Synthesis of β-substituted α-amino acids via Lewis acid promoted enantioselective radical conjugate additions

Article abstract of DOI:10.1021/jo051334f

Lewis acid promoted radical conjugate additions to β-substituted αβ-unsaturated α-nitro esters and amides were investigated. With achiral Lewis acids, there was competition between the desired radical conjugate addition and undesired alkene reduction medi

Full text of DOI:10.1021/jo051334f

Synthesis of â-Substituted r-Amino Acids via Lewis Acid Promoted
Enantioselective Radical Conjugate Additions
Liwen He, G. S. C. Srikanth, and Steven L. Castle*
Department of Chemistry and Biochemistry, Brigham Young University, Provo, Utah 84602
scastle@chem.byu.edu
Received June 28, 2005
Lewis acid promoted radical conjugate additions to â-substituted R,â-unsaturated R-nitro esters
and amides were investigated. With achiral Lewis acids, there was competition between the desired
radical conjugate addition and undesired alkene reduction mediated by Bu3SnH. Zinc Lewis acids
provided the greatest amounts of addition products with both substrate classes. Studies with Bu3-
SnD indicated that the acidic R-stereocenter of the R-nitro ester products does not racemize under
controlled workup conditions. The corresponding R-nitro amides racemized significantly during
chromatography, but this problem could be greatly minimized by subjecting the crude adducts to
subsequent transformations. Indium-mediated reduction of the nitro group followed by acylation
of the resulting amine provided good yields of â-substituted R-amino acid derivatives with mimimal
levels of racemization. Attempts to use chiral Lewis acids in a stereoselective variant of this process
revealed that Kanemasa’s DBFOX/Ph ligand (14a) was uniquely effective. Moderate to good ee’s
and low dr’s were obtained with amide substrates. Determination of the absolute configurations of
the syn and anti isomers of adduct 7b showed that the hydrogen atom abstraction step was
significantly more stereoselective than the radical conjugate addition step. A model for substrate
binding to the chiral Lewis acid is presented.
Introduction
jugate additions4 are compatible with unprotected amides.
Indeed, additions of nucleophilic radicals to terminally
unsubstituted R,â-dehydroamino acid derivatives have
been used extensively to construct R-amino acids.5 Nev-
ertheless, to the best of our knowledge, Renaud’s work
is the only report of â-substituted R-amino acid synthesis
via intermolecular radical conjugate addition to a â-sub-
stituted dehydroamino acid.6
We reasoned that R,â-unsaturated R-nitro esters and
amides would function as extremely reactive acceptors
in radical conjugate additions. In fact, we felt that the
electrophilicity of these substrates, particularly when
complexed to a Lewis acid, would be sufficient to allow
additions to â-substituted acceptors to proceed, thereby
producing â-substituted R-amino acid derivatives. We
disclose herein full details of our investigations into
radical conjugate additions of R,â-unsaturated R-nitro
esters and amides,7 including the results of attempts to
render this process stereoselective by employing chiral
Lewis acids.
â-Substituted R-amino acids are present in several
peptide natural products;1 additionally, they are of inter-
est as conformationally constrained analogues of R-amino
acids.2 Accordingly, several methods have been devised
for their preparation,3 including some which involve
conjugate additions to R,â-unsaturated amino acid
precursors.3p-t This strategy is useful for the synthesis
of individual amino acids but becomes less practical with
complex peptide substrates. The latter compounds con-
tain multiple amides with acidic protons that are incom-
patible with the basic organometallic reagents typically
employed in conjugate additions. However, radical con-
(1) (a) Suzuki, H.; Morita, H.; Shiro, M.; Kobayashi, J. Tetrahedron
2004, 60, 2489. (b) Suzuki, H.; Morita, H.; Iwasaki, S.; Kobayashi, J.
Tetrahedron 2003, 59, 5307. (c) Kobayashi, J.; Suzuki, H.; Shimbo, K.;
Takeya, K.; Morita, H. J. Org. Chem. 2001, 66, 6626. (d) Leung, T.-W.
C.; Williams, D. H.; Barna, J. C. J.; Foti, S.; Oelrichs, P. B. Tetrahedron
1986, 42, 3333.
(2) (a) Hruby, V. J. J. Med. Chem. 2003, 46, 4215. (b) Gibson, S. E.;
Guillo, N.; Tozer, M. J. Tetrahedron 1999, 55, 585.
10.1021/jo051334f CCC: $30.25 © 2005 American Chemical Society
Published on Web 09/03/2005
8140
J. Org. Chem. 2005, 70, 8140-8147

Enantioselective Radical Conjugate Additions
TABLE 1. Radical Conjugate Additions to Esters 1
Results and Discussion
To begin our studies, we prepared R,â-unsaturated
R-nitro esters 1 via Knoevenagel condensations of methyl
nitroacetate with the requisite aldehydes8 and examined
their behavior as acceptors in Lewis acid promoted9
isopropyl radical conjugate additions (Table 1). We
employed conditions developed by Sibi for analogous
reactions with R,â-unsaturated N-acyloxazolidinone ac-
ceptors.10 Reactions conducted with p-methoxyphenyl-
substituted nitro ester 1a revealed that 1,4-reduction of
the acceptor by Bu3SnH11 was competitive with the
desired radical reaction. Use of the Lewis acid MgBr2‚
OEt2 delivered reduced compound 3a as the sole product.
Although we have yet to investigate this issue thor-
oughly, our observations suggest that reduction of 1a is
at least partially a radical process, as attempts to reduce
1a in the absence of i-PrI returned unreacted starting
material.12 In contrast, Nagano has demonstrated that
Bu3SnH-mediated reductions of aryl acrylates in the
substrate
Lewis acid
i-PrI (equiv) product yielda (%)
1a
1a
1a
1a
1a
1a
1a
1a
1b
1b
1b
1b
1b
1c
1c
1d
1d
MgBr2‚OEt2
ZnCl2
ZnCl2
Zn(OTf)2
Yb(OTf)3
Yb(OTf)3
Cu(OTf)2
none
MgBr2‚OEt2
ZnCl2
La(OTf)3
Sm(OTf)3
Yb(OTf)3
ZnCl2
MgBr2‚OEt2
MgBr2‚OEt2
ZnCl2
25
5
12
25
5
3a
60
60
85
85
60
70
80
81
3a
2a
2a
3a
12
12
20
5
2a
2a
2a
nrb
2b
2b
2b
nrb
2c
5
85
58
65
5
5
5
15
15
5
85
(3) (a) Yu, S.; Pan, X.; Lin, X.; Ma, D. Angew. Chem., Int. Ed. 2005,
44, 135. (b) Tsunoda, T.; Tatsuki, S.; Shiraishi, Y.; Akasaka, M.; Itoˆ,
S. Tetrahedron Lett. 1993, 34, 3297. (c) Roff, G. J.; Lloyd, R. C.; Turner,
N. J. J. Am. Chem. Soc. 2004, 126, 4098. (d) Burk, M. J.; Bedingfield,
K. M.; Kiesman, W. F.; Allen, J. G. Tetrahedron Lett. 1999, 40, 3093.
(e) O’Donnell, M. J.; Cooper, J. T.; Mader, M. M. J. Am. Chem. Soc.
2003, 125, 2370. (f) Wang, W.; Wang, J.; Li, H. Tetrahedron Lett. 2004,
45, 7243. (g) Marigo, M.; Kjærsgaard, A.; Juhl, K.; Gathergood, N.;
Jørgensen, K. A. Chem. Eur. J. 2003, 9, 2359. (h) Notz, W.; Tanaka,
F.; Watanabe, S.; Chowdari, N. S.; Turner, J. M.; Thayumanavan, R.;
Barbas, C. F., III. J. Org. Chem. 2003, 68, 9624. (i) Co´rdova, A.; Notz,
W.; Zhong, G.; Betancort, J. M.; Barbas, C. F., III. J. Am. Chem. Soc.
2002, 124, 1842. (j) Ferraris, D.; Young, B.; Cox, C.; Dudding, T.; Drury,
W. J., III; Ryzhkov, L.; Taggi, A. E.; Lectka, T. J. Am. Chem. Soc. 2002,
124, 67. (k) Bull, S. D.; Davies, S. G.; Garner, A. C.; Mujtaba, N. Synlett
2001, 781. (l) Gu, X.; Ndungu, J. M.; Qiu, W.; Ying, J.; Carducci, M.
D.; Wooden, H.; Hruby, V. J. Tetrahedron 2004, 60, 8233. (m)
Soloshonok, V. A.; Tang, X.; Hruby, V. J.; Van Meervelt, L. Org. Lett.
2001, 3, 341. (n) Tamura, O.; Yoshida, S.; Sugita, H.; Mita, N.; Uyama,
Y.; Morita, N.; Ishiguro, M.; Kawasaki, T.; Ishibashi, H.; Sakamoto,
M. Synlett 2000, 1553. (o) Medina, E.; Moyano, A.; Perica`s, M. A.; Riera,
A. Helv. Chim. Acta 2000, 83, 972. (p) Hara, S.; Makino, K.; Hamada,
Y. Tetrahedron 2004, 60, 8031. (q) Acevedo, C. M.; Kogut, E. F.; Lipton,
M. A. Tetrahedron 2001, 57, 6353. (r) Han., G.; Lewis, A.; Hruby, V.
J. Tetrahedron Lett. 2001, 42, 4601. (s) Liang, B.; Carroll, P. J.; Joullie´,
M. M. Org. Lett. 2000, 2, 4157. (t) Burtin, G.; Corringer, P.-J.; Young,
D. W. J. Chem. Soc., Perkin Trans. 1 2000, 3451.
2c,3c
nrb
nrb
7,49
5
a For reactions of 1a, yields were calculated from 1H NMR due
to persistent tin byproducts. All other yields are for isolated
materials. b Neither product observed.
presence of MgBr2‚OEt2 proceed by an ionic mechanism.13
Accordingly, the reduction of 1a and related substrates
will be the subject of future studies. Fortunately, switch-
ing to Zn, Yb, or Cu Lewis acids and increasing the
amount of i-PrI to 12 equiv resulted in exclusive forma-
tion of conjugate addition product 2a. Radical conjugate
addition to 1a also occurred in the absence of Lewis acid,
but larger amounts of i-PrI (20 equiv) were required.
Phenyl-substituted nitro ester 1b afforded adduct 2b
with Zn, La, and Sm Lewis acids; curiously, this sub-
strate gave neither 2b nor 3b when MgBr2‚OEt2 or Yb-
(OTf)3 were used. ZnCl2-promoted additions to p-fluo-
rophenyl-substituted nitro ester 1c were also successful.
Interestingly, the MgBr2‚OEt2-mediated reaction of 1c
produced a minor amount of adduct 2c along with the
expected reduced product 3c. To date, our attempts to
perform radical conjugate additions to isopropyl-substi-
tuted nitro ester 1d have been unsuccessful. The R,â-
unsaturated esters were employed as ca. 2:1 mixtures of
olefin isomers, and adducts 2 were each isolated in 1:1
dr. Although isomerically pure Z-1a-c could be obtained
via recrystallization,14 use of this material did not
improve the dr.
(4) For a review, see: Srikanth, G. S. C.; Castle, S. L. Tetrahedron
Articles in Press (doi:10.1016/j.tet.2005.07.077)
(5) Selected examples: (a) Miyabe, H.; Asada, R.; Takemoto, Y.
Tetrahedron 2005, 61, 385. (b) Sibi, M. P.; Asano, Y.; Sausker, J. B.
Angew Chem., Int. Ed. 2001, 40, 1293. (c) Kabat, M. M. Tetrahedron
Lett. 2001, 42, 7521. (d) Chai, C. L. L.; King, A. R. J. Chem. Soc., Perkin
Trans. 1 1999, 1173. (e) Axon, J. R.; Beckwith, A. L. J. J. Chem. Soc.,
Chem. Commun. 1995, 549. (f) Gasanov, R. G.; Il’inskaya, L. V.;
Misharin, M. A.; Maleev, V. I.; Raevski, N. I.; Ikonnikov, N. S.; Orlova,
S. A.; Kuzmina, N. A.; Belokon,’ Y. N. J. Chem. Soc., Perkin Trans. 1
1994, 3343. (g) Kessler, H.; Wittmann, V.; Ko¨ck, M.; Kottenhahn, M.
Angew. Chem., Int. Ed. Engl. 1992, 31, 902. (h) Crich, D.; Davies, J.
W.; Negro´n, G.; Quintero, L. J. Chem. Res., Synop. 1988, 140. For a
complete listing, see ref 4.
(6) Renaud, P.; Stojanovic, A. Tetrahedron Lett. 1996, 37, 2569.
(7) Preliminary communication: Srikanth, G. S. C.; Castle, S. L.
Org. Lett. 2004, 6, 449.
(8) Fornicola, R. S.; Oblinger, E.; Montgomery, J. J. Org. Chem.
1998, 63, 3528.
(9) For a review of Lewis acid promoted radical reactions, see:
Renaud, P.; Gerster, M. Angew. Chem., Int. Ed. 1998, 37, 2562.
(10) Sibi, M. P.; Ji, J.; Sausker, J. B.; Jasperse, C. P. J. Am. Chem.
Soc. 1999, 121, 7517.
(11) Nozaki, K.; Oshima, K.; Utimoto, K. Bull. Chem. Soc. Jpn. 1991,
64, 2585
(12) Treatment of 1a with ZnCl2 (1.1 equiv) and Bu3SnH (2 equiv)
resulted in recovery of starting material. Repeating this experiment
with the addition of Et3B (5 equiv) also yielded recovered 1a. When
this reaction was conducted in the presence of i-PrI (5 equiv), reduced
product 3a was obtained in 60% yield (see the second entry of Table
1).
Having established the viability of esters 1a-c as
radical conjugate addition acceptors, we wished to ex-
amine the performance of the corresponding amides in
this reaction. The synthesis of these substrates is detailed
in Scheme 1. Direct amide bond formation between
(13) Hirasawa, S.; Nagano, H.; Kameda, Y. Tetrahedron Lett. 2004,
45, 2207.
(14) The olefin stereochemistry of Z-1a and Z-1b was determined
by X-ray crystallography, and the identities of the isomers of 1c were
assigned by comparison of its 1H NMR spectra to those of 1a and 1b.
The olefinic proton of the E-isomers resonates at 8.1-8.0 ppm, whereas
in the Z-isomers it resonates at 7.6-7.5 ppm.
J. Org. Chem, Vol. 70, No. 20, 2005 8141

He et al.
SCHEME 1. Synthesis of r,â-Unsaturated r-Nitro
Amides 6
SCHEME 2. Conversion of 2 and 7 into Protected
Amino Acids
TABLE 3. Incorporation and Retention of Deuterium in
2 and 7
TABLE 2. Radical Conjugate Additions to Amides 6
substrate
time (h)
workup
H2O
0.25 N HCl
0.25 N HCl
0.25 N HCl
1 N HCl
H incorporationa (%)
1c
1c
1b
1a
6a
6b
6c
3
3
3
3
1
3
3
10
4
substrate
Lewis acid
time (h)
7/8
yielda (%)
ndb
ndb
4
6a
6a
6a
6a
6a
6a
6a
6b
6b
6b
6c
6c
6c
MgBr2‚OEt2
Zn(OTf)2
Cu(OTf)2
Yb(OTf)3
Sm(OTf)3
La(OTf)3
noneb
Zn(OTf)2
Cu(OTf)2
Yb(OTf)3
Zn(OTf)2
Cu(OTf)2
Yb(OTf)3
2
1
1
1
1
1
4
3
3
3
4
4
4
0:100
86:14
52:48
28:72
25:75
25:75
100:0
84:16
44:56
25:75
89:11
44:56
25:75
82
92
70
62
1 N HCl
1 N HCl
ndb
11
54 (64)
50 (64)
72
a Measured by 1H NMR of d-2a-c- or d-10a-c (see text). b Not
detected.
84
52
addition to 6a in the absence of Lewis acid was observed
which produced 7a exclusively. However, this process
required 20 equiv of i-PrI to deliver a reasonable yield of
7a in 4 h. Comparison of the yields and times of each
reaction indicated that p-methoxyphenyl-substituted amide
6a was the best substrate for the radical conjugate
addition, followed by phenyl-substituted amide 6b and
p-fluorophenyl-substituted amide 6c. Adducts 7 were
isolated as 1:1 mixtures of diastereomers despite the fact
that a single olefin isomer of 6 was used in each reaction.
Radical conjugate addition products 2 and 7 were
successfully converted into N-protected amino acid de-
rivatives 9 and 10, respectively, via catalytic hydrogena-
tion and N-Cbz protection (Scheme 2). Thus, we had
accomplished our initial goal of synthesizing â-substi-
tuted R-amino acids via radical conjugate additions to
R,â-unsaturated R-nitro esters and amides. Accordingly,
we turned our attention to development of a stereoselec-
tive variant of this method. A requirement for such a
process is inhibition of racemization of the sensitive
R-stereocenters of 2 and 7.17 To determine if this would
be possible, we performed radical conjugate additions to
1 and 6 with Bu3SnD. Absence of an R-hydrogen signal
58
80
46
30
a Sum of the isolated yields of 7 and 8. Yields in parentheses
are based on recovered 6. b 20 equiv of i-PrI was used.
methyl nitroacetate (4)15 and benzylamine afforded N-ben-
zyl-2-nitroacetamide (5). Amide 5 then participated in
Knoevenagel condensations with aryl aldehydes, provid-
ing R,â-unsaturated R-nitro amides 6 in moderate yields
(49-66%). In contrast to esters 1, the amides were
obtained as single alkene diastereomers possessing the
E configuration.16
Our studies of isopropyl radical conjugate additions to
amides 6 are summarized in Table 2. We employed the
same conditions used in additions to esters 1 with the
exception of a change in the CH2Cl2/Et2O ratio from 4:1
to 1:1 in order to facilitate solubility of the substrate-
Lewis acid complexes. Additionally, we held the amount
of i-PrI constant at 5 equiv; greater amounts of this
reagent did not lead to alterations in product ratios. In
contrast to the additions to esters, the reactions with
amide substrates 6 afforded reduced products 8 in
varying amounts depending on the Lewis acid. Lan-
thanide triflate-mediated reactions delivered 8 as the
major product, whereas the use of Cu(OTf)2 resulted in
roughly equimolar amounts of 7 and 8. The best results
(7:8 ) 5-8:1, 80-92% yield) were obtained with Zn(OTf)2
as Lewis acid. As with ester 1a, a slow radical conjugate
1
in the H NMR spectra of d-2 and d-7 would constitute
evidence that epimerization could be prevented.
The results of these experiments are contained in Table
3. When we conducted this reaction with electron-poor
ester 1c (R1 ) p-FPh), we observed modest levels of
R-proton incorporation (10%) when water was used in the
workup. Fortunately, when this reaction was worked up
with 0.25 N HCl instead of water, the amount of D-H
exchange dropped to 4%. Significantly, R-protons were
(15) Zen, S.; Koyama, M.; Koto, S. Organic Syntheses; Wiley: New
York, 1988; Collect. Vol. VI, p 797.
(16) The E olefin stereochemistry of 6a was determined by X-ray
crystallography, and the configurations of 6b and 6c were assigned
by comparison of their 1H NMR spectra to that of 6a. The olefinic
protons of these compounds resonate at 8.02-8.01 ppm, consistent with
the chemical shifts of the corresponding protons in E-1a-c.
1
undetectable in the H NMR spectra of d-2b (R ) Ph)
(17) The pKa of ethyl nitroacetate has been measured as 5.62. For
a review of the chemistry of R-nitro esters, see: Shipchandler, M. T.
Synthesis 1979, 666.
8142 J. Org. Chem., Vol. 70, No. 20, 2005

Enantioselective Radical Conjugate Additions
TABLE 4. Nitro Reduction of Amides 7
substrate
conditions
T (°C)
time (h)
yielda (%)
H incorporation (%)
6a
6a
6a
6a
6a
6b
6c
Zn, 6 N HCl, EtOH
rt
80
80
rt
rt
rt
rt
12
11
11
40
12
12
12
24
27
49
14
72
71
73
24
17
16
10
10
4
Fe, concd HCl, EtOH
Sn, concd HCl, EtOH
SnCl2, 6 N HCl, EtOH
In, concd HCl, THF/H2O
In, concd HCl, THF/H2O
In, concd HCl, THF/H2O
4
a Isolated yield of d-12 from 6.
SCHEME 3. Indium Reduction of Nitro Esters 2
and d-2a (R ) p-MeOPh) when this weakly acidic workup
was employed. On the other hand, radical conjugate
additions to amides 6a-c with Bu3SnD were character-
ized by extensive D-H exchange (32-57%, data not
shown). NMR spectra implicated the amide hydrogen in
this exchange.18 Moreover, examination of spectra of
crude d-7 revealed that loss of the R-deuterium was
occurring during SiO2 chromatography. Consequently, we
removed this step from our procedure and subjected
crude d-7 to hydrogenation followed by N-Cbz protection
and chromatography of carbamates 10. We were pleased
to discover that this modification greatly attenuated the
D-H exchange, as R-protons were not detected in the 1H
NMR spectrum of d-10b and were observed at low levels
(4% and 11%) in the spectra of d-10a and d-10c. It is
noteworthy that of the six substrates examined in this
study, only one exhibited >4% proton incorporation at
the R-position.
Although hydrogenating crude amide adducts d-7
allowed us to determine the extent of D-H exchange, the
reductions were sluggish and difficult to reproduce due
to the presence of tin byproducts and required excess
amounts of 10% Pd/C. Therefore, we sought an alterna-
tive method that would facilitate high-yielding, reproduc-
ible nitro reduction of crude d-7 while keeping R-proton
incorporation to a minimum. Our survey of the various
methods for nitro group reduction19 is presented in Table
4. Of the reagents examined, only In/HCl20 provided
acetamides d-12 in good yield with little D-H exchange.
These conditions were also applicable to crude adducts
d-2 derived from esters 1 (Scheme 3). Accordingly, we
selected this reduction protocol for use in our investiga-
tions of stereoselective radical conjugate additions to both
6 and 1 due to the fact that chromatographic removal of
tin byproducts from the adducts was not required.
At this stage, we commenced investigations into ste-
reoselective radical conjugate additions promoted by
chiral Lewis acids. We chose R,â-unsaturated R-nitro
amide 6a as our test substrate, and we analyzed each
reaction by purifying adduct 7a and subjecting it to chiral
HPLC. From our previous work, we recognized that this
survey would be impacted by epimerization of the R-ste-
reocenter. Nevertheless, this assay permitted us to
identify promising chiral Lewis acids in rapid fashion.
Each reaction was conducted in CH2Cl2; ether cosolvents
were unnecessary due to the solubilizing effect of the
chiral ligands. Unfortunately, most combinations of chiral
ligands (bisoxazoline,21 indan-pybox,22 prolinol,23 salen24)
and Lewis acids (Mg, Zn, Ni, Cu, Al, lanthanides)
examined gave both diastereomers of 7a as racemic
mixtures. Moreover, many of the reactions were sluggish
and did not proceed to completion, a consequence of
replacing small, electron-withdrawing achiral ligands
with bulky, electron-rich chiral ligands. Our only promis-
ing results came with the DBFOX/Ph ligands developed
by Kanemasa and Curran (Figure 1).25 Use of the parent
ligand 14a with Mg(NTf2)226 afforded 7a as a 1:1 mixture
(18) 1H NMR spectra recorded immediately after isolation of d-7a-c
exhibited amide N-H signals attenuated by an amount consistent with
the extent of proton incorporation at the R-position. Over time, the
N-bound deuterium exchanged with protons derived from adventitious
moisture and the amide signals returned to their normal levels of
intensity. Attempts to block this exchange by performing radical
conjugate additions on protected or tertiary amides were unsuccessful;
the N-Boc derivative of 6a afforded reduction product exclusively,
whereas N-PMB and N-Me versions were unreactive.
(21) Sibi, M. P.; Ji, J. J. Org. Chem. 1997, 62, 3800.
(22) Davies, I. W.; Gerena, L.; Lu, N.; Larsen, R. D.; Reider, P. J. J.
Org. Chem. 1996, 61, 9629.
(23) (a) Kobayashi, S.; Ogawa, C.; Kawamura, M.; Sugiura, M.
Synlett 2001, 983. (b) Sibi, M. P.; Manyem, S. Org. Lett. 2002, 4, 2929.
(24) Taylor, M. S.; Jacobsen, E. N. J. Am. Chem. Soc. 2003, 125,
11204.
(25) (a) Kanemasa, S.; Oderaotoshi, Y.; Yamamoto, H.; Tanaka, J.;
Wada, E.; Curran, D. P. J. Org. Chem. 1997, 62, 6454. (b) Iserloh, U.;
Curran, D. P.; Kanemasa, S. Tetrahedron: Asymmetry 1999, 10, 2417.
(c) Iserloh, U.; Oderaotoshi, Y.; Kanemasa, S.; Curran, D. P. Org. Synth.
2003, 80, 46.
(19) Larock, R. C. Comprehensive Organic Transformations, 2nd ed.;
Wiley-VCH: New York, 1999; pp 821-828.
(20) Lee, J. G.; Choi, K. I.; Koh, H. Y.; Kim, Y.; Kang, Y.; Cho, Y. S.
Synthesis 2001, 81.
J. Org. Chem, Vol. 70, No. 20, 2005 8143

He et al.
TABLE 5. Optimization of Stereoselective Radical Conjugate Addition
entry
Lewis acid
Et3B
concn of 6a (M)
yield (%)
syn/anti
% ee (syn, anti)a
1
2
3
4
5
6
7
8
MgClO4
1.0 M in hexane
1.0 M in hexane
3.5 M in CH2Cl2
3.5 M in CH2Cl2
3.5 M in CH2Cl2
3.5 M in CH2Cl2
3.5 M in CH2Cl2
3.5 M in CH2Cl2
0.05
0.05
0.05
0.05
0.1
76
44
55
49
76
62
70
57
2.0:1
2.0:1
1.5:1
1.2:1
1.4:1
1.2:1
1.4:1
1.2:1
19, 13
28, 29
70, 29
77, 72
88, 76
68, 72
84, 64
41, 30
Mg(NTf2)2
MgClO4
Mg(NTf2)2
b
Mg(NTf2)2
c
Mg(NTf2)2
0.1
d
Mg(NTf2)2
0.1
0.1
e
Mg(NTf2)2
a Determined by chiral HPLC (see the Supporting Information for details). b Average of two runs. c 4 Å MS added. d 0.5 equiv of Lewis
acid and 14a was used. e 0.3 equiv of Lewis acid and 14a was used.
increases in ee, with Mg(NTf2)2 delivering the best results
(entries 3 and 4). Elimination of hexane from the reaction
mixture allowed us to increase the substrate concentra-
tion to 0.1 M, further improving the yield and ee’s (entry
5). Addition of 4 Å MS was deleterious to the reaction
(entry 6),28 and substoichiometric quantities of chiral
Lewis acid afforded lower ee’s (entries 7 and 8). The latter
result was attributed to competition from the uncatalyzed
radical conjugate addition caused by inefficient turnover
of the chiral Lewis acid. Although each diastereomer
could be obtained in good ee, the diastereoselectivity was
disappointingly low, with only a slight preference ob-
served for the syn isomer.29
FIGURE 1. DBFOX/Ph ligands.
SCHEME 4. Synthesis of Enantioenriched 7a
We then proceeded to investigate the substrate scope
of the stereoselective reaction using the conditions shown
in Table 5, entry 5 (Table 6). Low ee’s were observed for
all three esters tested (1a-c) despite the fact that they
were used as the pure Z-isomers. In contrast, amide
substrates 6b and 6c delivered protected â-amino acid
derivatives 10b and 10c with fair ee’s. Since the carbonyl
groups of amides are generally more electron-rich than
those of esters, they may bind more tightly to the chiral
Lewis acid, leading to better selectivity.30 In support of
this hypothesis, amide 6a possessing an electron-rich â-p-
methoxyphenyl group afforded significantly better ee’s
than any other substrate tested (Table 5, entry 5).
Unfortunately, all of the acceptors examined exhibited
only modest (1.4-2.6:1) preferences for the syn diaste-
reomer.
of diastereomers in 56 and 47% ee (Scheme 4). Similar
results were obtained with Mg(ClO4)2. Increasing the
ligand bulk was detrimental, as Me-DBFOX/Ph (14b)
delivered 7a in low ee’s (11, 20%), and n-Bu-DBFOX/Ph
(14c) provided racemic products.27
Having identified Mg/14a as the best chiral Lewis acid,
we set out to optimize the reaction conditions and more
accurately determine the stereoselectivity. Thus, we
isolated and analyzed carbamate 10a in order to avoid
the R-epimerization inherent in purification of 7a. The
results of our investigation of the effects of several
parameters on the yield and ee’s are summarized in Table
5. All reactions were conducted for 4-5 h with three
initiation cycles consisting of addition of Et3B (5 equiv),
i-PrI (5 equiv), and Bu3SnH (2.5 equiv) performed at 1.5
h intervals. The need for extra initiation likely reflects
the weaker strength of the chiral Lewis acids relative to
their achiral congeners. We discovered that solvent and
concentration had a profound effect on the process. When
Et3B was introduced as a 1.0 M solution in hexane, the
ee’s were low and the concentration of 6a could not be
increased above 0.05 M due to insolubility of the Lewis
acid-substrate complex (entries 1 and 2). Switching to
a 3.5 M solution of Et3B in CH2Cl2 led to significant
In an effort to understand the origins of the stereose-
lectivity, we converted adducts 7b into the known amino
acids syn- and anti-15 (Scheme 5).29 Measurement of the
(28) For a discussion of the effect of molecular sieves on chiral Lewis
acids, see ref 23b.
(29) The relative stereochemistry of ester adduct 2b (see Table 6)
was determined by reduction to the amine (In, concd HCl), preparative
TLC separation of the diastereomers, and ester hydrolysis (6 N HCl).
1H NMR spectra of the diastereomers matched the literature data for
the known amino acids: Liao, S.; Shenderovich, M. D.; Lin, J.; Hruby,
V. J. Tetrahedron 1997, 53, 16645. Relative stereochemistry assign-
ments of all other adducts are based on comparison of their 1H NMR
spectra to those of 2b. Specifically, the R- and â-protons of the syn
isomers each resonated 0.15-0.20 ppm upfield from the corresponding
protons of the anti isomers, and the syn-isopropyl methine resonated
ca. 0.20 ppm downfield from the anti-isopropyl methine.
(30) In competition experiments, Sato has observed a preference for
Lewis acids to complex with acrylamides rather than acrylates: Urabe,
H.; Yamashita, K.; Suzuki, K.; Kobayashi, K.; Sato, F. J. Org. Chem.
1995, 60, 3576.
(26) Sibi, M. P.; Petrovic, G. Tetrahedron: Asymmetry 2003, 14,
2879.
(27) Shirahase, M.; Kanemasa, S.; Hasegawa, M. Tetrahedron Lett.
2004, 45, 4061.
8144 J. Org. Chem., Vol. 70, No. 20, 2005

Enantioselective Radical Conjugate Additions
TABLE 6. Substrate Scope of Stereoselective Radical
Conjugate Additiona
FIGURE 2. Postulated reactive complex.
TABLE 8. Conjugate Additions of Various Alkyl
Radicals to 6a
substrate
yield (%)
syn/anti
% ee (syn, anti)b
1a
1b
1c
6b
6c
75
60
66
66
59
2.6:1
1.7:1
1.7:1
1.4:1
1.6:1
21, 28
9, 17
6, 12
64, 62
72, 50
a For reaction conditions, see Table 5, entry 5. b Determined by
chiral HPLC of 9, 10, or the corresponding acetates 12 (see the
Supporting Information for details).
SCHEME 5. Absolute Configuration
Determination
R
yield (%)
syn/anti
% ee (syn, anti)a
Et
52
46
b
1:1.9
1.4:1
57, 64
80, 81
c-C6H11
t-Bu
a Determined by chiral HPLC of 17 (anti isomer) or the
corresponding acetate (syn isomer); see the Supporting Information
for details. b Complex mixture (see text).
the â carbon of 6. The R ee was greatest with p-
methoxyphenyl-substituted amide 6a, a fact consistent
with our hypothesis that more electron-rich substrates
form tighter complexes with the chiral Lewis acid.
Finally, the relatively similar ee’s obtained for syn and
anti diastereomers (Tables 5 and 6) coupled with the fact
that the absolute configuration at the R-carbon is identi-
cal for both major enantiomers (Scheme 5) indicates that
R-stereocenter formation is under reagent rather than
substrate control, an outcome predicted by our prelimi-
nary investigations.7
Curran and Kanemasa have proposed that enantiose-
lective radical conjugate additions to N-enoyloxazolidi-
nones mediated by Mg(ClO4)2-DBFOX/Ph proceed
through an octahedral complex.25b This geometry can also
explain our stereochemical results (Figure 2). Thus,
complexation of the substrate carbonyl oxygen in the
plane of the chiral ligand and one of the nitro oxygens
perpendicular to this plane leads to shielding of the re
face of the olefin, with the magnitude of this effect being
greatest at the R-carbon. Then, the major product of the
reaction is produced by radical conjugate addition and
hydrogen atom abstraction occurring on the si face of the
olefin. This substrate-Lewis acid binding model is purely
empirical; it is unclear to us why the amide carbonyl and
nitro oxygen would prefer to bind in the orientation
shown. Accordingly, this is a subject worthy of future
study.
Finally, we performed additions of various alkyl radi-
cals to 6a in order to more fully determine the scope of
the stereoselective radical conjugate addition (Table 8).
In comparison to isopropyl radical (Table 5, entry 5), the
smaller ethyl radical added to 6a with lower ee. Interest-
ingly, this addition slightly favored the anti diastereomer;
the reason for this reversal is unclear. The attenuated
ee is also puzzling since the data in Tables 5 and 6
suggest that the â-stereocenter has little effect on the
TABLE 7. Selectivity of r and â Stereocenter
Formation
substrate
R ee (%)
â ee (%)
6a
6b
6c
83
63
64
20
12
25
optical rotations of these enantiomerically enriched
samples indicated that the major enantiomers possessed
the (2R,3S)- and (2R,3R)-configurations, respectively.
With knowledge of the absolute configurations and rela-
tive amounts of all four isomers produced in the chiral
Lewis acid mediated radical conjugate addition to amide
6b, we were able to determine the level of selectivity in
formation of the R and â stereocenters. By assuming that
the absolute configurations of adducts 7a and 7c were
identical to those of 7b, we performed analogous calcula-
tions for these substrates derived from additions to 6a
and 6c (Table 7). These data show that the initial alkyl
radical addition step is only marginally selective, whereas
the subsequent hydrogen atom abstraction proceeds with
fair to good stereoselectivity. This suggests that the chiral
Lewis acid provides reasonable shielding of the R but not
J. Org. Chem, Vol. 70, No. 20, 2005 8145

He et al.
added a Lewis acid (0.14 mmol). The resultant mixture was
stirred at rt for 30 min. The clear solution was cooled to -78
°C and treated first with i-PrI (71 µL, 106 mg, 0.63 mmol),
then immediately with Et3B (1.0 M solution in hexanes, 0.66
mL, 0.63 mmol), and finally with Bu3SnH (77 µL, 81 mg, 0.25
mmol). O2 was added via syringe (2 mL) every 30 min, and
the reaction was stirred at -78 °C for 3 h. The mixture was
diluted with CH2Cl2 (10 mL) followed by 0.25 N HCl (5 mL),
and the layers were separated. The organic layer was dried
(Na2SO4) and concentrated in vacuo. Flash chromatography
(SiO2, 2 × 12 cm, 0-0.75% EtOAc in hexanes gradient elution)
afforded the adducts as colorless oils (2a-c) or the reduced
compounds (3a, 3c) along with tin impurities. Analytically
pure samples were prepared by subjecting the compounds to
further chromatography (same conditions).
Methyl 3-(4-Methoxyphenyl)-4-methyl-2-nitropentan-
oate (2a). Prepared with ZnCl2 or Zn(OTf)2 as Lewis acid and
obtained in 85% yield as a 1:1 mixture of diastereomers: 1H
NMR (CDCl3, 500 MHz) δ 7.16 and 7.13 (2d, J ) 9.0 Hz, 2H),
6.82 (d, J ) 8.5 Hz, 2H), 5.60 and 5.56 (2d, J ) 10.5 and 11
Hz, 1H), 3.85 and 3.79 (2s, 3H), 3.78 and 3.52 (2s, 3H), 3.62-
3.54 (m, 1H), 2.10-2.02 (m, 1H), 0.92-0.84 (m, 6H); 13C NMR
(CDCl3, 125 MHz) δ 164.8 and 164.0, 159.2, 132.5, 130.8, 130.4,
127.0 and 126.8, 113.9, 91.2 and 90.8, 55.4, 53.8 and 53.4, 51.8,
29.5 and 29.4, 21.7 and 21.2, 18.1 and 18.0; IR (film) νmax 3240,
1656, 1545 cm-1; HRMS (EI) m/z 281.1256 (M+, C14H19NO5
requires 281.1263).
Methyl 2-Deuterio-3-(4-methoxyphenyl)-4-methyl-2-ni-
tropentanoate (d-2a). 1H NMR (CDCl3, 500 MHz) was
identical to 2a except for the absence of the two doublets at δ
5.60 and 5.56. Also, the multiplet at δ 3.62-3.54 appeared as
two doublets centered at δ 3.60 and 3.56 (J ) 6.5 Hz).
Methyl 4-Methyl-2-nitro-3-phenylpentanoate (2b). Pre-
pared with ZnCl2 as Lewis acid and obtained in 85% yield as
a 1:1 mixture of diastereomers: 1H NMR (CDCl3, 500 MHz) δ
7.38-7.30 (m, 3H), 7.18-7.11 (m, 2H), 5.66 and 5.62 (2d, J )
11 and 10.5 Hz, 1H), 3.93 and 3.54 (2s, 3H), 3.68-3.59 (m,
1H), 2.06-2.00 (m, 1H), 0.86 and 0.79 (2d, J ) 7.0 Hz, 6H);
13C NMR (CDCl3, 125 MHz) δ 164.7 and 163.9, 135.2 and 135.1,
132.7 and 132.5, 130.6 and 130.0, 129.7 and 129.6, 129.3 and
129.0, 128.4 and 127.9, 91.0 and 90.8, 53.8 and 53.6, 52.5 and
52.4, 29.6 and 29.4, 21.7 and 21.1, 18.1 and 18.0; IR (film) νmax
3242, 1651, 1555 cm-1; HRMS (FAB) m/z 274.1054 (MNa+,
C13H17NO4Na requires 274.1055).
stereochemistry of hydrogen atom abstraction. Additions
of cyclohexyl radical proceeded with similar stereoselec-
tivity but lower yields than those of isopropyl radical.
However, no adducts could be obtained with tert-butyl
radical despite the use of excess reagents (a total of 30
equiv of t-BuI, 30 equiv of Et3B, and 15 equiv of n-Bu3-
SnH added in six portions). Acceptor 6a was completely
consumed, and no reduction product could be detected,
suggesting that addition of the bulky tertiary radical to
the substituted â-carbon is slower than polymerization
of the alkene or other decomposition pathways. Conse-
quently, it appears that the stereoselective reaction
performs best with secondary alkyl radicals. Neverthe-
less, a more extensive study of additions of various
nucleophilic radicals to 6a is required in order to better
understand the scope and limitations of this process.
Conclusions
We have demonstrated that R,â-unsaturated R-nitro
esters and amides undergo Lewis acid promoted radical
conjugate additions, affording â-substituted R-amino acid
derivatives. With achiral Lewis acids, competitive reduc-
tion is more pronounced with amides, presumably due
to stronger substrate-Lewis acid complexes. Neverthe-
less, good yields of adducts can be obtained from both
classes of acceptors when Zn Lewis acids are employed.
Through the use of Bu3SnD, we discovered a workup,
nitro reduction, and isolation protocol that greatly mini-
mizes epimerization of the extremely acidic R-stereo-
center present in adducts 2 and 7. This finding led us to
survey several chiral Lewis acids in an attempt to develop
a stereoselective process. The best chiral promoter ex-
amined to date is Kanemasa’s DBFOX/Ph ligand (14a)
combined with Mg(NTf2)2 or Mg(ClO4)2. This complex
affords protected amino acid derivatives 10 with fair to
good ee’s but disappointingly low dr’s. Determination of
the absolute configuration of the major enantiomers
produced in additions to 6b revealed that hydrogen atom
abstraction proceeds with significantly better selectivity
than does the radical addition. Accordingly, a chiral
Lewis acid capable of providing improved enantiofacial
discrimination at the acceptor â-carbon is necessary to
render these stereoselective radical conjugate additions
practical.
Methyl 2-Deuterio-4-methyl-2-nitro-3-phenylpentanoate
(d-2b). 1H NMR (CDCl3, 500 MHz) was identical to 2b except
for the absence of the two doublets at δ 5.66 and 5.62. Also
the multiplet at δ 3.68-3.59 appeared as two doublets centered
at δ 3.66 and 3.61 (J ) 8.5 Hz).
General Procedure for Radical Conjugate Additions
to Amides 6a-c Promoted by Achiral Lewis Acids. To a
solution of 6 (0.080 mmol) in CH2Cl2-THF (1:1, 1.5 mL) was
added a Lewis acid (0.088 mmol). The resultant mixture was
stirred at rt for 30 min. The clear solution was cooled to -78
°C and treated with first with i-PrI (47 µL, 68 mg, 0.40 mmol),
then immediately with Et3B (1.0 M solution in hexanes, 0.41
mL, 0.40 mmol), and finally with Bu3SnH (50 µL, 52 mg, 0.17
mmol). O2 was added via syringe (2 mL) every 30 min, and
the reaction was stirred at -78 °C for 1-4 h. The mixture
was diluted with CH2Cl2 (15 mL) followed by 1 N HCl (5 mL),
and the layers were separated. The organic layer was dried
(Na2SO4) and concentrated in vacuo. Flash chromatography
(SiO2, 2 × 12 cm, 2-8% EtOAc in hexanes gradient elution)
afforded mixtures of 7 and 8 as white solids. Analytically pure
samples of 7a-c were prepared by washing the solid with cold
hexanes.
Although the additions controlled by chiral Lewis acids
require further development, the corresponding reactions
with achiral Lewis acids have great potential in peptide
synthesis. The utility of amides 6 as acceptors suggests
that peptides containing the R,â-unsaturated R-nitro
amide moiety may also be good substrates for this
reaction. The stereochemical information present in these
acceptors could provide the necessary stereocontrol.
Moreover, the ability to selectively epimerize the R-nitro
amide stereocenter and obtain a thermodynamically more
stable product would be extremely useful provided a
means of controlling the radical addition step could be
devised. Efforts along these lines in the context of natural
products total synthesis are underway and will be
reported in due course.
N-Benzyl-3-(4-methoxyphenyl)-4-methyl-2-nitropen-
tanamide (7a). Prepared with Zn(OTf)2 as Lewis acid, ob-
tained as an 86:14 mixture with 8a in 82% combined yield,
and isolated as a 1:1 mixture of diastereomers: 1H NMR
(CDCl3, 500 MHz) δ 7.36-7.28 (m, 2H), 7.26-7.18 (m, 3H),
7.02 (d, J ) 8.5 Hz, 2H), 6.87 and 6.40 (2br s, 1H), 6.80 (d, J
Experimental Section
General Procedure for Radical Conjugate Additions
to Esters 1a-c Promoted by Achiral Lewis Acids. To a
solution of 1 (0.12 mmol) in CH2Cl2-Et2O (4:1, 1 mL) was
8146 J. Org. Chem., Vol. 70, No. 20, 2005

Enantioselective Radical Conjugate Additions
) 8.5 Hz, 2H), 5.59 and 5.57 (2d, J ) 10.0 and 10.5 Hz, 1H),
4.46 and 4.20 (dd and d, J ) 5.5, 15 Hz and 6.0 Hz, 1H), 4.45
and 4.20 (dd and d, J ) 6.0, 12.5 Hz and 6.0 Hz, 1H), 3.81 and
3.77 (2s, 3H), 3.50 and 3.44 (2dd, J ) 5.0, 11 Hz and 5.0, 10.0
Hz, 1H), 2.08-2.00 and 1.98-1.92 (2m, 1H), 0.94 and 0.90 (2d,
J ) 8.0 Hz, 3H), 0.91 (d, J ) 8.5 Hz, 3H); 13C NMR (CDCl3,
125 MHz) δ 162.9 and 162.2, 159.4, 137.0 and 136.9, 130.9
and 130.3, 129.1 and 128.8 (2C), 128.1, 127.9 and 127.85 (2C),
127.80 (2C), 126.8 and 126.2, 114.0 and 113.9, 93.1 and 92.9,
55.4, 54.0 and 53.3, 44.2 and 44.0, 29.5 and 28.9, 21.8 and 21.3,
18.2 and 17.6; IR (film) νmax 3259, 1659, 1552 cm-1; HRMS
(FAB) m/z 379.1647 (MNa+, C20H24N2O4Na requires 379.1634).
N-Benzyl-4-methyl-2-nitro-3-phenylpentanamide (7b).
Prepared with Zn(OTf)2 as Lewis acid, obtained as an 84:16
mixture with 8b in 84% combined yield, and isolated as a 1:1
mixture of diastereomers: 1H NMR (CDCl3, 500 MHz) δ 7.36-
7.24 (m, 6H), 7.20 (d, J ) 10.5 Hz, 2H), 7.10 and 6.79 (2d, J )
7.0 and 8.0 Hz, 2H), 6.85 and 6.37 (2br s, 1H), 5.64 and 5.62
(2d, J ) 7.5 and 10.5 Hz, 1H), 4.51-4.41 (m, 1H), 4.25-4.15
(m, 1H), 3.57-3.50 (m, 1H), 2.12-2.04 and 2.03-1.97 (2m, 1H),
0.98-0.85 (m, 6H); 13C NMR (CDCl3, 125 MHz) δ 162.8 and
162.1, 137.0 and 136.8, 134.9 and 134.5, 129.9 and 129.3, 129.1
and 128.8 (2C), 128.7, 128.6 and 128.5 (2C), 128.18 and 128.10,
127.9 and 127.7, 128.0 and 127.8 (2C), 93.0 and 92.8, 54.4 and
54.0, 44.3 and 44.0, 29.5 and 28.9, 21.8 and 21.3, 18.1 and 17.7;
IR (film) νmax 3221, 1650, 1564 cm-1; HRMS (FAB) m/z
327.1704 (MH+, C19H22N2O3H requires 327.1709).
and 5.11 (s, 2H), 5.00 (br s, 1H), 4.99 and 4.84 (2d, J ) 6 and
9.5 Hz, 1H), 4.86 and 4.82 (2dd, J ) 6.5, 9.0 Hz, 1H), 3.78 (s,
3H), 3.65 and 3.63 (2s, 3H), 2.84-2.79 and 2.62-2.58 (2m, 1H),
2.26-2.20 and 2.09-2.02 (2m, 1H), 1.26-1.19 and 0.79-0.72
(m, 6H); 13C NMR (CDCl3, 125 MHz) δ 172.7 and 171.7, 158.9,
156.8 and 155.9, 136.5 and 136.4, 130.5, 130.1 (2C), 129.7 (2C),
128.8 and 128.7, 128.4 and 128.3 (2C), 127.9 and 127.2, 114.2
and 113.9, 67.3 and 67.1, 56.2 and 56.0, 55.3, 54.5, 52.3 and
52.1, 29.9 and 29.0, 21.5, 21.1; IR (film) νmax 3346, 2959, 1726,
1602, 1222, 1048 cm-1; HRMS (FAB) m/z 386.1975 (MH+,
C22H27NO5H requires 386.1967).
syn-9a was obtained in 21% ee, as analyzed by HPLC
(Chiralcel OD-H, 98:2 hexane/i-PrOH, 1 mL/min; tR ) 16.9 min
(major), 27.3 min). anti-9a was obtained in 28% ee, as analyzed
by HPLC under identical conditions (tR ) 11.4 min (major),
25.2 min).
Methyl 2-Benzyloxycarbonylamino-2-deuterio-3-(4-
methoxyphenyl)-4-methylpentanoate (d-9a). (Produced by
combination of the general procedure for radical conjugate
additions to esters promoted by achiral Lewis acids with the
reduction and protection protocols contained in the chiral
Lewis acid general procedure.) 1H NMR (CDCl3, 500 MHz) was
identical to 9a except for the absence of the two signals at δ
4.99 and 4.84. Also, the multiplets at δ 2.84-2.79 and 2.62-
2.58 appeared as two doublets centered at δ 2.82 and 2.61
(J ) 10.0 and 10.5 Hz). HRMS (FAB) m/z 387.2018 (M+,
C22H26DNO5H requires 387.2012).
N-Benzyl 2-benzyloxycarbonylamino-3-(4-methoxy-
phenyl)-4-methylpentanamide (10a) (76% yield, 1.4:1 syn/
anti): 1H NMR (CDCl3, 500 MHz) δ 7.37-7.24 (m, 7H), 7.17
(d, J ) 10.5 Hz, 2H), 7.05 (dd, J ) 8.5, 14.5 Hz, 2H), 6.81-
6.73 (m, 3H), 6.25 and 6.01 (2 br s, 1H) 5.35 and 4.97 (2d, J )
7.0 and 5.5 Hz, 1H), 5.12 and 5.09 (2s, 2H), 4.72 and 4.61 (2t,
J ) 4.5 and 9.5 Hz, 1H), 4.45-4.24 and 4.37 (m and dd, J )
7.0, 15.5 Hz, 1H), 4.03 (dd, J ) 4.5, 15.5 Hz, 1H), 3.87 (s, 3H),
3.19-3.08 and 2.87-2.77 (2m, 1H), 2.38-2.22 and 2.10-1.94
(2m, 1H), 1.05 and 0.91 (2d, J ) 6.0 Hz, 3H), 0.80 and 0.77
(2d, J ) 6.2 Hz, 3H); 13C NMR (CDCl3, 125 MHz) δ 170.6,
159.4, 156.3, 138.4 and 137.5, 136.3 (2C), 129.8, 128.79 and
128.72 (2C), 128.5 (2C), 128.4, 128.36, 128.32 (2C), 127.8,
127.6, 127.4 and 127.3 (2C), 113.7, 67.4, 57.2, 55.7 and 55.6,
54.8 and 53.4, 43.5, 28.5, 21.7, 18.7; IR (film) νmax 3281, 2930,
1649, 1520 cm-1; HRMS (FAB) m/z 483.2252 (MNa+, C28H32N2-
O4Na requires 483.2260).
syn-10a was obtained in 88% ee, as analyzed by HPLC
(Chiralcel OD-H, 98:2 hexane/i-PrOH, 1 mL/min; tR ) 20.3 min
(major), 27.8 min). anti-10a was obtained in 76% ee, as
analyzed by HPLC under identical conditions (tR ) 27.6 min
(major), 38.4 min).
N-Benzyl 2-Benzyloxycarbonylamino-2-deuterio-3-(4-
methoxyphenyl)-4-methylpentanamide (d-10a). (Pro-
duced by combination of the General Procedure for radical
conjugate additions to amides promoted by achiral Lewis acids
with the reduction and protection protocols contained in the
chiral Lewis acid general procedure.) 1H NMR (CDCl3, 500
MHz) was identical to 10a except for the absence of the signals
at δ 4.72 and 4.61. Also, the two multiplets at δ 3.19-3.08
and 2.87-2.77 appeared as two doublets centered at δ 3.11
and 2.80 (J ) 8.5 and 9.0 Hz). HRMS (FAB) m/z 484.2308
(MNa+, C28H31DN2O4Na requires 484.2322).
General Procedure for Radical Conjugate Additions
Promoted by Chiral Lewis Acid. A solution of Mg(NTf2)2
(116.9 mg, 0.20 mmol) and ligand 14a (91.7 mg, 0.20 mmol)
in anhydrous CH2Cl2 (1.5 mL) was stirred at rt for 1-2 h and
then treated with Z-1 or E-6 (0.20 mmol). The walls of the
reaction vessel were washed with CH2Cl2 (0.5 mL), and the
mixture was stirred at -78 °C for 30 min. Alkyl iodide (1.0
mmol), Bu3SnH (134 µL, 145 mg, 0.50 mmol), Et3B (3.45 M
solution in CH2Cl2, 290 µL, 1.0 mmol), and O2 (10 mL) were
added sequentially, and identical quantities of these reagents
were added twice more at 1.5 h intervals. The mixture was
stirred at -78 °C for an additional 1.5 h (4.5 h total since
initiation of radical reaction), treated with 2 N HCl (10 mL),
and extracted with CH2Cl2 (2 × 10 mL). The combined organic
layers were concentrated in vacuo.
The crude adduct 2 or 7 was treated with concd HCl (0.60
mL, 7.2 mmol), H2O (3 mL), THF (3 mL), and indium powder
(184 mg, 1.6 mmol). The resulting mixture was stirred at rt
for 12 h, the solids were removed, and the volatiles were
removed in vacuo. The residue was diluted with 1 N HCl (10
mL), and tin byproducts were removed by extraction with
hexanes (3 × 10 mL). The aqueous layer was treated with Na2-
CO3 (added until pH ≈ 8) and satd aq sodium potassium
tartrate solution (10 mL) and then extracted with CH2Cl2 (3
× 20 mL). The combined organic layers were dried (Na2SO4)
and concentrated in vacuo. Thirty percent of the crude amine
mixture was subjected to flash chromatography (SiO2, 25-60%
EtOAc in hexanes gradient elution for amines derived from 2
or 1-4% MeOH in CH2Cl2 gradient elution for amines derived
from 7) to afford samples of the pure syn and anti diastereo-
mers suitable for chiral HPLC analysis after further deriva-
tization.
The remaining 70% of the crude amine mixture was treated
with benzyl chloroformate (22.8 µL, 27.6 mg, 0.15 mmol), Na2-
CO3 (16.3 mg, 0.15 mmol), and THF (3 mL). The resulting
mixture was stirred at rt for 12 h, concentrated in vacuo, and
purified by flash chromatograghy (SiO2, 7-13% EtOAc in
hexanes gradient elution for carbamates derived from 2 or 20-
30% EtOAc in hexanes gradient elution for carbamates derived
from 7), affording carbamates 9 or 10 as white solids that were
mixtures of diastereomers.
Acknowledgment. We thank Brigham Young Uni-
versity and the National Institutes of Health (GM70483)
for financial support and Dr. John F. Cannon for
performing X-ray crystallography.
Supporting Information Available: General experimen-
tal details, synthetic procedures, spectral data, and 1H and
13C NMR spectra for all new compounds. This material is
available free of charge via the Internet at http://pubs.acs.org.
Methyl
2-benzyloxycarbonylamino-3-(4-methoxy-
phenyl)-4-methylpentanoate (9a) (75% yield, 2.6:1 syn/
anti): 1H NMR (CDCl3, 500 MHz) δ 7.38-7.34 (m, 5H), 6.98
and 6.94 (2d, J ) 9.0 and 8.5 Hz, 2H), 6.84-6.81 (m, 2H), 5.14
JO051334F
J. Org. Chem, Vol. 70, No. 20, 2005 8147