Effect of Ligand Structure in the Bisoxazoline Mediated Asymmetric Addition of Methyllithium to Imines

Full text of DOI:10.1021/jo0007175

J . Org. Chem. 2000, 65, 5875-5878
5875
Sch em e 1
Effect of Liga n d Str u ctu r e in th e
Bisoxa zolin e Med ia ted Asym m etr ic
Ad d ition of Meth yllith iu m to Im in es
Scott E. Denmark* and Cory M. Stiff
Roger Adams Laboratory, Department of Chemistry,
University of Illinois, Urbana, Illinois 61801
denmark@scs.uiuc.edu
Received May 11, 2000
In tr od u ction
The increasing importance of enantiomerically en-
riched amines in organic chemistry derives from their
extensive use as auxiliaries,1 resolving agents,2 and inter-
mediates3 in the synthesis of both natural and unnatural
compounds. Moreover, the need for a wider range of
structurally diverse and enantiopure amines for solid-
phase (combinatorial) synthesis of organic compounds
constitutes a continuing challenge. Of the many strate-
gies available for the synthesis of amines, one of the most
attractive is the addition of organometallic reagents to
CdN functionalities.4 In comparison to the many suc-
cesses reported for the enantioselective addition to car-
bonyl compounds,5 alkylations of the corresponding azome-
thine compounds are more limited, but steadily growing.6
Several years ago, we showed that bisoxazolines7 1 and
(-)-sparteine 2, are excellent promoters for the addition
of lithium reagents to aromatic, olefinic, and aliphatic
aldimines (Scheme 1).8 Ligands 1a and 1b promote the
addition of methyl- and vinyllithium to imines 3-6, to
provide the desired amines in high yield (79-97%) and
enantioselectivity (71-91% ee). For highly reactive spe-
cies such as n-BuLi and PhLi, a more basic ligand, 2,
was necessary. With 2 as the promoter, the addition of
n-BuLi and PhLi to imine 6 proceeded smoothly to afford
the desired amines in 90% and 99% yield and 91% and
82% ee, respectively. Additionally, substoichiometric
amounts of the ligands could be used, however with a
attendant decrease in selectivity. Clearly, the structure
and gross architecture of the ligand play a central role
in the activation and stereoselection aspects of this
process. In our original report, the primary variation in
structure of the bisoxazoline was of the substituents at
the C(4) stereogenic center. It was found that the bulkier
groups gave enhanced stereoselectivity. In continuation
of these studies we have undertaken a systematic ex-
amination of the influence of the bridging group by way
of bond angle modulation and steric bulk.
Resu lts a n d Discu ssion
Effect of Liga n d Bite An gle. A correlation between
ligand geometry, specifically chelate bite angle, and
enantioselectivity has been reported by several research
groups.9 Davies et al. found that variation of the ring size
in spirocyclic bisoxazolines i provided different ligand
geometries.9b As the spiro ring size decreased, the angle,
Φ, correspondingly increased. The resulting change in
the geometry influenced the stereochemical course of the
copper-catalyzed, Diels-Alder reaction. We surmised
that a similar structural change in 11 should affect the
coordination angle of lithium, θ, which in turn, should
have a direct effect on the course of the addition of
alkyllithium reagents to aldimines.
(1) (a) Whitesell, J . K. Chem. Rev. 1989, 89, 1581. (b) Seyden-Penne,
J . Chiral Auxiliaries and Ligands in Asymmetric Synthesis; Wiley:
New York, 1995; pp 306-340.
(2) J acques, J .; Collet, A.; Wilen, S. H. Enantiomers, Racemates, and
Resolution; Wiley: New York, 1981.
(3) Moser, H.; Rihs, G.; Santer, H. Z. Naturforsch. 1982, 376, 451.
(4) (a) Volkmann, R. A. In Comprehensive Organic Synthesis,
Additions to C-X π-Bonds, Part 1; Schreiber, S. L., Ed.; Pergamon
Press: Oxford, 1991; Vol. 1, Chapter 1.12. (b) Kleinman, E. F.;
Volkmann, R. A. In Comprehensive Organic Synthesis, Additions to
C-X π-Bonds, Part 2; Heathcock, C. H., Ed.; Pergamon Press: Oxford,
1991; Vol. 2, Chapter 4.3. (c) Klein, J . In The Chemistry of Double-
bonded Functional Groups: Supplement A; Patai, S., Ed.; Wiley:
Chichester, 1989; Vol. 2, Part 1, Chapter 10.
(5) (a) Soai, K.; Niwa, S. Chem. Rev. 1992, 92, 833. (b) Soai, K.;
Shibata, T. In Comprehensive Asymmetric Catalysis; J acobsen, E. N.,
Pfaltz, A., Yamamoto, H., Eds.; Springer-Verlag: Berlin, 1999; Chapter
26.1.
(6) (a) Denmark, S. E.; Nicaise, O. J .-C. J . Chem. Soc., Chem.
Commun. 1996, 999. (b) Enders, D.; Reinhold: U. Tetrahedron:
Asymmetry 1997, 8, 1895. (c) Bloch, R. Chem. Rev. 1998, 98, 1407. (d)
Kobayashi, S.; Ishitani, H. Chem. Rev. 1999, 99, 1069. (e) Denmark,
S. E.; Nicaise, O. J .-C. In Comprehensive Asymmetric Catalysis;
J acobsen, E. N., Pfaltz, A., Yamamoto, H., Eds.; Springer-Verlag:
Berlin, 1999; Chapter 26.2. (f) Merino, P.; Franco, S.; Merchan, F. L.;
Tejero, T. Synlett 2000, 442.
Because of the wide application of bisoxazolines as
bidentate ligands in numerous reactions, several simple
methods for the synthesis of these compounds have been
reported.7,10 Given the basic structural features of the
desired spirocyclic bisoxazolines 11a -d , a route involving
(7) (a) Bolm, C. Angew. Chem., Int. Ed. Engl. 1991, 30, 542. (b)
Pfaltz, A. Acc. Chem. Res. 1993, 26, 339. (c) Reiser, O. Angew. Chem.,
Int. Ed. Engl. 1993, 32, 547. (d) Ghosh, A. K.; Mathivanan, P.;
Cappiello, J . Tetrahedron: Asymmetry 1998, 9, 1.
(8) Denmark, S. E.; Nakajima, N.; Nicaise, O. J .-C. J . Am. Chem.
Soc. 1994, 116, 8797.
(9) (a) Trost, B. M.; Van Vranken, D. L.; Bingel, C. J . Am. Chem.
Soc. 1992, 114, 9327. (b) Davies, I. W.; Gerena, L.; Castonguay, L.;
Senanayake, C. H.; Larsen, R. D.; Verhoeven, T. R.; Reider, P. J . J .
Chem. Soc., Chem. Commun. 1996, 1753. (c) Davies, I. W.; Deeth, R.
J .; Larsen, R. D.; Reider, P. J . Tetrahedron Lett. 1999, 40, 1233.
10.1021/jo0007175 CCC: $19.00 © 2000 American Chemical Society
Published on Web 08/16/2000

5876 J . Org. Chem., Vol. 65, No. 18, 2000
Notes
Ta ble 1. Asym m etr ic Ad d ition of MeLi to Im in esa
a common intermediate was envisioned (Scheme 2). All
of the spirocyclic bisoxazolines in this study were derived
from the unsubstituted parent compound 1411 which is
easily accessible from the condensation of L-tert-leucinol,
12, and diethyl malonimidate 13.12 Treatment of 14 with
lithium diisopropylamide (LDA) at -78 °C followed by
addition of an R,ω-diiodo- or dibromoalkane and warming
to room temperature afforded the desired spirocyclic
bisoxazolines in usable yields (41-72%).
entry
imine
R
ligand
yield,b %
ee,c %
1
2
3
3
3
3
5
5
5
5
6
6
6
6
Ph
Ph
Ph
Ph
11a
11b
11c
11d
11a
11b
11c
11d
11a
11b
11c
11d
82
97
94
87
93
87
92
81
81
82
75
92
51
73
71
70
2
84
90
91
44
90
91
75
3
Sch em e 2
4
5
PhCHdCH
PhCHdCH
PhCHdCH
PhCHdCH
PhCH2CH2
PhCH2CH2
PhCH2CH2
PhCH2CH2
6
7
8
9
10
11
12
a
MeLi (low halide, ether). All reactions performed on a 1.0 mmol
scale. Yield of chromatographically homogeneous material. c De-
b
termined by CSP HPLC analysis.
To better correlate the ligand geometries with the
reaction selectivities, we carried out computational mod-
eling of the ground state of the free ligand and the
corresponding ligand/MeLi complexes. As noted before,
it was predicted that as ring size increased the bridge
angle would decrease. However from the results of both
molecular mechanics (MM2) and semiempirical (PM3)
calculations this trend did not manifest in the results of
the geometry optimizations (Table 2). In particular, the
spiro cyclopentyl compound 11c gave an unexpectedly
large bite angle of 111.8°. For comparison, the calculated
angle and stereoselectivities for acyclic bisoxazoline 1c
are also provided (vide infra).
With the desired spiro-bisoxazolines in hand, the addi-
tion of MeLi to imines 3, 5, and 6 promoted by 11a -d
could be surveyed, and the results are collected in Table
1. In our previous studies, we found that optimal yields
and enantioselectivities were obtained when 2 equiv of
MeLi was combined with 1 equiv of the bisoxazoline in
toluene for 30 min at low temperature prior to addition
of the imine. For this study, the imine was allowed to
react with the preformed complex at -75 °C and the
reaction was then quenched with methanol at that
temperature.
Ta ble 2. Cor r ela tion of En a n tioselectivity a n d Liga n d
Geom etr y (Br id ge An gle)
Benzaldehyde imine 3 was the first substrate to be
examined (Table 1, entries 1-4). The addition product 7
was isolated in high yield (82-97%) from each reaction;13
however, only modest enantioselectivities were seen (51-
73%). Interestingly three of the four ligands provided the
amine with approximately the same ee; only the cyclo-
propyl ligand 11a gave a significantly different result. A
similar trend in enantioselectivity was seen in the
reactions with cinnamyl and hydrocinnamyl imines, 5
and 6. Again the desired amines were isolated in high
yield (75-94%) and varying levels of enantioenrichment
(4-94%) (Table 1, entries 5-12). As was seen in the
reaction with 3, cyclopropyl bisoxazoline 11a afforded 9
and 10 in low ee, 2 and 44% respectively. The other
bisoxazolines all promoted the reaction with a high
degree of selectivity. If the geometric constraints of the
spirocyclic bisoxazolines do produce geometrically unique
complexes, the range of enantioselectivities seen suggests
that the bite angle can have a dramatic effect on the
course of the reaction.
ligand
11a
11b
11c
11d
1c
bridge angle (deg)
112.7
110.0
111.8
106.8
109.7
3
5
6
51
2
44
73
84
90
71
90
91
70
91
75
67
94
93
In all cases examined, the cyclopropyl-based ligand 11a
gave the lowest enantioselectivity (Table 2). This low
selectivity may be due to a steric resistance to the com-
plexation of MeLi. None of the many low energy confor-
mations produced in the geometry minimization with
molecular mechanics had the nitrogens in a coplanar or
nearly coplanar relationship. Coplanarity of the nitrogens
is believed to facilitate the complexation of Li by both
sp2 lone pairs, thereby increasing the rigidity of the sys-
tem and, presumably, the selectivity of the ligand. From
the remaining data it would appear that the optimal
angle around the bridging carbon is between 109° and
111°. It seems that if the bridge angle is too large or small
the enantioselectivity decreases, although the effect is
(10) (a) Denmark, S. E.; Nakajima, N.; Nicaise, O. J .-C.; Faucher,
A.-M.; Edwards, J . P. J . Org. Chem. 1995, 60, 4884. (b) Evans, D. A.;
Peterson, G. S.; J ohnson, J . S.; Barnes, D. M.; Campos, K. R.; Woerpel,
K. A. J . Org. Chem. 1998, 63, 4541.
(11) Mu¨ller, D.; Umbricht, G.; Weber, B.; Pfaltz, A. Helv. Chim. Acta
1991, 74, 232.
(12) (a) McElvain, S. M.; Schroeder, J . P. J . Am. Chem. Soc. 1949,
71, 40. (b) Hall, J .; Lehn, J .-M.; DeCian, A.; Fischer, J . Helv. Chim.
Acta 1991, 74, 1.
(13) In all cases the ligand was recovered after chromatography in
81-99% yield.

Notes
J . Org. Chem., Vol. 65, No. 18, 2000 5877
more pronounced at larger angles. There does not appear
to be a dramatic correlation between enantioselectivity
and any ligand geometry predicted by these calculations.
Indeed it seems that the enantioselectivity of the reaction
is rather more substrate dependent. Moreover, the nature
of the complexation of MeLi is still unclear.
The synthesis of ligand 1h began with the alkylation
of 14 with methallyl chloride (Scheme 4). Simmons-
Smith cyclopropanation14 using either the traditional
zinc-copper couple or Furakawa conditions afforded 17
in 50% yield. Although catalytic hydrogenation of cyclo-
propane rings is well documented,15 the rings in 17
proved unreactive even under extreme pressures of H2
(5000 psi) and with a variety of metal catalysts. Conse-
quently, we decided to employ 17 instead of 1g in the
reaction with imines 3, 5, and 6 and methyllithium.
The preparation of unsymmetrical bisoxazoline 1g
began by combination of tert-butyl malonyl dichloride 18
with tert-leucinol 12 to afford bishydroxyamide 19 in 80%
yield. Activation of the terminal hydroxyl groups with
methanesulfonyl chloride, followed by treatment with
base provided tert-butyl bisoxazoline 20. Not surprisingly,
20 proved impossible to alkylate with methyl iodide (or
methallyl halides as well).
Ch a r t 1
Sch em e 4
Effect of Br id ge Gr ou p Size. In the initial survey
of bisoxazolines, we investigated the effect of the size of
the substituents in a cursory fashion by examining both
ethyl (1a ) and isobutyl groups (1b). This change had a
small influence on the enantioselectivity which was also
substrate dependent; i.e., 1b was better with aromatic
imines while 1a was better with olefinic and aliphatic
imines. To further explore this variable, we prepared the
simplest member of the group, dimethyl substituted
bisoxazoline 1c, the bulkier representative 1d and the
hybrid of these two, the unsymmetrical bisoxazoline 1e
(Chart 1). The most congested member in this series, 1f,
was considered to be impracticably difficult to prepare;
thus again, the hybrid analogue 1g was targeted. Finally,
in view of the success with 1b, we also sought the more
flexible neopentyl analogue 1h .
Sch em e 5
The dimethyl, diisopropyl, and mixed bisoxazolines 1c,
1d , and 1e were synthesized as described previously for
compounds 11 (Scheme 3). Alkylation of 14 with methyl
iodide in the presence of LDA afforded 1c uneventfully
in good yield. Analogous alkylation of 14 with LDA
followed by isopropyl iodide afforded intermediate 15,
which was subsequently alkylated under the same condi-
tions with either isopropyl iodide or methyl iodide to
provide 1d and 1e in 35% and 57% overall yields,
respectively.
The results from methyllithium addition to the three
different imines in the presence of these variously
substituted bisoxazolines are collected in Table 3 along
with the results from our previous studies. The trends
identified with 1a and 1b were followed at both ends of
the spectrum. For benzaldehyde imine 3 (wherein bulkier
group seemed to be beneficial), the smaller ligand 1c gave
the poorest selectivity, whereas the largest ligand 1d
gave the highest. The hybrid 1e gave an intermediate
selectivity, though somewhat closer to the lower end.
Interestingly, the neopentyl and isobutyl ligands, 1h and
1b, gave the same selectivity. For the cinnamaldehyde
(5) and hydrocinnamaldehyde (6) imines, the opposite
trend was observed. In these cases, the smaller ligand
1c gave the best selectivities whereas the larger ones (1d
and 17) gave the lowest. It is striking that the simple
change from ethyl to methyl groups has such a beneficial
effect.
Sch em e 3
Con clu sion s
Investigations into the effect that ligand architecture
has on the bisoxazoline promoted addition of MeLi to
(14) (a) Simmons, H. E.; Carins, T. L.; Vladuchick, S. A.; Hoiness,
C. M. Org. React. 1972, 20, 1. (b) Furukawa, J .; Kawabata, N.;
Nishimura, J . Tetrahedron Lett. 1966, 3353. (c) Furukawa, J .; Kawa-
bata, N.; Nishimura, J . Tetrahedron 1968, 24, 53.
(15) Newham, J . Chem. Rev. 1963, 63, 123.

5878 J . Org. Chem., Vol. 65, No. 18, 2000
Notes
Ta ble 3. Cor r ela tion of En a n tioselectivity to Liga n d
Su bstitu en t
chromatography (SiO2, hexane/EtOAc, 1/1) to give 338 mg (41%)
of 11a as white solid after Kugelrohr distillation. For 11a : bp
125-130 °C (0.4 mmHg); mp 79.0-80.0 °C; [R]D -83.8° (CHCl3,
c ) 1.115); 1H NMR (CDCl3, 400 MHz) δ 4.19 (dd, J ) 10.0, 8.6,
2H), 4.10 (dd, J ) 8.6, 7.4, 2H), 3.82 (dd, J ) 10.0, 7.2, 2H),
1.50 (m, 2H), 1.27 (m, 2H), 0.86 (s, 18H); 13C NMR (CDCl3, 100
MHz) δ 165.38, 75.22, 69.13, 33.82, 25.66, 18.19, 15.11; IR
(CHCl3) ν 2959, 2904, 2869, 1667, 1552, 1478, 1393, 1363. Anal.
Calcd for C17H28N2O2: C, 69.82; H, 9.65; N, 9.58. Found: C,
69.92; H, 9.56; N, 9.81.
1,1′-Bis[2-((4S)-(1,1-d im et h ylet h yl)-1,3-oxa zolin yl)]cy-
clobu ta n e (11b): white solid; bp 125-130 °C (0.4 mmHg); mp
60.0-61.0 °C; [R]D -69.8° (CHCl3, c ) 1.02); 1H NMR (CDCl3,
400 MHz) δ 4.19 (dd, J ) 10.1, 8.7, 2H), 4.11 (dd, J ) 8.6, 7.2,
2H), 3.89 (dd, J ) 10.1, 7.1, 2H), 2.71 (m, 2H), 2.50 (m, 2H),
2.02 (pentet, J ) 7.9, 2H), 0.89 (s, 18H); 13C NMR (CDCl3, 100
MHz) δ 167.19, 75.37, 69.27, 42.14, 33.92, 30.29, 25.67, 16.76;
IR (CHCl3)
18H30N2O2: C, 70.55; H, 9.87; N, 9.14. Found: C, 70.74; H,
10.06; N, 9.22.
ν 2957, 2903, 2869, 1664. Anal. Calcd for
C
1,1′-Bis[2-((4S)-(1,1-d im eth yleth yl)-1,3-oxa zolin yl)]cyclo-
p en ta n e (11c): bp 125-130 °C (0.4 mmHg); mp 51.0-52.0 °C;
[R]D -63.4° (CHCl3, c ) 1.195); 1H NMR (CDCl3, 400 MHz) δ
4.15 (dd, J ) 10.1, 8.6, 2H), 4.08 (dd, J ) 8.5, 7.1, 2H), 3.85 (dd,
J ) 10.0, 7.1, 2H), 2.38 (m, 2H), 2.12 (m, 2H), 1.71 (m, 4H), 0.87
(s, 18H); 13C NMR (CDCl3, 100 MHz) δ 167.96, 75.25, 69.04,
49.07, 35.38, 33.90, 25.66, 24.98; IR (CHCl3) ν 2958, 2904, 2870,
1661. Anal. Calcd for C19H32N2O2: C, 71.21; H, 10.06; N, 8.74.
Found: C, 71.28; H, 10.17; N, 8.83.
1,1′-Bis[2-((4S)-(1,1-d im eth yleth yl)-1,3-oxa zolin yl)]cyclo-
h exa n e (11d ): bp 140-150 °C (0.4 mmHg); mp 79.0-80.0 °C;
[R]D -55.9° (CHCl3, c ) 1.045); 1H NMR (CDCl3, 400 MHz) δ
4.12 (dd, J ) 10.1, 8.8, 2H), 4.05 (dd, J ) 8.6, 7.2, 2H), 3.87 (dd,
J ) 10.2, 7.1, 2H), 2.10 (m, 2H), 1.94 (m, 2H), 1.67 (m, 2H), 1.47
(m, 4H), 0.87 (s, 18H); 13C NMR (CDCl3, 100 MHz) δ 167.40,
75.51, 68.50, 43.13, 33.86, 32.50, 25.82, 25.41, 22.48; IR (CHCl3)
ν 2957, 2904, 2868, 1659. Anal. Calcd for C20H34N2O2: C, 71.81;
H, 10.24; N, 8.37. Found: C, 71.94; H, 10.36; N, 8.33.
Gen er a l P r oced u r e for th e Ad d ition of MeLi to Im in es:
(R )-N -(4-M e t h o x y p h e n y l)-R-m e t h y lb e n z e n e m e t h a n -
a m in e (7). A solution of 1c (294 mg, 1.0 mmol) in toluene (10
mL) in a flame-dried, 50 mL, three-neck, round-bottom flask
equipped with a stir bar, thermometer, septum, and nitrogen
inlet was cooled in a CO2/2-propanol bath (-75 °C internal).
MeLi (1.6 mL of a 1.27 M solution in low halide Et2O, 2.0 mmol)
was added to the solution via syringe and the resulting mixture
was stirred at -75 °C for 30 min. A solution of imine 3 (211 mg,
1.0 mmol) in toluene (10 mL) was slowly added to the reaction
mixture, maintaining an internal temperature below -70 °C.
The resulting yellow solution was stirred at -75 °C for 2 h. The
reaction was quenched at low temperature with MeOH (2.5 mL).
The reaction was warmed to room temperature and diluted with
water (10 mL). The mixture was extracted with Et2O (3 × 50
mL) and the combined organic layers were washed with brine
(1 × 50 mL), dried over MgSO4, and concentrated to a yellow
oil. Purification by chromatography (SiO2, hexane/EtOAc, 4/1)
afforded the amine 204 mg (90%) after Kugelrohr distillation.
Further elution gave recovered 1c (270 mg, 92%). Data for 7:
bp 150-160 °C (0.5 mmHg); 1H NMR (CDCl3, 400 MHz) δ 7.17-
7.38 (m, 5H), 6.65-6.71 (m, 2H), 6.42-6.49 (m, 2H), 4.41 (q, J
) 6.7, 1H), 3.79 (br s, 1H), 3.69 (s, 3H), 1.50 (d, J ) 6.7, 3H);
HPLC tR(R)-7, 23.60 (83.5%); tR(S)-7, 27.87 (16.5%) 67% ee
(hexane/EtOH, 99.2/0.8, 0.5 mL/min).
ligand, enantioselectivity, % (yield, %)
imine
1a
1b
1c
1d
1e
17
3
5
6
75 (95) 85 (95) 67 (90) 89 (87) 74 (97) 85 (85)
85 (79) 73 (90) 94 (73) 81 (77) n/a
91 (96) 87 (97) 93 (81) 87 (77) n/a
77 (87)
77 (92)
a
MeLi (low halide, ether). All reactions performed on a 1.0 mmol
scale. Yield of chromatographically homogeneous material. c De-
b
termined by CSP HPLC analysis.
imines revealed that while ligand bite angle plays only
a small role in the course of the reaction, the size of the
bridging substituents has a dramatic effect on the reac-
tion selectivity. From the data presented two generaliza-
tions can be made: (1) for reactions with aromatic imines,
bisoxazolines with large groups provide amines in high
enantiomeric excess, and conversely (2) high enantiose-
lectivity is seen in reactions with olefinic and aliphatic
imines when the promoters have small substituents.
Further studies on the contribution of electronic effects
on the stereochemical course of the addition as well as
mechanistic investigation of the origin of catalysis and
enantioselection are in progress.
Exp er im en ta l Section
Gen er a l Exp er im en ta l Da ta (See th e Su p p or tin g In for -
m a tion ). Gen er a l P r oced u r e for th e Syn th esis of Sp ir o-
cyclic Bisoxa zolin es. 1,1′-Bis[2-((4S)-(1,1-d im eth yleth yl)-
1,3-oxa zolin yl)]cyclop r op a n e (11a ) To a solution of 14 (750
mg 2.8 mmol) in THF (50 mL) in a flame-dried, 100 mL, three-
necked, round-bottom flask equipped with a stir bar, septum,
thermometer, and nitrogen inlet were added TMEDA (850 µL,
5.6 mmol, 2.0 equiv) and i-Pr2NH (400 µL, 2.8 mmol, 1.0 equiv).
The solution was cooled in a dry ice/2-propanol bath (-75 °C
internal). n-BuLi (3.8 mL of a 1.5 M solution in hexane, 5.6
mmol, 2.0 equiv) was added to the cold solution via syringe. The
reaction mixture was warmed to -20 °C and stirred at that
temperature for 30 min. The solution was cooled back below -70
°C, and 1,2-dibromoethane (243 µL, 2.8 mmol, 1.0 equiv) was
added via syringe. After the addition, the cold bath was removed
and the mixture was allowed to stir at room temperature for an
additional 16 h. The reaction mixture was quenched with
saturated aqueous NH4Cl solution (25 mL) and diluted with
water (10 mL) to dissolve the resulting salts. The mixture was
extracted with Et2O (3 × 75 mL). The combined organic layers
were washed with brine (1 × 75 mL), dried over MgSO4, and
concentrated to a thick oil. The residue was purified by column
Ack n ow led gm en t. This work was supported by the
Pharmacia and Upjohn Company.
Su p p or tin g In for m a tion Ava ila ble: Preparation and
characterization of compounds 1c-e,h , 11a -d , 14-17, 19, and
20 is provided along with experimental details of all data
presented in Tables 1 and 3. This material is available free of
charge via the Internet at http://pubs.acs.org.
J O0007175