Relative basicities of ortho-, meta-, and para-substituted aryllithiums

Article abstract of DOI:10.1021/jo8020083

(Chemical Equation Presented) The relative basicities of aryllithiums bearing methoxy, chlorine, fluorine, trifluoromethyl and trifluoromethoxy substituents at the ortho, meta, and para positions have been assessed. To this end, two aryllithiums of compar

Full text of DOI:10.1021/jo8020083

Relative Basicities of ortho-, meta-, and para-Substituted
Aryllithiums
Joanna Gorecka-Kobylinska and Manfred Schlosser*
Institute of Chemical Sciences and Engineering, Ecole Polytechnique Fe´de´rale (EPFL-BCh),
CH-1015 Lausanne, Switzerland
manfred.schlosser@epfl.ch
ReceiVed September 11, 2008
The relative basicities of aryllithiums bearing methoxy, chlorine, fluorine, trifluoromethyl and trifluo-
romethoxy substituents at the ortho, meta, and para positions have been assessed. To this end, two
aryllithiums of comparable basicity were equilibrated together with the corresponding bromo- or iodoarenes
in a 1: 2 mixture of pentanes with tetrahydrofuran at -50, -75, or -100 °C. The “basicity”
(protodelithiation) increments ∆∆G derived from the equilibrium constants are linearly correlated with
the relative protonation enthalpies of the corresponding aryl anions in the gas phase. However, the
correlation factor proves to be position-dependent. Compared with “naked” aryl anions, the basicity of
aryllithiums mirrors the effects of ortho, meta, and para substituents to the extent of 36%, 30%, and
25%, respectively.
Introduction
the relative basicities of the benzenides corresponding to the
haloaryllithiums generated) only to the extent of approximately
12%.2,3
An amazingly good free energy relationship was found to
exist between the gas-phase acidity of fluorobenzene, the three
difluoro-, three trifluoro-, and three tetrafluorobenzenes, and
pentafluorobenzene on the one hand and their rates of metalation
by sec-butyllithium in tetrahydrofuran at -751 or -100 °C2 on
the other hand. The same relationship holds between the gas-
phase acidity of oligochlorobenzenes and their reactivity toward
sec-butyllithium at -100 °C.3 Obviously, the stabilization
resulting from electrostatic and polarizing interactions with the
substituents is proportional at the level of the “naked” haloaryl
anions and of the transition states that mediate the countercurrent
transfer of a proton and the metal, although the kinetics (i.e.,
the free energies of activation) mirror the thermodynamics (i.e.,
The free energy relationship between gas-phase acidities and
reactivities toward metalation reactivities within the oligofluo-
robenzene and oligochlorobenzene series was by no means
foreseeable. It would be unreasonable to expect such a relation-
ship to cover also the comparison between substituted arenes
that belong to different structural families. In fact, although
fluorobenzene is slightly less acidic than chlorobenzene in the
gas phase,4 the former substrate reacts in tetrahydrofuran 7.9
times more rapidly with sec-butyllithium at -100 °C and more
than 20 times more rapidly with lithium 2,2,2,6-tetramethylpi-
peridide (LITMP).2,3 More generally speaking, there is no simple
(1) Bu¨ker, H. H.; Nibbering, N. M. M.; Espinosa, D.; Mongin, F.; Schlosser,
M. Tetrahedron Lett. 1997, 38, 8519–8522.
(2) Schlosser, M.; Marzi, E. Chem. Eur. J. 2005, 11, 3449–3454.
(3) Schlosser, M.; Marzi, E.; Cottet, F.; Bu¨ker, H. H.; Nibbering, N. M. M.
Chem. Eur. J. 2001, 7, 3511–3516.
(4) Andrade, P. B. M.; Riveros, J. M. J. Mass Spectrom. 1996, 31, 767–
770; Chem. Abstr. 1996, 125, 247111.
(5) Davico, G. E.; Bierbaum, V. M.; DePuy, C. H.; Ellison, C. B.; Squires,
R. R. J. Am. Chem. Soc. 1995, 117, 2590–2599.
222 J. Org. Chem. 2009, 74, 222–229
10.1021/jo8020083 CCC: $40.75 2009 American Chemical Society
Published on Web 11/26/2008

ortho-, meta-, and para-Substituted Aryllithiums
TABLE 1. Absolute (∆Hg) and Relative (∆∆Hg) Deprotonation
Enthalpies (in Parentheses) of Heterosubstituted Benzenes and Their
Statistically Corrected Rates (krel) of sec-Butyllithium Promoted
Metalation Relative to Benzene
ions and the corresponding aryllithiums (generated by metalation
of the aromatic substrates) as this time only ground-state species
are compared with each other. We decided to embark on such
a study.
Results
The acid-base equilibration involving two simple arenes and
the corresponding aryllithiums is not feasible. The ortho-
lithiation of anisole by phenyllithium requires long reaction
times at 100 °C,13 in other words, conditions causing extensive
decomposition of the reagent by solvent attack. The “uphill”
reverse reaction between benzene and 2-anisyllithium would
be even a hundred to a thousand times slower. Moreover, no
meta- or para-substituted aryllithiums would become accessible
by this transmetalation method. To assess the relative basicities
of typical aryllithiums, we have therefore established the
equilibrium between selected pairs of aryllithiums and the
corresponding bromo- or iodoarenes by permutational halogen/
metal interconversion (Scheme 2). If we assume the aryl-Br and
a Reference 5. b This number is considerably error-afflicted.1 c Computed
at an MP2 level of theory: 378 kcal/mol relative to 391 for benzene.6
d Reference 1. e Computed at an MP2 level of theory: 390.8 kcal/mol
relative to 399.8 for benzene.7 f See also the experimental data of 391
kcal/mol rel. to the identical value of water.8 g See also ref 9.
h Reference 3. i Computed at an MP2 level of theory: 383.2 kcal/mol
relative to 399.8 for benzene.7 j Reference 10. k See also refs 3 and 10.
SCHEME 2. Assessment of Relative Aryllithium Basicitiesa
correlation between the thermodynamic acidity of a substituted
benzene and its rate of metalation (Table 1).
This lack of agreement between thermodynamic and kinetic
acidities was to be expected. In this context it suffices to recall
the influence of the reagent on the regioselectivity of metalation.
The optional attack of the organometallic base at a site adjacent
to the more electronegative or the better coordinating substituent
in O-methoxymethoxy protected 4-fluorophenol and the N-tert-
butoxycarbonyl protected 4-fluoroaniline and 4-anisidine (4-
methoxyaniline) offers a striking illustration of this fact (Scheme
1).11,12
a By their halogen/metal permutational equilibration with the corre-
sponding bromo- or iodoarenes followed by trapping with a suitable
electrophile El (e.g., El ) I).
aryl-I bond strengths to be substituent-invariant, the equilibrium
constants can be directly translated into differential basicities
(i.e., thermodynamic stabilities). The postulated approximate
stability invariance of bromoarenes and iodoarenes is arguable.
Although at variance with some experimental findings,14 a DFT
study15 has revealed o-dihalobenzenes to be significantly less
stable (up to 2.6 kcal/mol in typical cases) than their meta- and
para-isomers. On the other hand, if the energies of all of our
ortho-substituted bromoarenes and iodoarenes were raised to
some extent, the relative aryllithium stabilities would be only
marginally affected. In view of other ambiguities, we consider
an error of (1 kcal/mol to be tolerable for any o-aryllithium
data and likewise (0.5 kcal/mol for m- and p-aryllithiums.
The equilibrations were carried out in tetrahydrofuran or
diethyl ether at -50, -75, or -100 °C approaching from both
sides the final mixture of four components. Analogous equili-
bration reactions have been reported previously.16,17 However,
they cover not more than a total of three examples and are
numerically unreliable as a result of inadequate reaction
conditions and analytical procedures.
SCHEME 1. Optionally Site-Selective Metalation of O- or
N-Protected 4-Fluorophenol, 4-Fluoroaniline, and
4-Anisidine Depending on the Reagent Employed
Thus one definitively cannot expect to find any far reaching
interdependence of the gas-phase acidity of electronegatively
substituted arenes and the rates with which they react with
organometallic bases. However, a correlation may be possible
between the relative stabilities (basicities) of the naked carban-
(6) Schlosser, M.; Mongin, F.; Porwisiak, J.; Dmowski, W.; Bu¨ker, H. H.;
Nibbering, N. M. M. Chem. Eur. J. 1998, 7, 1281–1286.
(7) Kovacˇevic´, B.; Maksic´, Z. B.; Primorac, M. Eur. J. Org. Chem. 2003,
3777–3783.
(8) Dahlke, G. D.; Kass, S. R. Int. J. Mass Spectrom. Ion Proc. 1992, 117,
633–645.
(9) Maggi, R.; Schlosser, M. Tetrahedron Lett. 1999, 40, 8797–8800.
(10) Castagnetti, E.; Schlosser, M. Chem. Eur. J. 2002, 8, 799–804.
(11) Mongin, F.; Maggi, R.; Schlosser, M. Chimia 1996, 50, 650–652.
(12) Marzi, E.; Mongin, F.; Spitaleri, A.; Schlosser, M. Eur. J. Org. Chem.
2001, 2911–2915.
(13) Wittig, G.; Pockels, U.; Dro¨ge, H. Ber. Dtsch. Chem. Ges. 1938, 71,
1903–1912.
(14) (a) Hubbard, W. N.; Knowlton, J. W.; Huffman, H. M. J. Phys. Chem.
1954, 58, 396–402. (b) Platonov, V. A.; Simulin, Y. N. Zh. Fiz. Khim. 1984,
58, 2682–2686; Russ. J. Phys. Chem. (Engl. Transl.) 1984, 58, 1630-1632;
Chem. Abstr. 1985, 102, 33733.
(15) Taskinen, E. Struct. Chem. 2000, 11, 293–301; Chem. Abstr. 2000, 134,
41897.
(16) Gilman, H.; Jones, R. G. J. Am. Chem. Soc. 1941, 63, 1441–1443.
(17) Winkler, H. J. S.; Winkler, H. J. Am. Chem. Soc. 1966, 88, 964–969.
J. Org. Chem. Vol. 74, No. 1, 2009 223

Gorecka-Kobylinska and Schlosser
SCHEME 4. Halogen/Metal Permutational Equilibration of
4-, 3-, and 2-Chlorophenyllithium with Phenyllithium, Other
Chlorophenyllithiums, or 2-Anisyllithium
The concentrations of the bromo- or iodoarenes were quanti-
fied by gas chromatography using a calibrated internal standard.
In the same way the amounts of products derived from the
aryllithium intermediates by electrophilic trapping (using mo-
lecular iodine if the exchange partners were bromoarenes and
1,2-dibromotetrafluoroethane if they were iodoarenes; see
Experimental Section) were determined. The two sets of data
have to be complementary to be meaningful. In concrete terms,
if two aryllithiums were found to coexist in a 90:10 ratio, the
corresponding bromoarene or iodoarene permutation partners
(introduced in equimolar quantities) have to be present in a
10:90 ratio, and consequently the equilibrium constant K
amounts to 81. Such an ideal component balance ordinarily
prevailed. Noticeable deviations were encountered when 2-chlo-
rophenyllithium was involved. This intermediate tends to slowly
undergo lithium chloride elimination accompanied by aryne
formation even at -100 °C.
We started our work by juxtaposing phenyllithium to each
of the three isomeric anisyllithiums (Scheme 3). The intercon-
SCHEME 3. Halogen/Metal Permutational Equilibration of
2-, 3-, and 4-Anisyllithium with Phenyllithium
mol if the gap between 2-chlorophenyllithium and phenyllithium
is bridged by 2-anisyllithium. Assuming the latter number to
be more reliable, we assign 4.4 kcal/mol as a nonarithmetic
average value. Such bona fide adjustments were made in all
following cases where a direct equilibration with phenyllithium
was not possible. For clarity, the bromoarenes and iodoarenes
exchange partners are omitted from the equilibration schemes
shown below.
The interconversion of 4-fluorophenyllithium with phenyl-
lithium, 3-anisyllithium, and 3- and 4-chlorophenyllithium gave
coinciding relative basicities of -1.5 to -1.7, i.e., on average
1.6 kcal/mol (Scheme 5). Upon equilibration with 3-fluorophe-
SCHEME 5. Relative Basicities of 4-, 3-, and
2-Fluorophenyllithium As Revealed by Halogen/Metal
Permutational Equilibrations
version with the corresponding iodoarenes proceeded in tet-
rahydrofuran even at -50 °C fairly slowly, equilibrium being
attained after 2-30 h. The basicities of the organometallic
species involved being not too different, it was possible to
determine all equilibrium constants K (see preceding paragraph)
and, as a corollary, the relative protodelithiation free energies
∆∆G (in kcal/mol) with sufficient accuracy. The introduction
of a methoxy group into a meta position of phenyllithium was
found to have virtually no effect on the basicity. In contrast it
destabilizes the aryllithium when attached to the para position
and stabilizes it considerably when located at an ortho position.
Next we turned to the chlorophenyllithiums. Whereas the
equilibrium position of the 4-isomer relative to phenyllithium
could be readily probed, the 2- and 3-isomers proved to be too
much stabilized to coexist with phenyllithium any more.
Therefore, 3-chlorophenyllithium was compared with 4-chlo-
rophenyllithium, and 2-chlorophenyllithium was examined in
competition with both 3-chlorophenyllithium and 2-anisyllithium
(Scheme 4).
As mentioned above, the chloroarene data are less reliable
than all others because of ꢀ-elimination of lithium chloride
occurring to a minor extent even at -100 °C. The two last
equilibria (in Scheme 4) enable us to check on the accuracy of
our approach. If the basicity of 2-chlorophenyllithium relative
to phenyllithium is evaluated using 3- and 4-chlorophenyllithium
as relay species, one counts a difference of 2.0 + 1.1 + 1.9 )
5.0 kcal/mol. This number shrinks to 4.2 () 2.8 + 1.4) kcal/
nyllithium, the latter was found to be 1.0 kcal/mol more stable.
Together with the permutation data involving 3- and 4-chlo-
rophenyllithium, the averaged relative basicity of -2.5 kcal/
mol was deduced (Scheme 5). Finally, when 2-fluorophenyl-
lithium was assessed relative to 3-methoxyphenyllithium and
3-chlorophenyllithium, the same value of -5.3 kcal/mol resulted
(Scheme 5).
224 J. Org. Chem. Vol. 74, No. 1, 2009

ortho-, meta-, and para-Substituted Aryllithiums
SCHEME 7. Relative Basicities of 4-, 3-, and
2-(Trifluoromethoxy)phenyllithium As Revealed by Halogen/
Metal Permutational Equilibrations
The assessment of the relative basicities of the (trifluoro-
methyl)phenyllithium isomers also allowed for experimental
redundancy as a control and check on the accuracy of the
equilibration method. The comparison of 3-(trifluorometh-
yl)phenyllithium with phenyllithium, 2-anisyllithium, 2-fluo-
rophenyllithium, and 4-(trifluoromethyl)phenyllithium (at -1.8
kcal/mol) gave a median value of -2.4 kcal/mol (Scheme 6).
SCHEME 6. Relative Basicities of 4-, 3-, and
2-(Trifluoromethyl)phenyllithium As Revealed by Halogen/
Metal Permutational Equilibrations
SCHEME 8. Basicity of
4-(2,2-Difluoro-1,3-benzodioxolyl)lithium Relative to
2-(Trifluoromethoxy)phenyllithium
According to the data collected with 2-anisyllithium, 2-fluo-
rophenyllithium, and 4-(trifluoromethyl)phenyllithium, 2-(tri-
fluoromethyl)phenyllithium was placed at -5.0 kcal/mol on the
basicity scale (Scheme 6).
TABLE 2. Increments ∆∆G Quantifying the Proton Affinity of
para-, meta-, and ortho-Substituted Aryllithiumsa
at the para
at the meta
at the ortho
substituent
position
position
position
The “exotic” trifluoromethoxy substituent attracts more and
more curiosity.18 Quantum chemical calculations7 have ranked
it, as far as its acidifying effect is concerned, on the top of all
substituents having no double bonds. For this reason we deemed
it indispensable to include it into the present investigation. 2-,
3-, and 4-(Trifluoromethoxy)phenyllithium were found to be
indeed less basic than any of the other ortho-, meta-, and para-
substituted species investigated by us so far (Scheme 7). Thus,
4-(trifluoromethoxy)phenyllithium exhibited a relative basicity
of -2.8 kcal/mol. In the same way, 3- and 2-(trifluoromethoxy)-
phenyllithium proved to be clearly less basic than 3- and
2-(trifluoromethyl)phenyllithium, relative basicity increments of
-4.0 and -6.0 kcal/mol being assigned.
We have drawn attention to the exceptional acidity of 2,2-
difluoro-1,3-benzodioxole already on previous occasions and
have demonstrated how this property can be advantageously
exploited for synthetic purposes.10,19-22 This heterocyclic unit
sets indeed a new record mark (∆∆G ) -8.7 kcal/mol) as the
lithiated species is still 2.5 kcal/mol less basic than 2-(trifluo-
romethoxy)phenyllithium (Scheme 8).
H3CO
F
Cl
F3C
F3CO
+0.3 (+0.2)
-1.6 (-4.3)
-1.9 (-7.6)
-1.8 (-11.3)
-2.8 (-9.8)
0.0 (-1.6)
-2.6 (-6.5)
-3.0 (-9.8)
-2.4 (-11.0)
-4.0 (-11.1)
-2.7 (-9.4)
-5.1 (-12.7)
-4.4 (-14.5)
-4.6 (-13.9)
-6.0 (-17.1)
a Present work. In tetrahydrofuran solution, relative to that of
phenyllithium and, in parentheses, differential protonation enthalpies
∆∆Hg of the corresponding aryl anions,7,23 in the gas phase, relative to
the parent benzenide (phenyl) ion. All numbers are given in kcal/mol.
protonation enthalpies of the corresponding aryl anions (Table
2). In the case of aryl anions, we quote computational (ab initio7
and density functional23) rather than experimental protonation
data as the latter are not always available, at least not with the
required reliability.
A comparison of enthalpies with free energies is acceptable
in the given case. The ∆H values of deprotonation in the gas
phase exceed the corresponding ∆G values by an almost
constant factor of 8 kcal/mol (representing the entropy change
due to proton detachment and the change in rotational sym-
metry).6 The equilibrium entropy in solution is assumed to be
small as substituent effects on the aggregation and solvation of
the examined aryllithium should be minute.
Discussion
The differences in the basicities of the substituted aryllithiums
relative to phenyllithium are listed together with the gas-phase
The aryllithium basicity increments ∆∆G were plotted
separately for the para-, meta-, and ortho-substituted species
against the protonation enthalpies ∆Hg of the corresponding
naked aryl anions in the gas phase. In all three regioisomeric
(18) Leroux, F.; Jeschke, P.; Schlosser, M. Chem. ReV. 2005, 105, 827–856.
(19) Castagnetti, E.; Schlosser, M. Eur. J. Org. Chem. 2001, 691–695.
(20) Schlosser, M.; Gorecka, J.; Castagnetti, E. Eur. J. Org. Chem. 2003,
452–462.
(21) Gorecka, J.; Leroux, F.; Schlosser, M. Eur. J. Org. Chem. 2004, 64–
68.
(22) Leroux, F.; Gorecka, J.; Schlosser, M. Synthesis 2004, 326–328.
(23) Hyla-Kryspin, I.; Grimme, S.; Bu¨ker, H. H.; Nibbering, N. M. M.; Cottet,
F.; Schlosser, M. Chem. Eur. J. 2005, 11, 1251–1256.
J. Org. Chem. Vol. 74, No. 1, 2009 225

Gorecka-Kobylinska and Schlosser
FIGURE 1. para-Substituted aryllithiums: their exchange-monitored
basicities relative to phenyllithium plotted against the protonation
enthalpies ∆Hg of the corresponding aryl anions in the gas phase (all
numbers being given in kcal/mol).
FIGURE 3. ortho-Substituted aryllithiums: their exchange-monitored
basicities relative to phenyllithium plotted against the protonation
enthalpies ∆Hg of the corresponding aryl anions in the gas phase (all
numbers being given in kcal/mol).
tion of the metal with the heteroelements oxygen, fluorine, or
chlorine in solution. This conclusion does not hold for gas-
phase structures or transition states (unless independent evi-
dence25 becomes available). Even in solution a different ground-
state behavior may be encountered when flexible electron-donor
substituents such as dimethylaminomethyl26,27 are examined.
The substitutent effects are substantially attenuated when one
moves from the naked anions to aryllithiums in solution, and
more intriguingly, the degree of this attenuation is position-
dependent. The differences in the slopes of 0.25, 0.30, and 0.36
(Figures 1-3) cannot be attributed alone to a field effect as
purely electrostatic interactions would level off with a higher
exponential power of the distance. To rationalize the findings
we proceed stepwise. We start with a working hypothesis that
will be tested and refined in future investigations. How is the
electron density distribution in a benzene molecule perturbed
upon proton abstraction? Having lost its second binding partner,
the lone pair will come closer to the sole remaining one, the
ipso-carbon atom of the benzenide ion (1), and expand spheri-
cally. To minimize electron-electron repulsion, the neighboring
σ-bonds become polarized by accumulating electron density at
the vicinal carbon atoms (C-2 and C-6) and by depleting the
negatively charged center (C-1). At the same time σ-π
coupling3,10,28 causes the π-electron cloud to drift away from
C-1 and move toward C-4.
FIGURE 2. meta-Substituted aryllithiums: their exchange-monitored
basicities relative to phenyllithium plotted against the protonation
enthalpies ∆Hg of the corresponding aryl anions in the gas phase (all
numbers being given in kcal/mol).
series, the points were reasonably well accommodated by
straight lines having slopes of 0.25, 0.30, and 0.36 with
correlation coefficients R of 0.93, 0.90, and 0.86, respectively
(Figures 1-3). The quasi-coincidence of our increments for p-,
m-, and o-fluorine (-1.6, -2.6, -5.1) and p-, m-, and o-chlorine
(-1.9, -3.0, -4.4) substitution with those derived by Streit-
wieser et al.24 from polyhalophenylcesium equilibrium basicities
(F -1.4, -3.0, -5.2; Cl -2.1, -2.7, -4.2) is noteworthy.
The linear correlation established between gas phase anion
and solution organolithium basicities are unprecedented and
unpredicted. The relatively moderate deviations from linearity,
in particular regarding fluoro and methoxy substituents in the
meta and ortho series, may, but must not, be meaningful.
Possibly they merely reflect numerical errors made in the kinetic
determination of the ∆∆G increments (estimated as (0.2 kcal/
mol) or the computation of gas-phase protonation enthalpies
∆Hg (probably exceeding 1.0 kcal/mol).
SCHEME 9. Electronic Reorganization upon Deprotonation
of Benzene
An electron-withdrawing substituent X placed at the ortho
position will accentuate the through-bond σ-polarization (of the
C-1/C-2 and C-2/C-X bonds) and also attract the electron excess
The linear relationship found between aryl anions and ortho-
substituted aryllithiums rules out any intramolecular coordina-
(25) Chadwick, S. T.; Rennels, R. A.; Rutherford, J. L.; Collum, D. B. J. Am.
Chem. Soc. 2000, 122, 8640–8647.
(26) Jones, F. N.; Vaulx, R. L.; Hauser, C. R. J. Org. Chem. 1963, 28, 3461–
3465.
(24) Stratakis, M.; Wang, P. G.; Streitwieser, A. J. Org. Chem. 1996, 61,
3145–3150.
226 J. Org. Chem. Vol. 74, No. 1, 2009

ortho-, meta-, and para-Substituted Aryllithiums
at C-1 through space by its effective nuclear charge. Thus, the
substituent exerts simultaneously σ-inductive and field effects,
both of them stabilizing the carbanion (2). Analogously the same
happens between the negatively charged center and an electron-
donating substituent, although this time both effects are
destabilizing. Such electrostatic effects level off with distance
and should vanish for substituents accommodated at the para
position. However, the location of the π-cloud, little affected
by an ortho substituent, must respond very sensitively to para
substituents irrespective of whether they act as electron acceptors
(3) or donors (4). Finally, meta substituents should exhibit a
blend of σ-inductive, field, and π-polarizing effect if all of them
are in weakened form (Scheme 10).
FIGURE 4. Equilibration between p-anisyllithium and phenyllithium
by halogen/metal permutation with the corresponding iodoarenes.
SCHEME 10. σ-Polarization by ortho Substituents X and
π-Polarization by para Substituents X
Combining benzenide (the phenyl anion) with a lithium cation
will attenuate the electron excess of the original carbanion to a
considerable extent. This must weaken the σ-coupled π-drift
significantly. As a consequence, the stabilizing or destabilizing
effect of para substituents shrinks to 25% of the value
established for the metal-fee (naked) anions. Qualitatively the
same is encountered with meta and ortho substituents. Their
stabilizing effects on the aryllithiums decrease to, respectively,
30% and 36% of the carbanion numbers.
FIGURE 5. Equilibration between o-anisyllithium and phenyllithium
by halogen/metal permutation with the corresponding iodoarenes.
In this context one may feel tempted to ask how closely
phenyllithium resembles the phenyl anion as far as their basicity
is concerned. A crude guess can be based on a thermochemical
cycle that leads to an estimate of the protodelithiation enthalpy
of phenyllithium in the gas phase (Scheme 11). Making the
plausible29-33 assumption that the C-Li bond strength (ho-
SCHEME 11. “Carbanion-Like” Nature of Phenyllithium
FIGURE 6. Equilibration between o-anisyllithium and o-chlorophe-
nyllithium by halogen/metal permutation with the corresponding
bromoarenes.
or even 260 kcal/mol, if the lithium cation set free combines
with the benzene formed to afford an π-arene complex.34-36
These numbers correspond to 60% or 65% of the protonation
enthalpy of the naked phenyl anion.
The basicity of phenyllithium will be further lowered in the
condensed phase where aggregation and solvation provide extra
stability. Our rule-of-thumb according to which organolithiums
retain roughly 40% of carbanion character thus seems to match
reality fairly well.37
molytic dissociation energy) approximates 60 kcal/mol, the
protolysis of phenyllithium should be exothermic by some 240
(33) van Eikema Hommes, N. J. R.; Schleyer, P. V. R. Angew. Chem. 1992,
104, 768–771; Angew. Chem., Int. Ed. Engl. 1992, 31, 755-758.
(34) Woodin, R. L.; Beauchamp, J. L. J. Am. Chem. Soc. 1978, 100, 501–
508.
(35) Kaufmann, L.; Vitze, H.; Bolte, M.; Lerner, H.-W.; Wagner, M.
Organometallics 2007, 26, 1771–1776.
(27) Klein, K. P.; Hauser, C. R. J. Org. Chem. 1967, 32, 1479–1483.
(28) Schlosser, M.; Ginanneschi, A.; Leroux, F. Eur. J. Org. Chem. 2006,
2956–2969.
(29) Fowell, P. A.; Mortimer, C. T. J. Chem. Soc. 1961, 3793–3796.
(30) Holm, T. J. Organomet. Chem. 1974, 77, 27–30.
(36) Rodgers, M. T.; Armentrout, P. B. Int. J. Mass Spectrom. 2007, 267,
167–182.
(37) Schlosser M. In Organometallics in Synthesis: A Manual, 2nd ed.;
Schlosser, M., Ed.; Wiley: Chichester, 2002; pp 1-352; spec. 47-49.
(31) Klumpp, G. W. Recl. TraV. Chim. Pays-Bas 1986, 105, 1–21.
(32) Arnett, E. M.; Moe, K. D. J. Am. Chem. Soc. 1991, 113, 7068–7069.
J. Org. Chem. Vol. 74, No. 1, 2009 227

Gorecka-Kobylinska and Schlosser
FIGURE 7. Equilibration between m-anisyllithium and p-fluorophe-
nyllithium by halogen/metal permutation with the corresponding
iodoarenes.
FIGURE 8. Equilibration between o-anisyllithium and m-(trifluorom-
ethyl)phenyllithium by halogen/metal permutation with the correspond-
ing bromoarenes.
Experimental Section
Generalities. Solutions of tert-butyllithium in pentanes (1.7
M) were supplied by Chemetall (D-60487 Frankfurt, Germany).
Air- and moisture-sensitive materials were stored in Schlenk
tubes or Schlenk burettes. They were protected by and handled
under an atmosphere of 99.995% pure nitrogen, using appropriate
glassware (Glasgera¨tebau Pfeifer, D-98711 Frauenwald, Ger-
many). Diethyl ether and tetrahydrofuran were dried by distil-
lation from sodium wire after the characteristic blue color of in
situ generated sodium biphenyl ketyl (benzophenone-sodium
“radical-anion”) had been found to persist. The temperature of
dry ice/methanol baths is consistently indicated as -75 °C.
Further details concerning experimental routine can be found
in previous reports from this laboratory.2,38-40
FIGURE 9. Equilibration between m-(trifluoromethyl)phenyllithium
and p-(trifluoromethyl)phenyllithium by halogen/metal permutation with
the corresponding bromoarenes.
Starting Materials and Products. All isomers of methoxy-,
chloro-, fluoro-, trifluoromethyl-, or trifluoromethoxy-substituted
bromobenzenes and iodobenzenes are commercial (Acros Organ-
ics, B-2440 Geel, Aldrich-Fluka, CH-9479 Buchs, and Apollo
Scientific, Stockport UK-SK6 2QR). The same holds for
4-bromo-2,2-difluoro-1,3-benzodioxole, whereas the 4-iodo ana-
log had to be prepared according to a literature procedure.20,21
Equilibration Experiments. At -50, -75, or -100 °C, tert-
butyllithium in pentanes (5.9 mL, 1.7 M, 10 mmol) was added
to a solution of a brominated substrate (5.0 mmol) in tetrahy-
drofuran (10 mL). After 45 min, a solution of the second
brominated substrate (5.0 mmol) and biphenyl (approximately
0.3 g, 2 mmol), as an “internal standard” for gas chromatographic
quantification, in tetrahydrofuran (4.0 mL) was added to the
reaction mixture. Aliquots (of approximately 2.0 mL) were
withdrawn regularly after 3, 15, 45, 180, 360 min and, if needed,
20 and 60 h and added to a solution of iodine (0.51 g, 2.0 mmol)
in diethyl ether (4.0 mL). The organic phase was washed with
a saturated aqueous solution (1.0 mL) of sodium thiosulfate
before being subjected to gas chromatographic analysis (2 m,
5%, SE-30, 100 °C (10 min) f 200 °C, heating rate 25 °C/min;
2 m, 5%, C-20 M, 100 °C [10 min] f 200 °C, heating rate 25
°C/min; 30 m, DB-Wax, 75 °C [10 min] f 200 °C, heating rate
25 °C/min; 30 m, DB-23, 75 °C [10 min] f 200 °C, heating
rate 25 °C/min). The amounts of iodo and bromo compounds
were calculated from the products/standard ratios and corrected
using separately determined calibration factors. Some reactions
started from iodinated rather than bromo-substituted substrates.
In such cases the samples withdrawn were trapped with a solution
of 1,2-dibromo-1,1,2,2-tetrafluoroethane (0.24 mL, 0.52 g, 2.0
mmol) in diethyl ether (4.0 mL).
FIGURE 10. Equilibration between p-(trifluoromethoxy)phenyllithium
and p-(trifluoromethyl)phenyllithium by halogen/metal permutation with
the corresponding bromoarenes.
(38) Masson, E.; Schlosser, M. Eur. J. Org. Chem. 2005, 4401–4405.
(39) Leroux, F.; Lefebvre, O.; Schlosser, M. Eur. J. Org. Chem. 2006, 3147–
3151.
(40) Heiss, C.; Marzi, E.; Mongin, F.; Schlosser, M. Eur. J. Org. Chem.
2007, 669–675.
FIGURE 11. Equilibration between o-anisyllithium and 4-(2,2-difluoro-
1,3-benzodioxolyl)lithium by halogen/metal permutation with the
corresponding bromoarenes.
228 J. Org. Chem. Vol. 74, No. 1, 2009

ortho-, meta-, and para-Substituted Aryllithiums
A few representative equilibration diagrams are shown below
(Figures 4). They feature various couples of aryllithiums. Some of
those have almost identical basicities, whereas others have very
divergent ones (e.g., 54:46 vs 96:4, Figures 8 and 5, respec-
tively).
Forschung, Bern (grant 20-100′336-02). The authors are
indebted to Profs. Gernot Frenking (Marburg), Herbert Mayr
(Mu¨nchen), and Andrew Streitwieser (Berkeley) for valuable
comments.
Acknowledgment. This work was supported by the Schweiz-
erische Nationalfonds zur Fo¨rderung der wissenschaftlichen
JO8020083
J. Org. Chem. Vol. 74, No. 1, 2009 229