Aggregation and Reactivity of the Cesium Enolate of 6-Phenyl-α -tetralone: Comparison with the Lithium Enolate

Article abstract of DOI:10.1021/jo034543d

The cesium enolate of 6-phenyl-α-tetralone (CsPAT) has a λ_max in THF at about 387 nm, but the variation with concentration is too small for application of singular value decomposition. Protontransfer studies with several indicators show that CsPAT forms monomer-tetramer mixtures with a tetramerization equilibrium constant, K _1,4 = 2.3 × 10¹¹ M^-3. The pK of the monomer is 23.39 on a scale where fluorene is assigned 22.9 (per hydrogen). For comparison, the lithium enolate, LiPAT, is also a monomer-tetramer with K _1,4 = 4.7 × 10¹⁰ M^-3 and a monomer pK = 14.22. HMPA in large amounts promotes dissociation to monomer with both enolates. Ion-pair SN2 initial rates were measured for CsPAT with several alkyl halides and with methyl tosylate and compared with other rates with LiPAT. In all cases, the enolate monomers are much more reactive than the aggregates. Reaction of CsPAT with alkyl halides is generally C-alkylation but HMPA promotes increasing amounts of O-alkylation. A new indicator, 11-methyl-11H-benzo[b]fluorene, has a pK on the cesium scale of 23.39.

Full text of DOI:10.1021/jo034543d

Aggr ega tion a n d Rea ctivity of th e Cesiu m En ola te of
6-P h en yl-r-tetr a lon e: Com p a r ison w ith th e Lith iu m En ola te1
Daniel Zerong Wang† and Andrew Streitwieser*
Department of Chemistry, University of California Berkeley, Berkeley, California 94720-1460
astreit@socrates.berkeley.edu
Received April 28, 2003
The cesium enolate of 6-phenyl-R-tetralone (CsPAT) has a λmax in THF at about 387 nm, but the
variation with concentration is too small for application of singular value decomposition. Proton-
transfer studies with several indicators show that CsPAT forms monomer-tetramer mixtures with
a tetramerization equilibrium constant, K1,4 ) 2.3 × 1011 M-3. The pK of the monomer is 23.39 on
a scale where fluorene is assigned 22.9 (per hydrogen). For comparison, the lithium enolate, LiPAT,
is also a monomer-tetramer with K1,4 ) 4.7 × 1010 M-3 and a monomer pK ) 14.22. HMPA in
large amounts promotes dissociation to monomer with both enolates. Ion-pair SN2 initial rates
were measured for CsPAT with several alkyl halides and with methyl tosylate and compared with
other rates with LiPAT. In all cases, the enolate monomers are much more reactive than the
aggregates. Reaction of CsPAT with alkyl halides is generally C-alkylation but HMPA promotes
increasing amounts of O-alkylation. A new indicator, 11-methyl-11H-benzo[b]fluorene, has a pK
on the cesium scale of 23.39.
In recent papers, we have described the aggregation
and reactivity of the lithium and cesium enolates of
several ketones: p-phenylisobutyrophenone, LiPhIBP2
and CsPhIBP,3 2-phenylcyclohexanone, LiPCH and
CsPCH,4 2-(p-biphenylyl)cyclohexanone, LiBPCH5 and
CsBPCH,6 2,6-diphenyl-R-tetralone, LiPhPAT and CsPh-
PAT,7 and the lithium enolates of p-phenylsulfonylisobu-
tyrophenone, LiSIBP,8 6-phenyl-R-tetralone, LiPAT, and
2-benzyl-6-phenyl-R-tetralone, LiBnPAT.9 In the present
paper, we extend the study to the cesium enolate, CsPAT,
of 6-phenyl-R-tetralone, PAT. This system was chosen
because it has a close relationship to p-phenylisobuty-
rophenone, PhIBP; in both, the enolate function is not
conjugated to the ring but the p-phenyl group provides a
sufficient chromophore for UV-vis spectral study. In the
tetralone system, however, the enolate function is in a
conformationally fixed ring compared to the greater
mobility in LiPhIBP and CsPhIBP. Moreover, the car-
bonyl R-position in PAT is secondary rather than tertiary
as in PhIBP and permits comparisons with R-substitu-
ents. The object is a better quantitative understanding
of the effect of the steric environment of the enolate on
its aggregation and reactivity. The present study supple-
ments our previous report on the aggregation and
reactivity of the corresponding lithium enolate, LiPAT.9
and additional kinetic studies are compared. For com-
parison, J ackman and Bortiatynski10 have reported that
the lithium salt of R-tetralone is predominantly tet-
rameric in THF based on NMR analysis.
† Current address: School of Science and Computer Engineering,
University of Houston Clear Lake, Houston, TX 77058. E-mail:
wang@cl.uh.edu.
Alkali enolates had been well-known to be aggregated
in THF solution before our work10-13 but the equilibrium
constants for aggregate formation were unknown and
little was known about the relative roles of monomer and
aggregates in reactions. We have shown that in alkyla-
tion reactions of all of the enolates that we have studied
the second-order rate constants for the monomers are
greater than for the corresponding aggregates. In the
(1) Carbon Acidity. 119.
(2) Abbotto, A.; Leung, S. S.-W.; Streitwieser, A.; Kilway, K. V. J .
Am. Chem. Soc. 1998, 120, 10807-10813.
(3) Streitwieser, A.; Krom, J . A.; Kilway, K. A.; Abbotto, A. J . Am.
Chem. Soc. 1998, 120, 10801-10806.
(4) Wang, D. Z.-R.; Streitwieser, A. Can. J . Chem. 1999, 77, 654-
658.
(5) Streitwieser, A.; Wang, D. Z. J . Am. Chem. Soc. 1999, 121, 6213-
6219.
(6) Streitwieser, A.; Wang, D. Z.-R.; Stratakis, M. J . Org. Chem.
1999, 64, 4860-4864.
(10) J ackman, L. M.; Bortiatynski, J . Adv. Carbanion Chem 1992,
1, 45-87.
(11) J ackman, L. M.; Lange, B. C. Tetrahedron 1977, 33, 2737-2769.
(12) Bauer, W.; Seebach, D. Helv. Chim. Acta 1984, 67, 1972-1988.
(13) Seebach, D. Angew. Chem., Int. Ed. Engl. 1988, 27, 1624-1654.
Heathcock, C. H. In Comprehensive Synthetic Chemistry; Trost, B., Ed.;
Pergamon Press: New York, 1991; pp 181-238.
(7) Wang, D. Z.-R.; Kim, Y.-J .; Streitwieser, A. J . Am. Chem. Soc.
2000, 122, 10754-10760.
(8) Abu-Hasanayn, F.; Stratakis, M.; Streitwieser, A. J . Org. Chem.
1995, 60, 4688-4689.
(9) Streitwieser, A.; Kim, Y.-J .; Wang, D. Z.-R. Org. Lett. 2001, 3,
2599-2601.
10.1021/jo034543d CCC: $25.00 © 2003 American Chemical Society
Published on Web 10/15/2003
8936
J . Org. Chem. 2003, 68, 8936-8942

Reactivity of the Cesium Enolate of 6-Phenyl-R-tetralone
present work, a further objective was to learn more about
the effect of enolate structure on reactivity.
We have used two complementary methods for deter-
mining the stoichiometry and equilibrium constants for
aggregation of alkali enolates, the effect of concentration
on the UV-vis spectra, and the use of coupled equilibria,
usually the effect of aggregation on proton-transfer
equilibria.
Sp ectr a a n d p K of CsP AT
Solutions of CsPAT were prepared by deprotonating
the ketone with diphenylmethylcesium and UV-vis
spectra were taken on the solutions after successive
dilutions with THF. A series of five runs were carried
out in 1 mm cells, and two runs were made at higher
dilution in 1 cm cells. One of the latter runs gave results
inconsistent with the others and was discarded. Results
of the other six runs are summarized in Table S1
(Supporting Information). Within any given run, the λmax
varied by a few nanometers but the normalized spectra
show an isosbestic point at 387 nm. A plot of the
absorbance divided by cell path length at the isosbestic
point vs the formal concentration of CsPAT is linear over
a concentration range of more than 2 orders of magnitude
to give ꢀ ) 6627 ( 8 as shown in Figure S2 (Supporting
Information).
The variation in λmax is too small for application of
singular value decomposition, an analysis applied in
several previous examples of enolate aggregation.2,4-6
Thus, we could use only the method of coupled equilibria
for determining the aggregation equilibrium constants;
however, the presence of an isosbestic point for CsPAT
shows that only two components are present. A similar
isosbestic point had been found for LiPAT. The stoichi-
ometry and equilibrium constants for aggregate forma-
tion were determined by the effect of concentration on
proton-transfer equilibria of the type in eq 1 whose
observed equilibrium constant is given by eq 2.
F IGURE 1. Observed pK of PAT vs the indicator Me23BF.
The slope of the linear section shown (circles) is -0.663
corresponding to an average aggregation number n˜ ) 2.98.
they do not react significantly with the ketones studied
under these conditions.
The cesium salt of Me23BF has three absorption
maxima at 692.5 (ꢀ ) 1272), 634.5 (ꢀ ) 1652), and 431.0
nm (ꢀ ) 23 909). The data for three runs are given in
Table S2 (Supporting Information) and Lambert-Beer’s
Law plots leading to the ꢀ values are shown in Figure
S2 (Supporting Information). The pK of Me23BF on the
cesium scale in THF was measured relative to 45MP; two
runs with a total of 25 measurements gave pK ) 23.39
( 0.01.
A typical plot of the observed pK vs log{CsPAT} is
shown in Figure 1 with Me23BF as the indicator. The
slope of the linear portion, at concentrations greater than
{CsPAT} ) 10-4 M, -0.663, corresponds to an average
aggregation number of 3.0.16 Similar plots for the other
four runs are given as Figures S3-S5 (Supporting
Information). These runs give average aggregation num-
bers of 2.95 to 3.21 and suggest that CsPAT is a mixture
of mostly tetramer at these concentrations.
RH + M+ In- a R- M+ + In H
{R- M+}[ln H]
(1)
(2)
We have shown that for a monomer-tetramer mixture
a convenient analysis of the data is to plot Kobs with an
indicator vs the quantity ({CsPAT}/Kobs)3. A monomer-
tetramer mixture gives a straight line with an intercept
Ko, corresponding to the equilibrium constant with the
Kob
)
[RH]{M+ ln-}
In eq 2, {R- M+} denotes the formal concentration of
the metal enolate. The indicators were cesium salts of
highly delocalized carbanions whose relative ion pair
acidities are referenced for convenience to the ionic
acidity of fluorene in DMSO, pK ) 22.9 per hydrogen.14
Extensive tables of pKs are available for cesium salts of
various indicators in THF.15 In the present work, the
cesium pK of CsPAT was measured against 11H-benzo-
[b]fluorene (2,3-benzofluorene, 23BF, pK ) 23.63) and
9H-benzo[def]fluorene (4,5-methylenephenanthrene, 45MP,
pK ) 22.91).15 In addition, in this work the pK of another
indicator, 11-methyl-11H-benzo[b]fluorene (methyl-2,3-
benzfluorene, Me23BF), was determined and used for a
further comparison of CsPAT. The use of indicators with
tertiary acidic protons is particularly convenient because
4 17
enolate monomer, and a slope ) 4K1,4Ko . Such a plot
for the results in Figure 1 is shown as Figure 2 and gives
Ko ) 0.97 (pK ) 23.40), K1,4 ) 3.21 × 1011 M-3. Corre-
sponding plots for the other runs are shown in Figures
S6-S8 and give pK values of 23.43, 23.39. 23.39, 23.35
and K1,4 values of 2.29 × 1011, 2.43 × 1011, 1.71 × 1011,
1.82 × 1011. The average values are: pK of the monomer,
23.39 ( 0.02, K1,4 ) (2.3 ( 0.4) × 1011 M-3. Figure 3 shows
the combined results of all of the runs with different
indicators put on a common scale.
As shown in Table S1 λmax at different concentrations
of CsPAT in 1 mm cells varied from 372 to 379 nm. At
the higher dilutions in a 1 cm cell λmax ranged up to 387.5
nm. Although consistent results could not be obtained
(16) Kaufman, M. J .; Streitwieser, A., J r. J . Am. Chem. Soc. 1987,
109, 6092-6097.
(17) Krom, J . A.; Petty, J . T.; Streitwieser, A. J . Am. Chem. Soc.
1993, 115, 8024-8030.
(14) Bordwell, F. G. Acc. Chem. Res. 1988, 21, 456-463.
(15) Streitwieser, A.; Ciula, J . C.; Krom, J . A.; Thiele, G. J . Org.
Chem. 1991, 56, 1074-1076.
J . Org. Chem, Vol. 68, No. 23, 2003 8937

Wang and Streitwieser
dissociation constant is much lower than those of the
lithium indicator salts studied.18 CsPAT is more ag-
gregated than LiPAT whose K1,4 ) 4.7 × 1010 M-3 but
the difference in ∆G° is only 1 kcal mol-1. An important
driving force for dissociation of the tetramer to monomer
is the resulting increased solvation of the metal cation.
This principle has been shown generally for 1:1 salts,19
and has been demonstrated not only for enolates20 but
also for lithium phenolates21 and computationally.22 The
present results show the increased solvation of the
monomeric cesium enolate is almost as great as that of
lithium enolate.
This effect is also shown by the effects of HMPA, which
had been shown earlier to convert lithium enolate ag-
gregates to monomers.23 This effect of HMPA is well-
known for other lithium compounds, for example, lithium
phenolates,,24 lithiated nitriles,25 lithium amides,26-28 (but
note)29 lithium chloride,30 and aryllithiums.31-33 Small
amounts of HMPA, up to 0.1 M, have little effect on the
λmax of CsPAT, but larger amounts, up to 1 M, cause
bathochromic shifts to 385 nm (Figure S11, Supporting
Information), the λmax estimated above for the monomer.
Thus, even the large cesium cation is sufficiently solvated
by HMPA to stabilize the monomer. Similarly, HMPA
causes a shift of λmax of LiPAT to 353 nm (Figure S12,
Supporting Information), only slightly greater than that
estimated above for the monomer. These changes can be
compared to the effects of HMPA on the spectra of some
indicators. Addition of HMPA to the cesium salt of
p-biphenylyldiphenylmethane causes a progressive ba-
thochromic shift, probably a change from the contact ion
pair cesium salt to increasing amounts of SSIP. However,
HMPA has no effect on the spectrum of the lithium salt
of trimethyldihydroanthracene, undoubtedly because this
lithium salt is an SSIP to begin with.
F IGURE 2. The intercept, 0.97, is Ko for the equilibrium
between the monomer of CsPAT and the indicator Me23BF;
the slope, 1.15 × 1012, ) 4K1,4Ko .
4
F IGURE 3. Composite of all of the pK runs of CsPAT with
different indicators: circles and squares, Me23BF; diamonds
and triangles, 23BF; crosses, 45MP. The dotted line is that
calculated for the average values: pK of the monomer, 23.39;
(18) Kaufman, M. J .; Gronert, S.; Streitwieser, A., J r. J . Am. Chem.
Soc 1988, 110, 2829-2835.
K1,4 ) 2.3 × 1011 M-3
.
(19) Chabanel, M. Pure Appl. Chem. 1990, 62, 35-46.
(20) Streitwieser, A.; J uaristi, E.; Kim, Y.-J .; Pugh, J . K. Org. Lett.
2000, 2, 3739-3741.
in applying SVD to these data, using the value deter-
mined above for K1,4 the amounts of monomer and
tetramer present in each of the solutions in Table S1
could be evaluated. Plotting the observed λmax vs the mole
fraction of monomer, N(monomer), in each solution gives
a linear correlation with some scatter (Figure S9, Sup-
porting Information) but extrapolation to N ) 0 and N
) 1 gives the λmax of tetramer and monomer, 372 and
385 nm, respectively. A similar plot for LiPAT gives λmax
of tetramer and monomer, 340 and 347 nm, respectively,
a substantially smaller difference between monomer and
aggregate.
The pK of CsPAT, 23.39, is substantially higher than
that of the corresponding lithium enolate, LiPAT, 14.22.9
We emphasize again that these values are not pKa but
are values referred to an arbitrary (but convenient)
standard. All of the cesium salts are contact ion pairs
(CIP). The lithium enolate is referred to a solvent-
separated ion pair (SSIP) standard; the much lower value
of LiPAT relative to CsPAT indicates that, in common
with all of the other lithium enolates we have studied,
LiPAT is present in THF as a contact ion pair whose
(21) J ackman, L. M.; DeBrosse, C. W. J . Am. Chem. Soc. 1983, 105,
4177-4184.
(22) Pratt, L. M.; Streitwieser, A. J . Org. Chem. 2003, 68, 2830-
2838.
(23) Leung, S. S.-W.; Streitwieser, A. J . Org. Chem. 1999, 64, 3390-
3391.
(24) J ackman, L. M.; Chen, X. J . Am. Chem. Soc. 1992, 114, 403-
411.
(25) Carlier, P. R.; Lo, C. W.-S. J . Am. Chem. Soc 2000, 122, 12819-
12823.
(26) J ackman, L. M.; Scarmoutzos, L. M.; Smith, B. D.; Williard, P.
G. J . Am. Chem. Soc. 1988, 110, 6058-6063.
(27) Sato, D.; Kawasaki, H.; Shimada, I.; Arata, Y.; Okamura, K.;
Date, T.; Koga, K. J . Am. Chem. Soc. 1992, 114, 761-763.
(28) Romesberg, F. E.; Bernstein, M. P.; Gilchrist, J . H.; Harrison,
A. T.; Fuller, D. J .; Collum, D. B. J . Am. Chem. Soc. 1993, 115, 3475-
3483.
(29) Romesberg, F. E.; Gilchrist, J . H.; Harrison, A. T.; Fuller, D.
J .; Collum, D. B. J . Am. Chem. Soc. 1991, 113, 5751-5757.
(30) Reich, H. J .; Borst, J . P.; Dykstra, R. R. J . Am. Chem. Soc. 1993,
115, 8728-8741.
(31) Reich, H. J .; Gudmundsson, B. O. J . Am. Chem. Soc. 1996, 118,
6074-6075.
(32) Reich, H. J .; Green, D. P.; Medina, M. A.; Goldenberg, W. S.;
Gudmundsson, B. O¨ .; Dykstra, R. R.; Phillips, N. H. J . Am. Chem. Soc.
1998, 120, 7201-7210.
(33) Reich, H. J .; Goldenberg, W. S.; Gudmundsson, B. O.; Sanders,
A. W.; Kulicke, K. J .; Simon, K.; Guzei, I. A. J . Am. Chem. Soc. 2001,
123, 8067-8079.
8938 J . Org. Chem., Vol. 68, No. 23, 2003

Reactivity of the Cesium Enolate of 6-Phenyl-R-tetralone
TABLE 1. Su m m a r y of Ra te Con sta n ts
RX
k2, M-1 s-1
n-hexyl chloride
n-hexyl bromide
n-hexyl iodide
methyl tosylate
p-t-Bu-benzyl chloride
(1.15 ( 0.03) × 103
0.292 ( 0.004
4.07 ( 0.05
4.30 ( 0.09
0.59 ( 0.01
should be noted again that any significant reaction via
an aggregate would show up as upward curvature of plots
such as in Figure 4.
n-Hexyl chloride is much less reactive, and a constant
concentration of enolate was used with varying excess
amounts of halide. These results are summarized in
Table S5 and Figure S15 (Supporting Information).
Methyl tosylate was included but gave more scatter than
the n-hexyl halides. Its results are summarized in Table
S6 and Figure S16 (Supporting Information). Finally,
tert-butylbenzyl chloride (BBC) is shown in Table S7 and
Figure S17 (Supporting Information). The rate constants
found are summarized in Table 1. The errors shown are
the standard deviations of the linear slopes, but the true
probable errors are closer to 10%.
F IGURE 4. Initial rates of n-hexyl iodide and bromide
compared to the monomer concentration of CsPAT times the
concentration of RX. Slopes shown are as follows: HexI, k2 )
4.07 ( 0.05 M-1 s-1, R2 ) 0.933; HexBr, k2 ) 0.294 ( 0.004
M-1 s-1, R2 ) 0.982.
The reactivity order of the hexyl halides is typical of
SN2 reactions.41 In particular, the relative reactivity of
the iodide is essentially normal. If its reaction involved
single electron transfer a much larger rate constant
would have been expected. This result agrees with that
found previously for the reaction of the cesium enolate
of 2,6-diphenyl-R-tetralone (CsPhPAT) with n-hexyl io-
dide.7 Tosylates are usually somewhat more reactive than
the corresponding bromides and that is found here, also
in agreement with results with CsPhPAT.7 Similarly,
benzyl halides are typically 2 orders of magnitude more
reactive in SN2 reactions than ethyl halides; in the
present case, BBC is 500 times more reactive than
n-hexyl chloride and fits the typical pattern well.
Alkylation kinetics were also run with the lithium
enolate, LiPAT. Results of alkylation and dialkylation
with benzyl bromide were published earlier.9 Additional
runs were made with some substituted benzyl bromides
and with methyl brosylate (MeOBs). In most cases, there
was an initial rapid decrease in the enolate absorbance
on adding the alkylating agent followed by a slower
decrease. In these cases, the initial rate was determined
after the first 20-30 s and following the reaction for an
additional approximately 10%. This procedure is some-
what different from that used with the reaction of LiPAT
with benzyl bromide reported earlier;9 a reanalysis of this
reaction gives a rate constant about 10% greater, k2 )
0.105 M-1 s-1. Results for o-chlorobenzyl bromide
(oClBnBr) and o-methylbenzyl bromide (oMeBnBr) are
given in Tables S8 and S9 (Supporting Information) and
displayed in Figure 5. Similar results for m-chlorobenzyl
bromide and methyl brosylate are summarized in Tables
S10 and S11 and Figures S18 and S19 (Supporting
Information). The rate constants are summarized in
Table 2.
Alk yla tion Kin etics
We had shown previously that the alkylation reactions
of several lithium and cesium enolates occur dominantly
with the enolate monomers.2,5-9 In the present work, we
include additional examples of such alkylations because
these are ion-pair SN2 reactions and it would be valuable
to have examples of structural effects for comparison with
ionic SN2 reactions in more polar solvents. Earlier studies
of relative reactivities of different enolates in ether are
less useful for this purpose because the role of aggrega-
tion was not determined.34-38 Lithium halides have been
shown to form mixed aggregates with lithium eno-
lates;39,40 although such mixed aggregates have not been
demonstrated for cesium salts, we followed our usual
procedure of measuring initial rates. A large excess of
the alkylating agent was added to a solution of the
enolate in THF, and the absorbance of the enolate was
followed at λmax. The slope of about the first 10-15%
reaction was used to determine the rate constant of the
pseudo-first-order reaction using the calculated concen-
tration of the enolate monomer. Using this slope as an
approximation to the logarithm will give numbers about
5% low, but no correction was made because the experi-
mental error is probably somewhat greater than this.
Examples of the results with n-hexyl bromide and iodide
are shown in Figure 4. The data are summarized in
Tables S3 and S4 (Supporting Information). Each point
is a separate kinetic run using different concentrations
of enolate and/or halide. Hexyl halides were used because
of volatility considerations in our glovebox system. It
(34) Zook, H. D.; Rellahan, W. L. J . Am. Chem. Soc. 1957, 79, 881-
886.
(35) Zook, H. D.; Gumby, W. L. J . Am. Chem. Soc. 1960, 82, 1386-
1389.
(36) Zook, H. D.; Russo, T. J .; Ferrand, E. F.; Stotz, D. S. J . Org.
Chem. 1968, 33, 2222-2228.
(37) Caine, D.; Huff, B. J . L. Tetrahedron Lett. 1966, 4695-4700.
(38) Caine, D.; Huff, B. J . L. Tetrahedron Lett. 1967, 3399-3403.
(39) J ackman, L. M.; Haddon, R. C. J . Am. Chem. Soc. 1973, 95,
3687-3692.
(40) Abu-Hasanayn, F.; Streitwieser, A. J . Am. Chem. Soc. 1996,
118, 8136-8137.
Earlier, we found that the reaction of MeOBs with
CsPhPAT is 6.8 times that of MeOTs. If that same ratio
applies to the present case, CsPAT monomer is found to
be 3400 times as reactive as that of LiPAT. We had also
(41) Streitwieser, A. Chem. Rev. 1956, 56, 571.
J . Org. Chem, Vol. 68, No. 23, 2003 8939

Wang and Streitwieser
F IGURE 5. Initial rates (M-1 s-1) of reaction of o-chlorobenzyl
bromide (circles) and o-methylbenzyl bromide (squares) com-
pared to [RBr][LiPAT monomer]. The two slopes shown are
as follows: oClBnBr, k2 ) 0.229 ( 0.005 (R2 ) 0.83), oMeBnBr,
k2 ) 0.186 ( 0.003 (R2 ) 0.82).
F IGURE 7. Rates of lithium enolate monomers with m-
chlorobenzyl bromide compared to the enolate pKs. Points
shown are as follows: BPCH, 2-p-biphenylylcyclohexanone;
PCH, 2-phenylcyclohexanone; PhIBP, p-phenylisobutyrophe-
none; PAT. 6-phenyl-R-tetralone; PhPAT, 2,6-diphenyl-R-te-
tralone.
with mClBnBr are known. As shown in Figure 7, these
data give no Brønsted correlation. Apparently, the elec-
trophilic character of the lithium cation is as important
as the basicity of the enolate; that is, these rates are
dependent to an important extent on two independent
properties and not just one.
In previous work, HMPA was shown to increase rates
of reaction with alkylating agents not only by deaggre-
gating enolate aggregates to monomers but also by
increasing the reactivity of the monomers.23 It was shown
above that it requires about 1 M HMPA to completely
deaggregrate LiPAT to the monomer. Accordingly, the
reaction of LiPAT with methyl brosylate in the presence
of 0.25 M HMPA is still of fractional order in {LiPAT}
but is about 70-fold faster than in the absence of HMPA
(Figure S20, Supporting Information). We did not obtain
an accurate measure of the reactivity of CsPAT in the
presence of HMPA but did find that its rate of reaction
with n-hexyl chloride and 0.36 M HMPA is roughly 40
times faster than without HMPA.
F IGURE 6. Brønsted plot of log k2 with MeOTs of the
monomer ion pair pKs of cesium enolates in THF at 25 °C.
Points shown are as follows: PhIBP, p-phenylisobutyrophe-
none; PAT, 6-phenyl-R-tetralone; BPCH, 2-p-biphenylylcyclo-
hexanone; PhPAT, 2,6-diphenyl-R-tetralone. Slope shown is
0.278 ( 0.019; R2 ) 0.99.
P r od u ct Stu d ies
TABLE 2. Secon d -Or d er Ra te Con sta n ts for Rea ction of
LiP AT Mon om er w ith Alk yla tin g Agen ts in THF a t 25 °C
In previous product studies with the cesium and
lithium enolates of 2,6-diphenyl-R-tetralone, we reported
that reactions with alkyl halides gave almost exclusively
C-alkylation but that reactions of the cesium enolate with
methyl sulfonates gave substantial amounts of O-alkyl-
ation. These observations were further extended in the
present study.
RX
benzyl bromide
o-methylbenzyl bromide
o-chlorobenzyl bromide
m-chlorobenzyl bromide
methyl p-bromobenzenesulfonate
k 2, M-1 s-1
0.105
0.186
0.229
0.187
0.0085
PAT has two R-hydrogens and can form mono- and
dialkylation products (Scheme 1). The initial O-alkylation
product cannot undergo further reaction, but the product
of C-alkylation can undergo further alkylation to give CC
and CO dialkylation. The role of aggregation in deter-
mining the amount of dialkylation was reported recently.9
Reaction products were determined for a number of
kinetic solutions allowed to stand for a prolonged period.
Products were identified by GC-MS, and the mixtures
were analyzed using the uncalibrated GC areas. Accord-
ingly, the results will not be completely accurate but the
shown that the rates of reaction with MeOTs of three
cesium enolates correlated well with their pKs. We can
now add CsPAT to this correlation as a fourth point and
find that it also fits well (Figure 6) This relation amounts
to a Brønsted plot with a slope having a relatively small
value of 0.28. Thus, the effective basicity of a cesium
enolate is dominant in its reactivity with MeOTs but
apparently with a relatively early transition state.
The corresponding lithium enolates give no such cor-
relation. The addition of LiPAT from the present study
gives five lithium enolates whose monomer pKs and rates
8940 J . Org. Chem., Vol. 68, No. 23, 2003

Reactivity of the Cesium Enolate of 6-Phenyl-R-tetralone
TABLE 4. C- a n d O-Alk yla tion P r od u cts of Rea ction of
CsP AT w ith n -Hexyl Ha lid es
SCHEME 1. Mon o- a n d Dia lk yla tion P r od u cts
HMPA,
M
1st alk 2nd alk
RX
f(O)a f(C)b f(OC)c f(CC)d
f(C)e
f(C)f
HexCl
HexBr
HexI
HexCl
HexBr
HexI
HexCl
HexBr
HexI
HexCl
HexBr
HexI
HexCl
HexBr
HexI
HexCl
HexBr
HexI
HexCl
HexBr
HexI
HexCl
HexBr
HexI
0.0
0.29 0.26
0.11 0.26
0.10 0.36
0.73 0.07
0.52 0.14
0.18 0.40
0.81 0.02
0.62 0.09
0.24 0.35
0.85 0.01
0.65 0.12
0.29 0.46
0.87 0.01
0.69 0.12
0.35 0.47
0.87 0.03
0.69 0.14
0.32 0.51
0.92 0.01
0.75 0.13
0.39 0.52
0.94 0.03
0.75 0.17
0.39 0.54
0.84 0.04
0.77 0.01
0.36 0.44
0.26
0.36
0.23
0.19
0.28
0.25
0.16
0.27
0.26
0.14
0.22
0.17
0.12
0.19
0.14
0.10
0.17
0.13
0.07
0.12
0.09
0.02
0.09
0.07
0.11
0.20
0.14
0.19
0.27
0.31
0.01
0.05
0.18
0.00
0.03
0.15
0.00
0.02
0.08
0.00
0.00
0.03
0.00
0.00
0.04
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.02
0.05
0.71
0.89
0.90
0.27
0.48
0.82
0.19
0.38
0.76
0.15
0.35
0.71
0.13
0.31
0.65
0.13
0.31
0.68
0.08
0.25
0.61
0.06
0.25
0.61
0.16
0.23
0.64
0.42
0.43
0.58
0.04
0.15
0.43
0.03
0.10
0.37
0.00
0.07
0.31
0.00
0.00
0.18
0.00
0.00
0.22
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.08
0.27
0.50
1.00
1.42
1.89
2.83
4.02
4.78
5.69
TABLE 3. C- a n d O-Alk yla tion P r od u cts of Rea ction of
LiP AT w ith n -Hexyl Ha lid es
HMPA,
M
1st alk 2nd alk
RX
f(O)a f(C)b f(OC)c f(CC)d
f(C)e
f(C)f
HexCl
HexBr
HexI
HexCl
HexBr
HexI
HexCl
HexBr
HexI
HexCl
HexBr
HexI
HexCl
HexBr
HexI
HexCl
HexBr
HexI
HexCl
HexBr
HexI
HexCl
HexBr
HexI
0
0.00 1.00
0.00 1.00
0.00 1.00
0.00 1.00
0.00 0.98
0.00 0.88
0.00 0.98
0.00 0.90
0.00 0.78
0.00 0.98
0.00 0.86
0.00 0.85
0.00 0.97
0.00 0.89
0.00 0.85
0.00 0.91
0.00 0.88
0.00 0.85
0.00 0.87
0.00 0.87
0.00 0.84
0.00 0.89
0.00 0.91
0.00 0.92
0.07 0.85
0.34 0.62
0.00 0.81
0.00
0.00
0.00
0.00
0.00
0.00
0.00
0.06
0.06
0.00
0.08
0.05
0.00
0.07
0.05
0.08
0.09
0.05
0.11
0.10
0.05
0.10
0.08
0.03
0.04
0.04
0.07
0.00
0.00
0.00
0.00
0.02
0.12
0.02
0.04
0.17
0.02
0.05
0.10
0.03
0.04
0.10
0.02
0.03
0.09
0.02
0.03
0.10
0.01
0.01
0.05
0.03
0.00
0.12
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
1.00
0.93
0.66
1.00
0.50
0.99
1.54
1.99
2.98
3.96
4.88
5.69
1.00
1.00
1.00
1.00
0.41
0.75
1.00
0.39
0.68
1.00
0.34
0.66
0.18
0.27
0.65
0.13
0.25
0.65
0.11
0.15
0.57
0.43
0.00
0.62
HexCl
HexBr
HexI
a
b
O-RPAT/total area of four products. C-RPAT/total area of
four products. c CO-R2PAT/total area of four products. CC-R2PAT/
d
total area of four products. e Fraction of C-alkylation for the first
alkylation ) C-RPAT + OC-R2PAT + CC-R2PAT/total area of four
products f Fraction of C-alkylation for the second alkylation ) CC-
R2PAT/(OC-R2PAT + CC-R2PAT).
The results with CsPAT in Table 4 are significantly
different. Even without HMPA, CsPAT gives substantial
O-alkylation, but again, the second alkylation is still
greater on oxygen. With addition of even small amounts
of HMPA, the extent of O-alkylation is increased. O-
Alkylation is also more important for RCl than for RI
with RBr being intermediate. In the reaction of CsPAT
with p-tert-butylbenzyl chloride, only the mono- and
dialkylation products of C-alkylation were found and
identified.
HexCl
HexBr
HexI
a
b
O-RPAT/total area of four products. C-RPAT/total area of
four products. c OC-R2PAT/total area of four products. CC-R2PAT/
d
total area of four products. e Fraction of C-alkylation for the first
alkylation ) C-RPAT + OC-R2PAT + CC-R2PAT/total area of four
products. f Fraction of C-alkylation for the second alkylation ) CC-
R2PAT/(OC-R2PAT + CC-R2PAT).
These results on C- vs O-alkylation agree well with the
generalizations summarized by le Noble several decades
ago.42
trends are clear. Tables 3 and 4 give the results for
alkylation of LiPAT and CsPAT, respectively, with the
n-hexyl halides in the presence and absence of HMPA.
As mentioned above, the O-alkylation product, O-RPAT,
cannot undergo further reaction, but the C-alkylation
product can undergo further alkylation to give both CC-
and OC-dialkylation products. Thus, for the first alkyl-
ation, the relative amount of C-alkylation is determined
by the sum of C-RPAT and the dialkylation products. The
results in Table 3 show that LiPAT gives exclusively
C-alkylation with the hexyl halides. Small amounts of
HMPA do not change this result although with HMPA,
the second alkylation is increasingly at oxygen. Only at
the highest amount of HMPA used, 5.69 M, is significant
O-alkylation found for the first alkylation. But the
amounts of O-alkylation are generally small and trends
among the different halides are not consistent.
Con clu sion s
CsPAT forms monomer-tetramer mixtures in THF as
does LiPAT. Aggregation of an ion pair is electrostatically
favorable but the driving force is reduced for large cations
such as cesium compared to lithium. Solvation of the
cation preferentially stabilizes the monomer and such
solvation is more important for lithium than for the
larger cesium. These two opposing effects are apparently
of comparable importance for these enolates in THF
because the tetramerization equilibrium constant for
CsPAT is only slightly greater than for LiPAT. Solvation
of the cesium cation in CsPAT is still important as shown
by the effect of HMPA which shifts the equilibrium
toward monomer as it does for LiPAT.
(42) le Noble, W. Synthesis 1970, 1-6.
J . Org. Chem, Vol. 68, No. 23, 2003 8941

Wang and Streitwieser
To a solution of 4.33 g of benzo[b]fluorenone45 in 400 mL of
dry ether was added 10 mL of 3.0 M methylmagnesium
bromide. The solution was refluxed and quenched with HCl,
and the product was extracted with ether. Removal of solvent
from the washed and dried extract gave 4.98 g of 7-methyl-
7H-benzo[b]fluoren-7-ol. This material was reduced with HI/
AcOH because in other work we found that hydrogenolysis of
carbinols containing a naphthalene ring gave some reduction
at naphthalene to give impurities difficult to remove. The
above carbinol was refluxed with 40 mL of HI in 240 mL of
acetic acid until reaction was completed as indicated by TLC.
The product was extracted with ether, washed, dried, and
distilled. Purification by column chromatography (hexane/ethyl
acetate ) 50:1), and sublimation under vacuum gave 0.98 g
of pure compound (23% yield): mp 116-118 °C; 1H NMR (300
MHz, CDCl3) δ 8.19 (s, 1H), 7.87 (m, 4H), 7.54-7.57 (m, 1H),
7.36-7.52 (overlapping multiplets, 4H), 4.16 (q, J ) 7.5 Hz,
1H), 1.64 (d, J ) 7.5 Hz, 3H); 13C NMR (400 MHz, CDCl3) δ
149.5, 147.2, 140.0, 139.5, 133.2, 128.2, 127.9, 127.8, 127.2,
125.5, 125.4, 124.3, 122.5, 120.6, 117.9, 71.2, 42.0, 19.3. Anal.46
Calcd for C18H14: C, 93.87; H, 6.13. Found: C, 93.73; H, 6.07.47
Rea ction P r od u ct: CsP AT + n -Hexyl Br om id e. Prod-
ucts were separated by preparative TLC. O-Alkylation product
(4-hexyloxy-7-phenyl-1,2-dihydronaphthalene): 1H NMR (CDCl3,
300 MHz) δ 8.11 (d, J ) 8.19 Hz, 1H), 7.61 (d, J ) 7.6 Hz,
2H), 7.54-7.46 (m, 3H), 7.43-7.36 (m, 3H), 3.02 (t, J ) 6.3
Hz, 2H), 2.09-2.05 (m, 2H), 1.69-1.66 (m, 2H), 1.34-1.26 (m,
8H), 0.85 (t, J ) 6.8 Hz, 3H); MS (m/z, rel intensity) 308 (4),
307 (23), 306 (100), 305 (3), 236 (7), 235 (34), 222 (5), 221 (3),
207 (6), 194 (8), 165 (7). C-Alkylation product (2-hexyl-6-
phenyl-R-tetralone): 1H NMR (CDCl3 300 MHz) δ 8.10 (d, J
) 8.15 Hz), 1H), 7.63-7.51 (m, 2H), 7.49-7.30 (m, 4H), 7.28-
7.17 (m, 1H), 3.12-3.02 (m, 1H), 2.54-2.46 (m, 1H), 2.32-
2.23 (m, 1H), 2.00-1.82 (m, 2H), 1.58-1.26 (m, 10H), 0.97-
0.87 (m, 3H); MS (m/z, rel intensity) 307 (3), 306 (31), 222
(100), 221 (14), 207 (8), 194 (5), 166 (8), 165 (18).
In alkylation reactions, the monomer of CsPAT is much
more reactive than the tetramer. The n-hexyl halides
show a normal reactivity order: RI > RBr > RCl. Methyl
tosylate is about as reactive as n-hexyl iodide and p-tert-
butylbenzyl chloride is twice as reactive as n-hexyl
bromide. The monomer of CsPAT is about 3000 times
more reactive than the monomer of LiPAT. Together with
several other cesium enolates, the reaction of CsPAT with
methyl tosylate gives a Brønsted correlation of log k vs
pK, indicating that the basicity of the enolate moiety is
an important driving force in this alkylation. However,
the relatively small Brønsted slope of 0.28 suggests an
early transition state. Lithium enolates give no such
Brønsted correlation suggesting that the electrophilic
character of the lithium cation in the transition state (or
some other independent property) is of comparable
importance to the nucleophilicity of the enolate group.
These observations have additional significance since
these are ion pair SN2 reactions instead of the more
traditional reactions with anions.
The alkylation reactions of LiPAT with alkyl halides
are exclusively those of C-alkylation. Only with the
addition of large amounts of HMPA does O-alkylation
compete. This observation suggests that alkylation reac-
tions of lithium enolates could be enhanced by adding
HMPA with little danger of loss to O-alkylation. CsPAT,
however, gives substantial O-alkylation even in the
absence of HMPA. Addition of HMPA further increases
the role of O-alkylation. Coordination of the enolate
oxygen clearly favors C-alkylation. Weakening this in-
teraction, either by using a large cation or by making
the cation effectively larger with solvation, enhances the
role of O-alkylation. Increased steric hindrance at the
carbon center, however, such as by a substituent, also
enhances O-alkylation.
Rea ction P r od u ct: CsP AT + p-ter t-Bu tylben zyl Ch lo-
r id e. Monoalkylated (2-(p-tert-butylbenzyl)-6-phenyl-R-tetral-
one): MS (m/z, rel intensity) 370 (4), 369 (28), 368 (100), 367
(45), 353 (20), 320 (6), 311 (4), 234 (5), 222 (3), 221 (17), 194
(5), 193 (4), 178 (3), 165 (9). Dialkylated product (2,2-bis(p-
tert-butylbenzyl)-6-phenyl-R-tetralone: MS (m/z, rel intensity)
514 (1), 499 (1), 369 (7), 368 (41), 367 (100), 353 (3), 311 (4),
310 (1), 295 (6), 294 (22), 221 (3); HRMS found 515.3324, calcd
515.3314.
Exp er im en ta l Section
All UV measurements were carried out in a glovebox under
argon atmosphere at a constant temperature of 25.0 ( 0.1 °C,
maintained by a cooling bath. The sample compartment
located in the floor of the glovebox was connected to a
Shimadzu 3801 spectrometer with fiber optic cables. THF was
purified as described previously.43 Most indicators were avail-
able from previous work. The alkylating agents and indicators
were purified by vacuum sublimation or distillation. The
preparation of PAT was described previously.7
Ack n ow led gm en t. This work was supported in part
by NSF Grant No. 9980367.
Su p p or tin g In for m a tion Ava ila ble: Experimental de-
tails (12 tables and 20 figures). This material is available free
of charge via the Internet at http://pubs.acs.org.
11-Meth yl-11H-ben zo[b]flu or en e. This hydrocarbon has
J O034543D
been reported in the literature but without characterization.44
(45) Streitwieser, A.; Brown, S. M. J . Org. Chem. 1988, 53, 904-
906.
(46) Analysis by Analytical Services Laboratory, University of
California, Berkeley.
(43) Gronert, S.; Streitwieser, A., J r. J . Am. Chem. Soc. 1986, 108,
7016-7022.
(44) Lavoie, E. J .; Tulley, L.; Bedenko, V.; Hoffmann, D. Mutat. Res.
1981, 91, 167-176.
(47) Characterizations by Dr. Arlene McKeown.
8942 J . Org. Chem., Vol. 68, No. 23, 2003