The reaction of n-butyllithium with benzoic acid: Is nucleophilic addition competitive with deprotonation?

Article abstract of DOI:10.1002/(SICI)1099-1395(199707)10:7<537::AID-POC910>3.0.CO;2-Y

An evaluation of a branching vs sequential mechanism for the reaction of benzoic acid with n-butyllithium favors the latter.

Full text of DOI:10.1002/(SICI)1099-1395(199707)10:7<537::AID-POC910>3.0.CO;2-Y

JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 10, 537–541 (1997)
THE REACTION OF n-BUTYLLITHIUM WITH BENZOIC ACID: IS
NUCLEOPHILIC ADDITION COMPETITIVE WITH DEPROTONATION?†
PETER BEAK* AND LANCE A. PFEIFER
Department of Chemistry, University of Illinois, Urbana, Illinois 61801, USA
An evaluation of a branching vs sequential mechanism for the reaction of benzoic acid with n-butyllithium favors the
latter. © 1997 John Wiley & Sons, Ltd.
J. Phys. Org. Chem. 10, 537–541 (1997) No. of Figures: 2 No. of Tables: 3 No. of References: 12
Keywords: n-butyllithium; benzoic acid; nucleophilic addition; deprotonation
Received 10 October 1996; revised 4 December 1996; accepted 6 December 1996
INTRODUCTION
were claimed to be competitive. Our experiments showed,
however, that deprotonation precedes bromine lithium
exchange at the molecular level.4 Because of our interest in
this type of reaction, we have reinvestigated the reaction of
n-butyllithium with benzoic acid. We report that we are
unable to find the compelling evidence that we believe
would be needed to support the branching mechanism.
The branching pathway for reaction of benzoic acid and
n-butyllithium which was proposed in 1991 is shown in
Scheme 1.3 In one branch acid 1 reacts with n-butyllithium
to give carboxylate salt 2, which subsequently reacts slowly
with n-butyllithium to give 7. In the other branch,
nucleophilic addition of n-butyllithium to benzoic acid
gives adduct 3, which is also a precursor of the ketone 4.
The ketone reacts with n-butyllithium before quenching to
provide 5, the precursor to alcohol 6. Under this mecha-
nism, ketone 4 that is produced from 7 arises only on
exposure to aqueous acid. Consequently, alcohol 6 should
be produced only from 3 under the reaction conditions. In
the paper proposing this mechanism, it was stated that ‘18%
of n-butyllithium has been used in the nucleophilic
pathway’ and that control experiments have shown that ‘the
The reaction of carboxylic acids with organolithium rea-
gents to provide ketones and alcohols is well known.1 The
first step in the reaction has been considered to be formation
of a lithium carboxylate, which subsequently undergoes
nucleophilic addition by the organolithium reagent.2 In
1991, however, it was proposed that nucleophilic addition to
benzoic acid by n-butyllithium is competitive with deproto-
nation.3 The classic mechanism is sequential whereas the
more recent mechanism involves branching from benzoic
acid. If the later mechanism were to be correct, we believe
it would be the first case of transfer of a highly acidic proton
being slower than nucleophilic addition in organolithium
chemistry.
We have investigated reactions of organolithium reagents
for which deprotonation and bromine lithium exchange
* Correspondence to: P. Beak.
† Dedicated to Frederick G. Bordwell for his many and continuing
contributions to physical organic chemistry.
Contract grant sponsor: National Science Foundation.
Contract grant sponsor: National Institutes of Health.
Scheme 1
© 1997 by John Wiley & Sons, Ltd.
CCC 0894–3230/97/070537–05 $17.50

538
P. BEAK AND L. A. PFEIFER
Scheme 2
latter (n-butyllithium) adds to the carboxylate (2 in Scheme
1) with a relatively slow rate in comparison to other
reactions and we can consider that the bis-lithium alkoxide
(7 in Scheme 1) is virtually absent from the reaction
mixture.’3
The sequential reaction pathway is shown in Scheme 2
with the lithium carboxylate 2 being the precursor of 4 and
5.
We sought experimental data would distinguish the
branching and sequential pathways. In evaluating the
alternative mechanisms, we made the assumption that
Scheme 2 is the simpler process. The reaction of Scheme 1
has more species and more steps than Scheme 2. Under the
logic of processing experimental facts by Ockham’s razor,
Scheme 2 should be preferred unless there is compelling
experimental evidence to rule it out.5
butyllithium before the reaction was quenched after 30 s
with methanol and the ratio of ketone 4 and alcohol 6
determined. The results of these experiments are given in
Table 1.
The results in Table 1 show that the reaction of the
lithium carboxylate 2 with butyllithium leads to both the
ketone 4 and the alcohol 6. The salt 2 was fairly reactive
when the reactions were carried out at room temperature
and under conditions where the carboxylate was more
soluble. When either the temperature was lowered or the
amount of solvent was decreased, the amounts of addition
products decreased substantially, although they were still
detectable.
2 (1) n-BuLi
→
4+6
(2) H3O+
We note that neither our experiments nor other control
experiments which would use preformed 2 may be perfect
mimics of the reaction. The lithium carboxylate that is
produced in situ from n-butyllithium and benzoic acid could
be part of an aggregate that could have different reactivity
than exogenous lithium benzoate. Indeed 2, 3, 5 and 7
formed in the presence of n-butyllithium might be parts of
aggregates which would also involve n-butyllithium that
would have a different reactivity from n-butyllithium added
to preformed 2. Although we have not been able to make
rate comparisons, we note that if these results are taken to
show that 2 can give rise to 4 and 6 under the reaction
conditions in which 1 provides 4 and 6, the pathway in
Scheme 1 should be rejected. Under Scheme 1 the alcohol
6 can arise only via 3, which is not available from 2.
RESULTS AND DISCUSSION
The reaction of benzoic acid with n-butyllithium was
carried out in THF and found to provide the acid, the ketone
4 and the alcohol 6 as expected.3 The yields and ratios of
products were a function of the reaction conditions as
described (see below).3 We note that the reactions can be
heterogeneous, a factor which may complicate the inter-
pretation of both the reaction and control experiments.
The necessity for the branching pathways of Scheme 1 is
stated to be based in part of the assessment (see above) that
the carboxylate salt 2 reacts with n-butyllithium too slowly
to account for the products.3 In particular, if Scheme 1 is
correct, the alcohol 6 should not be produced from 2 under
the reaction conditions.
However, this interpretation is open to challenge. Rubot-
tom and Kim6 have shown that the amounts of tertiary
alcohols arising from the reaction of methyllithium with
carboxylic acids can be a function of the quenching reagent
We investigated the reaction of n-butyllithium with
lithium benzoate (2). The lithium carboxylate was prepared
from 1 and 1·15 equiv. of lithium hydride and exposed to n-
Table 1. Reaction of 2 with n-butyllithium
n-BuLi
(equiv.)
Temperature
(°C)
THF
Addition as
solvent (ml) % of n-BuLi
4:6 ratio
0·5
0·5
0·5
0·2
Ϫ78
20
20
40
40
5
7·8
73
18
18
0·7
1·5
0·2
0·6
20
5
© 1997 by John Wiley & Sons, Ltd.
JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 10, 537–541 (1997)

REACTION OF n-BUTYLLITHIUM WITH BENZOIC ACID
539
Table 3. Reactions of 1 and 1-d1 with n-butyllithium in THF at
Ϫ78 °C to give 4 and 6
(quenching with trimethylsilyl chloride greatly reduced, but
did not necessarily eliminate, the amount of tertiary alcohol
relative to quenching with aqueous ammonium chloride).
Such results suggest that the alcohol 6 might be an artifact
arising after the initial reaction and during the quenching.
This criticism could be applicable to a work-up with any
quenching reagent.6
n-BuLi
n-BuLi
4:6
Reactant
(equiv.)
addition (%)
ratio
1
1-d1
1·0
1·0
4·5
4·2
1:2·4
1:2
Accordingly, we carried out a second series of experi-
ments in which the amount of alcohol formed during the
reaction should be different under the reaction pathways in
Schemes 1 and 2. The relative rates of the reaction of
benzoic acid (1) and of benzoic acid-d1 (1-d1 ) with n-
butyllithium could be different. In this case, under the
mechanism in Scheme 2, the reaction would proceed via 2
from either 1 and 1-d1 and the ratios of products 4 and 6
would be the same from either reactant. However, the
mechanism in Scheme 1 should give different product ratios
for the reactions of 1 and 1-d1 . In this case a hydrogen–
deuterium isotope effect would favor the formation of 3
over 2 for the reaction of 1-d1 relative to 1. The subsequent
reactions of 3 should give a product profile for 4 and 6
which is different from that in which 2 provides 4. The
relative amount of 6 would be expected to be greater for the
reaction of 1-d1 than for 1 since in Scheme 1 3 provides the
only pathway to 6.
In order for this experimental test to be valid, the reaction
of 1-d with n-butyllithium must show a significant primary
hydrogen–deuterium isotope effect (the hydrogen deuter-
ium isotope effect for deprotonation of a carboxylic acid by
an alkyl lithium could be small if the reaction was diffusion
controlled or the transfer highly asymmetric7 ). We eval-
uated the isotope effect by the reaction of mixtures of
fivefold excesses of 1 and of 1-d with 3-phenylpropyl
lithium (8).4 The products of the reaction, 1-phenylpropane
(9) and 1-phenylpropane-3-d1 (9-d1 ), were analyzed by
GC–MS and by MS.
nominal isotope effect for the reaction of the 1:4 ratio than
for the 1:1 ratio of 1:1-d1 , however, extraction of a
quantitative isotope effect from these data is not warranted.
Reaction variables include the presence of 9 in 8 and
possible exposure to small amounts of atmospheric water.
The organolithium reagent 8 was shown by reaction with
deuterium oxide to contain about one third 9 (we presume
this results from the reaction of 8 with tert-butyl bromide
formed in the preparation of 8 from 1-phenyl-3-bromopro-
pane with tert-butyllithium). Nonetheless the present data
are sufficient to suggest that if 8 is a model for n-
butyllithium a hydrogen–deuterium isotope effect of at least
5:1 can be expected for the reaction of benzoic acid with n-
butyllithium under these reaction conditions. Thus, if 18%
of the reaction of 1 were to proceed by addition of n-
butyllithium to give 3 from 1 and n-butyllithium, then at
least 90% of the reaction of 1-d1 with n-butyllithium should
proceed via 3.
The products 4 and 6 from the reactions of 1 and 1-d1
with n-butyllithium are shown in Table 3. The results show
that these reactants afforded product yields and 4:6 ratios
which are very similar. Indeed, the slight differences, a
higher yield of addition products which contained a larger
amount of the alcohol 6 relative to 4 from 1 is the opposite
of what would be expected if 3 is the only precursor of 6 and
if more 3 is formed from 1-d1 than from 1. Accordingly,
these results discount Scheme 1 as compared with Scheme
2.
PhCO2H
Ph(CH2)2CH3
In summary, we have carried out two experiments to
evaluate the possibility that nucleophilic addition to benzoic
acid by n-butyllithium is competitive with deprotonation.
We do not find compelling evidence to support the
mechanism of Scheme 1 and suggest the data to be
consistent with the more prosaic mechanism of deprotona-
tion preceding nucleophilic addition. This preferred reaction
pathway as illustrated in Scheme 2 is as originally
suggested.1, 2
1
9
Ϫ78 °C
+PhCO2Li
+Ph(CH2 )3Li →
THF
8
2
PhCO2D
Ph(CH2 )2CH2D
1-d1
9-d1
The product ratios in Table 2 show there is an
experimentally significant isotope effect for the reaction of
1 and 1-d1 with 8. There appears to be a substantially greater
EXPERIMENTAL
General. 1H NMR spectra were obtained using a QE 300
MHz spectrometer. Chemical shifts are reported in parts per
million downfield from either tetramethylsilane (TMS) or
relative to CH2Cl2 as an internal standard and CDCl3 or
CD2Cl2 as solvent. When peak multiplicities are given, the
following abbreviations are used: s, singlet; d, doublet; t,
triplet; m, multiplet; and br, broadened. Mass spectra were
measured by Dr R. Milberg and co-workers (Noyes
Table 2. Reactions of mixtures of 1 and 1-d1 to give 1-d1 and 9-d1
Reactant ratio:
Product ratio:
1:1-d1
9:9-d1
1:4
1:1
4:1a
6:1a
a Corrected for natural isotopes and for the presence of 9 in 8.
© 1997 by John Wiley & Sons, Ltd.
JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 10, 537–541 (1997)

540
P. BEAK AND L. A. PFEIFER
Laboratory, University of Illinois, Urbana, IL, USA) on a
Finnigan MAT 731 mass spectrometer. Elemental analysis
was performed by the University of Illinois Microanalytic
Laboratory. Analytical gas chromatography (GC) was
performed on a Hewlett-Packard Model 5890 gas chroma-
tographer with a Hewlett-Packard Model 3396A integrator
and a 25 m HP-5 capillary column. GC–electron impact
ionization mass spectrometry (GC–MS) was carried out on
a Hewlett-Packard model 5890 GC–MS workstation. For
reactions with air-sensitive reagents, all glassware was oven
dried prior to use and all reactions were performed under a
nitrogen atmosphere.
silica column with 10% ethyl acetate–hexane as eluent to
give 0·380 g (82%, 95% pure) of 6. The alcohol 6 was a
clear, viscous oil. 1H NMR: (TMS, CDCl3 ), ␦ 7·36 (m, 3H,
ArH), 7·22 (m, 2H, ARH), 1·80 (m, 4H, COHCH2 ), 1·28 (m,
4H, CH2 ), 1·05 (m, 4H, CH2 ), 0·92 (m, 6H, CH3 ). GC
retention times: 4·26 min (impurity, 5%) and 7·07 min
(alcohol, 95%).
A T1 relaxation experiment for a known mixture of
ketone 4 and alcohol 6 was carried out to establish that the
relaxation times would allow accurate detection of the
product ratios by integration of an NMR spectrum of the
mixture. The experiment was performed using a known
1:1·3 ratio of the ketone to alcohol, and integration of the
spectrum after the T1 experiment showed a 1:1·2 ratio of the
ketone to alcohol. This experiment showed that accurate
(<10% error) integrations of the ketone and alcohol
products could be obtained under standard NMR data
acquisition conditions.
Materials. Diethyl ether (Et2O), pentane and tetra-
hydrofuran (THF) were distilled from sodium and
benzophenone under a nitrogen atmosphere immediately
preceding their use. 3-Phenylpropyl bromide was distilled
from calcium chloride at 58–60 °C and 0·15 Torr. n-
Butyllithium and tert-butyllithium were purchased from
Aldrich (1·5
M
) and titrated prior to use.8
Preparation of deuterated benzoic acid (1-d1 ). Benzoic
acid was stirred in 99·8% pure D2O for 1 h under nitrogen.
The solution was then concentrated until a small amount of
D2O remained (ca 1 ml). The deuterated acid was dried
under vacuum for 1 h at 50 °C, at which time the amount of
deuterated acid was reweighed (some sublimation had
occurred) and used immediately in the following steps.
Analysis: calculated for C6H5O2D, C 68·29, H 4·91; found,
C 68·28, H 4·87%; calculated for C6H5O2H, C, 68·84, H
4·95; found, C, 68·85, H 4·96%.
Isotopic ratios. The isotopic distribution was calculated
by the standard matrix method with respect to the molecular
ion region in the GC–MS trace of the corresponding
unlabeled compound.9 The isotopic distribution for an
intermolecular reaction was calculated based on the actual
spectra obtained minus the background spectra.
Preparation of 1-phenylpentanone (4).10 To a solution of
benzonitrile (4·85 mmol) in diethyl ether under nitrogen at
Ϫ10 °C was added 1·0 equiv. of n-butyllithium dropwise
over 5 min. The reaction mixture was stirred at this
temperature for 5 min, then warmed to room temperature
and stirred for an additional 3 h. The reaction mixture was
then poured into 50 ml of an ice–water mixture and warmed
to room temperature. The aqueous layer was extracted three
times with diethyl ether. The organic extracts were then
combined and washed once with brine. The organic layer
was dried over anhydrous sodium sulfate and concentrated,
giving a yellow oil. The crude product was purified on a
silica column with 10% ethyl acetate–hexane as eluent to
give 0·490 g (62%) of the product 4 as a non-viscous
slightly yellow oil of >95% purity. 1H NMR (TMS,
CDCl3 ): ␦ 7·98 (d, 2H, ARH), 7·35–7·60 (m, 3H, ArH),
2·97 (t, 2H, COCH2 ), 1·70 (m, 2H, CH2 ), 1·4 (m, 2H, CH2 ),
0·95 (t, 3H, CH3 ). GC retention times: 2·78 min (benzoni-
trile, 2%), 5·43 min (product, 95%) and 11·05 min
(impurity, 3%).
Reaction of benzoic acid with n-butyllithium (؊78 °C).
To a solution of the acid (5 mmol) in 40 ml of THF under
nitrogen was added n-butyllithium (5 mmol) at a rate of
1 ml/minϪ1. The solution was stirred at Ϫ78 °C for 30 min,
then quenched with a few milliliters of methanol at
Ϫ78 °C. The solution was concentrated, producing a white
solid, which was dissolved in a mixture of ethyl acetate and
0·1
M
KOH. The oranic layer was extracted three times with
40 ml of the potassium hydroxide solution, then the aqueous
layers were combined and extracted once with ethyl acetate.
The organic layers were combined and washed twice with
water and once with brine. The organic layer was dried over
anhydrous sodium sulfate and concentrated, giving a light
yellow oil. 1H NMR (CD2Cl2 ), 4+6: 4, ␦ 7·98 (d, 2H, ArH),
7·35–7·60 (m, 3H, ArH), 2·97 (t, 2H, COCH2 ), 1·70 (m, 2H,
CH2 ), 1·4 (m, 2H, CH2 ), 0·95 (t, 3H, CH3 ); 6, ␦ 7·36 (m,
3H, ArH), 7·22 (m, 2H, ArH), 1·80 (m, 4H, COHCH2 ), 1·28
(m, 4H, CH2 ), 1·05 (m, 4H, CH2 ), 0·92 (m, 6H, CH3 ). GC
retention times: 5·45 min (ketone) and 7·07 min (alcohol).
The ratios of the ketone and alcohol were determined by
measuring the relative peak heights of the respective
protons. Control experiments in which known amounts of
the ketone and alcohol were mixed and analyzed showed a
maximum error of 10% using this method of analysis. The
aqueous layer was reacidified with 6–HCL and extracted
with ethyl acetate twice. The organic layers were combined
and washed twice with water and once with brine, yielding
recovered benzoic acid. 1H NMR (TMS, CDCl3 ): ␦
Preparation of dibutylphenyl alcohol (6).11 To a solution
of nonan-5-one (2·13 mmol) in 25 ml of THF at 0 °C under
nitrogen was added 1·1 equiv. of phenylmagnesium bromide
over 2 min. The solution turned light yellow and was stirred
at 0 °C for 1 h, then warmed to room temperature and
stirred for another 30 min. The solution was quenched in
1
M
HCl and immediately extracted three times with diethyl
ether. The organic extracts were combined and washed once
each with water and brine. The organic layer was dried
under anhydrous sodium sulfate and chromatographed on a
© 1997 by John Wiley & Sons, Ltd.
JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 10, 537–541 (1997)

REACTION OF n-BUTYLLITHIUM WITH BENZOIC ACID
541
12·2–12·8 (s, br, 1H, COOH), 8·12 (d, 2H, ArH), 7·6
(t, ArH), 7·47 (t, ArH).
2 mmol of 8. The reagent became colorless immediately.
The recovered deuterated phenylpropane was analyzed by
GC–MS and MS. GC–MS data: background of phenyl-
propane standard, m/z 120 (100), 121 (10), 122 (0·5)l;
product for D2O: m/z 120 (57), 121 (100), 122 (10) for the
phenylpropane peak. MS data: background of phenyl-
propane standard, m/z 120 (100), 121 (10), 122 (0·4);
product for D2O, m/z 118 (7), 119 (23), 120 (69), 121 (100),
122 (10).
Preparation and reaction of lithium carboxylate (2). To a
solution of 5 mmol of benzoic acid in 40 ml of THF at room
temperature under nitrogen was added 1·15 equiv. of lithium
hydride. The solution was heated at 50–60 °C on a hot
water-bath for 1 h, during which time a white precipitate
formed. Bubbling of hydrogen gas was vigorous for the first
several minutes but was undetectable after 30 min. The
solution was concentrated, THF was added and the solution
was cooled to the desired temperature for reaction of the
carboxylate salt (20 or Ϫ78 °C) before the appropriate
amount of n-butyllithium (0·2 or 0·5 equiv.) was added at a
rate of 1 ml minϪ1. The solution was stirred for 30 s, at
which time it was quenched with 5 ml of methanol, stirred
for several minutes and then warmed, if needed, to room
temperature. The solution was concentrated, partitioned
Reaction of 3-phenylpropyllithium (8) with benzoic acid
(1). To the mixture of 1-d1 and 1 totaling 5 mmol in 40 ml
of THF was added 1 mmol of 8 at Ϫ78 °C. The solution
was stirred for 30 min, then quenched with methanol at
Ϫ78 °C. The solution was then partitioned between 0·1
M
KOH and diethyl ether. The aqueous layer was extracted
twice with diethyl ether. The ether extracts were combined
and washed twice with water and once with brine to afford
9 and 9-d1 . GC–MS and MS data were obtained. GC–MS
data: 80:20 D:H, m/z 120 (100), 121 (26), 122 (2) for the
phenylpropane peak; 50:50 D:H, m/z 120 (100), 121 (19),
122 (1) for the phenylpropane peak. MS data: 50:50 D:H,
m/z 119 (1), 120 (100), 121 (17), 122 (1).
between 0·1
M
KOH and ethyl acetate and separated. The
organic layer was extracted twice more with 30 ml of the
hydroxide solution. The aqueous layers were combined and
extracted once with ethyl acetate (40 ml). The organic
layers were combined and washed once with water and
twice with brine, dried over anhydrous sodium sulfate and
concentrated, giving a light yellow oil. NMR data were
taken on the mixture of ketone 4 and alcohol 6 by
comparing the integration of the ␣-protons on the ketone
with the phenyl protons. 1H NMR (CD2Cl2 ), 4+6: 4, ␦ 7·98
(d, 2H, ArH), 7·35–7·60 (m, 3H, ArH), 2·97 (t, 2H,
COCH2 ), 1·70 (m, 2H, CH2 ), 1·4 (m, 2H, CH2 ), 0·95 (t, 3H,
CH3 ); 6, ␦ 7·36 (m, 3H, ArH), 7·22 (m, 2H, ArH), 1·80 (m,
4H, COHCH2 ), 1·28 (m, 4H, CH2 ), 1·05 (m, 4H, CH2 ), 0·92
(m, 6H, CH3 ). GC retention times: 5·51 min (ketone) and
7·05 min (alcohol). The aqueous layer was reacidified using
ACKNOWLEDGMENTS
We are grateful to the National Science Foundation and the
National Institutes of Health for supporting this work.
REFERENCES
1. B. J. Wakefield, The Chemistry of Organic Compounds.
Pergamon Press, New York (1974); M. J. Jorgenson, Org.
React. 18 (1970).
2. R. Levine and M. Karten, J. Org. Chem. 41, 1176 (1976); T.
Cohen and Y. Ahn, Tetrahedron Lett. 35, 203 (1994).
3. C. Einhorn, J. Einhorn and J. L. Luche, Tetrahedron Lett. 32,
2771 (1991).
4. P. Beak, T. J. Musick, C. Liu, T. Cooper and D. J. Gallagher, J.
Org. Chem. 58, 7330 (1993); D. J. Gallagher and P. Beak, J.
Am. Chem. Soc. 113, 7984 (1991); P. Beak, T. J. Musick and C.
W. Chen, J. Am. Chem. Soc. 110, 3538 (1988).
5. R. Hoffman, V. I. Minkin and B. K. Carpenter, Bull. Soc. Chim.
Fr. 113, 117 (1996).
6. G. M. Rubottom and C. Kim, J. Org. Chem. 48, 1550 (1983).
7. L. Melander and W. H. Saunders, Jr, Reaction Rates of Isotopic
Molecules. Wiley, New York (1980); F. H. Westheimer, Chem.
Rev. 61, 265 (1961).
8. J. Suffert, J. Org. Chem. 54, 509 (1989).
9. J. B. Lambert, H. F. Shurvell, D. A. Lightner and R. G. Cook,
Introduction to Organic Spectroscopy, p. 351. Macmillan, New
York (1987).
10. L. S. Cook and B. J. Wakefield, J. Chem. Soc., Perkin Trans. 1,
2392 (1980).
11. P. Maslak, J. N. Narvaez and M. Parvez, J. Org. Chem. 56, 602
(1991).
12. J. Quick, P. Herlihy and J. F. Howes, J. Med. Chem. 27, 632
(1984).
5 ml of 6
M
HCl and extracted twice with ethyl acetate. The
organic layers were combined and washed twice with water
1
and once with brine, yielding recovered benzoic acid. H
NMR (TMS, CDCl3 ), recovered benzoic acid (1): ␦
12·2–12·8 (s, br, 1H, COOH(, 8·12 (d, 2H, ArH), 7·6 (t,
ArH), 7·47 (t, ArH).
Preparation of 3-phenylpropyllithium (8).12 To 25 ml of
diethyl ether at Ϫ78 °C were added 31·5 mmol (2·1 equiv.)
of tert-butyllithium over 8 min. The solution was stirred for
10 min, during which time a faint yellow color developed.
Neat 3-phenylpropyl bromide (15 mmol) was added by
syringe over 3 min, turning the solution bright yellow. The
solution was stirred for 45 min at Ϫ78 °C. A white
precipitate formed after 30 min of stirring. The diethyl ether
was evaporated using the Schlenk technique at Ϫ40 °C and
0·2 Torr (1 Torr=133·3 Pa) for 2 h, then 25 ml of pentane
were added by the Schlenk technique and the solution was
filtered through a dry Celite column, yielding an orange
solution of the lithium reagent.
Standardization of 3-phenylpropyllithium (8). To 10 ml of
99·8% D2O at room temperature under nitrogen were added
© 1997 by John Wiley & Sons, Ltd.
JOURNAL OF PHYSICAL ORGANIC CHEMISTRY, VOL. 10, 537–541 (1997)