Acyclic diastereoselection in prochiral radical addition to prochiral olefins

Article abstract of DOI:10.1021/ja017510t

The stereochemical preference (syn or anti) when prochiral radicals add to prochiral acceptors is of fundamental interest. The primary focus of this research was to determine which factors influence the relative stereochemistry between the β and γ chiral

Full text of DOI:10.1021/ja017510t

Published on Web 03/01/2002
Acyclic Diastereoselection in Prochiral Radical Addition to
Prochiral Olefins
Mukund P. Sibi,* Tara R. Rheault,† Sithamalli V. Chandramouli, and
Craig P. Jasperse
Contribution from the Department of Chemistry, North Dakota State UniVersity,
Fargo, North Dakota 58105-5516
Received November 10, 2001
Abstract: The stereochemical preference (syn or anti) when prochiral radicals add to prochiral acceptors
is of fundamental interest. The primary focus of this research was to determine which factors influence the
relative stereochemistry between the â and γ chiral centers when these are formed concurrently. While
moderate diastereoselectivity was found for addition of alkyl (6a-d) and R-alkoxy radicals (16a-c) (e6:1
syn) to acceptors 4, 7, 8, 10, and 14, consistently high selectivity was observed with less reactive halogenated
radicals (6f,g) (>15:1 anti). Steric influence in alkyl radical additions was difficult to evaluate due to decreased
reactivity when using bulky reaction partners; however, more reactive R-alkoxy radicals, it was found that
increasing steric bulk leads to moderate increases in selectivity. In addition, higher selectivity was observed
when employing lanthanide Lewis acids whose environment (reactivity) was modified using achiral additives,
suggesting a potentially simple means for selectivity enhancements in radical reactions. Overall these results
indicate that significant stereoelectronic effects are necessary to achieve high levels of selectivity in prochiral
radical additions to prochiral acceptors.
Introduction
determine whether the relative stereochemistry at the γ carbon
could also be controlled in situations where a prochiral radical
In recent years effective methods for Lewis acid-activated
intermolecular addition of alkyl radicals to enoyl oxazolidinones
have been developed.1 We have shown that radical addition to
acrylates as well as nonterminal alkenes in which the acceptor
is prochiral proceeds efficiently.2 Convenient methods are at
hand for controlling the absolute configuration at the â center,
either diastereoselectively (using a chiral auxiliary)3 or enan-
tioselectively (using a Lewis acid/chiral ligand combination).4
In addition, we have developed very selective tandem reactions
in which addition at the â center followed by trapping at the R
center results in the highly selective formation of two new
acyclic stereocenters.5 Given our ability to control the stereo-
chemistry at both the â and R centers, we were curious to
2 adds to a nonterminal alkene 1 (eq 1).6
At the outset of our study we were surprised at the lack of
information available regarding fundamental diastereoselectivity7
when prochiral radicals add to prochiral acceptors. Occasionally,
doubly diastereoselective intermolecular radical processes have
surfaced in the literature; however, only limited examples have
been provided, and the systematic acquisition of essential
information regarding factors impacting selectivity has not been
pursued.8 One exception is the intermolecular addition of ketyl
radical anions to electron-deficient olefins.9 Fukuzawa, for
example, has reported highly diastereoselective samarium
iodide-mediated ketyl radical addition to chiral enoates.10
Samarium ketyls are ionic compounds, however, and due to
their charge and the coordinating ability of the samarium
counterion they are not representative of ordinary, nonionic
radicals.
* To whom correspondence should be addressed. E-mail: Mukund.Sibi@
ndsu.nodak.edu.
† NSF graduate Fellow (1998-2001). ACS division of Organic Chemistry
Fellow (2001-2002).
(1) For general information on radical chemistry, see: Radicals in Organic
Synthesis; Renaud, P., Sibi, M. P., Eds.; Wiley-VCH: Weinheim, 2001;
Vols. 1 and 2.
(2) The rate constants for radical additions to alkenes are highly sensitive to
steric and polar effects of both the radical and the acceptor. For an excellent
discussion including rate data, see: Free Radicals in Organic Chemistry;
Fossey, J., Lefort, D., Sorba, J., Eds.; Wiley: Chichester, 1995; Chapter
12.
(3) (a) Sibi, M. P.; Jasperse, C. P.; Ji, J. J. Am. Chem. Soc. 1995, 117, 10779.
(b) Sibi, M. P.; Ji, J. J. Org. Chem. 1996, 61, 6090. (c) Sibi, M. P.; Ji, J.;
Sausker, J. B.; Jasperse, C. P. J. Am. Chem. Soc. 1999, 121, 7517.
(4) (a) Sibi, M. P.; Porter, N. A. Acc. Chem. Res. 1999, 32, 163. (b) Sibi, M.
P.; Ji, J.; Wu, J.-H.; Gu¨rtler, S.; Porter, N. A. J. Am. Chem. Soc. 1996,
118, 9200. (c) Sibi, M. P.; Ji, J. J. Org. Chem. 1997, 62, 3800. (d) Sibi, M.
P.; Shay, J. J.; Ji, J. Tetrahedron Lett. 1997, 38, 5955. (e) Also see:
Murakata, M.; Tsutsui, H.; Hoshino, O. Org. Lett. 2001, 3, 299. (f) Iserloh,
U.; Curran, D. P.; Kanemasa, S. Tetrahedron: Asymmetry 1999, 10, 2417.
(5) Sibi, M. P.; Chen, J. J. Am. Chem. Soc. 2001, 123, 9472.
In this paper we present our observations on diastereoselec-
tivity in the intermolecular additions of prochiral alkyl, R-ha-
(6) For discussion on stereoselective radical additions including intramolecular
cyclizations, see: Curran, D. P.; Porter, N. A.; Giese, B. Stereochemistry
of Radical Reactions; VCH: Weinheim, 1995.
9
2924 VOL. 124, NO. 12, 2002 J. AM. CHEM. SOC.
10.1021/ja017510t CCC: $22.00 © 2002 American Chemical Society

Prochiral Radical Addition to Prochiral Olefins
A R T I C L E S
Table 1. Effect of Radical Precursor on Diastereoselectivity
loalkyl, and R-alkoxyalkyl radicals to prochiral enoate acceptors.
Our study finds that R-haloalkyl radicals add with remarkably
high selectivity and R-alkoxyalkyl radicals with moderate to
good selectivity. In contrast, R-alkyl radicals add with low
diastereoselectivity, even at low temperature. Issues of steric
bulk, electronics, and Lewis acid additives will also be ad-
dressed.
Results
Our work began with the examination of simple diastereo-
selectivity in the addition of prochiral radicals to oxazolidinone
(7) (a) For examples and discussions on ionic reactions involving prochiral
reagents and prochiral substrates, see: Heathcock, C. H. In Asymmetric
Syntheses; Morrison, J. D., Ed.; Academic: New York, 1984; Vol. 3, Part
B, Chapter 2. (b) Perlmutter, P. Conjugate Addition Reactions in Organic
Synthesis; Pergamon: Oxford, 1992. (c) For some seminal work, see: Oare,
D. A.; Henderson, M. A.; Sanner, M. A.; Heathcock, C. H. J. Org. Chem.
1990, 55, 132. (d) Oare, D. A.; Heathcock, C. H. J. Org. Chem. 1990, 55,
157. (e) For conjugate additions of R-alkoxy organometallics to conjugated
systems, see: Chong, J. M.; Mar, E. K. Tetrahedron Lett. 1990, 31, 1981.
(f) Linderman, R. L.; McKenzie, J. R. Tetrahedron Lett. 1988, 29, 3911.
Selected examples of prochiral nucleophile addition to prochiral acceptors
under ionic conditions: (g) Lim, S. H.; Curtis, M. D.; Beak, P. Org. Lett.
2001, 3, 711. (h) Nishwaki, N.; Knudsen, K. R.; Gothelf, K. V.; Jorgensen,
K. A. Angew. Chem., Int. Ed. 2001, 40, 2992. (i) Juhl, K.; Gathergood, N.;
Jorgensen, K. A. Angew. Chem., Int. Ed. 2001, 40, 2995. (j) Liang, B.;
Carroll, P. J.; Joullie, M. M. Org. Lett. 2000, 2, 4157. Selected examples
of prochiral nucleophile addition to prochiral acceptors under neutral
conditions: (k) Evans, D. A.; Rovis, T.; Kozlowski, M. C.; Downey, C.
W.; Tedrow, J. S. J. Am. Chem. Soc. 2000, 122, 9134. (l) Evans, D. A.;
Willis, M. C.; Johnston, J. N. Org. Lett. 1999, 1, 865. (m) Evans, D. A.;
Scheidt, K. A.; Johnston, J. S.; Willis, M. C. J. Am. Chem. Soc. 2001, 123,
4480. (n) Kitajima, H.; Ito, K.; Katsuki, T. Tetrahedron 1997, 53, 17015.
(o) Johnson, J. S.; Evans, D. A. Acc. Chem. Res. 2000, 33, 325. (p) Bernardi,
A.; Colombo, G.; Scolastico, C. Tetrahedron Lett. 1996, 37, 8921. (q) For
prochiral radical addition to an aldehyde, see: Ohno, T.; Ishino, Y.;
Tsumagari, Y.; Nishiguchi, I. J. Org. Chem. 1995, 60, 458.
(8) For nitroxyl radical reactions, see: (a) Braslau, R.; Naik, N.; Zipse, H. J.
Am. Chem. Soc. 2000, 122, 8421. (b) Radical addition to substituted
stannanes: Damm, W.; Hoffmann, U.; Macko, L.; Neuberger, M.; Zehnder,
M.; Giese, B. Tetrahedron 1994, 50, 7029. (c) Hamon, D. P. G.; Massy-
Westropp, R. A.; Razzino, P. Tetrahedron 1995, 51, 4183. (d) Fliri, H.;
Mak, C.-P. J. Org. Chem. 1985, 50, 3438. (e) Easton, C. J.; Scharfbling,
I. M. J. Org. Chem. 1990, 55, 384. (f) Addition of an arenechromiumtri-
carbonyl prochiral radical to methyl crotonate: Merlic, C. A.; Xu, D. J.
Am. Chem. Soc. 1991, 113, 9855. (g) Merlic, C. A.; Walsh, J. C. J. Org.
Chem. 2001, 66, 2265. (h) Addition of prochiral radicals to prochiral carbene
complexes: Merlic, C. A.; Xu, D.; Nguyen, M. C. Tetrahedron Lett. 1993,
34, 227. (i) Addition of prochiral radicals to furanones: Bertrand, S.;
Hoffman, N.; Pete, J.-P. Eur. J. Org. Chem. 2000, 2227. (j) Bertrand, S.;
Glapski, C.; Hoffmann, N.; Pete, J.-P. Tetrahedron Lett. 1999, 40, 3169.
(k) Marinkovi, S.; Hoffmann, N. Chem. Commun. 2001, 1576. For other
examples of intermolecular prochiral radical addition to acceptors, see: (l)
Mikami, K.; Yamaoka, M. Tetrahedron Lett. 1998, 39, 4501. (m) Ahn, J.
H.; Lee, D. W.; Juong, M. J.; Lee, K. H.; Yoon, N. M. Synlett 1996, 1224.
(n) Mero, C. L.; Porter, N. A. J. Am. Chem. Soc. 1999, 121, 5155. For
some selected examples of intramolecular radical reactions, see: (o) Sasaki,
M.; Inoue, M.; Noguchi, T.; Takeichi, A.; Tachibana, K. Tetrahedron Lett.
1998, 39, 2783. (p) Andres, C.; Duque-Soladana, J. P.; Pedrosa, R. J. Org.
Chem. 1999, 64, 4282. (q) White, J. D.; Shin, H. Tetrahedron Lett. 1997,
38, 1141. (r) Lee, E. In Radicals in Organic Synthesis; Renaud, P., Sibi,
M. P., Eds.; Wiley-VCH: Weinheim, 2001; Vol. 2, Chapter 4.2. (s) Hart,
D. J. In Radicals in Organic Synthesis; Renaud, P., Sibi, M. P., Eds.; Wiley-
VCH: Weinheim, 2001; Vol. 2, Chapter 4.1. (t) For an example on
intermolecular radical coupling using chiral Lewis acids, see: Nguyen, P.
Q.; Scha¨fer, H. J. Org. Lett. 2001, 3, 2993.
entry
radical precursor
product
yielda (%)
ratio (syn:anti)
1
2
3
4
5
6
7
6a
6bd
6c
6d
6e
6f
5a
5b
5c
5d
5e
5f
80
40
83
93
88
90b
70c
1.8:1
1.3:1
2.0:1
2.0:1
1.2:1
1:15
6g
5g
1:17
a Reactions were run for 4 h using Yb(OTf)3, Bu3SnH, and Et3B/O2 in
2:1 CH2Cl2:THF unless otherwise noted. Yields are for the purified product.
b Five initiation cycles over 24 h, 10% ethyl addition. c Five initiation cycles
over 24 h, 30% ethyl addition. d Reaction using a radical precursor (R1 )
Me, R2 ) t-Bu, X ) I) gave no desired product.
crotonate (Table 1, eq 2). The radical additions were conducted
according to our standard reported procedure,3c involving (1)
tin hydride as reducing agent/chain carrier, (2) Et3B/O2 as low-
temperature radical initiator, and (3) Yb(OTf)3 as an activating
Lewis acid.11,12 We chose Yb(OTf)3 for reasons of convenience,
because its stability to water obviated the need for drybox
techniques or careful exclusion of air. In addition, Yb(OTf)3 is
sufficiently mild to be compatible with the other reactants and
can be used in substoichiometric quantities. THF serves to
dissolve the Yb(OTf)3 (as well as other lanthanide Lewis acids).
The addition of simple alkyl radicals (Table 1, entries 1-4)
proceeded with modest levels of diastereoselectivity. A slight
preference for syn addition was observed, but even at -78 °C
maximum selectivities of only 2:1 were observed.13 Results
indicated that increasing the bulk of the R2 alkyl group had
little influence on the diastereoselectivity of the addition. Simple
alkyl radicals are relatively nucleophilic, and their additions are
uncomplicated by possible electronic effects or chelation. The
highly reactive and highly nucleophilic methoxy-substituted
radical derived from 6e also showed low selectivity (entry 5)
(see eq 5 for synthesis of 6e, vide infra).
On the other hand, halogenated radicals derived from 6f and
6g added with remarkably high levels of diastereoselectivity
(15:1 and 17:1; entries 6, 7). The major diastereomer in each
case was determined to be anti by lactonization of the product
halides. Due to the electrophilic nature of these radicals,
however, several technical problems are encountered in these
reactions. Radicals generated from 6f and 6g are considerably
less reactive than the simple alkyl radicals, which necessitated
(9) (a) For an excellent recent review, see: Molander, G. A. In Radicals in
Organic Synthesis; Renaud, P., Sibi, M. P., Eds.; Wiley-VCH: Weinheim,
2001; Vol. 1, Chapter 2.1. For other reviews, see: (b) Molander, G. A.;
Harris, C. I. Chem. ReV. 1996, 96, 307. (c) Gansa¨uer, A.; Bluhm, H. Chem.
ReV. 2000, 100, 2771. For some selected examples, see: (d) Otsubo, K.;
Inanaga, J.; Yamaguchi, M. Tetrahedron Lett. 1986, 27, 5763. (e) Molander,
G. A.; Harris, C. I. J. Org. Chem. 1997, 62, 7418. (f) Taniguchi, N.;
Uemura, M. Tetrahedron Lett. 1997, 38, 7199. (g) Fukuzawa, S.; Nakanishi,
A.; Fujinami, T.; Sakai, S. J. Chem. Soc., Chem. Commun. 1986, 624. (h)
Fukuzawa, S.; Nakanishi, A.; Fujinami, T.; Sakai, S. J. Chem. Soc., Perkin
Trans. 1 1988, 1669. (i) Inanaga, J.; Ujikawa, O.; Handa, Y.; Otsubo, K.;
Yamaguchi, M. J. Alloys Compd. 1993, 192, 197. (j) Kawatsura, M.;
Matsuda, F.; Shirahama, H. J. Org. Chem. 1994, 59, 6900. (k) Enholm, E.
J.; Trivellas, A. Tetrahedron Lett. 1994, 35, 1627. For reactions with stannyl
ketyls, see: (l) Enholm, E. J.; Kinter, K. S. J. Org. Chem. 1995, 60, 4850.
(10) (a) Fukuzawa, S.-i.; Seki, K.; Tatsuzawa, M.; Mutoh, K. J. Am. Chem.
Soc. 1997, 119, 1482. (b) Matsuda, F.; Kawatsura, M.; Dekura, F.;
Shirahama, H. J. Chem. Soc., Perkin Trans. 1 1999, 2371. (c) Kawatsura,
M.; Dekura, F.; Shirahama, H.; Matsuda, F. Synlett 1996, 373. (d) Reference
8l.
(11) For an excellent review on Lewis acid-mediated radical reactions, see:
Renaud, P.; Gerster, M. Angew. Chem., Int. Ed. 1998, 37, 2562.
(12) Each of these components was important to the success of the additions,
as demonstrated by reactions using crotonate 4 and bromide 16c to give
18c. In the absence of either Et3B or tributyltin hydride, no product 18c
formed. These experiments confirm that a radical mechanism is operative.
(13) The relative stereochemistry was determined by an independent synthesis
of syn-3,4-dimethylhexanoic acid and comparison with hydrolyzed 5a. See
Supporting Information for details.
9
J. AM. CHEM. SOC. VOL. 124, NO. 12, 2002 2925

A R T I C L E S
Sibi et al.
Table 3. Effect of Lewis Acid Additives on Diastereoselectivity
Table 2. Effect of Olefin Substituent on Diastereoselectivity
c
entry
Lewis acid
equiv
solvent
CH2Cl2
2:1 CH2Cl2:THF
2:1 CH2Cl2:THF
CH2Cl2
2:1 CH2Cl2:THF
2:1 CH2Cl2:THF
CH2Cl2
yielda,b (%)
ratio (anti:syn)
1
2
3
4
5
6
7
none
<10
95
90
<10
95
1:1
4:1
8:1
1:1
7:1
5:1
1:1
Yb(OTf)3
Sm(OTf)3
Sm(OTf)3
Sm(OTf)3
Sm(OTf)3
MgBr2
1.0
1.0
1.0
0.3
2.0
2.0
90
43
a Reactions were run for 4 h using Lewis acid, Bu3SnH, and Et3B/O2.
b Isolated yields for column purified materials. c Product ratios were
1
determined by H NMR integration (400 MHz).
entry
substrate
radical precursor
product
yielda,f (%)
ratiog (syn:anti)
fumarate substrates 9 and 10 are orders of magnitude more
reactive than their crotonate or cinnamate analogues).15 Bromo-
ethyl radical additions to the fumarates 9 and 10 proceeded
cleanly and efficiently, giving products uncontaminated by
starting material or ethyl addition. However, as observed with
the sec-butyl radical, the more reactive fumarate acceptors gave
much reduced diastereoselectivity (compare entry 5 with entries
8 and 9).
Past results on optimizing chloromethyl radical addition to
fumarate substrates indicated that Sm(OTf)3 was the optimal
Lewis acid for halogenated radical additions to these types of
substrates.16 With this information in hand, a brief study of the
effect of the Lewis acid on the diastereoselectivity of bromoethyl
radical addition to fumarates was undertaken (Table 3, eq 4). It
was again found that Sm(OTf)3 gave the highest diastereo-
selectivities (8:1) when using substrate 10, doubling the result
previously obtained using the stronger Lewis acid Yb(OTf)3
(Table 3, entries 2 and 3). The effect of increasing temperatures
on the double diastereoselective additions was also studied. As
expected, increasing reaction temperatures from -78 to 0 °C
or room temperature for the reaction shown in eq 4 with
Sm(OTf)3 as a Lewis acid resulted in complete erosion of
selectivity from 8:1 to 1:1.
1
2
3
4
5
6
7
8
9
4
7
8
10
4
7
8
9
10
6a
6a
6a
6a
6g
6g
6g
6g
6g
5a
11a
12a
14a
5g
11g
12g
13g
14g
80
80
1.8:1
2:1
2:1
1:1
1:17
1:2
23 (71)e
85
70b
20c,d
<5 (95)e
95
1:2
1:4
95
a Reactions were run for 4 h using Yb(OTf)3, Bu3SnH and Et3B/O2 in
2:1 CH2Cl2:THF unless otherwise noted. b 30% ethyl addition, separation
problems. c 50% ethyl addition. d Isolated product as lactone. e Recovered
starting material. f Isolated yields for column purified materials. g Product
1
ratios were determined by H NMR integration (400 MHz).
multiple initiation cycles and large excesses of reagents in order
to get high yields. Also, the product bromide in 5g is prone to
radical reduction by excess tributyltin hydride and triethylborane,
resulting in substantial amounts of a conjugate ethyl addition
(R1 ) H) a side product. Finally, the presence of trace amounts
of unreacted starting material and ethyl addition side product
(R1 ) H) made chromatographic purification of halogenated
products 5f and 5g difficult.
The influence of the radical acceptor substituent R on
diastereoselectivity is shown in Table 2 (eq 3). Increasing the
size of R (R ) Ph, i-Pr; entries 2, 3) did not significantly affect
the selectivity for the addition of the sec-butyl radical.14 It is
noteworthy that the larger substituents also decreased the
reactivity of the acceptor toward radical additions (entries 3 and
7). In addition, increasing the reactivity of the acceptor alkene
by using the fumarate-derived substrates correspondingly re-
duced the selectivity (entry 4; compare entry 4 with entry 1
and entry 5 with entries 8 and 9).
Table 4 shows a series of reactions in which prochiral
methoxyalkyl radicals add to nonterminal alkenes.17 We have
found that methoxyalkyl bromides 16 are readily prepared from
acetals via reaction with acetyl bromide (eq 5),18 so it was
relatively easy to systematically vary the size of the R2 group.
Since the bromoethyl radical had shown higher selectivity
than the sec-butyl radical in Table 1, we also screened its ability
to add to alkenes 7-10. It was observed that the substrate
reactivity followed the order R ) CO2Et, CO2-t-Bu > Me >
Ph > i-Pr (in previous work we have likewise found that the
The R-bromoalkyl ethers 16 were conveniently prepared and
used in situ (eq 6), since their formation was quantitative and
the R-bromoalkyl ethers 16 were sensitive to workup. The
(14) We have attempted to add the radical derived form 3,3-dimethyl-2-
iodobutane, but alkylation occurred in <20% yield, presumably due to steric
hindrance, and we were unable to determine the selectivity.
(15) Sibi, M. P.; Ji, J. Angew. Chem., Int. Ed. Engl. 1997, 36, 274.
(16) (a) Reference 15. (b) Sibi, M. P.; Liu, P.; Ji, J.; Chen, J. Hajra, S. J. Org.
Chem., in press.
9
2926 J. AM. CHEM. SOC. VOL. 124, NO. 12, 2002

Prochiral Radical Addition to Prochiral Olefins
A R T I C L E S
Table 4. Steric Effect of R-Alkoxy Radicals on
Table 5. Effect of Lewis Acids on Yield and Diastereoselectivity
for R-Alkoxy Radical Additionsa
Diastereoselectivitya
c
entry
substrate
radical
product
ratio (syn:anti)
yieldb (%)
c
entry
Lewis acid
diastereoselectivityb (syn:anti)
yield (%)
1
2
3
4
5
6
4
4
4
7
7
16ad
16b
16c
16a
16c
16a
18ad
18b
18c
19a
19c
20a
1.2:1
2.3:1
6:1
1.5:1
6:1
88
90
89
94
90
35
1
2
3
4
5
6
7
8
9
10
11
12
13
none
Yb(OTf)3
Er(OTf)3
Gd(OTf)3
Sm(OTf)3
Pr(OTf)3
1:1
6:1
4:1
2:1
4:1
10d
89d
95
85d
90
17
5.5:1
1:1
2:1
45d
90
90
Y(OTf)3
a The general procedure described was followed. Preparative reactions
were conducted using 5 equiv of freshly prepared bromoalkyl ether, 5 equiv
of Bu3SnH, 0.1 equiv of Yb(OTf)3, and 2 equiv of Et3B. b Isolated yields
for column purified materials. c Product ratios were determined by 1H NMR
integration (400 MHz). d Table 4, entry 1, corresponds to Table 1, entry 5.
Note that 16a ) 6e and 18a ) 5e.
Sc(OTf)3
La(OTf)3
Yb(OTf)3 (0.1 equiv)
MgBr2(OEt2)
ZrCl4
1.5:1
1:1.5
6:1
1:1.2
1:3
30
89d
90e
85e
<10
TiCl4
a The general procedure described was followed. Preparative reactions
were conducted using 5 equiv of freshly prepared bromoalkyl ether, 5 equiv
of Bu3SnH, 1 equiv of Lewis acid (unless otherwise noted), and 2 equiv of
Et3B. b Diastereomer ratios were determined by 1H NMR integration (400
MHz). c Chemical yields were determined by 1H NMR integration (400
MHz, pentachloroethane as an internal standard) and GLC. d Isolated yields.
e No THF cosolvent.
methyl acetate side product was a harmless bystander in the
radical reactions. A control reaction also showed that R-bromo-
alkyl ether 16c (R2 ) t-Bu) is stable to Yb(OTf)3 in solution
for over 10 h at room temperature, confirming that cations are
not involved in the addition reaction under our low-temperature
reaction conditions. Alkoxyalkyl radicals derived from 16
showed high reactivity19 toward conjugate addition because of
the methoxy substituent, so yields in the addition reaction were
generally high (Table 4, entries 1-5), except when the R group
on the alkene became very large (entry 6).
In the addition of alkoxyalkyl radicals, the syn isomer was
consistently preferred over anti (Table 4).20 Increasing the size
of the R2 group on the radical increased the diastereoselectivity
in additions to both crotonate 4 (entries 1-3) and cinnamate 7
(entries 4 and 5). Increasing the size of the R group on the alkene
also increased the selectivity (compare entries 1, 4, and 6). Good
selectivity was observed only with a very large substituent on
either the radical (entries 3, 5) or on the alkene (entry 6).
Table 5 shows results in which the addition of the methoxy-
alkyl radical derived from 16c to crotonate 4 was conducted
using several different Lewis acids for activation of the acceptor.
The diastereoselectivity is surprisingly sensitive to the Lewis
acid (Table 5). We do not have a full explanation for the
complex relationship between selectivity and Lewis acid
structure, but some of the results are interesting. While simple
Yb(OTf)3 gave 6:1 selectivity (entry 2), other Lewis acids tested
showed lower syn selectivity or even anti selectivity. Among
the metal triflates tested, it is interesting that Yb(OTf)3 is
intermediate in both Lewis acid strength and ionic radius (entries
2-9), so there is no simple correlation between diastereo-
selectivity and either Lewis acid strength or size. Magnesium
bromide and zirconium tetrachloride actually give anti selectiv-
ity, although the selectivity is low (entries 11, 12). Ytterbium
triflate could be used catalytically (entry 10). The very strong
Lewis acid titanium tetrachloride provided only trace amounts
of product, due to incompatibility with the reactants (entry 13).
Results from a series of reactions in which Lewis basic
additives were added to Yb(OTf)3 for the formation of product
18c are presented in Table 6. Lewis basic additives have been
found to effect a number of lanthanide-mediated reactions,21
and the formation of 18c reflects a sensitive case study for the
(17) The addition of R-alkoxy radicals to acrylates has been reported. Intermo-
lecular addition: (a) Nishiyama, Y.; Yamamoto, H.; Nakata, S.; Ishii, Y.
Chem. Lett. 1993, 841. (b) Giese, B.; Hoch, M.; Lamberth, C.; Schmidt,
R. R. Tetrahedron Lett. 1988, 29, 1375. (c) Bimwala, R. M.; Vogel, P. J.
Org. Chem. 1992, 57, 2076. (d) Kessler, V. H.; Wittmann, V.; Kock, M.;
Kottenhahn, M. Angew. Chem., Int. Ed. Engl. 1992, 32, 902. (e) Garner,
P. P.; Cox, P. B.; Klippenstein, S, J. J. Am. Chem. Soc. 1995, 117, 4183.
(f) Garner, P.; Leslie, R.; Anderson, J. T. J. Org. Chem. 1996, 61, 6754.
(g) RajanBabu, T. V. Acc. Chem. Res. 1991, 24, 139. (h) RajanBabu, T.
V.; Nugent, W. A. J. Am. Chem. Soc. 1994, 116, 986.
(18) For conversion of a dimethyl acetal to an R-chloroacetal, see: Lokensgard,
J. P.; Fischer, J. W.; Batrz, W. J. J. Org. Chem. 1985, 50, 5609.
(19) Alkoxyalkyl radicals are “nucleophilic”. See: Giese, B. Radicals in Organic
Synthesis: Formation of Carbon-Carbon Bonds; Pergamon: Oxford, 1986;
Chapter 2. Giese, B.; Dupuis, J.; Hasskerl, T.; Meixner, J. Tetrahedron
Lett. 1983, 24, 703.
(20) The relative stereochemistry was established by conversion of products 5f
and 18c to known lactones (TMSI, CDCl3). Denmark, S. E.; Forbes, D. C.
Tetrahedron Lett. 1992, 33, 5037. We thank Prof. Scott Denmark for
providing us spectral data for the lactone derived from 18c. In 4,5-
disubstituted lactones, the chemical shift of the C5-H is consistently upfield
in the trans compounds relative to the cis, and this relationship was used
to assign the lactones derived from 18c. See: Fang, J.-M.; Liao, L.-F.;
Hong, B.-C. J. Org. Chem. 1986, 51, 2828. Ueno, Y.; Moriya, O.; Chino,
K.; Watanabe, M.; Okawara, M. J. Chem. Soc., Perkin Trans. 1 1986, 1351.
Carretero, J. C.; Rojo, J. Tetrahedron Lett. 1992, 33, 7407 and references
therein.
(21) For the use of additives in rare earth Lewis acid-mediated reactions, see:
(a) Shibasaki, M.; Vogl, E. M.; Groger, H. Angew. Chem., Int. Ed. 1999,
38, 1570. (b) Saito, T.; Kawamura, M.; Nishimura, J.-i. Tetrahedron Lett.
1997, 38, 3231. (c) Mikami, K.; Kotera, O.; Motoyama, Y.; Sakaguchi, H.
Synlett 1995, 975. (d) Kobayashi, S.; Hachiya, I. J. Org. Chem. 1994, 59,
3590. (e) Kobayashi, S.; Araki, M.; Hachiya, I. J. Org. Chem. 1994, 59,
3758. (f) Kobayashi, S.; Ishitani, H. J. Am. Chem. Soc. 1994, 116, 4083.
(g) For a review on lanthanide coordination with macrocyclic ligands, see:
Alexander, V. Chem. ReV. 1995, 95, 273. Modification of lanthanide
reactivity and or structure by addition of ligands: (h) Aspinall, H. C.;
Dwyer, J. L. M.; Greeves, N.; McIver, E. G.; Wooley, J. C. Organometallics
9
J. AM. CHEM. SOC. VOL. 124, NO. 12, 2002 2927

A R T I C L E S
Sibi et al.
Table 6. Effect of Additives to Yb(OTf)3 for R-Alkoxy Radical
selectivity at the â-carbon is completely controlled (g25:1) in
the conjugate addition.24 These reactions also demonstrate that
absolute as well as relative stereochemistry can be controlled
in the addition of prochiral radicals to chiral olefins.
Additionsa
entry
additive (equiv)b
Yb(OTf)3 (equiv) diastereoselectivityc,d (syn:anti)
1
2
3
4
5
6
7
8
9
none
NEt3 (3)
DMSO (3)
HMPA (3)
H2NCH2CH2OH (1)
HOCH2CH2OH (1, 2 or 3)
HO(CH2CH2O)2H (1)
HO(CH2CH2O)3H (1)
HO(CH2CH2O)4H (1)
1
1
1
1
1
1
1
1
1
1
1
1
1
0.1
6:1
5:1
8:1
10:1
6:1
10:1
8:1
10:1
14:1
11:1
9:1
9:1
6:1
Discussion
Prochiral alkyl radicals add to electron-deficient alkenes with
low levels of diastereoselectivity (Tables 1 and 2). It appears
as though steric factors alone are unable to afford substantial
diastereoselectivity in the additions of these nucleophilic alkyl
radicals to electrophilic alkenes. Radical additions are normally
understood to proceed via early transition states, so it is not
surprising that early bond formation makes these reactions
relatively insensitive to steric factors. The preferred syn products
likely arise predominantly via the transition state 25 shown in
Figure 1, where gauche interactions are minimized compared
to 23 or 24, also leading to syn products. Ordinarily the largest
substituents on a forming bond orient themselves anti to each
other, just as they do relative to ethane, so if R acts as the largest
substituent on the alkene carbon (as opposed to dCHCOX),
25 represents the situation where the “large” groups R and R2
are anti to each other. Transition state 25 also explains why
increasing the size of R decreases the reactivity of the acceptor
but does not provide substantially increased levels of selectivity,
due to unfavorable interactions with the methyl group on the
radical. A similar argument can be used to explain the decreasing
reactivity when bulky R2 groups are introduced. The minor anti
product may arise from a similar transition state 26 where the
methyl and the R2 group are interchanged, and which experi-
ences similar gauche interactions. The early transition state
combined with the modest effective difference in size between
R and the dCH(CO)X groups on the acceptor carbon may limit
the energy discrimination between 25 and 26.
In the case of methoxyalkyl radical additions (see Table 4),
we had originally thought that chelation of the methoxy group
to the Lewis acid might result in high selectivity (transition states
27 or 30, Figure 2). However, our results with ytterbium triflate
as Lewis acid show that the major diastereomer actually results
from an open transition state in which the methoxy substituent
on the radical is not complexed to substrate-bound Yb(OTf)3.
Although the substrate itself is chelated to the Yb(OTf)3, the
incoming methoxyalkyl radical is not. We believe the major
diastereomer forms via an open transition state for the following
reasons. First is the impact of additives on the reaction 4 +
10 HO(CH2CH2O)5H (1)
11 12-C-4 (1)
12 15-C-5 (1)
13 18-C-6 (1)
14 HOCH2CH2OH (0.2)
10:1
a The general procedure described was followed. Reactions were
conducted using 5 equiv of freshly prepared bromoalkyl ether, 5 equiv of
Bu3SnH, 1 equiv of Lewis acid (unless otherwise noted), the additive (see
table), and 2 equiv of Et3B. b Number of equivalents relative to substrate.
c Yields were generally >90% as determined by isolation or GLC or 1H
NMR. d Diastereomer ratios were determined by 1H NMR integration (400
MHz).
effect of additives.22 In all cases the chemical yields exceeded
90%. Addition of HMPA or ethylene glycol increased the syn
selectivity for 18c from 6:1 to 10:1 (entries 4, 5). With
tetraethylene glycol the selectivity increased to 14:1 (entry 9).
Our observations suggest that oxygenated Lewis bases influence
the selectivity but that amines have minimal influence. This is
in keeping with the known oxophilicity of lanthanides. However,
we are uncertain exactly how these additives impact the
stereoselectivity. A modest accelerating effect was also observed
in reactions using Lewis basic additives. This may suggest that
a later transition state may not be the sole contributor for
enhanced selectivity. It is possible that coordinating additives
suppress a minor competing chelation pathway. While we do
not have a clear explanation for these results, they may provide
some clues regarding the interaction of ytterbium triflate with
additives that may provide insight toward the development of
improved rare earth Lewis acid catalysts.23
The addition of the radical derived from 16c to the chiral
substrate 21 (eq 9) was also examined. We have previously
shown that isopropyl radical addition to 21 occurs with ∼25:1
diastereoselectivity.3a Radical addition of 16c to 21 gave 22 as
a 14:1 mixture of syn/anti compounds, with no trace of the other
two diastereomers. These experiments show that the facial
1998, 17, 1884. (i) Aspinall, H. C.; Greeves, N.; McIver, E. G. Tetrahedron
Lett. 1998, 39, 9283. (j) Greeves, N.; Aspinall, H. C.; Browning, A. F.;
Ravenscroft, P. Tetrahedron Lett. 1994, 35, 4639. (k) Lacote, E.; Renaud,
P. Angew. Chem., Int. Ed. 1998, 37, 2259.
(22) In contrast, addition of Lewis basic additives for the formation of 5a did
not lead to improvement in selectivity.
(23) NMR experiments involving Y(OTf)3 in acetonitrile have shown that
ethylene glycol binds preferentially over substrate 4 and indicate that yttrium
can accommodate two or three ethylene glycols.
(24) The absolute stereochemistry at the â-carbon is based on analogy from
our previous work involving conjugate radi cal additions to chiral substrate
21 (ref 3a).
9
2928 J. AM. CHEM. SOC. VOL. 124, NO. 12, 2002

Prochiral Radical Addition to Prochiral Olefins
A R T I C L E S
Figure 1.
Figure 2.
Figure 3.
16c f 18c (see Table 6). Additives such as HMPA and ethylene
glycol are strongly coordinating bases with high affinities for
lanthanides. If syn product resulted from a chelated transition
state, we expected that these additives should disrupt chelation
and reduce the syn/anti selectivity, but exactly the opposite
occurred. Second, a chelated transition state leading to syn
products must proceed via transition state 27. This chelated
transition state appears severely hindered because the R2 group
suffers two gauche interactions, and this should intensify as the
sizes of R and R2 increase. However, increasing the size of either
R or R2 gave enhanced syn selectivity.25 Third, complexation
of a methoxyalkyl radical to a Lewis acid should reduce its
nucleophilicity, so it is not surprising that alkoxyalkyl radicals
that are not bound to a Lewis acid are the reactive species.26 It
is possible that the anti product, which is modestly preferred
when zirconium tetrachloride or magnesium bromide was used
as Lewis acid, does result from a transition state 30, in which
the methoxy group is chelated to the Lewis acid.
Given an open transition state for the usual addition of
methoxyalkyl radicals, transition state 29, where gauche inter-
actions are minimized and in which the large groups R and R2
are anti to each other, again seems best, especially when R *
methyl. In terms of steric volume, the methoxy group is smaller
than even a methyl group, and obviously much smaller than
tert-butyl.27 That transition state 28 is not the source of the syn
products is evident from the dependence of diastereoselectivity
on R. As R gets larger, 28 should be increasingly destabilized,
but experimentally enlarging R increased rather than decreased
the syn selectivity (Table 4, compare entries 1, 4 and 6). That
transition state 29 is operative is also consistent with the
observation that syn product is preferred in all cases, and that
selectivity increased when the size of either R or R2 increased
(Table 4, entries 1-3, 4 and 6). The preference for transition
state 29 may also have a stereoelectronic component. In 29,
the oxygen lies anti relative to the alkene. Analogous anti
relationships of oxygen substituents to alkenes in radical
additions have been noted previously and attributed to stereo-
electronic effects, and effects of this type are probably key to
the high selectivity associated with addition of samarium ketyl
radicals to alkenes.28 The stereoelectronic effects are probably
much smaller in the case of methoxyalkyl radicals than in
samarium alkoxy radicals (samarium ketyls) because the oxygen
has greater negative charge density in the latter. The R2 group
and the carbonyl carbon have a pseudo syn pentane relationship
in 29, but because of the early transition state and the trigonal
(planar) geometry of the R-carbon, we suggest that the inside
position is not significantly encumbered.29
Haloalkyl radicals add with remarkably high anti selectivity,
especially in additions to crotonate 4. Of the three projections
31-33 that would lead to the anti product, transition state 31
seems the best, since 32 and 33 would likely experience more
allylic strain and more severe gauche interactions (Figure 3).
In contrast to alkyl and alkoxyalkyl radicals, haloalkyl radicals
are relatively electrophilic and much less reactive, so that the
transition state is probably relatively late. As is to be expected,
reactions involving a later transition state are much more
sensitive to both steric and electronic factors.
What is surprising is that the haloalkyl radicals give anti
selectivity, whereas alkyl and methoxyalkyl radicals give syn
(25) On the other hand, the R2 group has a syn pentane-type arrangement relative
to the carbonyl group in transition state 29. While R2 would also suffer
syn pentane relationships with R when R * Me (Table 1, entries 4, 5), it
is possible that when R ) Me the syn product arises from transition state
28, and when R * Me the syn product arises from transition state 29. While
it is possible that 18c forms via 28, the cinnamate 7 also underwent syn-
selective addition by 16c (6:1), which should proceed via 28.
(26) The involvement of an open transition state is not all that surprising. Even
in the absence of additives, the methoxy group is less basic than the
cosolvent THF, especially since the basicity of the oxygen atom is reduced
by electron donation to the radical. Also, if a methoxyalkyl radical was
coordinated to a Lewis acid, the radical should be less nucleophilic and
less reactive toward alkene addition.
(27) The A values for -Cl, -OCH3, -CHdCH2, -CH3, -CH(CH3)2, -C6H5,
and -C(CH3)3 are 0.5, 0.6, 1.7, 1.8, 2.1, 2.9, and >4.5, respectively.
Stereochemistry of Organic Compounds; Eliel, E., Wilen, S., Eds.; Wiley:
New York, 1993.
(28) Beckwith, A. L. J. Tetrahedron 1981, 37, 3073.
(29) An alternate explanation is that due to incipient syn pentane interactions,
the inside position is the most hindered, and that syn product is formed via
28. When R2 ) t-Bu, however, an even more severe syn pentane interaction
occurs with R.
9
J. AM. CHEM. SOC. VOL. 124, NO. 12, 2002 2929

A R T I C L E S
Sibi et al.
selectivity. The A values for -Cl and -OCH3 are very similar,
so if steric factors are the only consideration, haloalkyl and
methoxyalkyl radicals should both prefer the same isomer,
whether syn or anti. Why do haloalkyl radicals add via 31 rather
than via 34? Clearly, the differing outcomes between haloalkyl
and methoxyalkyl radicals have an electronic basis. The less
reactive haloalkyl radicals add with much later transition states.
Under these conditions, 34 may be destabilized due to develop-
ing syn pentane interactions between the methyl group and the
carbonyl carbon. Another possible explanation is that transition
state 31 benefits from favorable dipole/dipole interactions
(attractive interaction of the halogen with the carbonyl carbon
or the Lewis acid). The same dipolar advantages do not apply
in the reaction of alkoxyalkyl radicals because the much earlier
transition state and the shorter C-O bond length make too great
a spatial distance for significant dipole/dipole interactions.
Further, unlike chlorine and bromine, the methoxy group is a
strong π donor to the radical, so that the C-O bond has
drastically reduced or even reversed dipolar character.
be of limited synthetic value, but the results are of fundamental
interest. The alteration of the environment (reactivity) of the
lanthanide Lewis acids with achiral additives detailed in this
work suggests a potential and simple route for selectivity
enhancements in radical reactions.
Overall, our results show that additions of prochiral radicals
to prochiral alkenes via open transition states are not uniformly
highly diastereoselective. The selectivity is only slightly higher
than observed in anionic nucleophilic addition of R-alkoxy-
lithiums, for example.7e To be highly diastereoselective, it
appears that intermolecular addition of achiral prochiral radicals
to prochiral alkenes will require either strong stereoelectronic
effects30 or chelation control to provide sufficient organization.
Acknowledgment. This work was supported by the National
Institutes of Health (GM-54656). T.R.R. thanks NSF and the
ACS Division of Organic Chemistry for graduate fellowships.
We thank Megan Kirsch, Mei Liu, and Sarah Steffl for technical
assistance.
Supporting Information Available: Characterization data for
compounds and experimental procedures (PDF). This material
is available free of charge via the Internet at http://pubs.acs.org.
Conclusions
In this work we have examined factors that are fundamental
to understanding reactions between two prochiral fragments. We
have shown that alkoxyalkyl and haloalkyl radicals add to
enoates with moderate to good diastereoselectivity, and the
methodology could have synthetic potential. The diastereo-
selectivity in our study using alkyl radicals is modest, and may
JA017510T
(30) Fukuzawa, ref 10a, gets highly syn-selective addition of samarium ketyl
radicals to crotonate esters. Stereoelectronic and/or chelation plays a major
role in these reactions.
9
2930 J. AM. CHEM. SOC. VOL. 124, NO. 12, 2002