Kinetic enolate formation by lithium arylamide: Effects of basicity on selectivity

Article abstract of DOI:10.1021/jo0263465

Five ketones R₁COCH₂R₂ (1a-e) were enolized in tetrahydrofuran solvent employing lithium arylamides with different electron-withdrawing and -donating substituents on the phenyl ring (4a-e). Enolate selectivity is unaffecte

Full text of DOI:10.1021/jo0263465

SCHEME 1
Kin etic En ola te F or m a tion by Lith iu m
Ar yla m id e: Effects of Ba sicity on
Selectivity
Linfeng Xie,* Keith Vanlandeghem,
Kurt M. Isenberger, and Carolyn Bernier
Department of Chemistry, University of Wisconsin,
Oshkosh, Wisconsin 54901
xie@uwosh.edu
removal of the less hindered proton. In light of the
seemingly similar steric hindrance in 2 and 3, the
completely opposite E- and Z-selectivity is both surprising
and interesting. We hypothesized that the enhanced
Z-selectivity displayed by 3 was due to its much more
stabilized nitrogen as compared to 2, which resulted in
a highly loose transition state leading to the Z preference.
We now report further evidence supporting this hypoth-
esis.
To make unbiased comparisons on the electronic
nature of amide bases, we must maintain a consistent
steric effect among the amide bases used. We chose
various substituted lithium anilides (X-C6H4NHLi, 4a :
X ) H; 4b: X ) p-CH3O; 4c: X ) o-Cl; 4d : X ) m-Cl;
4e: X ) p-CH3CO2; 4f: X ) p-CN) for this purpose. The
steric influence on the reactive nitrogen center imposed
by the ring substituents, especially those on the meta and
para positions, is expected to be negligible. This allows
us to assess the electronic effects of substituents on
enolate selectivity. As a comparison, we also included in
our study two other highly stabilized nitrogen bases:
lithium 2,4,6-trichloroanilide (5) and lithium diphenyl-
anilide (6). 4-6 were prepared by treating their corre-
sponding amine precursors with butyllithium in THF.
They were then allowed to react with ketones 1a -e and
the resulting enolates were quenched with chlorotri-
methylsilane to yield trimethylsilyl enol ethers (Scheme
1). The enol ethers, after workup, were subject to GCMS
and NMR analysis and the isomeric enolate ratios
determined. The E/ Z ratios derived from 1a -e by
lithium arylamides 4a -f are listed in Table 1.
We noted from Table 1, extries 1, 7, and 13, that the
unsubstituted lithium anilide (4a ) gave better Z-selectiv-
ity than LDA or lithium N-isopropylanilide used in our
previous study in the same solvent. This is presumably
due to the less steric hindrance in 4a , and can be
explained by the Ireland’s transition state model (Scheme
2).3 In the transition states involving 4a , the less
hindered hydrogen is expected to favor the axial position
while the aryl group occupies the equatorial position. The
less sterically hindered interaction, H T R2, results in a
lower energy for the transition state Bq leading to
Z-enolate preference. It is also clear from Table 1 that
4b-d gave nearly identical E/ Z selectivity to that by 4a
for the three ketones (1a , 1b, and 1d ) studied. We have
previously concluded that E/ Z selectivity is a result of
balancing steric (bulkiness of the amide base and R1 and
R2) and electronic factors (electron density on the nitro-
Received August 23, 2002
Abstr a ct: Five ketones R1COCH2R2 (1a -e) were enolized
in tetrahydrofuran solvent employing lithium arylamides
with different electron-withdrawing and -donating substit-
uents on the phenyl ring (4a -e). Enolate selectivity is
unaffected by a moderate electron-releasing or -withdrawing
group, but significantly enhanced by strong electron-
withdrawing substituents to yield predominantly Z-enolate.
Outstanding selectivity was achieved with lithium trichloro-
anilide (5) and lithium diphenylamide (6). The results are
rationalized in terms of electronic effects on the tightness
of the transition states.
Enolates are an important reactive intermediate that
has found wide applications in organic synthesis.1 The
geometry of enolates and related silyl enol ethers (E or
Z) is essential to the control of stereochemistry (diastereo-
or enantioselectivity) in the subsequent reactions. In an
earlier effort to understand the role of steric and elec-
tronic effects of lithium amides in governing enolate E/ Z
selectivity, we investigated kinetic enolization of simple
ketones (1a : R1 ) Et, R2 ) Me; 1b: R1 ) i-Pr, R2 ) Me;
1c: R1 ) Ph, R2 ) Me; 1d : R1 ) Me, R2 ) Et; 1e: R1 )
i-Bu, R2 ) i-Pr) in tetrahydrofuran (THF) with several
alkyl-, aryl-, and trimethylsilyl-substituted lithium amide
bases.2 The kinetic E/ Z selectivity varied widely with
different substitution patterns on the nitrogen. Excellent
E-selectivity was achieved with lithium tert-butyltri-
methylsilylamide (2) at room temperature while very
high Z preference was obtained by using lithium tri-
methylsilylanilide (3) at -78 °C. We established that
these were the results of kinetic enolization, as the
enolization of unsymmetric 2-pentanone and 2-methyl-
3-pentanone yielded the major enolate derived from the
(1) For recent reviews, see: (a) Evans, D. A.; Dart, M. J .; Duffy, J .
L.; Yang, M. G. J . Am. Chem. Soc. 1996, 118, 4322. (b) Braun, M.;
Sacha, H. J . Prakt. Chem. Chem. Ztg. 1993, 335, 653. (c) Heathcock,
C. H. In Asymmetric Synthesis; Morrison, J . D., Ed.; Academic Press:
New York, 1984; Vol. 3, p 111. (d) Masamune, S.; McCarthy, P. A. In
Macrolide Antibiotics; Omura, S., Ed.; Academic Press: New York,
1984; p 127.
(2) Xie, L.; Isenberger, K. M.; Held, G.; Dahl, L. M. J . Org. Chem.
1997, 62, 7516.
(3) Ireland, R. E.; Mueller, R. H.; Willard, A. K. J . Am. Chem. Soc.
1976, 98, 2868.
10.1021/jo0263465 CCC: $25.00 © 2003 American Chemical Society
Published on Web 12/11/2002
J . Org. Chem. 2003, 68, 641-643
641

TABLE 1. Kin etic En ola te Selectivity fr om Keton es
TABLE 2. E/Z Selectivity fr om Keton es 1a -e by 3, 5,
1a -e by Ba ses 4a -f a t 0 °C in THF a
a n d 6 in THF a
entry
ketones
bases
E:Z
entry
ketones
bases
E:Z
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
1a
1a
1a
1a
1a
1a
1b
1b
1b
1b
1c
1c
1d
1d
1d
1d
1d
1d
1e
1e
4a
4b
4c
4d
4e
4f
4a
4b
4e
4f
21:79
22:78
24:76
25:75
10:90
8:92
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
1a
1b
1c
1d
1e
1a
1a
1b
1b
1c
1c
1d
1d
1e
1e
5
5
5
5
5
3
6
3
6
3
6
3
6
3
6
11:89b
2:98b
6:94b
24:76b
6:94b
7:93c
11:89b
8:92b
4:96b
2:98b
2:98
5:95c
5:95c
0:100c
0:100c
0:100c
10:90c
4:96c
4e
4f
2:98
4a
4b
4c
4d
4e
4f
50:50c
40:60c
44:56c
44:56c
26:74c
21:79c
14:86
9:91
2:98c
0:100c
a
In the case of unsymmetric ketones, deprotonation occurred
b
mainly on the less substituted carbon. The reactions were run
at 0 °C. c The reactions involving 3 and 6 were both carried out at
-78 °C to achieve the best selectivity.
4e
4f
a
All results were obtained with a base-to-ketone ratio of 1.1:1.
b
drawing effects of the chlorine diminish, thus resulting
in a very small selectivity change.
A small amount of the more substituted regioisomer was also
obtained. c Deprotonation from the primary carbon constitutes
g80% of the overall yield.
On the other hand, strong electron-withdrawing CH3-
CO2 (4e) and CN (4f) groups on the para position of the
phenyl ring enhance the Z-selectivity significantly. Since
these groups on the para position are expected to cause
negligible change in the steric hindrance of the base, the
enhanced Z-selectivity must come from the substantial
electronic delocalization of the anionic nitrogen through
the resonance effect. This is indeed consistent with our
previous report that substantial stabilization of the
electron density on the amide nitrogen leads to a signifi-
cantly loose transition state thus favoring the Z-enolate.2
Table 2 summarizes E/ Z selectivity for 1a -e by 5 and
6 in THF, along with the selectivity by 3 that we reported
previously. It is clear that both 5 and 6 produced very
high Z preference, with lithium diphenylamide giving the
best Z-selectivity. If the steric hindrance of the amide
bases were the only factor governing the E/ Z selectivity,
then 3, 5, and 6 would all seem more hindered than the
other bases used in this study and be expected to give
higher E-selectivitysopposite of what we observed. Other
examples of steric effect on enolate selectivity was
previously reported.4 The enhanced Z-selectivity is con-
sistent with the fact that the nitrogen in 5 and 6 is
extensively stabilized, just as in 3, by the strong electron-
withdrawing nature of the three Cl’s in 5 and the phenyl
groups in 6. Comparisons between selectivity by 4c, 4d ,
and 5 provide further evidence that sufficient electronic
delocalization on the nitrogen (three Cl’s vs one Cl) is
necessary to achieve significantly enhanced Z-selectivity.
Strong evidence also came from the comparisons of the
substituent effects on the E/ Z selectivity along the series
LDA f lithium N-isopropyltrimethylsilylamide f 3 and
the series LDA f lithium N-isopropylanilide f 6. While
the first isopropyl replacement in LDA by a trimethylsilyl
or phenyl group yields nearly identical E/ Z selectivity,
the second replacement by a phenyl group (leading to 3
and 6, respectively) results in a dramatic Z enhancement.
Although these results are in complete agreement with
SCHEME 2
gen) in the transition state, and strong electron-with-
drawing substituents on the nitrogen result in a loose
transition state favoring the Z-enolate. In the current
case, the similar selectivities given by 4b and 4a are not
particularly surprising, as the negative nitrogen would
resist any further electron donation from the p-OCH3.
Moreover, the p-methoxy group should cause no change
in steric hindrance. Thus, the perturbation in both steric
and electronic nature of 4b is negligibly small to change
the energy difference between transition states Aq and
Bq. Likewise, the effect of o- and m-Cl (4c and 4d ) on the
E/ Z ratio is relatively small (compare entries 1, 3, and
4, and 13, 15, and 16). Although a small increase in the
steric hindrance of 4c might be expected due to the o-Cl,
it is conceivable that this effect may be counter-balanced
by the electronic effect of the moderate electron-with-
drawing o-Cl (due to its proximity to the nitrogen and
its ability to coordinate to the lithium counterion). In the
case of m-Cl, both steric and inductive electron-with-
(4) Heathcock, C. H.; Buse, C. T.; Kleschick, W. A.; Pirrung, M. C.;
Sohn, J . E.; Lampe, J . J . Org. Chem. 1980, 45, 1066.
642 J . Org. Chem., Vol. 68, No. 2, 2003

molecular sieves. 1c and 1e were dried over 3 Å molecular sieves
and used without further purification. The identity and purity
of all ketones were confirmed by 1H NMR to indicate no
detectable amount of impurity or water.
a loose transition state due to the strong electronic
delocalization on the nitrogen, it is possible that they may
also result from the change in the aggregation state of
the lithium amides. It is possible that different ag-
gregates (monomers, dimers, or higher aggregates sol-
vated by a different number of solvent molecules) may
become the active deprotonating species when the elec-
tronic property (basicity) of our lithium amides is drasti-
cally altered. Extensive aggregation studies on alkyl-
lithium and lithium amides (particularly LDA) have been
reported by Collum and co-workers.5
Am in es. The precursor amines for the preparation of 4-6
are all commercially available. Unless otherwise noted, liquid
amines (4b-d ) were dried over molecular sieves and used
without further purification. They were kept from any light
1
source as much as possible and their purity was checked by H
NMR and GCMS. Solid amines (4e,f, 5, and 6) were used directly
after their mp was checked for purify and identify.
An ilin e. Aniline (4a ) was distilled under reduced pressure
and dried over molecular sieves before use.
In conclusion, we reported further evidence on the
effect of electronic delocalization in lithium amide bases
on the enolate stereoselectivity. Z-enolate selectivity can
be significantly enhanced by modifying amide bases with
strong electron-withdrawing substituents. Excellent Z-
selectivity can be achieved with lithium diphenylamide,
which is of practical value in organic synthesis.
Gen er a l P r oced u r e of En oliza tion . Lithium amide bases
4-6 were prepared by adding 1.1 mmol of n-butyllithium to a
stirring solution of 1.2 mmol of corresponding amine in 5 mL of
dry THF at 0 °C under nitrogen. The solution was allowed to
stir for 15 min. The solution was then kept at 0 °C or lowered
to -78 °C, depending on the conditions desired. Ketone (1 mmol)
was then slowly added by a gastight syringe, and the reaction
was stirred for an additional 15 min before the enolates were
quenched with 1.2 mmol of freshly distilled chlorotrimethylsilane
(over CaH2). After being stirred for 10 min, the reaction mixture
was allowed to warm to room temperature and poured into 10
mL of pentane in a small separatory funnel. The mixture was
washed twice with 5 mL each of saturated sodium bicarbonate
solution and once with brine solution, and the organic layer was
dried over anhydrous magnesium sulfate. The solution was
filtered and the product ratios determined on GC/MS. At least
two individual experiments (normally 3-4 experiments) for each
entry in Tables 1 and 2 were run to ensure reproducible results.
Usually 2-3 GC injections were made for each experimental run.
To ensure kinetically controlled conditions, we also carried out
the above reactions using a base/ketone molar ratio of 2:1. The
E/Z ratios showed no difference from those obtained with a base/
ketone ratio of 1.1:1.
Exp er im en ta l Section
Gen er a l. Glassware and syringe needles were dried at 140
°C overnight in an oven. Gastight syringes were dried in a
desiccator under vacuum (e0.1 Torr) for at least 1 h before use.
The enolization experiments were carried out in Schlenk flasks
1
under a nitrogen atmosphere. H NMR was obtained on a 270-
MHz FT-NMR spectrometer. The quantitative analysis of tri-
methylsilyl enol ethers was carried out on GC-MS and NMR.
Solven t. Tetrahydrofuran (99.5%) was dried over Na/benzo-
phenone under N2 until a purple color persists, and distilled
immediately prior to use.
n -Bu tyllith iu m . n-Butyllithium (1.6 M in hexane) was
titrated with reagent grade diphenylacetic acid for its exact
concentration. In most cases n-butyllithium concentration was
found to be within 10% of what was labeled on the bottle.
Keton es. 1a , 1b, and 1d were distilled on a B/R Model 8T
micro spinning band distillation system, and dried over 3 Å
Ack n ow led gm en t. We thank the University of
Wisconsin-Oshkosh Faculty Development Program for
the financial support of this project. We also acknowl-
edge the National Science Foundation and the College
of Letters and Science at UW-Oshkosh for the funding
of a 270-MHz FT-NMR spectrometer.
(5) For most recent studies, see: (a) Rutherford, J . L.; Collum, D.
B. J . Am. Chem. Soc. 2001, 123, 199. (b) Sun, X.; Collum, D. B. J . Am.
Chem. Soc. 2000, 122, 2452. (c) Aubrecht, K. B.; Lucht, B. L.; Collum,
D. B. Organometallics 1999, 18, 2981. (d) Remenar, J . F.; Lucht, B.
F.; Collum, D. B. J . Am. Chem. Soc. 1997, 119, 5567.
J O0263465
J . Org. Chem, Vol. 68, No. 2, 2003 643