Equilibrium constants and alkylation kinetics of two lithium enolates/LiHMDS mixed aggregates in THF

Article abstract of DOI:10.1021/ol017175d

LiEn + LiHMDS ?^Kagg Li₂EnHMDS (Formula Presented) Products Products Mixed aggregates between lithium enolates and lithium hexamethyldisilazide (LiHMDS) have been studied in THF using UV-vis spectroscopy. The equillibrium constants (K

Full text of DOI:10.1021/ol017175d

ORGANIC
LETTERS
2002
Vol. 4, No. 4
573-575
Equilibrium Constants and Alkylation
Kinetics of Two Lithium Enolates/
LiHMDS Mixed Aggregates in THF
Yeong-Joon Kim and Andrew Streitwieser*
Department of Chemistry, UniVersity of California, Berkeley, California 94720-1460
astreit@socrates.berkeley.edu
Received December 4, 2001
ABSTRACT
Mixed aggregates between lithium enolates and lithium hexamethyldisilazide (LiHMDS) have been studied in THF using UV−vis spectroscopy.
The equilibrium constants (Kagg) between monomeric LiEn and monomeric LiHMDS are 760 and 560 M-1 when LiEn are LiSIBP and LiBnPAT,
respectively. The alkylation kinetics of the reactions with benzyl bromide were studied at 25 °C. The rate constants for the mixed aggregates,
kMixed, are substantially smaller than those of the monomeric enolates.
Lithium amides such as lithium diisopropylamide (LDA) and
lithium hexamethyldisilazide (LiHMDS) are frequently used
to deprotonate ketones to form lithium enolates. The product
lithium enolates are generally aggregated and could form
mixed aggregates with the lithium amide. The importance
of such mixed aggregates and their possible role in subse-
quent reactions has been inadequately addressed. Mixed
aggregates between lithium amides and lithium enolates have
been reported both in the crystal state1-3 and in solution.4,5
Such mixed aggregates have been the subject of theoretical
study,6,7 but there has been virtually no quantitative study
of their equilibrium constants for formation and their
reactivities. In this paper we report the formation constants
of mixed aggregates in THF of LiHMDS with two lithium
enolates, LiSIBP and LiBnPAT, from p-phenylsulfonyl-
isobutyrophenone (SIBP) and 6-phenyl-2-benzyl-R-tetralone
(BnPAT), respectively, and a kinetics study of their reac-
tivities in an alkylation reaction. Lithium hexamethyldi-
silazine (LiHMDS) is widely used because it combines the
properties of a strong base with weak nucleophilicity and it
also has the advantage of solubility in both hydrocarbon and
polar solvents. We reported previously the use of UV-vis
spectroscopy for determining the equilibrium constants and
alkylation kinetics of a lithium enolate/lithium bromide
mixed aggregate in THF.8 This approach was applied in the
present cases. We have also reported that LiSIBP and
LiBnPAT in THF solution are mixtures of monomer and
dimer (K1,2 ) 5.0 104 M-1 for LiSIBP9 and K1,2 ) 3.8 103
M-1 for LiBnPAT).10
(1) Williard, P. G.; Hintze, M. J. J. Am. Chem. Soc. 1987, 109, 5539-
5541.
(2) Williard, P. G.; Hintze, M. J. J. Am. Chem. Soc. 1990, 112, 8602-
8604.
(3) Sun, C.; Williard, P. G. J. Am. Chem. Soc. 2000, 122, 7829-7830.
(4) Hall, P. L.; Gilchrist, J. H.; Harrison, A. T.; Fuller, D. J.; Collum,
D. B. J. Am. Chem. Soc. 1991, 113, 9575-9585.
(5) Sakuma, K.; Gilchrist, J. H.; Romesberg, F. E.; Cajthaml, C. E.;
Collum, D. B. Tetrahedron Lett. 1993, 34, 5213-5216.
(6) Romesberg, F. E.; Collum, D. B. J. Am. Chem. Soc. 1994, 116, 9187-
9197.
Incremental addition of LiHMDS to THF solutions of
LiSIBP causes a blue shift in the UV-vis spectra from λmax
) 391 nm to λmax ) 370.5 nm as shown in Figure 1. An
(8) Abu-Hasanayn, F.; Streitwieser, A. J. Am. Chem. Soc. 1996, 118,
8136-8137.
(9) Abu-Hasanayn, F.; Stratakis, M.; Streitwieser, A. J. Org. Chem. 1995,
60, 4688-4689.
(10) Streitwieser, A.; Kim, Y.-J.; Wang, D. Z.-R. Org. Lett. 2001, 3,
2599-2601.
(7) Pratt, L. M.; Streitwieser, A. Unpublished work.
10.1021/ol017175d CCC: $22.00 © 2002 American Chemical Society
Published on Web 01/22/2002

these concentrations and LiSIBP forms a 1:1 mixed aggregate
with LiBr, we assume that the mixed aggregates in the
present cases are also the 1:1 aggregates, (LiEn)(LiHMDS),
Scheme 1.
Scheme 1
The value of Kagg is given by eq 1 where [LiEn] and
[Li2-EnHMDS]
Kagg
)
(1)
[LiEn][LiHMDS]
[LiHMDS] are the concentrations of monomers in equilib-
rium. Since LiHMDS exists primarily as a monomer in
THF,11 its concentration is taken directly. Deconvolution of
the UV-vis spectra of the mixtures, making use of previ-
ously determined spectra for the enolate monomers and
dimers, gave the concentrations of monomer, dimer, and
mixed aggregate for each mixture (Tables S1 and S2,
Supporting Information).
Figure 1. Effect of incremental addition of LiHMDS to a solution
of 0.00208 M LiSIBP in THF at 25 °C. The plot with lowest
absorbance is without addition of LiHMDS. The plot with highest
absorbance is from addition of 0.077 M LiHMDS. The net effect
is a shift in λmax from 391 to 370.5 nm.
isosbestic point at 394.5 nm indicates the formation of a
single mixed aggregate species. A similar blue shift (λmax
)
Average values found for Kagg are taken as 560 ( 90 M-1
for LiBuPAT/LiHMDS and 760 ( 55 M-1 for LiSIBP/
LiHMDS, with the errors taken as the standard deviations.
The estimated errors are probably reasonable measures of
the accuracy as well as the precision. The values found for
the two Kagg are of similar magnitude and are somewhat
lower than that found for LiSIBP/LiBr, Kagg ) 3600 M-1.
There have been only a few reports of the reactivities of
enolate-amide mixed aggregates and then only for proton-
transfer reactions.4 In the present work we use the Kagg values
found above in a quantitative kinetic study of alkylation
reactions of the mixed aggregates. Reaction kinetics between
LiEn and a large excess of benzyl bromide (BnBr) were
measured in the presence of LiHMDS. Initial rates were
determined by the decrease of absorbance of LiEn at the
isosbestic point for the first 10-20% reaction. Since
LiHMDS can also react with BnBr and complicate the
reaction kinetics, the products were analyzed by GC-MS
after quenching the first 10-20% reaction. The desired
alkylation product was at least 2.5 times larger than the
LiHMDS reaction product, N,N-bis(trimethylsilyl)benzyl-
amine. Consequently, changes in concentration of BnBr due
to reaction with LiHMDS are negligible during the initial
stages of reaction.
from 361 to 345 nm) and isosbestic point at 363 nm were
observed for the addition of LiHMDS to a THF solution of
LiBnPAT (Figure 2). Since these enolates form dimers at
The total rate expression is given by eq 2. In previous
Figure 2. Effect of incremental addition of LiHMDS to a solution
of 0.00064 M LiBnPAT in THF at 25 °C. The plot with lowest
absorbance is without addition of LiHMDS. The plot with highest
absorbance is from addition of 0.0509 M LiHMDS. The net effect
is a shift in λmax from 361 to 345 nm.
-d{LiEn}
rate )
) kM [LiEn][BnBr] +
kMixed [Li2EnHMDS][BnBr] (2)
dt
work the dimers of LiSIBP and LiBuPAT were shown to be
574
Org. Lett., Vol. 4, No. 4, 2002

much less reactive in alkylation reactions than the mono-
mers;9,10 thus, the dimer reactions have been omitted from
eq 2. Simple rearrangement gives eq 3 in which the rate
run with 0.25 M LiSIBP and 0.5 M LiHMDS. Sixty percent
of the starting enolate is present as the mixed aggregate and
40% is present as the dimer. The amount of monomer is
less than 1% and would be invisible to NMR, yet 40% of
an initial alkylation reaction will involve the monomer and
only 60% is with the mixed aggregate. For a chiral lithium
amide with comparable numbers, much of the product would
derive from the achiral enolate monomer, and the product
ee, at least in THF, would necessarily be disappointing.
An extension of this work was attempted with LDA, but
the above lithium enolates were found to be unsuitable.
LiSIBP reacts in some manner with LDA and was not studied
further. LiBnPAT is deprotonated by LDA to form a dianion.
The lithium enolates LiPhAT (from 2-phenyl-R-tetralone)
and LiPhPAT (from 2,6-diphenyl-R-tetralone) were explored.
Incremental addition of LDA generated a blue shift as in
LiHMDS addition (Figure 4). Since LDA itself absorbs up
kM
rate
) kMixed
+
(3)
[Li2EnHMDS][BmBr]
Kagg [LiHMDS]
function is now a linear equation in 1/[[LiHMDS]. From the
initial rates and Kagg and K1,2, the quantities were calculated
for each kinetic point. The rate constants for monomer, kM,
and mixed aggregate, kMixed, were determined from plots of
1/[LiHMDS] vs rate/([Mixed][BnBr]) (Figure 3).
Figure 3. Plot for determining the rate constants for the reaction
of monomeric LiEn and Li2EnHMDS. The regression lines are:
LiSIBP (circles), (0.0024 ( 0.0005) + (0.00032 ( 4 × 10-6)x,
(R2 ) 0.999; Kagg ) 760); LiBnPAT (squares), (0.012 ( 0.003) +
(4.16 ( 0.33) × 10-4x (R2 ) 0.969; Kagg ) 560).
Figure 4. Effect of incremental addition of LDA to a solution of
LiPhAT in THF at 25 °C.
The initial rates give the linear plots in Figure 3. Using
Kagg ) 760 and 560 M-1, the slopes in Figure 3 yield kM )
0.241 ( 0.003 and 0.233 ( 0.019 M-1 s-1 for LiSIBP and
LiBnPAT, respectively. The accuracy is probably not as good
as the precision but the results are in good agreement with
those obtained in the absence of LiHMDS, 0.32 (for p-tert-
butylBnBr which is expected to be somewhat more reactive
than BnBr) and 0.174 M-1 s-1, respectively.8,10 The intercepts
give kMixed ) 0.0024 ( 0.0005 and 0.012 ( 0.003 M-1 s-1
for LiSIBP and LiBnPAT, respectively, which are much
smaller values, by factors of 100 and 20, respectively, than
kM.
to 350 nm, it was not possible to assign the λmax for the mixed
aggregates. It was also not possible to derive Kagg because
the K1,2 for LDA is not known, although it is known that
LDA exists primarily as a dimer in THF solution at 25 °C.
Tables S3 and S4 (Supporting Information) summarize the
experimental data. Approximate evaluation, however, indi-
cates that Kagg is much higher for LDA than for LiHMDS.
Acknowledgment. This material is based on work
supported by the National Science Foundation under grant
9980367.
Supporting Information Available: Tables of spectral
data. This material is available free of charge via the Internet
at http://pubs.acs.org.
The significance of these results can be demonstrated by
a hypothetical synthesis example of an alkylation reaction
(11) Kimura, B. Y.; Brown, T. L. J. Organomet. Chem. 1971, 26, 57.
Org. Lett., Vol. 4, No. 4, 2002
OL017175D
575