Kinetics of bromine-magnesium exchange reactions in substituted bromobenzenes

Article abstract of DOI:10.1021/jo802770h

Competition experiments have been performed to determine the relative reactivities of substituted bromobenzenes, bromonaphthalenes, and 9-bromoanthracene toward i-PrMgCl·;LiCl in THF at 0 °C. The rates of the bromine-magnesium exchange reactions are accelerated by electron-acceptor substituents, the activating efficiency of which increases in the order para < meta ortho. The activation free enthalpies of the bromine-magnesium exchange reactions correlate fairly (r² = 0.83) with the proton affinities of analogously substituted aryllithiums (slope 0.8). The kinetics of two representative bromoarenes with i-PrMgCl·LiCl were found to be first-order in both bromoarene and i-PrMgCl·LiCl. Combination of the resulting second-order rate constants with the k_rel values from competition experiments allowed us to calculate reaction times for the bromine-magnesium exchange reactions of a large variety of bromoarenes.

Full text of DOI:10.1021/jo802770h

Kinetics of Bromine-Magnesium Exchange Reactions in Substituted
Bromobenzenes
Lei Shi, Yuanyuan Chu, Paul Knochel, and Herbert Mayr*
Department Chemie und Biochemie der Ludwig-Maximilians-UniVersita¨t Mu¨nchen, Butenandtstrasse 5-13
(Haus F), 81377 Mu¨nchen, Germany
herbert.mayr@cup.lmu.de
ReceiVed December 23, 2008
Competition experiments have been performed to determine the relative reactivities of substituted
bromobenzenes, bromonaphthalenes, and 9-bromoanthracene toward i-PrMgCl· LiCl in THF at 0 °C.
The rates of the bromine-magnesium exchange reactions are accelerated by electron-acceptor substituents,
the activating efficiency of which increases in the order para < meta , ortho. The activation free enthalpies
of the bromine-magnesium exchange reactions correlate fairly (r2 ) 0.83) with the proton affinities of
analogously substituted aryllithiums (slope 0.8). The kinetics of two representative bromoarenes with
i-PrMgCl· LiCl were found to be first-order in both bromoarene and i-PrMgCl· LiCl. Combination of the
resulting second-order rate constants with the krel values from competition experiments allowed us to
calculate reaction times for the bromine-magnesium exchange reactions of a large variety of bromoarenes.
Introduction
uents on the rates of these reactions.4 Though it was evident
that electron-withdrawing substituents facilitated the exchange
process, the relative magnitude of these substituent effects, in
particular the ranking para < meta , ortho did not correlate
with Hammett’s σ values or related substituent constants.5
We have now extended this work to donor-substituted and
multiply substituted benzenes, as well as to naphthalene and
anthracene derivatives. In order to give synthetic chemists a
clue what reaction times to expect for certain exchange reactions,
we have also determined absolute rate constants for some of
these reactions that allowed us to convert the relative reactivities
determined in this and the earlier work into absolute reaction
times.
Bromine-magnesium exchange reactions provide a straight-
forward access to aromatic Grignard reagents including func-
tionalized derivatives.1,2 A particularly efficient way to achieve
this conversion is the treatment of bromoarenes with
i-PrMgCl · LiCl.3 In a recent communication we reported
competition experiments that revealed the influence of substit-
(1) (a) Knochel, P.; Dohle, W.; Gommermann, N.; Kneisel, F. F.; Kopp, F.;
Korn, T.; Sapountzis, I.; Vu, V. A. Angew. Chem. 2003, 115, 4438–4456; Angew.
Chem., Int. Ed. 2003, 42, 4302-4320. (b) Jensen, A. E.; Dohle, W.; Sapountzis,
I.; Lindsay, D. M.; Vu, V. A.; Knochel, P. Synthesis 2002, 565–569.
(2) (a) Silverman, G. S.; Rakita, P. Handbook of Grignard Reagents; Marcel
Dekker: New York, 1996. (b) Richey, H. G., Jr. Grignard Reagents: New
DeVelopments; Wiley: New York, 1999. (c) Yamamoto, H.; Oshima, K. Main
Group Metals in Organic Synthesis; Wiley: New York, 2004. (d) Knochel, P.
Handbook of Functionalized Organometallics; Wiley: Weinheim, 2005.
(3) (a) Krasovskiy, A.; Knochel, P. Angew. Chem. 2004, 116, 3396–3399;
Angew. Chem., Int. Ed. 2004, 43, 3333-3336. (b) Krasovskiy, A.; Straub, B. F.;
Knochel, P. Angew. Chem. 2006, 118, 165–169; Angew. Chem., Int. Ed. 2006,
45, 159-162. (c) Hauk, D.; Lang, S.; Murso, A. Org. Process Res. DeV. 2006,
10, 733–738. (d) Li, F.; Castle, S. L. Org. Lett. 2007, 9, 4033–4036. (e) Hirner,
S.; Panknin, O.; Edefuhr, M.; Somfai, P. Angew. Chem. 2008, 120, 1933–1935;
Angew. Chem., Int. Ed. 2008, 47, 1907-1909. (f) Huang, L.; Wu, S.; Qu, Y.;
Geng, Y.; Wang, F. Macromolecules 2008, 41, 8944–8947. (g) Rauhut, C. B.;
Melzig, L.; Knochel, P. Org. Lett. 2008, 10, 3891–3894. (h) Duan, X.-F.; Ma,
Z.-Q.; Zhang, F.; Zhang, Z.-B. J. Org. Chem. 2009, 74, 939–942. (i) For iodine-
magnesium exchange reactions using i-PrMgCl ·LiCl, see: Martin, R.; Buchwald,
S. L. J. Am. Chem. Soc. 2007, 129, 3844–3845.
Results and Discussion
Relative Reactivities by Competition Experiments. Com-
petition experiments were carried out by adding i-PrMgCl·LiCl
in THF to an excess of two differently substituted bromoben-
zenes (approximately 2 equiv). The ratio of the resulting
arylmagnesium chlorides was then derived from the gas-
chromatographically determined product ratio obtained after
quenching with methanol (Scheme 1).
(4) Shi, L.; Chu, Y.; Knochel, P.; Mayr, H. Angew. Chem. 2008, 120, 208–
210; Angew. Chem., Int. Ed. 2008, 47, 202-204.
(5) Hansch, C.; Leo, A.; Taft, R. W. Chem. ReV. 1991, 91, 165–195.
2760 J. Org. Chem. 2009, 74, 2760–2764
10.1021/jo802770h CCC: $40.75 2009 American Chemical Society
Published on Web 03/12/2009

Br-Mg Exchange Reactions in Substituted Bromobenzenes
SCHEME 1. Competition Experiment for Determining the
Relative Reactivities of Bromoarenes toward i-PrMgCl·LiCl
SCHEME 2. Assessment of Relative Aryllithium Basicities
by Schlosser9
SCHEME 3. Effect of 3-F Substitution on Relative Rates of
Br-Mg Exchange
The relative reactivity of R1 and R2 can be expressed by
the competition constant κ, which is defined by eq 1:6
ka
kb
log([R1]0/[R1]t)
log([R2]0/[R2]t)
κ )
)
(1)
SCHEME 4. Effect of o-Methyl Groups on Relative Rates
of Br-Mg Exchange
Substitution of [R1]0 and [R2]0 by the expressions in eqs 2
and 3 (mass balance) yields eq 4:
[R1]0 ) [R1]t + [P1]t
[R2]0 ) [R2]t + [P2]t
log(1 + [P1]t/[R1]t)
(2)
(3)
substituted compounds (r2 ) 0.97) and less for the para-
substituted compounds (r2 ) 0.92). From the reaction constants,
F ) 5.38 for the meta-substituted compounds, and F ) 2.65
for the para-substituted compounds, one can derive that the
bromine-magnesium exchange is predominantly accelerated by
inductively withdrawing substituents. Because the σm values
exclusively reflect inductive effects, the resulting F value is
larger than that derived from σp, which includes inductive as
well as mesomeric effects.
κ )
(4)
log(1 + [P2]t/[R2]t)
Competition constants calculated according to eq 4 from the
chromatographically determined ratios [P1]t/[R1]t and [P2]t/
[R2]t are used to evaluate the relative reactivities.
Because κ, as defined by eq 4, was found to be independent
of the reaction times, one can conclude that subsequent reactions
of the arylmagnesium halides with bromoarenes do not take
place, i.e., that the competition constants are the result of kinetic
control.
Schlosser and Gorecka-Kobylinska have just reported rela-
tive aryllithium basicities that were derived from equilibrium
constants for the reactions described in Scheme 2.9
Each of the 38 bromoarenes listed in Figure 1 was subjected
to competition experiments with several other bromoarenes as
illustrated in Scheme 1. The resulting 58 competition constants
κ, 22 of which (marked by asterisks) have been reported
previously,4 are summarized in Figure 1. Solving the resulting
overdetermined set of linear equations (eq 5) by least-squares
minimization7 yielded the krel values given in Figure 1.8
Figure 3 shows a fair correlation between log krel for the
bromine-magnesium exchange reactions and the relative aryl-
lithium basicities. When the slope of this correlation is multiplied
by -2.303RT ) -1.245, a ratio ∆∆Gq/∆∆G0 ) 0.80 is
obtained, indicating a transition state with a charge distribution
in the aryl ring that closely resembles that of the corresponding
aryllithium compounds.
From the regioselectivities of the exchange reactions in
dibromo-substituted benzenes, we had previously derived ap-
proximate additivities of the substituent effects.4 Rough addi-
tivity is also observed for 3-fluorine substituents as shown by
the comparison in Scheme 3, where the second fluorine atom
is demonstrated to cause a somewhat stronger acceleration than
the first fluorine (492 ) 2401).
Because of the small effects of methyl and methoxy groups in
meta- and para-positions, it is not surprising that also two or three
methyl or methoxy groups in meta- or para-positions relative to
bromine affect the reactivity by less than a factor of 3. Steric
shielding is probably responsible for the fact that two methyl groups
in positions ortho to bromine retard by a factor of 15, whereas
one o-methyl reduces the exchange rate of Br by a factor of only
2.3 (Scheme 4).
log κ ) log ka - log kb
(5)
From the fact that electron-withdrawing groups generally
accelerate the bromine-magnesium exchange but that 3-CN and
3-CF3 accelerate more than 4-CN and 4-CF3, respectively, we
had previously concluded that there is no correlation between
the effects of substituents on the rates of exchange and
Hammett’s substituent constants. This statement can now be
refined because data for donor substituents have now become
available. When log krel for all investigated meta- and para-
substituted bromobenzenes is plotted against σ, a rather poor
correlation is observed. Individual treatment of the meta- and
para-substituted bromobenzenes (Figure 2) leads to a significant
improvement of the correlations, particularly for the meta-
(6) Huisgen, R. Angew. Chem. 1970, 82, 783–794; Angew. Chem., Int. Ed.
As in electrophilic aromatic substitutions, naphthalene is more
reactive than benzene, and the 1-position of naphthalene is more
reactive than the 2-position. Annelation of a second benzene
Engl. 1970, 9, 751-762.
2
(7) κcalcd have been calculated by minimizing Σ∆2, where ∆2 ) (κ-κcalcd
using the program “What’s Best! 4.0 Commercial” by Lindo Systems Inc.
)
(8) Actually, some additional competition constants for brominated het-
eroarenes have also been included for the calculation of krel. In this way we
intend to avoid reoptimization when publishing exchange rates for further classes
of compounds.
(9) Gorecka-Kobylinska, J.; Schlosser, M. J. Org. Chem. 2009, 74, 222–
229.
J. Org. Chem. Vol. 74, No. 7, 2009 2761

Shi et al.
FIGURE 1. Relative reactivities of substituted bromobenzenes toward i-PrMgCl·LiCl in THF at 0 °C and τ1/2 for 1 M solutions (κ marked with
an asterisk were taken from ref 4). (a) krel not statistically corrected.
SCHEME 5. Relative Reactivities of Bromobenzene,
Bromonaphthalenes, and 9-Bromoanthracene toward
i-PrMgCl·LiCl
and [chlorobenzene]/[1-bromo-3-chlorobenzene], respec-
tively, was analyzed by GC, from which the concentration
of the reactant was calculated by eq 6:
[R]0
[R] )
(6)
1 + fRP(AP/AR)
[R] is the concentration of the reactant (bromoarene), [R]0
is the concentration of the bromoarene at t ) 0, i.e., [R]0 )
[R] + [P], and fRP ) (AR[P])/(AP[R]) is the response factor,
which was determined from the GC peak areas AR and AP of
artificial mixtures of reactant R and product P, respectively.
For the consumption of R, the general rate law (eq 7) is
considered:
ring to bromobenzene (f 9-bromoanthracene) has a slightly
larger effect than annelation of the first benzene ring (Scheme
5).
Determination of Absolute Rate Constants. To convert the
relative rate constants listed in Figure 1 into absolute rate
constants, we have determined the kinetics of the Br-Mg
exchange reactions of 4-bromobenzonitrile and 1-bromo-3-
chlorobenzene with i-PrMgCl · LiCl at 0 °C (Scheme 6).
For that purpose, THF solutions that were 0.01 M in
bromoarene and 0.1-0.5 M in i-PrMgCl · LiCl were prepared.
After appropriate times, aliquots of these solutions were taken
with gastight syringes, injected into methanol, and the ratio
[product]/[reactant], i.e., [benzonitrile]/[4-bromobenzonitrile]
-d[R]
dt
) k[R]n[i-PrMgCl·LiCl]m
(7)
Because the ratio [i-PrMgCl·LiCl]/[bromoarene] varied be-
tween 10 and 50, the concentration [i-PrMgCl·LiCl] is almost
constant, and k[i-PrMgCl·LiCl]m was replaced by kobs, i.e.,
2762 J. Org. Chem. Vol. 74, No. 7, 2009

Br-Mg Exchange Reactions in Substituted Bromobenzenes
-d[R]
dt
) kobs[R]n with kobs ) k[i-PrMgCl·LiCl]m (8)
From the linearity of the plots of -ln[R] versus t (Figure 4),
one can derive that the reactions are first order in bromoarenes R,
i.e., n ) 1.
FIGURE 4. Plot of -ln[R] versus time for the reaction of 1-bromo-
3-chlorobenzene ([R] ) 0.01 M) with variable concentrations of
i-PrMgCl · LiCl [(O) 0.1 M, (∆) 0.3 M, (b) 0.5 M; all three linear
correlations with r2 > 0.999] in THF at 0 °C.
FIGURE 2. Plot of the logarithm of relative rates, log krel, against
Hammett’s σ values.
FIGURE 5. Dependence of the pseudo-first-order rate constants kobs
for the reaction of 1-bromo-3-chlorobenzene with i-PrMgCl·LiCl on
the concentration of i-PrMgCl·LiCl (in THF at 0 °C).
Grignard reagents with carbonyl compounds10 and chlorosi-
lanes11 in THF. The slope of the correlation shown in Figure 5
gave the second-order rate constant for the reaction of
i-PrMgCl · LiCl with 1-bromo-3-chlorobenzene (8.37 × 10-4
M-1 s-1), and an analogous evaluation of the reaction of
4-bromobenzonitrile with i-PrMgCl·LiCl in THF gave a second-
order rate constant of 2.97 × 10-3 M-1 s-1. The ratio of the
directly determined rate constants (3.55) is 14% smaller than
the ratio derived from the competition experiments (4.14). Using
the average value, one can now convert the krel values of Figure
1 into second-order rate constants by multiplying krel with 6.0
× 10-6 M-1 s-1. Under typical reaction conditions,3a the THF
solutions are 1 M in both bromoarene and i-PrMgCl·LiCl. With
the assumption that the second-order rate law derived above
also holds at these concentrations, one can calculate the reaction
times needed for conversion. A conversion of 50% is then
reached within t ) 1/(k[R]0), i.e., within 16 s for 2-bromoben-
zonitrile, the most reactive compound of this series, and within
29 days for 2-bromo-1,3-dimethylbenzene, the least reactive
compound of this series. In order to reach 90% conversion, a
FIGURE 3. Correlation between log krel for the bromine-magnesium
exchange reactions (0 °C, THF, data from Figure 1) and ∆∆G0 for the
relative aryllithium basicities (Scheme 2). Data used for this correlation
are given in Table S14 of Supporting Information.
SCHEME 6. Determination of the Kinetics of
Bromine-Magnesium Exchange
(10) (a) Yamataka, H.; Matsuyama, T.; Hanafusa, T. J. Am. Chem. Soc. 1989,
111, 4912–4918. (b) Holm, T. J. Org. Chem. 2000, 65, 1188–1192. (c) Shimizu,
M.; Yamataka, H. Bull. Chem. Soc. Jpn. 2004, 77, 1757–1761.
(11) (a) Rudolph, S. E.; Charbonneau, L. F.; Smith, S. G. J. Am. Chem. Soc.
1973, 95, 7083–7093. (b) Tuulmets, A.; Nguyen, B. T.; Panov, D.; Sassian, M.;
Jaerv, J. J. Org. Chem. 2003, 68, 9933–9937. (c) Tuulmets, A.; Nguyen, B. T.;
Panov, D. J. Org. Chem. 2004, 69, 5071–5076.
Plots of the resulting first-order rate constants kobs versus
[i-PrMgCl·LiCl] were almost linear, indicating first-order
dependence on the concentration of the organomagnesium
reagent, i.e., m ) 1 in eqs 7 and 8. Obviously, the nature of the
organometallic does not alter in this concentration range.
Analogous observations have been made for reactions of
(12) This value comes from kt ) 1/[R]t-1/[R]0 for bimolecular reactions
with equal concentrations [R]0 for the two reactions partners.
J. Org. Chem. Vol. 74, No. 7, 2009 2763

Shi et al.
From the slope of 0.8 for the plot of log krel versus the relative
basicities (log K) of the corresponding aryllithiums, one can
derive a transition state where the charge distribution in the aryl
ring is comparable to that in the analogous aryllithium com-
pounds. Schlosser’s explanation of the large ortho effect by
σ-polarization9 may be applied analogously.
Experimental Section
Procedure for Determination of Relative Exchange Rates
K. A dry and nitrogen-flushed 25-mL flask, equipped with a
magnetic stirrer, was charged with 2-bromobenzonitrile R1 (1.3
mmol, 237 mg), 1-bromo-2-(trifluoromethoxy)benzene R2 (1.3
mmol, 313 mg), and internal standard (n-C14H30) in THF (4.20 mL).
The reaction mixture was cooled to 0 °C. i-PrMgCl ·LiCl (1.3 mmol,
1.00 mL solution in THF) was added in one portion. After certain
times (e.g., 60 min), ca. 0.2 mL of the reaction mixture was taken
with a gastight syringe and injected in methanol. The reaction
mixture was washed with saturated aqueous NaCl solution (30 mL).
The aqueous phase was extracted with diethyl ether (ca. 25-30
mL). The etheral solutions were dried over Na2SO4 and analyzed
by GC.
Procedure for Determination of Pseudo-First-Order Rate
Constants. A solution of 1-bromo-3-chlorobenzene (37.4 mg, 0.195
mmol) and n-C14H30 (internal standard, 9.8 mg, 0.0493 mmol) in
dry THF (12.31 mL) was combined with a 1.30 M solution of
i-PrMgCl·LiCl in THF (7.69 mL) at 0 °C to give a solution with
the concentration listed in Table S2 of Supporting Information. The
mixture was stirred under a nitrogen atmosphere at 0 °C, and after
variable times (typically every 3 min), 1.5 mL of this mixture was
taken with a gastight syringe and injected into 3 mL of methanol.
The mixture was extracted with diethyl ether and then analyzed by
GC.
FIGURE 6. Substituent effects on the reactivities of bromobenzene
derivatives toward i-PrMgCl·LiCl (0 °C, THF).
reaction time of 9/(k[R]0) is needed,12 i.e., the times listed in
Figure 1 have to be multiplied by a factor of 9.
For the reaction of i-PrMgCl·LiCl with 1-bromo-3-chlo-
robenzene in THF/1,4-dioxane (9/1), a second-order rate
constant of 3.08 × 10-3 M-1 s-1 was determined at 0 °C. The
addition of 10 vol % 1,4-dioxane thus accelerated the
bromine-magnesium exchange by a factor of 3.7, which allows
one to significantly reduce the reaction times listed in Figure 1.
Acknowledgment. We thank the Alexander von Humboldt
Foundation (research fellowship to L.S.), the Deutsche For-
schungsgemeinschaft (SFB 749), and the Fonds der Chemischen
Industrie for support of this work.
Conclusion
Bromine-magnesium exchange reactions in substituted bro-
mobenzenes with i-PrMgCl ·LiCl are strongly accelerated by
electron-withdrawing substituents. From the increasing effect in
the series para < meta , ortho (Figure 6), one can derive that it
is mostly the inductive substituent effect that is responsible for the
activation of the bromine-magnesium exchange, in line with a
transition state resembling the hypervalent halogen complex A.3b
Supporting Information Available: Experimental procedure
and data for the determination of the relative exchange rates κ
and determination of absolute rate constants. This material is
available free of charge via the Internet at http://pubs.acs.org.
JO802770H
2764 J. Org. Chem. Vol. 74, No. 7, 2009