Mechanism of the deprotonation reaction of alkyl benzyl ethers with n-butyllithium

Article abstract of DOI:10.1002/chem.201204467

Kinetic study of the α-lithiation of benzyl methyl ether (BME) by nBuLi has revealed that increasing the concentration of the organolithium compound does not necessarily increase the reactivity, and this is a consequence of the reactivities of the differe

Full text of DOI:10.1002/chem.201204467

FULL PAPER
DOI: 10.1002/chem.201204467
Mechanism of the Deprotonation Reaction of Alkyl Benzyl Ethers
with n-Butyllithium
M. Luz Raposo,[a] Fernando Fernꢀndez-Nieto,[b] Luis Garcia-Rio,*[a]
P. Rodrꢁguez-Dafonte,[a] M. Rita Paleo,[b] and F. Javier Sardina[b]
Abstract: Kinetic study of the a-lithia-
tion of benzyl methyl ether (BME) by
nBuLi has revealed that increasing the
concentration of the organolithium
compound does not necessarily in-
crease the reactivity, and this is a con-
sequence of the reactivities of the dif-
ferent nBuLi aggregates present in sol-
ution. We propose
mechanism, in which a pre-complexa-
tion step is a key process for substrates
a
dimer-based
bearing a donor oxygen atom that can
interact with the lithium cation to form
mixed dimers. For these studies, we
have developed a system based on UV/
Vis spectroscopy that allows kinetic
measurements to be conducted at
ꢀ808C under argon.
Keywords: aggregation · butyllithi-
um
· deprotonation · kinetics ·
mechanism
Introduction
understand their stability and reactivity.[7] A few studies
based on rapid injection NMR have provided insight into
the relative reactivities of different aggregates, from which it
was concluded that lower aggregates tend to be more reac-
tive.[8] However, the situation is often more complicated as
many organolithium compounds in ethereal solvents and in
the presence of diamines or halides are likely to form mix-
tures of coexisting homo- and mixed aggregates.[9] The pres-
ence of heteroaggregates can substantially modify the over-
all nature and reactivity of the organolithiums. Furthermore,
the outcomes of a wide variety of reactions can be explained
by the complexation of an organolithium reagent to the sub-
strate prior to the actual reaction.[10,11]
Three oligomeric structures (see Scheme 1) have been re-
ported for nBuLi: a hexamer in nonpolar hydrocarbon sol-
vents,[12] a tetramer in diethyl ether and in dimethoxyme-
thane,[13] and an equilibrium between the dimer and tetram-
er in THF.[3a,8,14] An important consequence of the existence
of several types of aggregates in solution and their different
reactivities is that fractional reaction orders have been
found, ranging from 0.25 to 1, in proton-transfer reactions
promoted by butyllithium.[15–17] Collum et al.[18] observed in
several studies carried out with lithium diisopropylamide
(LDA) that a reaction order of 0.5 implies the existence of
monomeric species in solution, whereas when the reaction
order is 1 the mechanism changes due to the appearance of
the dimer aggregate in solution.
Organolithium compounds are widely used in organic syn-
thesis[1] as they are good nucleophiles and strong bases and,
no less importantly, simple organolithiums are commercially
available and inexpensive. As organolithium reagents play a
central role in organic chemistry, their reactivity, structure,
and stability have been the subject of numerous studies.[2]
Most organolithiums are aggregated in solution, with a
degree of aggregation that is dependent on the temperature,
solvent, concentration, and the presence of chelating ligands
or additives.[3] Organolithium reagents assemble in solution
mainly as hexamers, tetramers, and dimers, besides the mon-
omers.[4] Higher aggregates are observed in hydrocarbon sol-
vents,[5] while donor solvents, such as ethers, and chelating
agents
(N,N,N’,N’-tetramethyl-1,2-ethane
N,N,N’,N’’,N’’-pentamethyldiethylenetriamine
(TMEDA),
(PMDTA),
N,N,N’,N’-tetramethyl-1,2-diaminocyclohexane (TMCDA))
favor lower aggregates.[6] For instance, PhLi in diethyl ether
exists as a mixture of tetramer and dimer, but conversion to
the dimer is observed upon the addition of TMEDA.[4a]
Due to aggregation and solvation processes, it is impor-
tant to know the structure of organolithiums in solution to
[a] Dr. M. L. Raposo, Prof. L. Garcia-Rio, Dr. P. Rodrꢀguez-Dafonte
Departamento de Quꢀmica Fꢀsica
Centro de Investigaciꢁn en Quꢀmica Biolꢁgica y Materiales
Moleculares (CIQUS), Universidad de Santiago
15782 Santiago (Spain)
In contrast to the amount of published papers concerning
the synthetic applications of nBuLi, to the best of our
knowledge there have only been a few published studies
about kinetics and reaction mechanisms. The lack of kinetic
data for processes in polar solvents such as THF is due to
the high reactivity of the organolithium species present in
such solvents under the usual kinetic working conditions.
Conventional spectrophotometers are generally unsuitable
for following reactions of organolithiums because it is neces-
Fax : (+34)981595012
E-mail: luis.garcia@usc.es
[b] Dr. F. Fernꢂndez-Nieto, Dr. M. R. Paleo, Prof. F. J. Sardina
Departamento de Quꢀmica Orgꢂnica, Centro de Investigaciꢁn
en Quꢀmica Biolꢁgica y Materiales Moleculares (CIQUS)
Universidad de Santiago, 15782 Santiago (Spain)
Supporting information for this article is available on the WWW
under http://dx.doi.org/10.1002/chem.201204467.
Chem. Eur. J. 2013, 19, 9677 – 9685
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9677

Information). To prove that the
yellow color was due to the
benzylic carbanion, the reaction
mixture was quenched with
MeOD. Subsequent 1H NMR
spectroscopic analysis of the
crude reaction mixture showed
only the presence of BME deu-
terated at the benzylic position;
Scheme 1.
therefore, the observed kinetic
behavior corresponded to the
sary to work at very low temperatures and under complete
exclusion of air and moisture to prevent undesirable reac-
tions. To gain a deeper understanding of the behavior of
nBuLi, we have studied the kinetics of the deprotonation of
several alkyl benzyl ethers (benzyl methyl, benzyl ethyl, and
benzyl tert-butyl ethers: BME, BEE, and BTE, respective-
ly), diphenylmethane (DPM), and triphenylmethane (TPM),
by using nBuLi in THF at ꢀ40 and ꢀ808C, with a low-tem-
perature probe and a UV/Vis spectrophotometer. The data
metalation process, and not to a Wittig rearrangement or a
b-elimination process.[20] After 24 h, however, the solution
1
had turned black and the H NMR spectrum, after quench-
ing with MeOD, was complex; no starting BME was ob-
served. Analysis of the reaction kinetics showed two well-
defined kinetic processes, the first one corresponding to the
anion formation (see the Supporting Information). The first
objective of this work was to study the proton-transfer proc-
ess between BME and nBuLi.
1
7
obtained were compared with the results of H and Li or
6Li NMR spectroscopic experiments conducted at the same
temperatures to ascertain the degree of aggregation of the
nBuLi reactive species. The influences of organolithium con-
centration, changes in the polarity of the medium, and the
addition of TMEDA on the equilibrium constants between
different aggregates and on the reaction rate have been in-
vestigated. Moreover, we have studied the dependence of
the reaction rate on the stability of the generated carbanion.
Figure 1 shows the influence of [nBuLi] on the pseudo-
first-order rate constant (kobs) for the deprotonation of
BME by this organolithium in anhydrous THF at ꢀ808C
under argon. The rate constant is seen to increase with
[nBuLi], but only up to a limiting value, above which
([nBuLi]>0.2m) a further increase in organolithium concen-
tration has no appreciable effect on the observed rate con-
stant.
Results and Discussion
Deprotonation of benzyl methyl ether (BME): We first stud-
ied the nBuLi-mediated a-lithiation reaction of benzyl
methyl ether shown in Scheme 2. nBuLi is a strong base
Scheme 2.
Figure 1. Plot of kobs versus [nBuLi] in THF/hexane [0.3 to 27% (v/v)] for
deprotonation of BME at ꢀ808C. [BME]=1.0ꢄ10ꢀ3 m. The solvent com-
position changed from 99.7% THF to 73% THF/27% hexanes on
increasing the n-butyllithium concentration.
with an estimated pKa of around 48, whereas the benzylic
proton of BME has an apparent pKa of 36.9—37.0,[19] hence
the a-lithiation reaction should be thermodynamically very
favored. It is well established in the literature that a carbon
acid will be deprotonated if there is a difference in pKa of
2–3 units between the substrate and the organolithium re-
agent.
The kinetic study was carried out under pseudo-first-order
conditions with [nBuLi] much higher than [BME]. When a
solution of BME (T=ꢀ808C) in anhydrous THF was treat-
ed with nBuLi, the reaction mixture turned yellow and so
the reaction could be followed spectrophotometrically by
monitoring the change in absorbance at a suitable wave-
length (475 nm) as a function of time (see the Supporting
To obtain the experimental results shown in Figure 1, we
used a commercial 2m solution of nBuLi in hexanes. As a
consequence, the composition of the medium changed from
99.7% THF to 73% THF/27% hexanes as the nBuLi con-
centration was increased. To assess whether the observed
profile might have been a result of decreasing polarity, we
studied the metalation of BME in 100% THF and in differ-
ent THF/hexane mixtures (v/v). Figure 2 shows that for
100% THF, 75% THF/25% hexanes, and 50% THF/50%
9678
www.chemeurj.org
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 9677 – 9685

Effect of Butyllithium Aggregation on its Reactivity
FULL PAPER
Figure 2. Plots of kobs versus [nBuLi] for deprotonation of BME in THF/
hexanes mixtures: 100% THF (*), 75% THF/25% hexanes (*), and
50% THF/50% hexanes (&) at ꢀ808C. [BME]=1.0ꢄ10ꢀ3 m.
hexanes, very similar behavior to that depicted in Figure 1
was observed. From these experiments, we concluded that
the experimental behavior (increase of kobs with [nBuLi] up
to a limiting value) could not be attributed to a kinetic
medium effect.
The observed behavior is characteristic of processes in-
volving an equilibrium stage prior to the rate-determining
step, and it is well-known that organolithium reagents are
prone to pre-complexation with substrates bearing different
donor atoms.[21] The possibility of a previous pre-equilibrium
stage before the rate-determining (deprotonation) step may
be investigated by comparing the behavior observed in the
metalation of BME with that of a compound without donor
atoms. To this end, we used triphenylmethane (TPM). Equi-
librium prior to the rate-determining step could be establish-
ed in the metalation of BME, but not in the case of TPM as
it lacks a donor oxygen atom. As is evident from Figure 3,
similar behavior was observed for the deprotonation of
these two different substrates by nBuLi, that is, an increase
in the rate constants with [nBuLi] up to a limiting value.
This indicates that a pre-complexation stage prior to the
proton-transfer step was not responsible for the observed
behavior in Figures 1 and 2, although it could have had an
Figure 3. Plots of kobs versus [nBuLi] in THF/hexanes (0.3 to 27%) for
deprotonations of BME (a) and TPM (b) at ꢀ408C. [BME]=1.0ꢄ10ꢀ3 m;
[TPM]=2.5ꢄ10ꢀ4 m. The solvent composition changed from 99.7% THF
to 73% THF/27% hexanes on increasing [nBuLi].
concentration of nBuLi and the polarity of the medium.
Some representative examples of 6Li NMR spectra as a
function of nBuLi concentration (Bu6Li in [D8]THF 0.06,
0.18, 0.36, and 1.2m) and solvent composition (different
THF/hexanes mixtures) are shown in the Supporting Infor-
6
mation. The Li NMR spectroscopic signal corresponding to
the BuLi tetramer increases both on increasing the BuLi
concentration and the percentage of hexanes in the solvent
mixture.
influence on the value of kobs
.
The observed behavior of kobs versus [nBuLi] may be ex-
plained in terms of the aggregation state of nBuLi and the
different reactivities of the aggregates. It is well known that
nBuLi in THF exists as a tetramer in equilibrium with a
dimer, an equilibrium that is affected by changes in concen-
tration and solvent composition. The experimental kinetic
results can be explained on the basis of this tetramer–dimer
equilibrium of nBuLi if on increasing the [nBuLi] the less
reactive aggregate, the tetramer, is favored,[8,22] resulting in
a nonlinear dependence of the reaction rate on concentra-
tion. Moreover, on increasing the nBuLi concentration, the
solvent composition changed because we used a 2m solution
of nBuLi in hexanes.
From a synthetic point of view, it is well established that
it is sometimes necessary to add a chelating agent to a reac-
tion mixture to increase the reactivity of an organolithium
reagent.[23,24] However, the role of TMEDA as a promoter
of deaggregation and therefore reactivity has been ques-
tioned, especially when it is deployed in the presence of
donor solvents.[23] TMEDA competes with THF in the solva-
tion of organolithium compounds, but, even at high concen-
trations, it is not able to transform all of the nBuLi tetramer
into the dimer. Although spectroscopic studies have shown
that nBuLi/TMEDA in toluene exists exclusively as
TMEDA-solvated dimers,[25] it has recently been shown that
in TMEDA/THF up to three different hetero- and homosol-
vated dimers coexist.[6a,26] 6Li NMR spectroscopic experi-
ments on nBuLi in THF and THF/TMEDA have shown
6
We used Li NMR spectroscopy to study changes in the
dimer–tetramer equilibrium as a result of changes in the
Chem. Eur. J. 2013, 19, 9677 – 9685
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9679

L. Garcia-Rio et al.
that, as expected, the proportion of the dimer increases in
the presence of the coordinating agent (see the Supporting
Information). Therefore, we decided to investigate the pres-
ence of TMEDA in our reaction media to confirm that the
aggregation equilibrium is responsible for the kinetic behav-
ior for the deprotonation of BME depicted in Figure 1.
The experiments were carried out by maintaining a
with that reported in the literature (K=47.6mꢀ1).[14a] From
these values, the aggregation equilibrium constants can be
interpolated for solvent compositions ranging from 100%
THF to 70% THF/30% hexanes, as used in most of our ki-
netic experiments.
Figure 5 shows the dimer and tetramer concentrations as
a function of the total nBuLi concentration, and as a func-
tion of the solvent composition. As can be observed, both
[TMEDA]/ACHTUNGTRENNUNG[nBuLi] ratio of 2. As shown in Figure 4, the ob-
served rate constant for the metalation of BME increased
with [nBuLi] in the presence of TMEDA. It is remarkable
Figure 4. Plots of kobs versus [nBuLi] in THF/hexanes (0.3 to 27%) for
deprotonation of BME at ꢀ808C in the absence of TMEDA (*) and
with [TMEDA]/ACHTUNGTRENNUNG
[nBuLi] (*). [BME]=1.0ꢄ10ꢀ3 m.
that, in the presence of [nBuLi]=0.52m, the value of kobs
was increased approximately tenfold due to the presence of
[TMEDA]=1.04m. This behavior allows us to propose that
the results shown in Figure 1 are basically due to a self-ag-
gregation process of the organolithium reagent, as a result
of which the percentage of the tetramer increases with in-
creasing [nBuLi].
Figure 5. a) Influence of the total butyllithium concentration on the
dimer (c) or tetramer (a) concentrations. b) Variation of the frac-
tion of nBuLi that is present as a dimer according to the [nBuLi] at
ꢀ808C and in different THF/hexanes compositions: blue, 100% THF;
black, 75% THF; red, 50% THF; and green, 25% THF.
Calculation of aggregation equilibrium constants: To ana-
lyze our kinetic results from a quantitative point of view, we
needed to obtain the values of the equilibrium constants for
the nBuLi aggregation (dimer to tetramer association) in
the different solvent compositions used. By defining the
equilibrium constant for the nBuLi aggregation as a dimer–
tetramer expression, we calculated K from the integrals of
the dimer and tetramer concentrations increase on increas-
ing the amount of butyllithium, this increase being more
pronounced for the tetramer than for the dimer. In fact, at
[nBuLi]>0.1m, the total concentration of the tetramer ex-
ceeds that of the dimer and the difference increases on in-
creasing the organolithium concentration. We consider that
increases in the butyllithium concentration induce the for-
mation of higher proportions of the tetramer (see Figure 5b
for the variation of the fraction of nBuLi that is present as
the dimer), which justifies the observed fact that the rate
constant of proton abstraction reaches a limiting value, since
the tetramer is less reactive than the dimer in proton-trans-
fer reactions. This reduction of the reaction rate is further
enhanced by the change in solvent polarity.
6
the Li NMR spectroscopic signals for the dimer and tetram-
er (see Figure S3, Supporting Information).
Table 1 shows the values of the aggregation equilibrium
constants as a function of solvent composition. As can be
observed, the values of K increase on increasing the per-
centage of hexanes in the solvent composition, with the ob-
tained value for 100% THF being in very good agreement
Table 1. Equilibrium constants for nBuLi aggregation in different THF/
hexanes mixtures at T=ꢀ808C.
% Hexanes (v/v)
0%
25%
50%
75%
K/mꢀ1
4.41ꢄ101
1.26ꢄ102
1.75ꢄ102
8.49ꢄ102
9680
www.chemeurj.org
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 9677 – 9685

Effect of Butyllithium Aggregation on its Reactivity
FULL PAPER
Reaction mechanism: The nonlinear kobs versus [nBuLi] pro-
file shown in Figure 1 can be explained by considering the
aggregation equilibrium of n-butyllithium, with the dimer
concentration being affected by the solvent composition and
the total organolithium concentration. In a first approxima-
tion, two reaction pathways may be considered involving the
dimer and tetramer of nBuLi. However, in a recent study,
Reich and co-workers demonstrated that the rate of acety-
lene deprotonation by the dimer is at least 106 times higher
than that by the tetramer.[8b] Consequently, we can neglect
the lithiation pathway by the tetramer and propose the
mechanism shown in Scheme 3. From this scheme, we pro-
Scheme 3.
pose a rate equation for which the observed rate constant
conforms to a linear dependence on dimer concentration.
kobs ¼ k2½ðnꢀBuLiðTHFÞ2Þ2ꢁ
ð1Þ
Figure 6. a) Plots of kobs versus [dimer]/m in 100 (*), 75 (*), and 50%
THF (&). b) Plot of kobs versus [dimer]/m, with the solvent composition
ranging from 99.7 to 73% THF/hexanes.
Considering the proton transfer as an elementary step, the
rate of the reaction can be written as the product of a “bi-
molecular” rate constant, k2, multiplied by the concentration
of the reacting species: nBuLi dimer and total BME concen-
Figure 6b shows an analogous linear plot for BME depro-
tonation by using a commercial solution of nBuLi in hex-
anes (kobs versus [nBuLi] data in Figure 1), for which the sol-
vent composition varied from 99.7% THF to 73% THF/
27% hexanes on increasing the butyllithium concentration.
The very good linear plot confirms that the nonlinear plot in
Figure 1 can be attributed to the nonlinear dependence of
the dimer concentration on total [nBuLi] as a consequence
of the aggregation equilibrium and the change in its equili-
brium constant due to changes in solvent composition.
Moreover, the slope of the plot (k2 =1.39ꢄ10ꢀ3 mꢀ1 sꢀ1) is in
very good agreement with previous values obtained for
other solvent compositions.
tration (rate=k2ACHTUNGTRENNUNG[BME]ACHTUGNTREN[NUGN dimer]). In cases for which the con-
centration of one of the reactants, BME under our experi-
mental conditions, is much lower than that of the other,
nBuLi in the present study, it can be assumed that the latter
concentration remains constant during the reaction. Assum-
ing that the nBuLi concentration remains constant during
the reaction is equivalent to assuming that the dimer con-
centration remains constant, and consequently the rate
equation can be rewritten as: rate=kobsACHTUNTRGNE[UGN BME], in which kobs
is expressed by Equation (1). If the nBuLi aggregation equi-
librium constant is available, we can use either the total
nBuLi concentration or the dimer concentration in the rate
equation.
The dimer concentration can be easily obtained by consid-
ering the mass balance for nBuLi and the equilibrium con-
stant for its dimer–tetramer aggregation. Figure 6a shows
the linear dependences obtained for BME deprotonation in
100, 75, and 50% THF. As can be observed, the slopes of
the plots (k2 =1.50ꢄ10ꢀ3, 1.67ꢄ10ꢀ3, and 1.16ꢄ10ꢀ3 mꢀ1 sꢀ1
for 100, 75, and 50% THF, respectively) are independent of
the solvent, which suggests that the deprotonation rate con-
stant is largely unaffected by solvent polarity under our ex-
perimental conditions. Small but perceptible intercepts can
be observed in Figure 6, which can be attributed to compet-
ing butyllithium decomposition.
Deprotonation reactions of DPM and TPM: To assess
whether the proton-transfer rate constant is affected by the
pKa difference between the donor and acceptor, we studied
the deprotonation of diphenylmethane (DPM) (pKa =
33.25)[27] and triphenylmethane (TPM) (pKa =31.26),[27] and
compared the results with those obtained for BME (pKa =
36.9–37). The reactions between DPM or TPM and nBuLi
in THF result in the formation of the corresponding anions,
which could be monitored at 510 and 570 nm, respectively,
under pseudo-first-order conditions with [substrate]=2.5ꢄ
10ꢀ4 m and changing [nBuLi] from 8.3ꢄ10ꢀ3 to 0.20m in dry
THF/hexanes mixtures at ꢀ408C under argon. Due to the
Chem. Eur. J. 2013, 19, 9677 – 9685
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9681

L. Garcia-Rio et al.
very low reactivities of DPM and TPM at ꢀ808C, the reac-
tions were carried out at ꢀ408C.
If the differences in pKa values between base and acid
were the driving force of the reaction, we would expect the
order of reactivity to be TPM>DPM>BME, but surpris-
ingly BME (highest pKa value in this series) reacts faster
than the other substrates (see Figure 7 for plots of kobs
Scheme 4.
The proposed structure for the mixed dimer has a 2:1:3
ratio of nBuLi/BME/THF. The formation of mixed dimers
with other stoichiometries, such as 2:2:2, is also possible, but
these have not been considered here because if formed they
should only be present in very small proportions. From the
structures of BME and THF, it can be concluded that the
donor abilities of the respective oxygen atoms should be
similar, such that both can contribute to the formation of a
mixed dimer. However, the concentration of THF in the re-
action mixture was much higher than that of BME (typically
12.3m versus 1ꢄ10ꢀ4 m), such that we can propose the for-
mation of a mixed dimer with the minimum amount of
BME.
Figure 7. Influence of [nBuLi] on the pseudo-first-order rate constant,
kobs, for lithiation of BEE (*), BME (*), BTE (&), TPM (&), and
DPM (~) in dry THF at ꢀ408C.
versus [BuLi]), followed by TPM and DPM as expected. It
should be noted that BME, despite being a thousand times
less acidic than DPM, reacts a hundred times faster. This
result is contrary to intuition and needed to be confirmed in
detail. Other factors besides the pKa difference between
donor and acceptor can affect the rate of proton-transfer re-
actions. Probably the best known is the existence of a non-
perfect synchronization between proton transfer and charge
development in the transition state. This phenomenon has
been well documented in the literature[28] and is related to
the stabilization of the resulting carbanion by resonance. A
decrease in reactivity with increased capacity of the carban-
ion for resonance stabilization may be expected. In such a
situation, we would expect DPM to be more reactive than
TPM, but this is not the observed behavior.
To explain the experimental reactivity sequence reported
in Figure 7, we propose that the heteroatom of BME must
play a role in the reaction mechanism as it can coordinate to
lithium by replacing one of the THF molecules in the solva-
tion sphere of the nBuLi reacting dimer, as shown in
Scheme 4. This process takes place in a previous step to the
proton-transfer reaction, such
Deprotonation of BEE and BTE: The rates of the deproto-
nation reactions of benzyl tert-butyl ether (BTE) and benzyl
ethyl ether (BEE) by nBuLi were investigated by UV/Vis
spectroscopy by monitoring the anion formation at 425 and
430 nm, respectively. As in the case of BME, pseudo-first-
order conditions were established by maintaining low con-
centrations of the ether ([BEE]=2.30ꢄ10ꢀ3 m and [BTE]=
3.20ꢄ10ꢀ3 m) and high concentrations of nBuLi (between
0.01 and 0.4m) in THF at ꢀ408C under argon.
Plots of kobs versus [nBuLi] for the deprotonations of all
of the substrates investigated so far indicate that the order
of reactivity is BEE>BME>BTE>TPM>DPM (see
Figure 7). This trend in the reaction rates cannot be the con-
sequence of the different acidities of the substrates,[29] but it
provides strong evidence that the presence of a donor heter-
oatom in the substrate has an impact on the mechanism, im-
plying that pre-complexation to the nBuLi must be impor-
tant (Scheme 5).[21] We propose that the oxygen of the sub-
that the latter can be consid-
ered as an “intramolecular re-
action”. To investigate the im-
portance of this pre-coordina-
tion, we studied the a-lithiation
of different alkyl benzyl ethers
with nBuLi, increasing the alkyl
chain size from methyl to ethyl
to tert-butyl.
Scheme 5.
9682
www.chemeurj.org
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 9677 – 9685

Effect of Butyllithium Aggregation on its Reactivity
FULL PAPER
strate competes with THF to form a hetero-aggregate dimer
prior to the deprotonation step. Pre-complexation between
lithium and oxygen can explain the different reactivities of
the ether substrates: steric hindrance within the coordina-
tion sphere at the lithium atom limits the mixed aggregate
formation for BTE, while induction effects would favor
BEE over BME, resulting in increased reactivity. Taking
into account the Li–O pre-complexation, we propose the re-
action mechanism shown in Scheme 5 for the a-alkoxy lith-
iation reaction.
The observed rate constant for the whole process thus in-
cludes the mixed dimer formation and the proton transfer,
as shown in Equation (2) (K2 is the equilibrium constant for
dimerQmixed dimer and k3 is the kinetic rate constant for
proton transfer).
K2k3½ðnBuLiðTHFÞ2Þ2ꢁ
ð2Þ
kobs
¼
1 þ K2½ðnBuLiðTHFÞ2Þ2ꢁ
Although the value of K2 is unknown, the dimer concentra-
tion is lower than 0.06m under our experimental conditions,
so we can simplify the expression for kobs to Equation (3).
kobs ¼ K2k3½ðnBuLiðTHFÞ2Þ2ꢁ
ð3Þ
Figure 8. a) 7Li NMR spectrum for [nBuLi]=0.32m. b) 7Li NMR spec-
trum of a mixture of [nBuLi]=0.32m and dimethylbenzyl methyl ether
(0.31m) in [D8]THF at ꢀ808C.
Formally, this equation is equal to Equation (1), and
therefore kobs is the result of multiplying both processes and
dimer concentration. Thus, we believe that for those com-
pounds without basic heteroatoms the process is controlled
by proton transfer, whereas pre-complexation is a key proc-
ess for those compounds bearing basic heteroatoms.
um compound does not necessarily increase the reactivity.
This behavior is not a consequence of the polarity of the
medium or of the substrate, but of the reactivity of the dif-
ferent aggregates of nBuLi present in the reaction media:
tetramer and dimer. The presence of TMEDA resulted in
an enhanced rate as a consequence of a higher proportion
of dimer in the reaction media, confirming the latter as the
reactive species.
Differences in the reaction rates for the various substrates
studied indicate that the widely accepted mechanism, in
which the nucleophilic carbon atom of nBuLi attacks the
acidic hydrogen atom, only explains the reactivity when
there are no basic heteroatoms in the molecule, whereupon
the pKa is the determining factor in the reaction. For those
compounds bearing a basic heteroatom that can coordinate
to the lithium, however, there is a step prior to deprotona-
tion, namely complexation of the lithium to the donor atom
to form the more reactive mixed dimer.
To obtain evidence supporting the formation of a mixed
7
aggregate, we carried out Li NMR spectroscopic studies at
ꢀ808C on two samples, one containing exclusively nBu7Li
(0.32m in [D8]THF) and the other containing a mixture of
nBu7Li (0.32m) and dimethylbenzyl methyl ether (0.31m), a
compound with no acidic benzylic hydrogen atoms capable
of reacting with the organolithium reagent. The 7Li NMR
spectrum in the presence of dimethylbenzyl methyl ether
shows a new signal at a higher magnetic field than those of
the tetramer and dimer (Figure 8). This signal can be attrib-
uted to the heterodimer, and the upfield shift it displays
may be due to the lithium supporting a higher charge densi-
ty, probably because of a cation–p interaction with the aro-
matic ring.
Conclusion
We have performed a kinetic study of the deprotonation re-
actions of alkyl benzyl ethers, diphenylmethane, and triphe-
nylmethane with nBuLi. The experimental setup using UV/
Vis spectroscopy to perform the kinetic study in the absence
of air and water, at low temperatures (ꢀ80 or ꢀ408C), al-
lowed us to obtain satisfactory quantitative data. Analysis of
the kinetics of the deprotonation reactions with nBuLi re-
vealed that increasing the concentration of the organolithi-
Experimental Section
Reagents and solvents: Solvents were distilled[30] under argon immediate-
ly before use: THF from Na/benzophenone and hexane from CaH2.
Commercial nBuLi was purchased from Aldrich (2m solution in hex-
anes). The diphenylacetic acid used to check solution titers was dried at
608C under vacuum for 24 h before use. TMEDA was distilled from
CaH2 and stored over 4 ꢅ molecular sieves. BEE, BTE, and nBu6Li were
prepared according to procedures described in the literature.[31] Air- and
Chem. Eur. J. 2013, 19, 9677 – 9685
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9683

L. Garcia-Rio et al.
moisture-sensitive reagents and solvents were manipulated under argon
by using standard vacuum-line and syringe techniques.
1998, 120, 7201–7210; b) T. F. Bates, M. T. Clarke, R. D. Thomas, J.
Am. Chem. Soc. 1988, 110, 5109–5112; c) W. Bauer, W. R. Winches-
ter, P. v. R. Schleyer, Organometallics 1987, 6, 2371–2379; d) G.
Fraenkel, M. Henrichs, J. M. Hewitt, B. M. Su, M. J. Geckle, J. Am.
Chem. Soc. 1980, 102, 3345–3350; e) G. Fraenkel, A. M. Fraenkel,
M. J. Geckle, F. Schloss, J. Am. Chem. Soc. 1979, 101, 4745–4747;
f) L. D. McKeever, R. Waack, M. A. Doran, E. B. Baker, J. Am.
Chem. Soc. 1969, 91, 1057–1061.
NMR sample preparation: All NMR tubes were prepared using stock
solutions and sealed under vacuum. Samples in 100% THF were concen-
trated to dryness by using a double-line vacuum/Ar manifold, and then
[D8]THF was carefully added at ꢀ808C and the solution was maintained
at this temperature until data acquisition. TMEDA was then added if
necessary.
[5] R. D. Thomas, R. M. Jensen, T. C. Young, Organometallics 1987, 6,
565–571.
Kinetic sample preparation: The reaction flask was purged under argon
and then dry THF was added. The flask was cooled to ꢀ808C and then
nBuLi and the corresponding ether were added. In the experiments re-
quiring the use of TMEDA, the order of addition was THF, ether,
TMEDA, and, after cooling, nBuLi.
[6] a) B. Qu, D. B. Collum, J. Am. Chem. Soc. 2006, 128, 9355–9360;
b) O. Eppers, H. Gꢈnther, Helv. Chim. Acta 1992, 75, 2553–2562;
c) T. Fox, H. Hausmann, H. Gꢈnter, Magn. Reson. Chem. 2004, 42,
788–794; d) J. M. Saꢂ, Helv. Chim. Acta 2002, 85, 814–840; e) G.
Carbone, P. O’Brien, G. Hilmersson, J. Am. Chem. Soc. 2010, 132,
15445–15450.
[7] a) W. Bauer, P. v. R. Schleyer, in Advances in Carbanion Chemistry
(Ed.: V. Snieckus), JAI Press, Greenwich, 1992, Vol. 1, pp. 89–175;
b) V. H. Gessner, C. Dꢆschlein, C. Strohmann, Chem. Eur. J. 2009,
15, 3320–3334.
Instruments: 1H and 7Li NMR spectra were recorded on
a Varian
500 MHz spectrometer. Kinetic data were recorded by using an Oceans
Optics USB4000 UV/Vis spectrophotometer with a T300-RT-UV-BX
probe, which was introduced into a two-necked round-bottomed flask.
The reaction flask was cooled with a Haake EK90 immersion cooler in
an acetone bath.
Kinetic analyses: For analysis of the kinetic data, we used the method of
integration, following the reaction for at least four half-lives and working
under pseudo-first-order conditions, in which one of the reactant concen-
trations was kept in large excess and therefore effectively constant with
respect to the other ([nBuLi] at least tenfold higher than [carbon acid]).
Applying an isolation method, the reaction order with respect to the re-
actant in deficit was determined. The absorbance versus time data were
fitted to a first-order equation:
[8] a) J. F. McGarrity, C. A. Ogle, Z. Brich, H.-R. Loosli, J. Am. Chem.
Soc. 1985, 107, 1810–1815; b) A. C. Jones, A. W. Sanders, M. J.
Bevan, H. J. Reich, J. Am. Chem. Soc. 2007, 129, 3492–3493.
[9] a) B. L. Lucht, D. B. Collum, Acc. Chem. Res. 1999, 32, 1035–1042;
b) R. A. Gossage, J. T. B. H. Jastrzebski, G. van Koten, Angew.
Chem. 2005, 117, 1472–1478; Angew. Chem. Int. Ed. 2005, 44, 1448–
1454.
[10] a) P. Beak, A. I. Meyers, Acc. Chem. Res. 1986, 19, 356–363; b) V.
Snieckus, Chem. Rev. 1990, 90, 879–933; c) W. Bauer, P. v. R.
Schleyer, J. Am. Chem. Soc. 1989, 111, 7191–7198; d) J. M. Saꢂ,
P. M. Deyꢂ, G. A. SuÇer, A. Frontera, J. Am. Chem. Soc. 1992, 114,
9093–9100; e) R. A. Rennels, A. J. Maliakal, D. B. Collum, J. Am.
Chem. Soc. 1998, 120, 421–422.
[11] a) H. J. Reich, W. S. Goldenberg, B. ꢇ. Gudmundsson, A. W. Sand-
ers, K. J. Kulicke, K. Simon, I. A. Guzei, J. Am. Chem. Soc. 2001,
123, 8067–8079; b) H. J. Reich, W. S. Goldenberg, A. W. Sanders,
K. L. Jantzi, C. C. Tzschucke, J. Am. Chem. Soc. 2003, 125, 3509–
3521.
[12] D. Margerison, J. P. Newport, Trans. Faraday Soc. 1963, 59, 2058–
2063.
[13] a) P. West, R. Waack, J. Am. Chem. Soc. 1967, 89, 4395–4399; b) K.
Bergander, R. He, N. Chandrakumar, O. Eppers, H. Gꢈnther, Tetra-
hedron 1994, 50, 5861–5868.
ðA1ꢀAtÞ
ð4Þ
Ln
¼ ꢀkobst
ðA1ꢀA0Þ
in which A1, At, and A0 represent the absorbances at “infinite” time, at
time “t”, and at “zero” time, respectively.
Acknowledgements
Financial support from the Ministerio de Economia y Competitividad of
Spain (Project CTQ2011–22436) and the Xunta de Galicia (PGIDIT10-
PXIB209113PR, PGIDIT10-PXIB209155PR, and 2007/085) is gratefully
acknowledged.
[14] a) J. F. McGarrity, C. A. Ogle, J. Am. Chem. Soc. 1985, 107, 1805–
1810; b) W. Bauer, T. Clark, P. v. R. Schleyer, J. Am. Chem. Soc.
1987, 109, 970–977; c) I. Keresztes, P. G. Williard, J. Am. Chem. Soc.
2000, 122, 10228–10229.
[1] a) J. Clayden, Organolithiums: Selectivity for Synthesis, Elsevier Sci-
ence, Oxford, 2002; b) D. M. Hodgson (Ed.), Organolithiums in
Enantioselective Synthesis, Topics in Organometallic Chemistry,
Vol. 5, Springer-Verlag, Berlin, 2003.
[2] a) Z. Rappoport, I. Marek (Eds.), The Chemistry of Organolithium
Compounds, John Wiley & Sons, Chichester, 2004, Parts 1 and 2;
b) Z. Rappoport, I. Marek (Eds.), The Chemistry of Organolithium
Compounds, John Wiley & Sons, Chichester, 2006, Vol. 2; c) A.-M.
Sapse, P. v. R. Schleyer (Eds.), Lithium Chemistry: A Theoretical and
Experimental Overview, John Wiley & Sons, New York, 1995; d) E.
Buncel, J. M. Dust, Carbanion Chemistry: Structures and Mecha-
nisms, Oxford University Press, New York, 2003.
[15] T. Holm, Acta Chem. Scand. Ser. B 1978, 32, 162–166.
[16] R. Waack, P. West, J. Organomet. Chem. 1966, 5, 188–191.
[17] a) P. West, R. Waack, J. I. Purmort, J. Am. Chem. Soc. 1970, 92,
840–845; b) H. Luitjes, F. J. J. de Kanter, M. Schakel, R. F. Schmitz,
G. W. Klumpp, J. Am. Chem. Soc. 1995, 117, 4179–4180.
[18] D. B. Collum, A. J. McNeil, A. Ramirez, Angew. Chem. 2007, 119,
3060–3077; Angew. Chem. Int. Ed. 2007, 46, 3002–3017.
[19] a) P. GraÇa, M. R. Paleo, F. J. Sardina, J. Am. Chem. Soc. 2002, 124,
12511–12514; b) P. Monje, P. GraÇa, M. R. Paleo, F. J. Sardina,
Chem. Eur. J. 2007, 13, 2277–2289.
[20] a) U. Azzena, L. Pilo, A. Sechi, Tetrahedron 1998, 54, 12389–12398;
b) P. T. Lansbury, V. A. Pattison, J. D. Sidler, J. B. Bieber, J. Am.
Chem. Soc. 1966, 88, 78–84.
[21] a) D. R. Hay, Z. Song, S. G. Smith, P. Beak, J. Am. Chem. Soc. 1988,
110, 8145–8153; b) J. Clayden, in The Chemistry of Organolithium
Compounds (Eds.: Z. Rappoport, I. Marek), John Wiley & Sons,
Chichester, 2004, Part 1, Chapter 10.
[22] J. Heinzer, J. F. M. Oth, D. Seebach, Helv. Chim. Acta 1985, 68,
1848–1862.
[23] D. B. Collum, Acc. Chem. Res. 1992, 25, 448–454.
[24] a) P. Beak, R. A. Brown, J. Org. Chem. 1977, 42, 1823–1824;
b) D. W. Slocum, A. Carroll, P. Dietzel, S. Eilerman, J. P. Culver, B.
[3] a) D. Seebach, R. Hꢆssig, J. Gabriel, Helv. Chim. Acta 1983, 66,
308–337; b) W. Bauer, D. Seebach, Helv. Chim. Acta 1984, 67,
1972–1988; c) H. J. Reich, J. P. Borst, R. R. Dykstra, D. P. Green, J.
Am. Chem. Soc. 1993, 115, 8728–8741; d) D. Li, I. Keresztes, R.
Hopson, P. G. Williard, Acc. Chem. Res. 2009, 42, 270–280; e) A.-C.
Pçppler, H. Keil, D. Stalke, M. John, Angew. Chem. Int. Ed. 2012,
51, 7843–7846; f) G. Barker, P. O’Brien, K. R. Campos, Org. Lett.
2010, 12, 4176–4179; g) M. Tait, M. Donnard, A. Minassi, J. Lefranc,
B. Bechi, G. Carbone, P. O’Brien, J. Clayden, Org. Lett. 2013, 15,
34–37.
[4] a) H. J. Reich, D. P. Green, M. A. Medina, W. S. Goldenberg, B. ꢇ.
Gudmundsson, R. R. Dykstra, N. H. Phillips, J. Am. Chem. Soc.
9684
www.chemeurj.org
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Chem. Eur. J. 2013, 19, 9677 – 9685

Effect of Butyllithium Aggregation on its Reactivity
FULL PAPER
McClure, S. Brown, R. W. Holman, Tetrahedron Lett. 2006, 47, 865–
868.
[29] The expected trend based on acidities would be: TPM>DPM>
BME>BEE>BTE.
[30] W. L. F. Armarego, D. D. Perrin, Purification of Laboratory Chemi-
cals, Butterworth-Heinemann, Oxford, 1998.
[25] a) D. Waldmꢈller, B. J. Kotsatos, M. A. Nichols, P. G. Williard, J.
Am. Chem. Soc. 1997, 119, 5479–5480.
[31] a) J. Barluenga, L. Alonso-Cires, P. J. Campos, G. Asensio, Synthesis
1983, 53–55; b) J. Barluenga, L. Alonso-Cires, G. Asensio, Synthesis
1979, 962–964; c) S. W. Wright, D. L. Hageman, A. S. Wright, L. D.
McClure, Tetrahedron Lett. 1997, 38, 7345–7348; d) G. Hilmersson,
O. Davidsson, J. Organomet. Chem. 1995, 489, 175–179.
[26] a) D. Hoffmann, D. B. Collum, J. Am. Chem. Soc. 1998, 120, 5810–
5811; b) J. L. Rutherford, D. Hoffmann, D. B. Collum, J. Am. Chem.
Soc. 2002, 124, 264–271, and references therein.
[27] a) A. Streitwieser, J. C. Ciula, J. A. Krom, G. Thiele, J. Org. Chem.
1991, 56, 1074–1076; b) D. A. Bors, M. J. Kaufman, A. Streitweiser,
J. Am. Chem. Soc. 1985, 107, 6975–6982.
Received: December 14, 2012
Revised: April 15, 2013
Published online: June 5, 2013
[28] a) C. F. Bernasconi, Adv. Phys. Org. Chem. 2010, 44, 223–324;
b) C. F. Bernasconi, Acc. Chem. Res. 1992, 25, 9–16.
Chem. Eur. J. 2013, 19, 9677 – 9685
ꢃ 2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemeurj.org
9685