not change the conclusions of the analysis.
A final piece of evidence is the absence of cyclization
products in the reduction of 9 under Li/NH3 conditions
(Scheme 3). We were not able to measure the optical purity
of 12 directly,25 but one might expect that it would have an
optical purity of about 30% on the basis of the Li/NH3
reduction of nitrile 1 under comparable conditions (Scheme
1). Similar enantiomeric excesses between acyclic product
12 and cyclic product 11 would dictate a similar lifetime of
the radical 15 in each reaction. Since one product is cyclic
and the other is not, the cyclization cannot be attributed to
the radical intermediate 15. Only in the reduction of 9 with
LiDBB, where intermediate alkyllithium reagents 16 and 16′
would have long lifetimes, is cyclization observed.
Acknowledgment. Support has been provided by the
University of California, Irvine. A.J.B. thanks Abbott
Laboratories and the Department of Education for fellowship
support. T.H. thanks the Japan Society for the Promotion of
Science for a predoctoral fellowship.
Supporting Information Available: Experimental pro-
cedures for the preparation and reduction of nitrile 9. Data
and analysis for the structural assignment of ester 11. This
material is available free of charge via the Internet at
http://pubs.acs.org.
(24) Johnson, C. C.; Horner, J. H.; Tronche, C.; Newcomb, M. J. Am.
Chem. Soc. 1995, 117, 1684-7.
(25) We were not able to determine the optical purity of 12 or any of a
number of derivatives by GC analysis on chiral columns. The optical rotation
of 12 has not been reported in the literature.
OL006866R
810
Org. Lett., Vol. 3, No. 6, 2001