Reductive elimination of aryl halides from palladium(II) [2]

Full text of DOI:10.1021/ja0034592

1232
J. Am. Chem. Soc. 2001, 123, 1232-1233
a
Table 1. The Yield of ArX and the Calculated Keq Value for the
Reactions in Eq 1
Reductive Elimination of Aryl Halides from
Palladium(II)
X
yield of ArX (%)
Keq
Amy H. Roy and John F. Hartwig*
1a (X)Cl)
1b (X)Br)
1c (X)I)
1d (X)Cl)
1e (X)Br)
70
70
39
30
75
9(3) × 10-2
Department of Chemistry, Yale UniVersity
P.O. Box 208107, New HaVen, Connecticut 06520-8107
ReceiVed September 22, 2000
2.3(3) × 10-3
3.7(2) × 10-5
not measuredb
3.3(6) × 10-4
The oxidative addition of aryl halides to palladium(0) com-
plexes1,2 is the initial step of catalytic processes such as Heck,3-5
Stille,6,7 and Suzuki coupling,8 as well as the recently developed
arylation of amines,9-12 ethers,13,14 and carbonyl compounds.15,16
Essentially any phosphine-ligated palladium(0) complex studied
previously undergoes oxidative addition of aryl bromides and
iodides. The opposite reaction, reductive elimination of aryl
halides, is rare because addition is favored thermodynamically.
A single example of aryl halide reductive elimination was
observed from a higher valent Pt(IV) species many years ago,17
but reductive elimination from low-valent centers has not been
observed directly.18
a Keq values are referenced to a 1 M standard state. b This reaction
appeared to consume all of the aryl chloride complex, but low yields
for the formation of free aryl chloride may prevent reversibility.
In general, complexes with increasingly electron-donating
ligands undergo faster oxidative addition because of the greater
driving force for oxidation of a more electron-rich metal.19 Indeed,
trialkylphosphine complexes add even chloroarenes,20 and some
of these and related ligands provide fast rates for coupling
processes.21-23 P(t-Bu)3 is the quintessential strongly electron
donating ligand. Its νCO value for {Ni[P(t-Bu)3](CO)3} is the
lowest of any phosphine ligand in Tolman’s classic study.24 Yet,
we report the surprising result that reductive elimination of aryl
halide is induced by addition of tri-tert-butylphosphine to arylpal-
ladium(II) halide complexes (eq 1) and is favored thermodynami-
cally over oxidative addition. We provide equilibrium constants
for the addition and elimination processes in eq 1 and show that
an unusual mechanism for ligand exchange occurs to initiate the
reductive elimination.
of P(t-Bu)3. All of the P(o-tolyl)3-ligated arylpalladium halides
formed {Pd[P(t-Bu)3]2} (2)26 upon addition of an excess of P(t-
Bu)3. Ortho-substituted aryl halide complexes 1a-c provided
higher yields of free aryl halide product than did complexes 1d-e
containing unhindered palladium-bound aryl groups. Biaryl and
arene were the predominant side products. The tert-butyl group
in 1a-c was employed to provide solubility. The amount of added
P(t-Bu)3 was crucial to obtain high yields of free aryl halide. The
highest yields were obtained when approximately 15 equiv of
P(t-Bu)3 per dimer were used. Although not limited by the values
of Keq when 15 equiv of P(t-Bu)3 were used, the yields for the
reductive elimination of compounds 1a-c paralleled the ther-
modynamic driving force. Yet, the rates did not. Reductive
elimination from chloride 1a was slower than that from bromide
1b, even though reductive elimination from 1a was more favored
thermodynamically. The low yield of chloroarene from reaction
of 1d is consistent with the slow rates for reaction of the more
hindered chloride 1a and with generally slow rates for the
microscopic reverse, oxidative addition of aryl chloride.
Keq values (Table 1) were obtained for the process in eq 1 by
initiating reactions from both sides of the equilibrium. All reaction
components of the equilibrium were observed by NMR spectros-
copy when less than 4 equiv of P(t-Bu)3 were added to a 10 mM
solution of chloride 1a, less than 10 equiv to a 10 mM solution
of bromide 1b, and less than 15 equiv to the same solution of 1c.
Quantitative data are provided in Table 1. An o-methyl group on
the aryl halide increased the value of Keq for reductive elimination
by a factor of roughly ten. Chloride 1a displayed the largest
driving force for reductive elimination and iodide 1c had the
smallest. The change in Keq that accompanied each successive
change of halide was roughly a factor of 100. Despite this trend,
oxidative addition of aryl halides 3c and 3e to {Pd[P(t-Bu)3]2}
was not observed in the absence of added P(o-tolyl)3.
Equation 1 and Table 1 summarize our data on the reductive
elimination of aryl halide from complexes 1a-e25 upon addition
(1) Stille, J. K.; Lau, K. S. Y. Acc. Chem. Res. 1977, 10, 434-442.
(2) Amatore, C.; Jutand, A. J. Organomet. Chem. 1999, 576, 254.
(3) Heck, R. F. Org. React. 1982, 27, 345-390.
(4) Crisp, G. T. Chem. Soc. ReV. 1998, 27, 427-436.
(5) Beletskaya, I. P.; Cheprakov, A. V. Chem. ReV. 2000, 100, 3009-
3066.
(6) Stille, J. K. Angew. Chem., Int. Ed. Engl. 1986, 25, 508-524.
(7) Farina, V.; Krishnamurthy, V.; Scott, W. J. Org. React. 1997, 50,
1-652.
(8) Miyaura, N.; Suzuki, A. Chem. ReV. 1995, 95, 2457-2483.
(9) Hartwig, J. F. In Modern Amination Methods; Ricci, A., Ed.; Wiley-
VCH: Weinheim, 2000.
(10) Hartwig, J. F. Angew. Chem., Int. Ed. 1998, 37, 2046-2067.
(11) Yang, B. H.; Buchwald, S. L. J. Organomet. Chem. 1999, 576, 125-
146.
(12) Wolfe, J. P.; Wagaw, S.; Marcoux, J.-F.; Buchwald, S. L. Acc. Chem.
Res. 1998, 31, 805-818.
(13) Mann, G.; Hartwig, J. F. J. Am. Chem. Soc. 1999, 121, 3224-3225.
(14) Aranyos, A.; Old, D. W.; Kiyomori, A.; Wolfe, J. P.; Sadighi, J. P.;
Buchwald, S. L. J. Am. Chem. Soc. 1999, 121, 4369-4378.
(15) Kawatsura, M.; Hartwig, J. F. J. Am. Chem. Soc. 1999, 121, 1473-
1478.
(16) Fox, J. M.; Huang, X.; Chieffi, A.; Buchwald, S. L. J. Am. Chem.
Soc. 2000, 122, 1360-1370.
(17) Ettorre, R. Inorg. Nucl. Chem. Lett. 1969, 5, 45-49.
(18) Echavarren, A. M.; Stille, J. K. J. Am. Chem. Soc. 1987, 109, 5478-
5486.
(19) Collman, J. P.; Hegedus, L. S.; Norton, J. R.; Finke, R. G. In Principles
and Applications of Organotransition Metal Chemistry, 2nd ed.; University
Science Books: Mill Valley, 1987; pp 279-354.
(20) Portnoy, M.; Milstein, D. Organometallics 1993, 12, 1665-1673.
(21) Littke, A. F.; Dai, C.; Fu, G. C. J. Am. Chem. Soc. 2000, 122, 4020-
4028.
The reductive elimination most likely occurs from a monomeric
complex, but the coordination sphere of this monomer and the
mechanism for its formation were unclear. Moreover, we could
not predict whether the generation or reaction of the monomer
was rate determining. To address these questions, we measured
the rate constants for reaction of 1b by 1H NMR spectroscopy at
55 °C. The concentration of P(t-Bu)3 was varied from 0.10 to
0.84 M, [P(o-tol)3] was varied from 0.030 to 0.35 M, and [1b]
was varied from 5.2 to 21 mM. Because the P(o-tol)3 product
(22) Hartwig, J. F.; Kawatsura, M.; Hauck, S. I.; Shaughnessy, K. H.;
Alcazar-Roman, L. M. J. Org. Chem. 1999, 64, 5575-5580.
(23) Wolfe, J. P.; Buchwald, S. L. Angew. Chem., Int. Ed. 1999, 38, 2413-
2416.
(25) Paul, F.; Patt, J.; Hartwig, J. F. Organometallics 1995, 14, 3030-
3039.
(26) Yoshida, T.; Otsuka, S. J. J. Am. Chem. Soc. 1977, 99, 2134-2140.
(24) Tolman, C. A. Chem. ReV. 1977, 77, 313-348.
10.1021/ja0034592 CCC: $20.00 © 2001 American Chemical Society
Published on Web 01/19/2001

Communications to the Editor
J. Am. Chem. Soc., Vol. 123, No. 6, 2001 1233
analyzed the products after addition of PPh3 to simplify the
product distribution. Although several potential pathways for
exchange are possible, this experiment was consistent with our
intuition that cleavage of the halogen bridges is more rapid than
the overall reaction at 55 °C. To probe for reversible reductive
elimination in the absence of added P(t-Bu)3, we added 4-BrC6H4-
t-Bu to 1b and {Pd[P(o-tolyl)3](p-tol)(Br)}2. Only small amounts
of exchange were observed after days at 55 °C.
Path B involves reversible or irreversible cleavage of the dimer,
followed by ligand exchange and reductive elimination. This
pathway again predicts a reaction that is either half order in dimer
first order in P(t-Bu)3 or, if first order in dimer, then zero order
in both P(t-Bu)3 and P(o-tol)3. Again, these predictions are
inconsistent with the kinetic data.
Figure 1. Plots of 1/kobs vs 1/[P(t-Bu)3] and 1/kobs vs [P(o-tol)3].
Scheme 1
Path C involves initial coordination of P(t-Bu)3 to palladium
and either reversible or irreversible formation of a three-coordinate
P(o-tol)3 complex and a four-coordinate mixed-phosphine com-
plex. The four-coordinate, P(t-Bu)3-ligated complex would then
reductively eliminate aryl halide either before or after dissociating
P(o-tolyl)3. This associative mechanism with reversible formation
of mononuclear species predicts a simple half-order dependence
of rate on [P(t-Bu)3], and this prediction is inconsistent with data
in Figure 1 displaying a first-order dependence and y-intercept.
Irreversible association of P(t-Bu)3 or dimer cleavage would
provide zero-order behavior in P(o-tolyl)3. The nonzero y-intercept
and the inverse rate dependence on [P(o-tolyl)3] in Figure 1 are
clearly inconsistent with this prediction. Path D shows an initial,
reversible associative phosphine substitution and irreversible
cleavage of the resulting dinuclear species. This mechanism would
show a simple first-order dependence on [P(t-Bu)3] and if the
liquid substitution were reversible a simple inverse dependence
on [P(o-tolyl)3]. These predictions are inconsistent with the
nonzero y-intercepts in Figure 1.
Instead, all of our data are consistent with Path E, involving
an irreversible dissociative ligand substitution, which is presum-
ably followed by cleavage of the dinuclear species prior to
reductive elimination. Of course, dissociative substitution is
unusual for square-planar geometries. The rate equation for
dissociative substitution predicts positive slopes and nonzero
y-intercepts for both plots in Figure 1. The first-order rate behavior
in [1b] indicates an irreversible formation of monomers from the
dinuclear species formed from ligand substitution. Moreover, the
nonzero y-intercept requires that the k2 step be irreversible. These
data demonstrate that each ligand substitution event leads to
reductive elimination. Thus, formation of monomers and reductive
elimination is faster than dissociation of P(t-Bu)3.
may be formed reversibly and can enter the rate equations, it was
added to the reactions to maintain a constant concentration.
Excellent fits to a first-order appearance of each product were
obtained. Figure 1 shows the dependence of kobs on the concentra-
tion of the two phosphines. A positive, linear dependence of 1/kobs
vs 1/[P(t-Bu)3] was observed, along with a nonzero y-intercept.
The inverse dependence of kobs on [P(o-tolyl)3] in Figure 1 also
showed a nonzero y-intercept.
The reductive elimination reported here uncovers several
fundamental principles. First, reductive elimination is often
induced by coordination of electron-accepting ligands, but is
induced here by coordination of a strongly electron-donating
ligand. Thus, severe steric crowding not only can accelerate the
rate for reductive elimination, but also can alter the thermody-
namics so severely that reductive elimination of aryl halide is
favored thermodynamically. Second, oxidative addition occurs
rapidly to P(t-Bu)3-ligated Pd(0) complexes in catalytic processes
despite the weak driving force for this reaction. Third, ligand
substitution in this square-planar system is dissociative, presum-
ably because of the steric properties of the ligands involved.
Finally, these studies suggest that a palladium-catalyzed aromatic
halide exchange process is feasible. Further studies that target
such a catalytic, aromatic Finkelstein reaction are in progress.
Qualitative mechanistic studies showed that the reaction was
not affected by light, and that anthracene was not formed as
product when the reactions were run in the presence of 9,10-
dihydroanthracene. Moreover, the rate constants were highly
reproducible. Thus, further studies to detect a radical pathway
were not conducted as part of our mechanistic evaluation.
Instead, the five classes of mechanism for the reductive
elimination in Scheme 1 were considered. Rate expressions are
provided as supporting information. Path A is initiated by
reversible or irreversible cleavage of the dimeric complex to form
two monomers, which undergo reductive elimination to form a
Pd(0) species that is trapped by P(t-Bu)3 to form 2. Although
reductive elimination from 1 would be endothermic, the trapping
of the Pd(0) intermediate by P(t-Bu)3 to form 2 would provide
the overall driving force. This mechanism predicts either half-
order rate behavior in dimer and first-order behavior in P(t-Bu)3
or first-order behavior in dimer and zero-order behavior in P(t-
Bu)3. Both of these predictions are inconsistent with our data.
Moreover, these mechanisms predict that dimer cleavage requires
55 °C to occur or that reductive elimination of aryl halide occurs
reversibly at 55 °C in the absence of P(t-Bu)3. To probe for
reversible dimer cleavage, we conducted the reaction in eq 2 and
Acknowledgment. We thank the NIH (R-01GM58108) for support
of this work, Johnson-Matthey for a gift of PdCl2, and Dr. Grace Mann
for the initial observation of this reaction.
Supporting Information Available: Spectroscopic and analytical data
on new compounds, procedures for kinetic studies, and kinetic expressions
for mechanisms A-E in Scheme 1 (PDF). This material is available free
of charge via the Internet at http://pubs.acs.org.
JA0034592