Origin of stereochemistry in the α-amino acid esters and amides generated from optically active zirconaaziridine complexes

Article abstract of DOI:10.1021/ja953893h

Methane elimination from ethylenebis(tetrahydroindenyl)(methyl)zirconium amides (R = alkyl or aryl, R' = aryl) (S,S)-26 gives a mixture of epimeric zirconaaziridines 25 and 30. When zirconaaziridines with R = alkyl are trapped with ethylene carbonate as t

Full text of DOI:10.1021/ja953893h

J. Am. Chem. Soc. 1996, 118, 3479-3489
3479
Origin of Stereochemistry in the R-Amino Acid Esters and
Amides Generated from Optically Active Zirconaaziridine
Complexes1
Daniel A. Gately and Jack R. Norton*
Contribution from the Department of Chemistry, Colorado State UniVersity,
Fort Collins, Colorado 80523
ReceiVed NoVember 20, 1995X
Abstract: Methane elimination from ethylenebis(tetrahydroindenyl)(methyl)zirconium amides (R ) alkyl or aryl,
R′ ) aryl) (S,S)-26 gives a mixture of epimeric zirconaaziridines 25 and 30. When zirconaaziridines with R ) alkyl
are trapped with ethylene carbonate as they are formed, methyl R-amino acid esters 16 are obtained in poor ee
(+14% to -56%); the ee’s of 16 reflect the kinetic ratio of 25 to 30. When epimeric zirconaaziridines with R )
aryl are allowed to equilibrate before ethylene carbonate is added, the esters 16 are obtained in >96% ee. When
epimeric zirconaaziridines with R ) alkyl are allowed to equilibrate before ethylene carbonate is added, 16 is obtained
in 21-97% ee. When isocyanates are added to the zirconaaziridine epimers 25 and 30, phenylglycinamides are
obtained in 80-99% ee. The ee’s of the esters and amides are better when R′ is o-anisyl than when R′ is Ph. A
Curtin-Hammett-Winstein-Holness analysis explains the stereochemistry in the esters and amides. When an
equilibrated mixture of epimers (R ) CH2Ph, R′ ) o-anisyl) 25g and 30g is treated with increasing ethylene carbonate
concentrations, the ee of the ester 16g increases from 53% to 89%. The ee of 16g at saturation with ethylene
carbonate implies that Keq (k25g/k30g) is 17.2 for the 30g h 25g equilibrium. A similar result (19.0) is obtained from
the de of the insertion product rac-38 when 30g h 25g is treated with an excess of t-BuNCO.
Introduction
The carboxylation of long-chain amines is an attractive route
for the synthesis of R-amino acids.2 Meyers and co-workers
have used formamidines to attach CO2 synthons at the R-carbon
of amines,3 e.g., the carboethoxylation of tetrahydroisoquinoline
(1) in eq 1.3a Treatment of the metalated formamidine 2 with
ethyl chloroformate gave the racemic ethyl R-amino acid ester
3.
Chong and Park have generated configurationally stable
R-aminoorganolithiums 11 from optically active R-amino-
organostannanes 10. Treatment of 11 with CO2 gave the N-Boc-
protected N-methyl-R-amino acids 12 with excellent retention
of stereochemistry (eq 4).6
Duhamel and co-workers have employed the optically active
base 5 to effect the asymmetric carboxylation of the imine 4.4
Acidic removal of the benzylidene fragment gave the phenyl-
glycine esters 6 in 0-41% enantiomeric excess (ee) (eq 2).
In an approach similar to the one in eq 2, Beak and
co-workers have used (-)-sparteine to effect the asymmetric
deprotonation of 7 to give the optically active adduct 8 and,
after electrophilic addition of CO2, (R)-9 in 88% ee (eq 3).5
X Abstract published in AdVance ACS Abstracts, March 15, 1996.
(1) Because organozirconocenes of this type behave with electrophiles
as if they have a Zr-C σ bond, we prefer to call them zirconaaziridines
rather than zirconium imines.
We have been exploring the use of zirconaaziridines7 13 to
effect reactions like eq 5. We recently reported the regioselective
(2) Williams has said that “asymmetric carboxylation [at the R position
of an amine] is a surprisingly rarely studied approach [and] ... a future area
of investigation”. Williams, R. M. Synthesis of Optically ActiVe R-Amino
Acids; Pergamon: Oxford, 1989; Vol. 7.
(3) (a) Meyers, A. I.; Hellring, S.; Hoeve, W. T. Tetrahedron Lett. 1981,
22, 5115. (b) Bolster, J. M.; Hoeve, W. T.; Vaalburg, W.; Van Dijk, T.
H.; Zijlstra, J. B.; Paans, A. M. J.; Wynberg, H.; Woldring, M. G. Int. J.
Appl. Radiat. Isot. 1985, 36, 339.
(4) Duhamel, L.; Duhamel, P.; Fouquay, S.; Eddine, J. J.; Peschard, O.;
Plaquevent, J.-C.; Ravard, A.; Solliard, R.; Valnot, J.-Y.; Vincens, H.
Tetrahedron 1988, 44, 5495.
insertion of CO2 into the Zr-C bond of 13b8 to give the
(6) Chong, J. M.; Park, S. B. J. Org. Chem. 1992, 57, 2220.
(7) Buchwald, S. L.; Watson, B. T.; Wannamaker, M. W.; Dewan, J. C.
J. Am. Chem. Soc. 1989, 111, 4486.
(8) X-ray analysis of 13b (R ) Ph, R′ ) o-anisyl) shows the oxygen of
the o-anisyl substituent bound to Zr: Gately, D. A.; Norton, J. R.; Goodson,
P. A. Unpublished results.
(5) Beak, P.; Kerrick, S. T.; Wu, S.; Chu, J. J. Am. Chem. Soc. 1994,
116, 3231.
0002-7863/96/1518-3479$12.00/0 © 1996 American Chemical Society

3480 J. Am. Chem. Soc., Vol. 118, No. 14, 1996
Gately and Norton
Scheme 1
Results and Discussion
Asymmetric Carbomethoxylation of RCH2NHR′ When R
Is Aromatic. One would expect elimination of methane from
the racemic zirconium amide rac-26a, generated from PhCH2N-
(Li)Ph and rac-[EBTHI]ZrMe(OTf), in THF (eq 7), to give the
single diastereomer rac-25a drawn in eqs 6 and 7. When rac-
26a was prepared at -40 °C in THF and warmed to room
temperature, 25a was indeed formed; only one diastereomer was
1
detectable by H NMR. However, as reported by Grossman
and Buchwald,10 rac-26a at room temperature in benzene gave
instead the zirconaisoindole rac-27 (eq 8).
zirconocene R-aminocarboxylate 14.9 Because we could not
obtain the desired R-amino acid from 14, we treated 13a,b with
ethylene carbonate (a CO2 synthon) to give 15a,b and, after
methanolysis in benzene, the racemic methyl R-amino acid esters
16a,b. Treatment of 13a,b with isocyanates (t-BuNCO and Me3-
SiNCO) gave the metallacycles 17, 18, 21, and 22 and, after
methanolysis, the racemic phenylglycinamides 19, 20, 23, and
24 (Scheme 1).
Grossman, Davis, and Buchwald10 used the C2-symmetric
ethylenebis(tetrahydroindenyl) (EBTHI) ligand11 to prepare
allylic amines in high ee from zirconaaziridines in noncoordi-
nating solvents,12 and concluded that the ligand orients the
aziridine substituent R away from the six-membered ring. We
have therefore investigated the stereochemistry of ethylene
carbonate (and isocyanate) insertion reactions with optically
active zirconaaziridines such as (S,S,R)-25; if stereoselective,
such insertions should lead to enantiomerically enriched (S)-16
(or (S)-amide) (eq 6).
Separate signals (like those we had seen9 with the unsubsti-
tuted cyclopentadienyl analog 13a) could not be seen for free
and coordinated THF in a solution of 25a. However, other
experiments imply that the THF is coordinated to the Zr in 25a
(as in 13a). Dissolving rac-27 in THF-d8 at room temperature
converts it to rac-25a (eq 9).
(9) Gately, D. A.; Norton, J. R.; Goodson, P. A. J. Am. Chem. Soc. 1995,
117, 986.
(10) (a) Grossman, R. B.; Davis, W. M.; Buchwald, S. L. J. Am. Chem.
Soc. 1991, 113, 2321. (b) Grossman, R. B. Ph.D. Thesis, Massachusetts
Institute of Technology, Cambridge, MA, 1991.
(11) Wild, F. R. W. P.; Wasiucionek, M.; Huttner, G.; Brintzinger, H.
H. J. Organomet. Chem. 1985, 288, 63.
(12) When they attempted to prepare the EBTHI analog of 13a in
benzene, Grossman and Buchwald obtained the zirconaisoindole A below.
Analogous species were formed in all cases when R ) aryl and no
coordinating atoms were present (see ref 10b).
Treatment of (S,S,R)-25a-c with ethylenecarbonate should
produce the spirocyclic complexes (S,S,S)-28a-c, and metha-
nolysis of (S,S,S)-28a-c should give the S-R-amino acid esters
16a-c (Scheme 2). When (S,S,R)-25a-c was treated with
ethylene carbonate, methanolysis, carried out in the presence
of added Cp2ZrMe2,13 gave the methyl R-amino acid esters (+)-
16a-c in 53-67% overall yield and >96% ee (Table 1, Scheme
2).
(13) With the EBTHI ligand we found Zr-promoted transesterification9
(like that of 29) to be extremely slow unless Cp2ZrMe2 was added.

Stereochemistry in R-Amino Acid Esters and Amides
J. Am. Chem. Soc., Vol. 118, No. 14, 1996 3481
Table 2. Product Yields, Configurations, and ee’s of (S)-16d-g
from the Reaction of Ethylene Carbonate with (S,S,R)-25d-ga
(Scheme 3)
Scheme 2
yield of
MeO2CCH(R)NHR′ configuration/ eec
product
R
R′
(16)b (%)
(optical sign) (%)
16d
16ed
16f
Me
Me
i-Bu
CH2Ph Ar
Ph
Ar
Ar
5
46
61
34
S/(-)
S/(-)
S/(-)
S/(+)
21
97
54
53
16g
a (S,S,R)-25d-f was prepared at 70 °C and then allowed to cool to
room temperature before addition of ethylene carbonate. b Isolated
yields, >98% pure by HPLC and 1H NMR. c Obtained from stationary
phase chiral HPLC. d Ar ) o-anisyl.
Table 3. Product Yields, Configurations, and ee’s of 16a,c-f
Obtained from Trapping 25a,c-fa with Ethylene Carbonate
(Schemes 4 and 5)
Table 1. Product Yields, Configurations, and ee’s of (S)-16a-c
from the Reaction of Ethylene Carbonate with (S,S,R)-25a-ca
(Scheme 2)
yield of
MeO2CCH(R)NHR′ configuration/
T
eec
(optical sign) (°C) (%)
product
R
R′
(16)b (%)
yield of
MeO2CCH(R)NHR′ configuration/
eec
(%)
16a
Ph
Ph
Ar
Ph
Ar
Ar
trace
43
68
58
51
S/(+)
S/(-)
S/(-)
R/(+)
S/(-)
R/(-)
-40
product
R
R′
(16)b (%)
(optical sign)
16cd Ar
-40 >99
70 22
70 56
70 14
70 52
16d
16e
16f
16g
Me
Me
i-Bu
CH2Ph Ar
16a
16bd
16c
Ph Ph
Ph Ar
Ar Ar
60
67
53
S/(+)
S/(-)
S/(-)
>98
96
98
49
a (S,S,R)-25a-c was prepared at room temperature. b Isolated yields,
>98% pure by HPLC and 1H NMR. c Obtained from stationary phase
chiral HPLC. d Ar ) o-anisyl.
a 25a,c were prepared from 26a,c at -40 °C; 25d-f were prepared
at 70 °C. b Isolated yields, >98% pure by HPLC and 1H NMR.
c Obtained from stationary phase chiral HPLC. d Ar ) o-anisyl.
Scheme 3
g) (R ) alkyl) formed by the same procedure (Scheme 3) is
also presumed to be S.
In Situ Trapping of 25d-g with Ethylene Carbonate at
70 °C. It seemed possible that the yields of the insertion
products (S,S,S)-28d-g and the methanolysis products (S)-
16d-g would improve if ethylene carbonate were present as
(S,S,R)-25d-g formed. (Buchwald and Grossman had obtained
good yields and high ee’s of allylic amines by generating
zirconaaziridines in the presence of alkynes.10,12
)
If methane elimination from (S,S)-26d-g gave (S,S,R)-25d-
g, trapping with ethylene carbonate (and retention of stereo-
chemistry in the insertion step) would give (S,S,S)-28d-g.
Methanolysis of (S,S,S)-28d-g should give (S)-16d-g. In fact
we obtained (S)-(-)-16d in 68% yield (improved from 5% in
Scheme 3) and (S)-(-)-16f in 51% yield, although in poor ee
(22% for (S)-(-)-16d and 14% for (S)-(-)-16f) (Table 3,
Scheme 4).
To our surprise, the products (R)-(+)-16e (from (S,S)-26e)
and (R)-(-)-16g (from (S,S)-26g) had R configurations, opposite
that (S) predicted by Scheme 3! If the insertion occurred with
retention, the R configuration must have come from (S,S,S)-
30e,g via (S,S,R)-31e,g (Scheme 4).16
In Situ Trapping of 25a,c with Ethylene Carbonate below
Room Temperature. We then examined the effect of trapping
25a,c with ethylene carbonate as it was formed. A THF solution
containing (S,S)-26c (prepared as in Scheme 2) was treated with
ethylene carbonate and Cp2ZrMe2 at -40 °C. (Solutions of
25a-c are red in the absence of ethylene carbonate; a persistent
yellow color suggested that (S,S,R)-25c was short-lived.)
Methanolysis gave (S)-(-)-16c in 43% yield and >99% ee
(Scheme 5, Table 3). However, similar treatment of (S,S)-26a
(R ) R′ ) Ph) gave only trace amounts of (S)-(+)-16a (Table
3).
The absolute stereochemistry of (+)-16a has been assigned
by comparing the sign of its rotation to that of its enantiomer
(-)-16a.14 The absolute configuration of (-)-16a is known to
be R, so the absolute configuration of (+)-16a must therefore
be S. A product (16) of the same configuration (S) is predicted
if ethylene carbonate inserts into the Zr-C bond of (S,S,R)-25
with retention of stereochemistry. The absolute stereochemistry
of the other products (16b,c) (R ) aromatic) formed by the
same procedure (Scheme 2) is also presumed to be S.
Asymmetric Carbomethoxylation of RCH2NHR′ When R
Is Alkyl. For R ) alkyl (26d-g) elimination of methane and
formation of 25d-g occurred during several h at 70 °C (Scheme
3). When the mixture containing (S,S,R)-25d-g was cooled
to room temperature and treated with ethylene carbonate,
methanolysis gave 16d-g in 5-61% yield and 21-97% ee
(Table 2). The absolute stereochemistry of (-)-16d has been
assigned by comparing the sign of its rotation to that of its
enantiomer (+)-16d.15 The absolute configuration of (+)-16d
is known to be R, so the absolute configuration of (-)-16d must
be S. The absolute stereochemistry of the other products (16e-
(15) Paradisi, M. P.; Romeo, A. J. Chem. Soc., Perkin Trans. 1 1977,
596.
(14) Christoskova, St.; Berova, B.; Simova, E.; Spassov, St.; Kurtev,
B.; Snatzke, G. Proceedings of the F.E.C.S. International Conference on
Circular Dichroism; VCH: Weinheim, 1987; pp 315-21.
(16) C-H epimerization of the esters 16 (or amides) was not detected
(via loss of optical rotation) under the conditions used in the workup
procedure.

3482 J. Am. Chem. Soc., Vol. 118, No. 14, 1996
Gately and Norton
Scheme 4
Scheme 6
configuration of allylic amines 32 in high ee17 and (2) we
obtained the S configuration of the methyl R-amino acid esters
16 in Schemes 2, 3, and 5 suggests that the zirconaaziridine in
both cases is largely (S,S,R)-25. However, if interconversion
of 25 and 30 is facile, the insertion reactions can be described
by eq 10 and analyzed by the Curtin-Hammett-Winstein-
Holness principle.18
Scheme 5
Curtin-Hammett-Winstein-Holness Principle. If the
enantiomeric purity of 16 in Schemes 2 and 3 is governed by
eq 10, it should depend on the first-order rate constants k25 and
k30 in the 30 h 25 equilibrium and the competing kR[ethylene
carbonate] and kS[ethylene carbonate], where kR and kS are
second-order rate constants.
Boundary Condition I.18a In reference to eq 10, if k25 and
k30 . kR[ethylene carbonate] and kS[ethylene carbonate], the
(S)-16/(R)-16 product ratio is given by eq 11 where Keq ) k25/
k30. The (S)-16/(R)-16 product ratio in eq 11 depends on both
the first-order (k25, k30) and second-order (kR, kS) rate constants
in eq 10.
N-Ph vs N-o-Anisyl. In Scheme 2, good yields and excellent
ee’s were obtained for (S)-16a-c regardless of whether R′ was
Ph or o-anisyl. However, in Schemes 4 and 5, when (S,S,R)-
25 was trapped by ethylene carbonate as it was formed, the
nature of R′ had a dramatic effect on the ee and yield of (S)-
16. (In Table 3, compare the ee’s and yields of 16a (R′ ) Ph)
vs those of 16c (R′ ) o-anisyl), and those of 16d (R′ ) Ph) vs
those of 16e-g (R′ ) o-anisyl).)
Whether ethylene carbonate was added after (Scheme 3) or
before (Scheme 4) (S,S,R)-25d was formed, the ee of 16d (R′
) Ph) was not affected. (Compare the 21% ee of 16d in
Scheme 3 vs 22% in Scheme 4.) In contrast, the ee’s of 16e-g
(R′ ) o-anisyl) declined, and in some cases reversed, between
the conditions used in Scheme 3 and those used in Scheme 4.
(Compare the 53-97% ee of 16e-g in Scheme 3 vs the ee’s
(14% S to 56% R) in Scheme 4.)
Origin of Stereochemistry in the Methyl r-Amino Acid
Esters 16. In an effort to understand how R′ and different
reaction conditions affect the stereochemistry in 16, we have
investigated the mechanisms of the key steps in Schemes 2-5.
Stereochemistry of Zirconaaziridine Intermediates. An
explanation for the poor ee’s of 16d-g in Scheme 4 can be
deduced from an elaborate deuterium labeling study performed
by Grossman.10b The ratio of MeH to MeD loss in Scheme 6
(and related experiments) reflects not only the isotope effect
but the stereoselectivity. Because an infinitely large (or small)
32h-d1/d0 isotope ratio did not result from this experiment,
Grossman concluded that the C-H activation step 26h f 25h/
30h was not very diastereoselective.
kS
eqkR
(S)-16 (S,S,S)-28
)
) K
(11)
(R)-16 (S,S,R)-31
Boundary Condition II.18a Alternatively, if k25 and k30
,
kR[ethylene carbonate] and kS[ethylene carbonate], the (S)-16/
(R)-16 product ratio is given by eq 12. The (S)-16/(R)-16
product ratio in eq 12 reflects the equilibrium ratio 30 h 25 in
eq 10 and depends only on the first-order rate constants k25 and
k30 there.
(S)-16 (S,S,S)-28
)
) Keq
(12)
(R)-16 (S,S,R)-31
A simple way of seeing if the ee in 16 reflects the operation
of eq 10 is to measure the (S)-16/(R)-16 ratio (ee) as the
concentration of added ethylene carbonate increases or de-
creases. An increase or decrease in the ee of 16 should occur
as we move from boundary condition I to boundary condition
II.
(17) The only exception was the ee (18%) of (S)-32a (R ) R′ ) Ph).
However, Grossman reported that the de of its precursor ((S,S,S)-41a) varied
“with the temperature and other conditions” (refs 10b and 12).
(18) (a) Seeman, J. I. Chem. ReV. 1983, 83, 83. (b) Eliel, E. L.; Wilen,
S. H. Stereochemistry of Organic Compounds; John Wiley and Sons: New
York, 1994; pp 648-655. (c) Carey, F. A.; Sundberg, R. J. AdVanced
Organic Chemistry, 3rd ed.; Plenum: New York, 1990; pp 215-216.
The fact that (1) Grossman and Buchwald obtained the S

Stereochemistry in R-Amino Acid Esters and Amides
J. Am. Chem. Soc., Vol. 118, No. 14, 1996 3483
Table 4. Product Yields, Configurations, and ee’s of MeO2CCH(R)NHR′ (16a,c,f,g) from Addition of Various Concentrations of Ethylene
Carbonate to 25a,c,f,ga
amount of ethylene carbonate (mmol)
in a volume of approximately 20 mL
entry
product
R
R′
yieldb (%)
configuration/(optical sign)
eec (%)
1
2
16a
16a
16c
16c
16f
16f
16g
16g
16g
16g
16g
16g
Ph
Ph
Ar
Ar
i-Bu
i-Bu
CH2Ph
CH2Ph
CH2Ph
CH2Ph
CH2Ph
CH2Ph
Ph
Ph
Ar
Ar
Ar
Ar
Ar
Ar
Ar
Ar
Ar
Ar
60
24
53
35
61
24
34
31
27
14
49
35
S/(+)
S/(+)
S/(-)
S/(-)
S/(-)
S/(-)
S/(+)
S/(+)
S/(+)
S/(+)
R/(-)
R/(-)
>98
>98
98
92
54
79
53
77
85
0.64
6.4
0.64
6.4
0.64
6.4
0.64
1.6
3d
4
5
6
7
8
9
6.4
10
11
12
89
52
54
19.2
0.64
6.4
a [25a,c,f,g] ) 0.32 mmol, ca. 8 mM. In entries 1-4, 25a,c were prepared at room temperature, followed by addition of ethylene carbonate in
benzene or THF at room temperature. In entries 5-10, 25f,g were prepared at 70 °C, followed by addition of ethylene carbonate at room temperature.
1
In entries 11-12, 25f,g were prepared at 70 °C in the presence of ethylene carbonate. b Isolated yields, >98% pure by HPLC and H NMR.
c Obtained from stationary phase chiral HPLC. d Ar ) o-anisyl.
carbonate] step in eq 10 is at least five times faster with the
minor diastereomer 30g than the kS[ethylene carbonate] step is
with the major diastereomer 25g.
In contrast, if the ee (92%) of (S)-(-)-16c reflects the
equilibrium ratio 30c h 25c in the upper limit (entry 4, Table
4) of ethylene carbonate concentration, then Keq ≈ 24.0 from
eq 12. The ee (98%) of (S)-(-)-16c at the lowest feasible
ethylene carbonate concentration (entry 3, Table 4) combined
with Keq (24.0) gives kS/kR > 4.1 from eq 11 for 25c and 30c.
The kS[ethylene carbonate] step in eq 10 is at least 4 times faster
with the major diastereomer 25c than the kR[ethylene carbonate]
step is with the minor diastereomer 30c.
Figure 1. Percent ee of (S)-(+)-16g vs amount of ethylene carbonate
(mmol) in a volume of approximately 20 mL. Data points were taken
from entries 7-10 in Table 4. (S,S,R)-25g (0.32 mmol) was prepared
at room temperature and then treated with ethylene carbonate.
Curtin-Hammett-Winstein-Holness Conditions Are Not
Achieved in Scheme 4. A Curtin-Hammett-Winstein-
Holness situation requires that a mixture of 30 h 25 be initially
equilibrated. Because the 25/30 ratio in Scheme 4 is kinetically
controlled, i.e., 25 and 30 have not been equilibrated, the product
ratio should be insensitive to the concentration of ethylene
carbonate. Indeed, when (S,S)-26g (R ) CH2Ph, R′ ) o-anisyl;
0.32 mmol) was heated to 70 °C in the presence of increasing
ethylene carbonate concentrations (from 0.64 to 6.4 mmol, in a
volume of approximately 20 mL, entries 11 and 12 in Table 4),
the ee of (R)-(-)-16g increased negligibly from 52% to 54%.
Direct Measurement of Keq in Eq 10? We then attempted
to measure the equilibrium ratio 30 h 25 directly by 1H NMR.
The required zirconaaziridine rac-25g was obtained by adding
Ph(CH2)2NLi(o-anisyl) to rac-[EBTHI]ZrMe(OTf) and heating
the solution in benzene to 70 °C overnight; during this time
methane was eliminated from rac-26g to give rac-25g in
Ethylene Carbonate Concentration vs the ee of 16. Indeed,
when 25g (0.32 mmol) was treated with increasing ethylene
carbonate concentrations (from 0.64 to 19.2 mmol, in a volume
of approximately 20 mL, entries 7-10 in Table 4), the ee of
(S)-(+)-16g increased from 53% to 89% (Figure 1) and then
leveled off. The ee (89%) of (S)-(+)-16g at saturation (Figure
1) therefore reflects the equilibrium ratio 30g h 25g; Keq
)
17.2 from eq 12. Similar results were obtained when 25f (0.32
mmol) was treated with increasing ethylene carbonate concen-
trations (from 0.64 to 6.4 mmol, in a volume of approximately
20 mL, entries 5 and 6 in Table 4); the ee of (S)-(+)-16f
increased from 54% to 79%.
The opposite trend was observed when 25c (0.32 mmol) was
treated with increasing ethylene carbonate concentrations (from
0.64 to 6.4 mmol, in a volume of approximately 20 mL, entries
3 and 4 in Table 4); the ee of (S)-(-)-16c decreased from 98%
to 92%. The latter may reflect the equilibrium ratio 30c h
25c.
However, when 25a (0.32 mmol) was treated with increasing
ethylene carbonate concentrations (0.64 f 6.4 mmol, in a
volume of approximately 20 mL, entries 1 and 2 in Table 4),
the ee of (S)-(+)-16a did not change. This result suggested
that boundary condition II was applicable even when only 0.64
mmol of ethylene carbonate was added. The 30a h 25a
equilibrium therefore lies far to the right; Keq ≈ 99 from eq 12.
Ratio of kS to kR for Various Zirconaaziridines (eq 11).
Because Keq (17.2) is known for the 30g h 25g equilibrium in
the upper limit (entry 10, Table 4) of ethylene carbonate
concentration, we can calculate what kS/kR would be if entry 7
(Table 4) represents the lower limit (boundary condition I). The
ee of 53% implies that kS/kR < 0.19 (eq 11). The kR[ethylene
1
quantitative yield. Unfortunately, direct H NMR observation
of the product mixture containing rac-25g in C6D6 at room
temperature (or at -105 °C in THF-d8 or -95 °C in toluene-
d8) showed only one compound and offered no evidence for
the equilibrium mixture rac-30g h rac-25g.
Ethylene Carbonate Concentration vs Insertion Product
Ratio (28/31). When rac-25g (ca. 19.6 µmol) was treated with
∼26 µmol of ethylene carbonate in a volume of approximately
0.5 mL for 1 h, an 84/16 rac-28g/rac-31g product ratio resulted
(t1/2 ≈ 12 min). When the concentration of ethylene carbonate
was increased to ∼190 µmol, in a volume of approximately
0.5 mL, a 94/6 rac-28g/rac-31g product ratio resulted after about
10 min (t1/2 ≈ 2 min) (Scheme 7). The diastereomeric excess
(de) of rac-28g increased from 68% to 88% (recall that a similar
change in ethylene carbonate concentration increased the ee of
its methanolysis product (S)-(+)-16g from 53% to 89%). This
result implied that the minor diastereomer rac-30g must have

3484 J. Am. Chem. Soc., Vol. 118, No. 14, 1996
Gately and Norton
Scheme 7
Scheme 9
(-)-24 (98% ee), also in low yield (Table 5, Scheme 8). The
absolute stereochemistry of (-)-23 and (-)-24 formed by the
same procedure (Scheme 8) was also presumed to be S.
Table 5. Product Yields, Configurations, and ee’s of Amides 19,
20, 23, and 24 from the Reaction of R′′NCO with (S,S,R)-25a and
(S,S,R)-25ba (Scheme 8)
Zr r O Chelation in 25b Slows Insertion Reactions. The
low yields of (S)-(-)-23 and (S)-(-)-24 reflect the slow insertion
reaction of R′′NCO with (S,S,R)-25b, presumably because of
the Zr r O interaction from the N-o-anisyl fragment in (S,S,R)-
25b. When rac-25b was treated with about 4.1 equiv of
t-BuNCO in THF-d8, 6 days (t1/2 ≈ 29 h) was required to obtain
the metallacycle rac-35 in quantitative yield (Scheme 9) (the
minor diastereomer rac-37 was not detected by 1H NMR).
Overall, inserting reagents react much faster when R′ is Ph in
25 (or when R′ is Ph in 13) than they do when R′ is o-anisyl.
yield of amide
configuration/ R′′NHCOCH(Ph)NHR′ b eec
product R′ R′′ (optical sign)
(%)
(%)
19
20
23d
24
Ph t-Bu
Ph
Ar t-Bu
Ar
S/(+)
S/(+)
S/(-)
S/(-)
62
51
31
25
92
80
>99
98
H
H
a (S,S,R)-25a,b were prepared at room temperature. b Isolated yields,
>98% pure by HPLC and 1H NMR. c Obtained from stationary phase
chiral HPLC. d Ar ) o-anisyl.
Origin of Stereochemistry in the Amides 19, 20, 23, and
24. As with the esters (S)-16 in eqs 10-12, the enantiomeric
purity of the amides 19, 20, 23, and 24 from Scheme 8 should
reflect the operation of the Curtin-Hammett-Winstein-
Holness equations (eqs 13-15).
Scheme 8
been present in equilibrium with rac-25g. (The absence of the
minor diastereomer rac-30g in the previous 1H NMR experiment
suggests that the rac-30g h rac-25g interconversion is fast
relative to the NMR time scale.)
Asymmetric Carboamidation of 25a. We then examined
the stereoselectivity of the isocyanate insertion reactions of 25.
Treatment of (S,S,R)-25a (prepared as in Scheme 2) with
t-BuNCO should give the metallacycle (S,S,S)-33 and, after
protonolysis, amide (S)-19. In fact, we obtained (S)-(+)-19 in
high ee (92%). Similar treatment of (S,S,R)-25a with Me3-
SiNCO gave, after protonolysis, (S)-(+)-20 in good ee (80%)
(Table 5, Scheme 8). The absolute configurations of (+)-19
and (+)-20 were assumed to be S, because they were from the
same (S,S,R)-25a that had given (S)-16a (see Table 1 and
accompanying discussion).
Effect of t-BuNCO Concentration on the Stereochemistry
of the Amides. The addition of a dilute19 solution of t-BuNCO
to excess (S,S,R)-25a gave a significant decrease in the ee of
(S)-(+)-19 from the one (92% ee) found above (Scheme 8, Table
5). When (S,S,R)-25a (0.320 mmol) was treated with t-BuNCO
(32 mM, 0.256 mmol), the ee of (S)-(+)-19 decreased to 36%
(Scheme 10).
(19) Low concentrations of isocyanates must be used to effect a
significant change in the ee’s of the phenyl-substituted amides 19 and 20;
the isocyanate must be added as a dilute solution so that mixing is complete
before the insertion reaction occurs. In contrast, because the o-anisyl-
substituted zirconaaziridines react more slowly, the concentrations of
isocyanate low enough to effect a change in the ee’s of the o-anisyl-
substituted amides 23 and 24 can be added neat to 25b.
Asymmetric Carboamidation of 25b. In contrast, treatment
of (S,S,R)-25b with t-BuNCO gave, after protonolysis, (S)-
(-)-23 in excellent ee (>99%) but low yield. Similar treatment
of (S,S,R)-25b with Me3SiNCO gave, after protonolysis, (S)-

Stereochemistry in R-Amino Acid Esters and Amides
J. Am. Chem. Soc., Vol. 118, No. 14, 1996 3485
Scheme 10
Scheme 13
Scheme 11
should reflect Keq (Scheme 12); the ee of the hydrolysis product
should therefore be high. The opposite effect is predicted when
an inserting reagent adds in a slow step (k[inserting reagent] ,
kforward or kback of the equilibrium).
The validity of Scheme 12 is evident when the rates and
stereoselectivities of the ethylene carbonate insertions of 25 are
compared with those of the isocyanate ones. Ethylene carbonate
reacts about 3 times more rapidly with 25g than does
t-BuNCO: when rac-25g (ca. 17.6 µmol) was treated at room
temperature with ca. 22.2 µmol of either reagent in a volume
of approximately 0.5 mL, the ethylene carbonate reaction
(Scheme 7) took only 1 h (t1/2 ≈ 12 min) whereas the t-BuNCO
reaction (Scheme 11) took 3 h (t1/2 ≈ 36 min). As predicted
by Scheme 12, the de was considerably higher (68% vs 28%)
with ethylene carbonate.
Scheme 12
Reagents less reactive than isocyanates put us closer to
boundary condition I and give results even further removed from
Keq. Treatment of rac-25a (∼20 µmol) with 2-butyne (ca. 83.0
µmol, in a volume of approximately 0.5 mL) gave rac-41a in
28% de (eq 16), whereas treatment of Me3SiNCO (ca. 24.0
Ratio of kS to kR for the Zirconaaziridines in Scheme 10.
The ee (36%) of (S)-(+)-19 in the limit of low t-BuNCO
concentration implies that kS/kR < 0.02 from eq 14. Thus, the
kR[t-BuNCO] step with the minor diastereomer 30a in Scheme
10 is at least 50 times faster than the kS[t-BuNCO] step with
the major diastereomer 25a.
Keq Is Independent of Inserting Reagent. Keq for the
30 h 25 equilibrium is independent of the nature of the inserting
reagent (ethylene carbonate or t-BuNCO). Thus, the product
ratio in eq 15 should be the same as that in eq 12. In fact,
when rac-25g (ca. 20.1 µmol) was treated with excess t-BuNCO
(ca. 454 µmol, in a volume of approximately 0.5 mL), a 95/5
rac-38/rac-39 ratio was obtained in the 1H NMR (Scheme 11).
This result gives Keq ) 19.0 from eq 15 for the 30g h 25g
equilibrium mixture, similar to the one (Keq ) 17.2) found (see
discussion below eq 12) from the addition of excess ethylene
carbonate to the equilibrium mixture 30g h 25g.
Influence of the Rate of Insertion on Stereoselectivity.
Two reagents that insert at different rates may give different
stereochemical results with the same zirconaaziridine. Equations
10 and 13 predict that the stereochemistry of the insertion
product should depend on the rate at which inserting reagents
add to the equilibrium mixture 30 h 25. If an inserting reagent
adds to the 30 h 25 equilibrium mixture in a fast step
(k[inserting reagent] . kforward or kback of the equilibrium) and
Keq is large, the stereochemical outcome of the insertion product
µmol, in a volume of approximately 0.5 mL) gave rac-40 in
80% de. However, sufficiently high concentrations of even a
relatively unreactive reagent such as 2-butyne move us back
toward the bottom of Scheme 12 (boundary condition II).
Increasing the concentration of 2-butyne from 0.03 to 12.8 mM
increased the de of rac-41a from 24% to 99%!
Mechanism of 30 h 25 Epimerization? In order for the
30 h 25 epimerization to occur, the Zr-C bond must cleave
from 30 to give either a Zr(II)-imine complex such as 42 or a
Zr(III) complex such as 43 with a carbon radical.20 After
rotation of the (Ph)CH-N bond and inversion of configuration
of the carbon (43′), recombination of the Zr-C bond in 43′
would give 25. With 42, recombination of its Zr-C bond is
all that is required to give 25 (Scheme 13).
(20) Grossman suggested that the 30 h 25 epimerization involved the
43 h 43′ equilibrium in Scheme 13 (see ref 10b).

3486 J. Am. Chem. Soc., Vol. 118, No. 14, 1996
Gately and Norton
tion. Cp2ZrMe2,22 rac-[EBTHI]ZrMe2,23 (S,S)-[EBTHI]ZrMe2,10b and
(S,S)-[EBTHI]ZrMe(OTf)10b were prepared by standard procedures. Cp2-
ZrCl2 was generously supplied by Boulder Scientific. Anilines were
prepared by reduction of the carboxamide with BH3‚SMe2/BF3‚Et2O
as described by Brown and co-workers.24
Scheme 14
1H NMR data were collected on a Bruker WNX 300-MHz FT
spectrometer; residual solvent proton shifts were used as internal
standards. Elemental analyses of air- and moisture-sensitive compounds
were performed by Analytische Laboratorien, Gummersbach, Germany;
those of all other compounds were performed by Midwest Laboratories,
Indianapolis, IN.
Chiral stationary phase HPLC data were collected on a Varian 9050
Star Detector. Compounds 16c, 19, 20, 23, and 24 were separated on
a Bakerbond OD chiralcel column; all others were separated on a
Bakerbond OJ chiralcel column. All compounds were detected at 254
nm. The presence of enantiomerically enriched esters 16 and amides
19, 20, 23, and 24 was confirmed by spiking the HPLC samples with
authentic (()-16 and (()-19, -20, -23, and -24. Optical rotations were
obtained on a Rudolph Research automatic polarimeter Autopol III.
The optical rotations ([R]measured) for 16 and 18 were obtained at 26
°C. Specific rotations ([R]D) are reported in deg/dm, and the concentra-
tion (c) is given in g/100 mL (THF). The ee (>99%) of (S,S)-[EBTHI]-
ZrMe2 was determined by treatment with excess (R)-(-)-O-acetylman-
delic acid in C6D6; the ratio of the resulting diastereomers was
In an unrelated study we found evidence of a Zr(II)-imine
complex like the one shown (42) in Scheme 13. We expected
treatment of the lithium amide (o-anisyl)CH2N(Li)CH2Ph with
Cp2ZrMe(OTf) to give the (methyl-)zirconium amide 44 an-
d,after regioselective C-H activation and loss of methane, the
zirconaaziridine 13i in eq 17.
1
determined by H NMR from the sp2 CH indenyl resonances.10b
General Procedure for the Preparation of (()-16b,c (MeO2CCH-
(R)NHR′). The following procedure is modified from the one described
earlier.9 A solution containing Cp2ZrMe2 (378 mg, 1.5 mmol) and THF
(10 mL) was cooled to -78 °C and treated with TfOH (127 µL, 1.44
mmol). The pale yellow solution was warmed to room temperature,
stirred for 1 h, and again cooled to -78 °C. In a separate flask BuLi
(1.6 M, 900 µL, 1.44 mmol) was added to a cold (0 °C) ether (10 mL)
solution containing RCH2NHR′ (1.44 mmol), and the solution stirred
for 5 min. The RCH2N(Li)R′ was transferred by cannula to the Cp2-
ZrMe(OTf) and stirred for 0.5 h at -78 °C; the solution was warmed
to room temperature and stirred overnight. The solvent was removed
and replaced with benzene (20 mL) containing ethylene carbonate (136
mg, 1.54 mmol). After stirring overnight, the solution was treated with
MeOH (1 mL) and stirred for an additional 8 h at 80 °C. The solvent
was removed, and the residue was treated with CH2Cl2 (20 mL) and
filtered. The residue from the filtrate was spotted on a Chromatotron
plate eluted with hexanes/ethyl acetate (25/1); further purification was
not needed.
The (methyl-)zirconium amide 44 was easily isolated. How-
ever, to our surprise, we obtained (E)-45 (84%) and (Z)-45
(16%) instead of 13i (Scheme 14)! The NMR of 45 shows a
1H methine singlet resonance (δ 8.09, δ 7.95) and a 13C methine
carbon resonance (δ 176.0), indicative of the CHdN imine
fragment in 45; these 1H and 13C resonances are well downfield
1
of the H methine proton resonance (δ 3-4 ppm) and 13C
methine carbon resonance (δ 60-70 ppm) expected in the
1
zirconaaziridine 13i. Moreover, the H resonance of the Cp’s
1
for (E)-45 (δ 5.71) and (Z)-45 (δ 5.85) and the H resonance
of the methylene signals for (E)-45 (δ 4.52) and (Z)-45 (δ 4.46)
were all singlets, indicative of a plane of symmetry bisecting
the CpZrCp and HCH planes.21
(()-16b (MeO2CCH(Ph)NH(o-anisyl)). Yield: 252 mg (64%). 1H
and 13C NMR data for (()-16b prepared by a different procedure were
reported earlier.9
(()-16c (MeO2CCH(o-anisyl)NH(o-anisyl)). Yield: 247 mg (52%).
1H NMR (CDCl3): δ 7.45 (d, 1 H), 7.32 (t, 1 H), 6.98 (t, 2 H), 6.88-
6.69 (m, 3 H), 6.58 (d, 1 H), 5.65 (s, 1 H), 5.47 (v br s, 1 H), 3.97 (s,
3 H), 3.88 (s, 3 H), 3.77 (s, 3 H). 13C NMR (CDCl3): δ 172.7, 157.0,
147.0, 136.2, 129.2, 127.8, 126.3, 121.0, 117.1, 111.0, 110.3, 109.5,
55.7, 55.3, 54.0, 52.4. Anal. Calcd for C17H19NO4: C, 67.76; H, 6.35;
N, 4.65. Found: C, 67.89; H, 6.38; N, 4.71.
Possible intermediates in the 30 h 25 interconversion are
thus 46 when R′ ) o-anisyl and 47 when R′ ) Ph.
General Procedure for the Preparation of (()-16d-g (MeO2-
CCH(R)NHR′). The procedure was the same as that described above
for (()-16b,c, but with the following modifications. After addition of
the RCH2N(Li)R′ to Cp2ZrMe(OTf), the solution was warmed to room
temperature. The solvent was removed and replaced with benzene (20
mL) containing ethylene carbonate (136 mg, 1.54 mmol); the solution
was transferred to a vacuum bulb, and the bulb was sealed and then
heated to 70 °C overnight.
Experimental Section
Materials. All air-sensitive compounds were prepared and handled
under a nitrogen atmosphere, using standard Schlenk and inert-
atmosphere-box techniques. Most of the solvents used were distilled
under N2 from sodium-benzophenone ketyl; hexanes were stirred over
H2SO4 and distilled from sodium-benzophenone ketyl in the presence
of tetraglyme. Trifluoromethanesulfonic acid (TfOH) was degassed
by three freeze/pump/thaw cycles at -196 °C, and finally transferred
into a flame-dried vacuum bulb. Isocyanates were stirred over P4O10
for 24 h and transferred by high vacuum into a flame-dried vacuum
bulb. All other reagents employed were used without further purifica-
(()-16d (MeO2CCH(Me)NHPh). Yield: 124 mg (48%). 1H NMR
(CDCl3): δ 7.17 (t, 2 H), 6.74 (t, 1 H), 6.62 (d, 2 H), 4.18 (CH q
overlapped with NH br s, 2 H), 3.76 (d, 3 H). 13C NMR (CDCl3): δ
175.1, 146.5, 129.3, 118.3, 113.8, 52.2, 51.9, 19.0. Anal. Calcd for
C10H13NO2: C, 67.02; H, 7.31; N, 7.82. Found: C, 67.29; H, 7.31; N,
7.78.
(22) (a) Samuel, E.; Rausch, M. D. J. Am. Chem. Soc. 1973, 95, 6263.
(b) References 7 and 9.
(21) Complex 45 failed to give crystals suitable for X-ray analysis; it
was unaffected by ethylene carbonate at room temperature or at 80 °C in
C6D6. Moreover, (o-anisyl)NdCH(Ph) did not displace the imine fragment
in 45 to give the known complex 13b.
(23) Waymouth, R. M.; Bangerter, F.; Pino, P. Inorg. Chem. 1988, 27,
758.
(24) Brown, H. C.; Choi, Y. M.; Narasimhan, S. J. Org. Chem. 1982,
47, 3153.

Stereochemistry in R-Amino Acid Esters and Amides
J. Am. Chem. Soc., Vol. 118, No. 14, 1996 3487
1
(()-16e (MeO2CCH(Me)NH(o-anisyl)). Yield: 198 mg (66%). 1H
NMR (CDCl3): δ 6.88-6.67 (m, 3 H), 6.54 (d, 1 H), 4.73 (v br s, 1
H), 4.17 (q, 1 H), 3.87 (s, 3 H), 3.72 (s, 3 H), 1.52 (d, 3 H). 13C NMR
(CDCl3): δ 174.9, 146.8, 136.3, 121.0, 117.3, 110.0, 109.6, 55.2, 52.0,
51.5, 18.8. Anal. Calcd for C11H15NO3: C, 63.14; H, 7.23; N, 6.69.
Found: C, 63.00; H, 7.27; N, 6.73.
overnight. (S)-(+)- or (S)-(-)-16d-g was >98% pure by H NMR
1
with the exception of (S)-(-)-16d (95% pure by H NMR).
Data for (S)-(-)-16d (MeO2CCH(Me)NHPh): yield ∼3 mg (5%),
21% ee, hexane/ethanol (95/5), flow rate 0.85 mL/min, retention time
(min) 24.4 and 37.3, [R]D ) -85° (c ) 0.040).
Data for (S)-(-)-16e (MeO2CCH(Me)NH(o-anisyl)): yield 39 mg
(58%), 97% ee, hexane/ethanol (95/5), flow rate 0.9 mL/min, retention
time (min) 17.0 and 21.3, [R]D ) -42.8° (c ) 0.150).
Data for (S)-(-)-16f (MeO2CCH(i-Bu)NH(o-anisyl)): yield 49 mg
(61%), 54% ee, hexane/ethanol (99/1), flow rate 0.9 mL/min, retention
time (min) 12.4 and 16.5, [R]D ) -70.9° (c ) 0.170).
(S)-(-)-16f (MeO2CCH(i-Bu)NH(o-anisyl)) (entry 6, Table 4) was
prepared by the general Scheme 3 procedure, except that 20 equiv (6.4
mmol) of ethylene carbonate was used (yield 19 mg (24%, 79% ee),
[R]D ) -69.2° (c ) 0.105)).
Data for (S)-(+)-16g (MeO2CCH(CH2Ph)NH(o-anisyl)): yield 31
mg (34%), 53% ee, hexane/ethanol (80/20), flow rate 0.85 mL/min,
retention time (min) 16.3 and 30.5, [R]D ) +9.6° (c ) 0.650).
(S)-(+)-16g (MeO2CCH(CH2Ph)NH(o-anisyl)) (entry 8, Table 4)
was prepared by the general Scheme 3 procedure, except that 5 equiv
(1.6 mmol) of ethylene carbonate was used (yield 28 mg (31%, 77%
ee), [R]D ) +10.6° (c ) 0.360)).
(S)-(+)-16g (MeO2CCH(CH2Ph)NH(o-anisyl)) (entry 9, Table 4)
was prepared by the general Scheme 3 procedure, except that 20 equiv
(6.4 mmol) of ethylene carbonate was used (yield 25 mg (27%, 85%
ee), [R]D ) +10.4° (c ) 0.590)).
(S)-(+)-16g (MeO2CCH(CH2Ph)NH(o-anisyl)) (entry 10, Table
4) was prepared by the general Scheme 3 procedure, except that 60
equiv (19.2 mmol) of ethylene carbonate was used (yield 13 mg (14%,
89% ee), [R]D ) +10.1° (c ) 0.310)).
Preparation of Methyl r-Amino Acid Esters in Scheme 4. After
addition of the RCH2N(Li)R′ to (S,S)-[EBTHI]ZrMe(OTf) (Scheme 2
procedure), the solution was warmed to room temperature. The solvent
was removed, and the residue containing (S,S)-26 was treated with
benzene (20 mL). The solution was transferred to a vacuum bulb
containing ethylene carbonate (56 mg, 0.640 mmol) and Cp2ZrMe2 (202
mg, 1.28 mmol) in benzene (20 mL); the bulb was sealed and the
solution heated to 70 °C overnight. (S)-(-)-16d,f and (R)-(+)- or (R)-
(()-16f (MeO2CCH(i-Bu)NH(o-anisyl)). Yield: 195 mg (54%).
1H NMR (CDCl3): δ 6.89-6.63 (m, 3 H), 6.56 (d, 2 H), 4.58 (br d, 1
H), 4.12 (q, 1 H), 3.87 (s, 3 H), 3.71 (s, 3 H), 1.83 (m, 1 H), 1.71 (t,
2 H), 1.04 (d, 3 H), 0.97 (d, 3 H). 13C NMR (CDCl3): δ 175.1, 147.0,
136.9, 121.1, 117.3, 110.1, 109.7, 55.4, 54.7, 51.9, 42.3, 24.9, 22.7,
22.1. Anal. Calcd for C14H21NO3: C, 66.91; H, 8.42; N, 5.57.
Found: C, 66.80; H, 8.39; N, 5.67.
(()-16g (MeO2CCH(CH2Ph)NH(o-anisyl)). Yield: 191 mg (46%).
1H NMR (CDCl3): δ 7.33-7.18 (m, 5 H), 6.87-6.68 (m, 3 H), 6.55
(d, 1 H), 4.79 (br s, 1 H), 4.36 (t, 1 H), 3.82 (s, 3 H), 3.65 (s, 3 H),
3.16 (d, 2 H). 13C NMR (CDCl3): δ 173.5, 147.0, 136.4, 136.2, 129.1,
128.3, 126.7, 121.0, 117.4, 110.3, 109.7, 57.6, 55.3, 51.8, 38.7. Anal.
Calcd for C17H19NO3: C, 71.56; H, 6.71; N, 4.91. Found: C, 71.35;
H, 6.65; N, 4.89.
Preparation of Methyl r-Amino Acid Esters by Scheme 2. A
Schlenk flask containing RCH2NHR′ (0.323 mmol) in cold (0 °C) ether
(5 mL) was treated with BuLi (1.6 M, 200 µL, 0.320 mmol) and stirred
for 5 min. The RCH2N(Li)R′ was transferred by cannula to a cold
(-40 °C) THF (5 mL) solution containing (S,S)-[EBTHI]ZrMe(OTf)
(169 mg, 0.325 mmol). After 5 min at -40 °C and 2 h at room
temperature, the red (or orange) solution was evaporated to dryness
and the residue was treated with benzene (20 mL). A separate solution
containing ethylene carbonate (56 mg, 0.640 mmol), Cp2ZrMe2 (202
mg, 1.28 mmol), and benzene (20 mL) was transferred by cannula to
the solution containing 25. After stirring overnight, the pale yellow
solution was treated with MeOH (300 µL) and heated to 80 °C for 4
h or until the â-hydroxyethyl ester 29 was consumed (as detected by
TLC). The solvent was filtered and evaporated to dryness. The residue
was spotted on a Chromatotron plate and eluted with hexanes/ethyl
acetate (25/1). RCH2NHR′ (Rf ≈ 0.6) was separated (typically 15-
30%) from the band that contained (S)-(+)- or (S)-(-)-16a-c (Rf ≈
1
0.3). (S)-(+)- or (S)-(-)-16a-c were >98% pure by H NMR.
1
(-)-16e,g were >98% pure by H NMR.
(S)-(+)-16a (MeO2CCH(Ph)NHPh) was prepared by the general
Scheme 2 procedure, but with the following modifications. After 5
min at -40 °C and 2 h at room temperature, the orange solution
changed immediately to cherry red upon addition of ethylene carbonate
(56 mg, 0.640 mmol) and Cp2ZrMe2 (201 mg, 0.800 mmol) in THF
(10 mL). After stirring overnight, the solvent was evaporated and the
residue was treated with benzene (20 mL) (yield 46 mg (60%), >98%
ee, hexane/ethanol (95/5), flow rate 0.85 mL/min, retention time (min)
42.4 and 47.5, [R]D ) +68.3° (c ) 0.315)).
(S)-(+)-16a (MeO2CCH(Ph)NHPh) (entry 2, Table 4) was pre-
pared by the general Scheme 2 procedure, except that 20 equiv (6.4
mmol) of ethylene carbonate was used (yield 18 mg (23%), >98% ee
(determined by [R]measured in THF), [R]D ) +70.3° (c ) 0.175)).
Data for (S)-(-)-16b (MeO2CCH(Ph)NH(o-anisyl)): yield 58 mg
(67%), 96% ee, hexane/ethanol (100/0) for 45 min, flow rate 1.0 mL/
min and then changed to hexane/ethanol (95/5), flow rate 0.85 mL/
min, retention time (min) 60.3 and 63.9, [R]D ) -35.7° (c ) 0.070).
Data for (S)-(-)-16c (MeO2CCH(o-anisyl)NH(o-anisyl)): yield 51
mg (53%), 98% ee, hexane/ethanol (95/5), flow rate 1.0 mL/min,
retention time (min) 11.6 and 25.4, [R]D ) -52.1° (c ) 0.165).
Data for (S)-(-)-16d (MeO2CCH(Me)NHPh): yield 39 mg (68%,
22% ee), [R]D ) -92.8° (c ) 0.118).
Data for (R)-(+)-16e (MeO2CCH(Me)NH(o-anisyl)): yield 39 mg
(58%, 56% ee), [R]D ) +38.1° (c ) 0.150).
Data for (S)-(-)-16f (MeO2CCH(i-Bu)NH(o-anisyl)): yield 41 mg
(51%, 14% ee), [R]D ) -72.7° (c ) 0.138).
Data for (R)-(-)-16g (MeO2CCH(CH2Ph)NH(o-anisyl)): yield 45
mg (49%, 52% ee), [R]D ) -11.7° (c ) 0.770).
(R)-(-)-16g (MeO2CCH(CH2Ph)NH(o-anisyl)) (entry 12, Table
4) was prepared by the general Scheme 4 procedure, except that 20
equiv (6.4 mmol) of ethylene carbonate was used (yield 31 mg (34%,
54% ee), [R]D ) -11.9° (c ) 0.530)).
Preparation of Methyl r-Amino Acid Esters by Scheme 5. After
addition of the RCH2N(Li)R′ to (S,S)-[EBTHI]ZrMe(OTf) (Scheme 2
procedure), the solution was stirred at -40 °C for 1 h. In a separate
flask, a cold (-40 °C) THF (20 mL) solution containing ethylene
carbonate (56 mg, 0.640 mmol) and Cp2ZrMe2 (202 mg, 1.28 mmol)
was transferred by cannula to the flask containing (S,S)-26; the solution
was stirred for 15 min at -40 °C before warming to room temperature
and stirring overnight. The solvent was removed, and the residue was
treated with benzene (20 mL) and MeOH (900 µL). (S)-(-)-16c was
(S)-(-)-16c (MeO2CCH(o-anisyl)NH(o-anisyl)) (entry 4, Table
4) was prepared by the Scheme 2 procedure, except that 20 equiv (6.4
mmol) of ethylene carbonate was used (yield 34 mg (35%, 92% ee),
[R]D ) -52.3° (c ) 0.065)).
1
>98% pure by H NMR.
Data for (S)-(-)-16c (MeO2CCH(o-anisyl)NH(o-anisyl)): yield 41
mg (43%, >99% ee), [R]D ) -57.6° (c ) 0.085).
Preparation of Methyl r-Amino Acid Esters by Scheme 3. After
addition of the RCH2N(Li)R′ to (S,S)-[EBTHI]ZrMe(OTf) (Scheme 2
procedure), the solution was warmed to room temperature. The solvent
was removed, and the residue containing (S,S)-26 was treated with
benzene (20 mL). The solution was transferred to a vacuum bulb and
heated to 70 °C overnight. The solution was cooled to room
temperature (1 h) and treated with ethylene carbonate (56 mg, 0.640
mmol) and Cp2ZrMe2 (202 mg, 1.28 mmol) in benzene (20 mL); the
bulb was sealed, and the solution was stirred at room temperature
Preparation of rac-25g. A solution containing rac-[EBTHI]ZrMe2
(318 mg, 0.820 mmol) and THF (10 mL) was cooled to -78 °C and
treated with TfOH (73 µL, 0.820 mmol). The pale yellow solution
was warmed to room temperature and stirred for 1 h, followed by
recooling to -78 °C. In a separate flask BuLi (1.6 M, 513 µL, 0.820
mmol) was added to a cold (0 °C) Et2O (5 mL) solution containing
Ph(CH2)2NH(o-anisyl) (189 mg, 0.820 mmol), and the solution was
stirred for 5 min. The Ph(CH2)2NLi(o-anisyl) was transferred by
cannula to the rac-[EBTHI]ZrMe(OTf); the yellow solution was stirred

3488 J. Am. Chem. Soc., Vol. 118, No. 14, 1996
Gately and Norton
for 0.5 h at -40 °C before warming to room temperature. The solvent
was evaporated, and the residue was dissolved in benzene and heated
to 70 °C overnight. The solution was filtered by cannula with benzene
washes (2 × 5 mL), and the filtrate was evaporated to yield an orange
solid. 1H NMR of the product mixture showed only one diastereomer.
Yield: 434 mg (90% pure by 1H NMR). An analytically pure sample
was obtained from benzene/THF/hexanes (5/1/100). 1H NMR
(C6D6): δ 7.52 (d, 2 H), 7.31 (t, 2 H), 7.18 (t overlapped with residual
C6H6, 1 H), 6.71 (t, 1 H), 6.32 (t, 1 H), 6.26 (d, 1 H), 6.13 (d, J ) 2.99
Hz, 1 H), 5.93 (d, 1 H), 5.30 (d, J ) 3.01 Hz, 1 H), 5.18 (d, J ) 2.77
Hz, 1 H), 4.42 (d, J ) 2.74 Hz, 1 H), 3.59-3.31 (m, 2 H), 3.20 (s, 3
H), 2.72-2.07 (m, 14 H), 1.99-1.78 (m, 2 H), 1.69-1.26 (m, 5 H).
Anal. Calcd for C35H39NOZr: C, 72.37; H, 6.77; N, 2.41. Found: C,
72.14; H, 6.61; N, 2.47.
1H NMR of rac-28g and rac-31g (Scheme 7). A 5 mm NMR tube
was charged with rac-25g (11.4 mg, 0.0196 mmol) and ∼0.5 mL of
C6D6. After complete dissolution of rac-25g, 1.3 equiv of ethylene
carbonate (2.3 mg, 0.0263 mmol) was added to the orange solution. In
<1 h, rac-25g was consumed and a pale yellow solution containing
rac-28g/rac-31g (84/16, 68% de) was obtained as shown by its 1H
NMR. Selected 1H NMR resonances of rac-31g (C6D6): δ 6.63
(aromatic d, J ) 7.93 Hz, 1 H), 5.82 (CH indenyl d, J ) 2.44 Hz, 1
H), 5.13 (CH indenyl d, J ) 2.38 Hz, 1 H), 4.91 (CH methine d, J )
8.05 Hz, 1 H). Similar treatment of rac-25g (11.7 mg, 0.0201 mmol)
with 9.3 equiv of ethylene carbonate (16.4 mg, 0.186 mmol) gave, after
<10 min, a 94/6 (88% de) rac-28g/rac-31g ratio. 1H NMR of rac-
28g (C6D6): δ 7.54 (d, 2 H), 7.24 (t, 2 H), 7.15 (aromatic m overlapped
with CH indenyl resonance and residual C6H6, 2 H), 6.96 (t, 1 H),
6.70 (d, J ) 7.74 Hz, 1 H), 6.57 (d, J ) 6.75 Hz, 1 H), 6.47 (t, 1 H),
5.55 (d, J ) 2.47 Hz, 1 H), 5.44 (d, J ) 2.94 Hz, 1 H), 5.13 (d, J )
2.56 Hz, 1 H), 5.03 (CH methine dd, J ) 11.3 Hz, 1 H), 3.90-3.72
(m, 2 H), 3.53 (OCH3 s overlapped with m, 3 H), 3.48-3.04 (m
overlapped with ethylene carbonate resonance, 4 H), 2.94-2.56 (m, 8
H), 2.47-2.06 (m, 9 H), 1.60-1.09 (m, 6 H).
(()-23 (t-BuNHCOCH(Ph)NH(o-anisyl)). A 100 mL Schlenk flask
was charged with 13b (217 mg, 0.5 mmol), benzene (20 mL), and
t-BuNCO (69 µL, 0.6 mmol). After stirring overnight, the solution
was treated with MeOH (1 mL) and stirred for 4 h. The solvent was
removed, and the residue was treated with CH2Cl2 (20 mL) and filtered.
The residue from the filtrate was spotted on a Chromatotron plate and
eluted with ethyl acetate/hexanes (2/1); evaporation of solvent afforded
a white solid that was washed with hexanes. Yield: 78 mg (52%). 1H
NMR (CDCl3): δ 7.48-7.31 (m, 5 H), 6.89-6.76 (m, 3 H), 6.73 (br
s, 1 H), 6.58 (d, 1 H), 5.01 (v br s, 1 H), 4.54 (s, 1 H), 3.82 (3 H, s),
1.31 (s, 9 H). 13C NMR (CDCl3): δ 170.3, 147.0, 139.1, 136.6, 128.9,
128.1, 127.2, 121.0, 118.4, 111.5, 109.2, 65.1, 55.2, 50.9, 28.4. Anal.
Calcd for C19H24N2O2: C, 73.05; H, 7.74; N, 8.97. Found: C, 72.69;
H, 7.80; N, 8.97.
(()-24 (H2NCOCH(Ph)NH(o-anisyl)). Preparation of (()-24 was
carried out as described above for (()-23, except that Me3SiNCO (160
µL, 1.0 mmol) replaced t-BuNCO. The residue from the filtrate was
spotted on a Chromatotron plate and eluted with ethyl acetate/hexanes
(7/1); evaporation of solvent afforded a white solid that was washed
with hexanes. Yield: 54 mg (42%). 1H NMR (CDCl3): δ 7.53-7.36
(m, 5 H), 6.89-6.66 (m, 3 H), 6.61 (br s, 1 H), 6.58 (d, 1 H), 5.53 (br
s, 1 H), 4.71 (s, 1 H), 3.82 (s, 3 H). 13C NMR (CDCl3): δ 174.1,
147.1, 138.5, 136.3, 129.2, 128.6, 127.4, 121.2, 118.7, 111.3, 109.5,
64.1, 53.4. Anal. Calcd for C15H16N2O2: C, 70.29; H, 6.29; N, 10.93.
Found: C, 70.24; H, 6.29; N, 10.87.
and the solution was filtered. The residue from the filtrate was spotted
on a Chromatotron plate and eluted with ethyl acetate/hexanes (7/1);
evaporation of solvent afforded a white solid that was washed with
hexanes (yield 264 mg (62%), 92% ee, hexane/2-propanol (90/10), flow
rate 0.6 mL/min, retention time (min) 14.1 and 21.8, [R]D ) +76.5° (c
) 0.780)). 1H and 13C NMR data for (()-19 were reported earlier;9
1
(S)-(+)-19 was >98% pure by H NMR.
(S)-(+)-20 (H2NCOCH(Ph)NHPh) (Scheme 8) was prepared like
(S)-(+)-19, except that neat Me3SiNCO (262 µL, 1.5 mmol) replaced
t-BuNCO. The residue was spotted on a Chromatotron plate and eluted
with hexanes/ethyl acetate (2/1); after evaporation, the white solid was
washed with hexanes (yield 173 mg (51%), 80% ee, hexane/2-propanol
(85/15), flow rate 0.6 mL/min, retention time (min) 37.6 and 48.2, [R]D
) 146.9° (c ) 0.770)). 1H and 13C NMR data for (()-20 were reported
1
earlier;9 (S)-(+)-20 was >98% pure by H NMR.
(S)-(-)-23 (t-BuNHCOCH(Ph)NH(o-anisyl)) (Scheme 8) was
prepared like (S)-(+)-19b, but with the following modifications. A
solution containing PhCH2NH(o-anisyl) (69 mg, 0.323 mmol) and Et2O
(5 mL) was cooled to 0 °C and treated with BuLi (1.6 M, 200 µL,
0.320 mmol), and the solution was stirred for 5 min. The PhCH2NLi-
(o-anisyl) was transferred by cannula to a cold (-40 °C) solution
containing (S,S)-[EBTHI]ZrMe(OTf) (169 mg, 0.325 mmol) in THF
(5 mL); the flask that contained PhCH2NLi(o-anisyl) was rinsed with
Et2O (2 mL). After 5 min at -40 °C and overnight at room
temperature, the orange-red solution was treated with neat t-BuNCO
(46 µL, 0.400 mmol) and stirred overnight. The pale yellow solution
was treated with MeOH (1 mL) and stirred for 4 h (yield 30 mg (31%),
>99% ee, hexane/ethanol (95/5), flow rate 0.9 mL/min, retention time
(min) 8.0 and 10.4, [R]D ) -40.0 (c ) 0.123)). (S)-(-)-23 was >98%
1
pure by H NMR.
(S)-(-)-24 (H2NCOCH(Ph)NH(o-anisyl)) (Scheme 8) was prepared
in the same way as (S)-(-)-23, except that neat Me3SiNCO (54 µL,
0.400 mmol) replaced t-BuNCO (yield 22 mg (25%), 98% ee, hexane/
ethanol (80/20), flow rate 0.8 mL/min, retention time (min) 16.6 and
19.5, [R]D ) -100.9° (c ) 0.095)). (S)-(-)-24 was >98% pure by
1H NMR.
Compound rac-25b was prepared like rac-25g, but with the
following modifications. PhCH2NH(o-anisyl) (175 mg, 0.820 mmol)
replaced Ph(CH2)2NH(o-anisyl). After the PhCH2NLi(o-anisyl) was
transferred by cannula to the rac-[EBTHI]ZrMe(OTf), the resulting
yellow solution was warmed to room temperature and the solution
turned red. After 1 h, the solvent was evaporated and the orange red
solid was dissolved in benzene (25 mL); the solution was filtered by
cannula, the solid was washed with benzene (2 × 5 mL), and the filtrate
was evaporated to yield crude rac-25b. 1H NMR of the product mixture
showed only one diasteromer. Yield: 440 mg (85% pure by 1H NMR).
An analytically pure sample was obtained from benzene/THF/hexanes
(5/1/100). 1H NMR (C6D6): δ 7.48 (d, 2 H), 7.39 (t, 2 H), 7.07 (t, 1
H), 6.98 (t, 1 H), 6.68 (d, 1 H), 6.53 (t, 1 H), 6.48 (d, 1 H), 5.41 (d, J
) 3.07 Hz, 1 H), 5.18 (d, J ) 2.80 Hz, 1 H), 4.84 (d, J ) 3.09 Hz, 1
H), 4.70 (d, J ) 2.76 Hz, 1 H), 3.78 (s, 1 H), 3.27 (s, 3 H), 2.67-1.94
(m, 15 H), 1.69-1.18 (m, 5 H).
Compound rac-25a was prepared like rac-25b, but with the
following modifications. PhCH2NHPh (59 mg, 0.320 mmol) replaced
PhCH2NH(o-anisyl). After the PhCH2NLiPh was transferred by cannula
to the rac-[EBTHI]ZrMe(OTf), the yellow solution turned red as it
warmed to room temperature. After 1 h, the solvent was evaporated
to yield an orange solid. Compound rac-25a was ∼38% pure (the
remaining 62% was a mixture of LiOTf, THF, and ca. 10-15%
PhNHCH2Ph) and was used without further purification. Selected 1H
NMR resonances of rac-25a (THF-d8): δ 6.28 (aromatic t, 1 H), 5.51
(CH indenyl d, J ) 2.49 Hz, 1 H), 5.40 (CH indenyl d, J ) 2.55 Hz,
1 H), 5.26 (CH indenyl d, J ) 2.61 Hz, 1 H), 4.28 (CH indenyl d, J )
2.61 Hz, 1 H), 3.55 (CH methine s, 1 H).
(S)-(+)-19 (t-BuNHCOCH(Ph)NHPh) (Scheme 8). A solution
containing (S,S)-[EBTHI]ZrMe2 (579 mg, 1.5 mmol) and THF (10 mL)
was cooled to -78 °C and treated with TfOH (133 µL, 1.5 mmol).
The pale yellow solution was warmed to room temperature and stirred
for 1 h followed by recooling to -78 °C. In a separate flask BuLi
(2.0 M, 750 µL, 1.5 mmol) was added to a cold (0 °C) Et2O (5 mL)
solution containing PhCH2NHPh (275 mg, 1.5 mmol), and the solution
was stirred for 5 min. The PhCH2NLi(Ph) was transferred by cannula
to the (S,S)-[EBTHI]ZrMe(OTf), and the solution was stirred for 0.5 h
at -78 °C and then warmed to room temperature and stirred overnight.
Neat t-BuNCO (177 µL, 1.5 mmol) was added, and the resulting red
solution was stirred for 1 h and then treated with MeOH (1 mL). The
solvent was removed, the residue was treated with CH2Cl2 (20 mL),
1H NMR of rac-35 (Scheme 9). A 5 mm NMR tube was charged
with rac-25b (12.0 mg, 0.0210 mmol), THF-d8 (∼0.5 mL), and 4.1
equiv of neat t-BuNCO (6.2 mg, 0.0625 mmol); after 6 days, rac-25b
was consumed. Selected 1H NMR resonances of rac-35 (THF-d8): δ
7.47 (aromatic d, 2 H), 6.53 (aromatic t, 1 H), 6.28 (aromatic d, 1 H),
5.98 (aromatic d, 1 H), 5.88 (CH indenyl d, J ) 2.61 Hz, 1 H), 5.84
(CH indenyl d, J ) 2.58 Hz, 1 H), 5.80 (CH indenyl d, J ) 2.65 Hz,
1 H), 5.61 (CH indenyl d, J ) 2.70 Hz, 1 H), 5.26 (CH methine s, 1

Stereochemistry in R-Amino Acid Esters and Amides
J. Am. Chem. Soc., Vol. 118, No. 14, 1996 3489
H), 4.33 (OCH3 s, 3 H), 1.19 (t-Bu s, 9 H). The minor diastereomer
rac-37 was not detected by H NMR.
Similar treatment of rac-25a (0.02 mmol) with n equiv of 2-butyne
(mg, mmol) gave the following de’s of rac-41a: 4.2 equiv (4.5 mg,
0.083 mmol), 28% de; 13 equiv (14 mg, 0.26 mmol), 46% de; 32 equiv
(35 mg, 0.64 mmol), 60% de.
1
(S)-(+)-19 (t-BuNHCOCH(Ph)NHPh) (Scheme 10) was prepared
like (S)-(+)-19, but with the following modifications. The solution
that contained (S,S,R)-25a (0.320 mmol, 20 mM) was treated with a
dilute benzene solution of t-BuNCO (0.256 mmol, 32 mM); metha-
nolysis and workup gave (S)-(+)-19 in 45% yield (36% ee, hexane/
2-propanol (92/8), flow rate 0.85 mL/min, retention time (min) 11.8
and 20.3).
Preparation of 44. A solution containing Cp2ZrMe2 (3.02 g, 12
mmol) and THF (40 mL) was cooled to -78 °C and treated with TfOH
(1.06 mL, 12 mmol). The pale yellow solution was warmed to room
temperature, stirred for 1 h, and again cooled to -78 °C. In a separate
flask BuLi (2.0 M, 6 mL) was added to a cold (0 °C) Et2O (40 mL)
solution containing (o-anisyl)CH2NHCH2Ph (2.73 g, 12 mmol); the
solution was stirred for 5 min. The pink solution containing (o-anisyl)-
CH2N(Li)CH2Ph was transferred by cannula to the Cp2ZrMe(OTf),
stirred for 0.5 h at -78 °C, and warmed to room temperature. The
solvent was evaporated from the red solution, the residue was treated
with benzene (75 mL), the solution was filtered by cannula, and the
residue was washed with benzene (25 mL). The filtrate was evaporated,
and the residue was treated with hexanes (100 mL) to give a tan
precipitate. The solid was filtered by cannula, washed with hexanes
(2 × 30 mL), and dried overnight under vacuum. Yield: 3.43 g (62%).
1H NMR (C6D6): δ 7.39 (d, 1 H), 7.31-7.09 (m, 7 H), 6.59 (d, 1 H),
5.75 (Cp s, 10 H), 4.53 (CH2 s, 2 H), 4.35 (CH2 s, 2 H), 3.25 (OCH3
s, 3 H), 0.32 (ZrCH3 s, 3 H). 13C NMR (C6D6): δ 158.1, 142.4, 129.5,
128.6, 128.4 (DEPT), 128.0 (DEPT), 127.4 (DEPT), 126.5, 120.4,
110.2, 110.0 (Cp), 59.5 (CH2, DEPT), 54.5 (OCH3, DEPT), 54.0 (CH2,
DEPT), 20.9 (ZrCH3, DEPT). Anal. Calcd for C26H29NOZr: C, 67.49;
H, 6.32; N, 3.03. Found: C, 67.20; H, 6.14; N, 2.99.
1H and 13C NMR of 45 (Scheme 14). A 5 mm NMR tube was
charged with 44 (50 mg, 0.108 mmol) and C6D6 (∼0.5 mL). The
sample was sealed, and the tan solution was heated to 80 °C for 1 h to
give a deep red solution. 1H NMR showed the major (E)-45 (84%)
and minor (Z)-45 (16%) isomers. 1H NMR of (E)-45 (C6D6): δ 8.09
(CHdN s, 1 H), 7.43 (d, 1 H), 7.23-7.07 (aromatic m overlapped
with aromatic resonances of (Z)-45), 5.71 (Cp s, 10 H), 4.56 (CH2 s,
2 H), 3.69 (OCH3 s, 3 H). Selected 1H NMR of (Z)-45 (C6D6): δ 7.95
(CHdN s, 1 H), 5.85 (Cp s, 10 H), 4.52 (CH2 s, 2 H), 3.79 (OCH3 s,
3 H). 13C NMR of (E)-45 (C6D6): δ 176.0 (CHdN, DEPT), 142.5,
141.7, 140.3, 129.5, 128.8, 128.2 (DEPT), 128.1 (DEPT), 127.7, 127.1,
123.2, 111.0 (Cp of (Z)-45), 109.9 (Cp of (E)-45), 61.6 (CH2, DEPT),
59.2 (OCH3, DEPT).
1H NMR of rac-38 and rac-39 (Scheme 11). A 5 mm NMR tube
was charged with rac-25g (11.7 mg, 0.0201 mmol) and THF-d8 (∼0.5
mL). After complete dissolution of rac-25g, 26 equiv of neat t-BuNCO
(45 mg, 0.454 mmol) was added to the orange solution. In <5 min,
the rac-25g was consumed and a 95/5 (90% de) rac-38/rac-39 product
1
1
ratio was shown by H NMR. Selected H NMR resonances of rac-
38 (THF-d8): δ 7.47 (d, J ) 7.14 Hz, 2 H), 7.12 (t, 2 H), 7.03 (t, 1 H),
6.94 (d, 1 H), 6.75 (d, 1 H), 6.38 (t, 1 H), 6.17 (d, 1 H), 5.69 (apparent
q, two overlapping d from CH indenyl resonances, J ) 2.71 Hz, and
J ) 2.71 Hz, 2 H), 5.57 (d, J ) 2.72 Hz, 1 H), 5.45 (d, J ) 2.70 Hz,
1 H), 4.44 (CH methine dd, J ) 7.68 Hz, 1 H), 4.23 (OCH3 s, 3 H),
3.46-3.40 (dd, 1 H), 1.41 (t-Bu s, 9 H). Similar treatment of rac-25g
(10.2 mg, 0.0176 mmol) with 1.3 equiv of t-BuNCO (2.2 mg, 0.0222
mmol) in THF-d8 gave, after 3 h, a 64/36 (28% de) rac-38/rac-39
1
product ratio. Selected H NMR resonances of rac-39 (THF-d8): δ
7.51 (aromatic d overlapped with rac-38 d, J ) 7.36 Hz, 2 H), 5.92
(CH indenyl d, J ) 2.86 Hz, 1 H), 5.79 (CH indenyl d, J ) 2.50 Hz,
1 H), 5.46 (CH indenyl d, J ) 2.71 Hz, 1 H), 4.34 (OCH3 s, 3 H), 1.27
(t-Bu s, 9 H).
1H NMR of rac-40 (Reaction 16). A 5 mm NMR tube was charged
with rac-25a (∼11 mg, 0.02 mmol), THF-d8 (∼0.5 mL), and neat Me3-
SiNCO (2.8 mg, 0.024 mmol). After 5 min, the 1H NMR showed rac-
1
40 (ca. 90% de). Selected H NMR resonances of rac-40 (THF-d8):
δ 7.47 (aromatic d, 2 H), 6.13 (CH indenyl d, 1 H), 5.93 (CH indenyl
d, 1 H), 5.88 (aromatic d, 2 H), 5.56 (CH indenyl resonance overlapped
with CH methine s, 2 H), 5.24 (CH indenyl d, 1 H), 0.02 (s, 9 H).
1H NMR of rac-41a (Reaction 16). A 5 mm NMR tube was
charged with rac-25a (∼11 mg, 0.02 mmol) and ∼0.5 mL of THF-d8.
After complete dissolution of rac-25a, 4.2 equiv of 2-butyne (4.5 mg,
0.083 mmol) was added to the orange-red solution. After 24 h, 2-butyne
was consumed and the red solution showed rac-41a in 24% de.
In a similar fashion, a 5 mm NMR tube was charged with rac-25a
(∼11 mg, 0.02 mmol) and neat 2-butyne (∼0.5 mL, 6.4 mmol). After
standing overnight, the excess 2-butyne was evaporated and the orange
Acknowledgment. We thank Dr. Patricia A. Goodson
(University of Wyoming) for the result in ref 8. We also thank
Professor Stephen L. Buchwald for a copy of Dr. Robert
Grossman’s Ph.D. thesis. We are also grateful to Professor
Albert I. Meyers for the use of his chiral HPLC apparatus. This
work was supported by NSF Grant CHE-9120454.
1
residue was dissolved in THF-d8. The H NMR showed rac-41a in
>99% de. Selected 1H NMR resonances of rac-41a (THF-d8): δ 6.79
(CH indenyl d, 1 H), 5.92 (CH indenyl d, 1 H), 5.78 (aromatic d, 2 H),
5.34 (CH indenyl d, 1 H), 5.05 (CH indenyl d, 1 H), 4.89 (CH methine
br s, 1 H), 1.92 (CH3 s overlapped with indenyl resonances), 1.24 (CH3
s overlapped with indenyl resonances).
JA953893H