Kinetics and mechanistic aspects of the Heck reaction promoted by a CN-palladacycle

Article abstract of DOI:10.1021/ja051834q

In the Heck reaction between aryl halides and n-butyl acrylate, the palladacycle {Pd[κ¹-C, κ¹-N-C=(C ₆H₅)C(Cl)CH₂NMe₂](μ-Cl)}2, 1, is merely a reservoir of the catalytically active Pd(0) species [1] (Pd colloids or highly active forms of low ligated Pd(0) species) that undergoes oxidative addition of the aryl halide on the surface with subsequent detachment, generating homogeneous Pd(II) species. The main catalytic cycle is initiated by oxidative addition of iodobenzene to [1], followed by the reversible coordination of the olefin to the oxidative addition product. All the unimolecular subsequent steps are indistinguishable kinetically and can be combined in a single step. This kinetic model predicts that a slight excess of alkene relative to iodobenzene leads to a rapid rise in the Pd(0) concentration while when using a slight excess of iodobenzene, relative to alkene, the oxidative addition product is the resting state of the catalytic cycle. Competitive experiments of various bromoarenes and iodoarenes with n-butyl acrylate catalyzed by 1 and CS, CP, and NCN palladacycles gave the same p value (2.4-2.5 for Ar-Br and 1.7-1.8 for Ar-I) for all palladacycles employed, indicating that they generate the same species in the oxidative addition step. The excellent fit of the slope with the σ₀ Hammett parameter and the entropy of activation of -43 ± 8 J mol_-1 K _-1 are consistent with an associative process involving the development of only a partial charge in the transition state for the oxidative step of iodobenzene.

Full text of DOI:10.1021/ja051834q

Published on Web 08/06/2005
Kinetics and Mechanistic Aspects of the Heck Reaction
Promoted by a CN-Palladacycle
Crestina S. Consorti, Fabr´ıcio R. Flores, and Jairton Dupont*
Contribution from the Laboratory of Molecular Catalysis, Institute of Chemistry, UFRGS
AVenida Bento Gonc¸alVes, 9500 Porto Alegre 91501-970 RS Brazil
Received March 22, 2005; E-mail: dupont@iq.ufrgs.br
Abstract: In the Heck reaction between aryl halides and n-butyl acrylate, the palladacycle {Pd[κ1-C, κ1-
N-Cd(C6H5)C(Cl)CH2NMe2](µ-Cl)}2, 1, is merely a reservoir of the catalytically active Pd(0) species [1]
(Pd colloids or highly active forms of low ligated Pd(0) species) that undergoes oxidative addition of the
aryl halide on the surface with subsequent detachment, generating homogeneous Pd(II) species. The main
catalytic cycle is initiated by oxidative addition of iodobenzene to [1], followed by the reversible coordination
of the olefin to the oxidative addition product. All the unimolecular subsequent steps are indistinguishable
kinetically and can be combined in a single step. This kinetic model predicts that a slight excess of alkene
relative to iodobenzene leads to a rapid rise in the Pd(0) concentration while when using a slight excess
of iodobenzene, relative to alkene, the oxidative addition product is the resting state of the catalytic cycle.
Competitive experiments of various bromoarenes and iodoarenes with n-butyl acrylate catalyzed by 1 and
CS, CP, and NCN palladacycles gave the same F value (2.4-2.5 for Ar-Br and 1.7-1.8 for Ar-I) for all
palladacycles employed, indicating that they generate the same species in the oxidative addition step. The
excellent fit of the slope with the σ0 Hammett parameter and the entropy of activation of -43 ( 8 J mol-1
K-1 are consistent with an associative process involving the development of only a partial charge in the
transition state for the oxidative step of iodobenzene.
Introduction
due their facile synthesis, thermal stability, and possibility to
modulate their steric and electronic properties.11 Turnover
Palladium-promoted C-C coupling reactions are one of the
most important and investigated classes of organometallic
reactions.1 These catalytic processes possess several advantages
when compared with more classical organic methods, such as
their superb functional group tolerance, mild conditions em-
ployed, less waste, and sometimes fewer steps than the original
stoichiometric routes. Among these catalytic processes, the
formation of a C-C bond between alkenes and an aromatic
ringsthe Heck-Mizoroki reaction2,3sis finding increasing use
for the industrial production of fine chemicals, such as
monomers for coatings, pharmaceuticals, sunscreen agents,
herbicides.4 The coupling of alkenes with aryl halides can be
accomplished by a plethora of palladium catalyst precursors
under various reaction conditions. In fact, almost any molecular
Pd(0) or Pd(II) catalyst precursor associated or not5 with
phosphorus, nitrogen, carbene and other ligands, colloids,6 and
supported7 or heterogeneous8,9 sources of palladium promotes
the Heck-Mirozoki reaction involving aryl iodides and bro-
mides. Among these catalytic processes, palladacycles including
CP, CN, CS, PCP, SCS, NCN, SeCSe types10 are one of the
most investigated classes of catalyst precursors, in particular
numbers (moles of product/moles of Pd) ranging from 104 to
106 are easily achieved, depending on the nature of the base,
solvent, temperature, and additive (usually N(nBu)4Br salt, the
so-called Jeffery conditions12) using these catalyst precursors.13
However, these procedures still suffer limitations, such as not
being applicable for less reactive and less expensive aryl
(5) See for example: (a) Gruber, A. S.; Pozebon, D.; Monteiro, A. L.; Dupont,
J. Tetrahedron Lett. 2001, 42, 7345-7348. (b) de Vries, A. H. M.;
Parlevliet, F. J.; Schmieder-van de Vondervoort, L.; Mommers, J. H. M.;
Henderickx, H. J. W.; Walet, M. A. M.; de Vries, J. G. AdV. Synth. Catal.
2002, 344, 996-1002. (c) de Vries, A. H. M.; Mulders, J. M. C. A.;
Mommers, J. H. M.; Henderickx, H. J. W.; de Vries, J. G. Org. Lett. 2003,
5, 3285-3288. (d) Alimardanov, A.; de Vondervoort, L. S. V.; de Vries,
A. H. M.; de Vries, J. G. AdV. Synth. Catal. 2004, 346, 1812-1817. (e)
Reetz, M. T.; de Vries, J. G. Chem. Commun. 2004, 1559-1563.
(6) (a) Beller, M.; Fischer, H.; Kuhlein, K.; Reisinger, C. P.; Herrmann, W.
A. J. Organomet. Chem. 1996, 520, 257-259. (b) Walter, J.; Heiermann,
J.; Dyker, G.; Hara, S.; Shioyama, H. J. Catal. 2000, 189, 449-455. (c)
Biffis, A.; Zecca, M.; Basato, M. J. Mol. Catal. A-Chem. 2001, 173, 249-
274. (d) Kogan, V.; Aizenshtat, Z.; Popovitz-Biro, R.; Neumann, R. Org.
Lett. 2002, 4, 3529-3532. (e) Moreno-Manas, M.; Pleixats, R. Acc. Chem.
Res. 2003, 36, 638-643. (f) Mandal, S.; Roy, D.; Chaudhari, R. V.; Sastry,
M. Chem. Mater. 2004, 16, 3714-3724. (g) Gniewek, A.; Trzeciak, A.
M.; Ziolkowski, J. J.; Kepinski, L.; Wrzyszcz, J.; Wlodzimierz, T. J. Catal.
2005, 229, 332-343.
(7) See, for example: (a) Djakovitch, L.; Heise, W.; Kohler, K. J. Organomet.
Chem. 1999, 584, 16-26. (b) Bergbreiter, D. E.; Osburn, P. L.; Wilson,
A.; Sink, E. M. J. Am. Chem. Soc. 2000, 122, 9058-9064. (c) Venkatesan,
C.; Singh, A. P. Catal.Lett. 2003, 88, 193-197. (d) Corma, A.; Garcia,
H.; Leyva, A. Tetrahedron 2004, 60, 8553-8560. (e) Venkatesan, C.; Singh,
A. P. J. Catal. 2004, 227, 148-163. (f) Yu, K. Q.; Sommer, W.; Weck,
M.; Jones, C. W. J. Catal. 2004, 226, 101-110. (g) Yu, K.; Sommer, W.;
Richardson, J. M.; Weck, M.; Jones, C. W. AdV. Synth. Catal. 2005, 347,
161-171. (h) Bergbreiter, D. E.; Osburn, P. L.; Frels, J. D. AdV. Synth.
Catal. 2005, 347, 172-184.
(1) (a) Negishi, E.; Ed. Handbook of Organopalladium Chemistry for Organic
Synthesis; John Wiley & Sons: New York, 2002. (b) Bellina, F.; Carpita,
A.; Rossi, R. Synthesis 2004, 2419-2440. (c) Zapf, A.; Beller, M. Chem.
Commun. 2005, 431-440.
(2) (a) Mizoroki, T.; Mori, K.; Ozaki, A. Bull. Chem. Soc. Jpn. 1971, 44, 581-
581. (b) Heck, R. F.; Nolley, J. P. J. Org. Chem. 1972, 37, 2320-2322.
(3) Beletskaya, I. P.; Cheprakov, A. V. Chem. ReV. 2000, 100, 3009-3066.
(4) de Vries, J. G. Can. J. Chem.-ReV. Can. Chim. 2001, 79, 1086-1092.
9
12054
J. AM. CHEM. SOC. 2005, 127, 12054-12065
10.1021/ja051834q CCC: $30.25 © 2005 American Chemical Society

Heck Reaction Promoted by a CN−Palladacycle
A R T I C L E S
chlorides,14 not being easily recyclable, and requiring high
reaction temperatures,15 thus limiting the scope of substrates
that can be used, i.e., those that are thermally stable. It is evident
that a detailed, intimate knowledge of the steps and the reaction
mechanism involved could provide new and/or more improved
procedures for these C-C coupling reactions.
Scheme 1. Chloropalladation of
N,N-Dimethyl-1-phenylpropargylamine
It is now accepted that in most of the cases, the catalytically
active species involved in these reactions are based on Pd(0)
and that the reaction proceeds through a Pd(0)/Pd(II) catalytic
cycle,16 whatever the nature of the catalyst precursor (ionic,
molecular, colloidal, supported or “heterogeneous”).17 Although
in some cases it was proposed that the Heck reaction can occur
at the metal surfaces of the colloidal or heterogeneous palladium-
based catalyst precursors,9 most experimental evidence points
to them serving as a reservoir of soluble catalytically active
species.8 Supported Pd(II) catalysts have apparently similar roles
in these coupling reactions, i.e., they are actually reservoirs of
soluble ill-defined forms of catalytically active Pd(0) species.7
There are various indications that in the arylation of alkenes
promoted by palladacycles, they act as a reservoir of catalytically
active Pd(0) species (closely akin to “ligandless” precursors5),
but there is also the involvement of colloidal palladium.11d In
this context, the question as to whether the reaction is promoted
by the nanoparticles themselves or that they serve as a reservoir
of soluble catalytically active Pd species still remains open.
However, it is not a simple task to distinguish between so-called
“homogeneous” and “heterogeneous” catalysts,18 in particular
in the reactions promoted by transition-metal compounds.
Notwithstanding, there are now several approaches, such as
poisoning experiments; intrinsic kinetics, related to the formation
of the nanoparticles and with the catalytic reaction itself; and
physical-chemical analysis (transmission electron microscopy,
for instance) that can be used as probes to determine the nature
of the “true” catalyst. Although some of these approaches have
been used to probe the species involved in the Heck reaction
promoted by palladacycles, only limited conclusions have
appeared so far, most of them indicating that they are precata-
lysts of ill-defined Pd(0) catalytically active species.5c,15b,17d,19
We have recently reported that the palladacycle 1 derived from
the chloropalladation of N,N-dimethyl-1-phenylpropragylamine
(Scheme 1) is an outstanding catalyst precursor for the coupling
of aryl iodides and bromides, and these reactions can be
conducted at room temperature.15b In this particular case, Hg
poisoning experiments indicated the involvement of low-ligated
Pd(0) as the catalytically active species. We have now employed
several of the most used physical-chemical probes to determine
not only the nature of the species involved in the Heck reaction
promoted by palladacycle 1 but also the kinetics and mechanism
of this process.
(8) (a) Djakovitch, L.; Kohler, K. J. Organomet. Chem. 2000, 606, 101-107.
(b) Zhao, F. Y.; Bhanage, B. M.; Shirai, M.; Arai, M. Chem. Eur. J. 2000,
6, 843-848. (c) Biffis, A.; Zecca, M.; Basato, M. J. Mol. Catal. A-Chem.
2001, 173, 249-274. (d) Biffis, A.; Zecca, M.; Basato, M. Eur. J. Inorg.
Chem. 2001, 1131-1133. (e) Kohler, K.; Wagner, M.; Djakovitch, L. Catal.
Today 2001, 66, 105-114. (f) Zhao, F. Y.; Murakami, K.; Shirai, M.; Arai,
M. J. Catal. 2000, 194, 479-483. (g) Kohler, K.; Heidenreich, R. G.;
Krauter, J. G. E.; Pietsch, M. Chem.Eur. J. 2002, 8, 622-631. (h)
Heidenreich, R. G.; Krauter, E. G. E.; Pietsch, J.; Kohler, K. J. Mol. Catal.
A-Chem. 2002, 182, 499-509. (i) Prockl, S. S.; Kleist, W.; Gruber, M. A.;
Kohler, K. Angew. Chem., Int. Ed. 2004, 43, 1881-1882.
(9) (a) Choudary, B. M.; Madhi, S.; Chowdari, N. S.; Kantam, M. L.; Sreedhar,
B. J. Am. Chem. Soc. 2002, 124, 14127-14136. (b) Dams, M.; Drijkonin-
gen, L.; Pauwels, B.; Van Tendeloo, G.; De Vos, D. E.; Jacobs, P. A. J.
Catal. 2002, 209, 225-236. (c) Corma, A.; Garcia, H.; Leyva, A.; Primo,
A. Appl. Catal. A-Gen. 2004, 257, 77-83. (d) Choudary, B. M.; Madhi,
S.; Kantam, M. L.; Sreedhar, B.; Iwasawa, I. J. Am. Chem. Soc. 2004,
126, 2292-2293.
(10) See for example: (a) Beller, M.; Fischer, H.; Herrmann, W. A.; Ofele, K.;
Brossmer, C. Angew. Chem., Int. Ed. Engl. 1995, 34, 1848-1849. (b) Ohff,
M.; Ohff, A.; vanderBoom, M. E.; Milstein, D. J. Am. Chem. Soc. 1997,
119, 11687-11688. (c) Albisson, D. A.; Bedford, R. B.; Scully, P. N.
Tetrahedron Lett. 1998, 39, 9793-9796. (d) Shaw, B. L.; Perera, S. D.;
Staley, E. A. Chem. Commun. 1998, 1361-1362. (e) Ohff, M.; Ohff, A.;
Milstein, D. Chem. Commun. 1999, 357-358. (f) Bergbreiter, D. E.;
Osburn, P. L.; Liu, Y. S. J. Am. Chem. Soc. 1999, 121, 9531-9538. (g)
Gruber, A. S.; Zim, D.; Ebeling, G.; Monteiro, A. L.; Dupont, J. Org. Lett.
2000, 2, 1287-1290. (h). (g) Alonso, D. A.; Najera, C.; Pacheco, M. C.
Org. Lett. 2000, 2, 1823-1826. (i) Rocaboy, C.; Gladysz, J. A. Org. Lett.
2002, 4, 1993-1996. (j) Yao, Q. W.; Kinney, E. P.; Zheng, C. Org. Lett.
2004, 6, 2997-2999.
(11) (a) Herrmann, W. A.; Bohm, V. P. W.; Reisinger, C. P. J. Organomet.
Chem. 1999, 576, 23-41. (b) Dupont, J.; Pfeffer, M.; Spencer, J. Eur. J.
Inorg. Chem. 2001, 1917-1927. (c) Bedford, R. B. Chem. Commun. 2003,
1787-1796. (d) Beletskaya, I. P.; Cheprakov, A. V. J. Organomet. Chem.
2004, 689, 4055-4082. (e) Dupont, J.; Consorti, C. S.; Spencer, J. Chem.
ReV. 2005, 105, 2427-2571.
(12) (a) Jeffery, T.; Galland, J. C. Tetrahedron Lett. 1994, 35, 4103-4106. (b)
Jeffery, T. Tetrahedron, 1996, 52, 10113-10130.
(13) Farina, V. AdV. Synth. Catal. 2004, 346, 1553-1582.
(14) Littke, A. F.; Fu, G. C. Angew. Chem., Int. Ed. 2002, 41, 4176-4211.
(15) Only two examples of Heck coupling at room temperature have been
reported so far: (a) Littke, A. F.; Fu, G. C. J. Am. Chem. Soc. 2001, 123,
6989-7000. (b) Consorti, C. S.; Zanini, M. L.; Leal, S.; Ebeling, G.;
Dupont, J. Org. Lett. 2003, 5, 983-986.
Experimental Section
General Methods. All catalytic reactions were carried out under
an argon or nitrogen atmosphere in an oven-dried resealable Schlenk
tube. All substrates were purchased from Acros and used without further
purification. N,N-dimethylacetamide was degassed and stored over
molecular sieves. Molten NBu4Br was exposed to vacuum for 1 h prior
to use. NMR spectra were recorded on a Varian Inova 300 MHz
spectrometer. In situ time-resolved IR analysis were performed in a
Bomem B-102 spectrometer adapted with an ATR probe Axiom Dipper
210 equipped with a ZnSe reflectance element. In situ time-resolved
UV analyses were performed in a Varian Cary 50 spectrophotometer.
Gas chromatography analyses were performed on a Hewlett-Packard-
5890 gas chromatograph with a FID detector and 30 m DB17 capillary
column. TEM and EDS analysis were performed in a JEOL-JEM 2010
electronic microscope (200 kV). The CN (1),20 CP (2),21 CS (3),22 and
(16) See, for example: (a) Amatore, C.; Jutand, A. J. Organomet. Chem. 1999,
576, 254-278. (b) Amatore, C.; Jutand, A. Acc. Chem. Res. 2000, 33, 314-
321. (c) Cruz, S. C.; Aarnoutse, P. J.; Rothenberg, G.; Westerhuis, J. A.;
Smilde, A. K.; Bliek, A. Phys. Chem. Chem. Phys. 2003, 5, 4455-4460.
(d) Buback, M.; Perkovic, T.; Redlich, S.; de Meijere, A. Eur. J. Org.
Chem. 2003, 2375-2382 and references therein.
(17) (a) Reetz, M. T.; Westermann, E.; Lohmer, R.; Lohmer, G. Tetrahedron
Lett. 1998, 39, 8449-8452. (b) Reetz, M. T.; Westermann, E. Angew.
Chem., Int. Ed. 2000, 39, 165-168. (c) de Vries, A. H. M.; Parlevliet, F.
J.; Schmieder-van de Vondervoort, L.; Mommers, J. H. M.; Henderickx,
H. J. W.; Walet, M. A. M.; de Vries, J. G. AdV. Synth. Catal. 2002, 344,
996-1002. (d) Rocaboy, C.; Gladysz, J. A. New J. Chem. 2003, 27, 39-
49. (e) Tromp, M.; Sietsma, J. R. A.; van Bokhoven, J. A.; van Strijdonck,
G. P. F.; van Haaren, R. J.; van der Eerden, A. M. J.; van Leeuwen, P. W.
N. M.; Koningsberger, D. C. Chem. Commun. 2003, 128-129. (f) Cassol,
C. C.; Umpierre, A. P.; Machado, G.; Wolke, S. I.; Dupont, J. J. Am. Chem.
Soc. 2005, 127, 3298-3299.
(18) (a) Lin, Y.; Finke, R. G. Inorg. Chem. 1994, 33, 4891-4910. (b) Widegren,
J. A.; Finke, R. G. J. Mol. Catal. A-Chem. 2003, 198, 317-341.
(19) (a) Nowotny, M.; Hanefeld, U.; van Koningsveld, H.; Maschmeyer, T.
Chem. Commun. 2000, 1877-1878. (b) Beletskaya, I. P.; Kashin, A. N.;
Karlstedt, N. B.; Mitin, A. V.; Cheprakov, A. V.; Kazankov, G. M. J.
Organomet. Chem. 2001, 622, 89-96. (c) Bedford, R. B.; Cazin, C. S. J.;
Hursthouse, M. B.; Light, M. E.; Pike, K. J. Wimperis, S. J. Organomet.
Chem. 2001, 633, 173-181. (d) Eberhard, M. R. Org. Lett. 2004, 6, 2125-
2128. (e) Consorti, C. S.; Ebeling, G.; Flores, F. R.; Rominger, F.; Dupont,
J. AdV. Synth. Catal. 2004, 346, 617-624.
(20) Dupont, J.; Basso, N. R.; Meneghetti, M. R.; Konrath, R. A.; Burrow, R.;
Horner, M. Organometallics 1997, 16, 2386-2391.
9
J. AM. CHEM. SOC. VOL. 127, NO. 34, 2005 12055

A R T I C L E S
Consorti et al.
Scheme 2 a
NCN (4)23 palladacycles were synthesized according to known proce-
dures. Dibenzobicyclo[a,e]cyclooctatetraene (DCT)24 was synthesized
starting from dibenzobicyclo[2.2.2]octatriene(dibenzobarrelene).25
In Situ UV Analysis. A quartz cuvette (b ) 1 cm) was charged
with palladacycle 1 (2 mL of a 4.76 × 10-4 mmol L-1 DMA solution)
and 1 mL of a DMA solution containing NaOAc (0.83 mmol/25 mL,
68 mg/25 mL), NBu4Br (0.12 mmol/25 mL, 38 mg/25 mL), and n-butyl
acrylate (0.71 mmol/25 mL, 91 mg/25 mL). The reaction progress was
followed by UV spectra.
In situ ATR-IR Analysis. A 50-mL glass reactor was evacuated
and back-filled with argon and charged with sodium acetate (5.75 mmol,
460 mg) and tetrabutylammonium bromide (0.8 mmol, 256 mg). The
flask was then connected to the ATR probe and then DMA (25 mL)
was added, the temperature stabilized, and the background spectra
acquired. n-Butyl acrylate (4.8 mmol, 687 µL) and palladacycle 1 (500
µL of a 8 × 10-3 mol L-1 DMA solution, 4 × 10-3 mmol) were added
to the reaction solution, and after a pretreatment time of 10 min,
iodobenzene (4 mmol, 448 µL) was added and the reaction monitored
by IR. Iodobenzene and trans-n-butyl cinnamate concentrations, as a
function of time, were obtained by integration of the 735 and 772 cm-1
bands corrected by calibration curves. GC analyses of the reaction were
performed at the end of the reaction and showed an excellent agreement
with IR data (1-2% deviation).
Typical Experiment for the Hammett Competition Reaction. A
10-mL resealable Schlenk flask was evacuated and back-filled with
argon and charged with sodium acetate (1.4 mmol, 112 mg) and
tetrabutylammonium bromide (0.2 mmol, 64 mg). The flask was then
evacuated and back-filled with argon and then N,N-dimethylacetamide
(5 mL), n-butyl acrylate (10 mmol, 1.5 mL), methyl benzoate as internal
standard (35 mg), 0.14 mmol of bromobenzene, and 0.14 mmol of one
of the substrates 4-MeOPhBr, 4-MePhBr, 3-MePhBr, 4-ClPhBr,
4-ClPhBr, 3-ClPhBr, 4-CF3PhBr, 3-CF3PhBr, 4-NCPhBr, 4-MeCOPhBr,
or 4-NO2PhBr were added. After the addition of the palladacycle 1 in
N,N-dimethylacetamide (42 µL of a 2.4 × 10-5 mol L-1 solution), the
reaction mixture was stirred at 150 °C. The reaction was monitored by
GC, and the initial relative rates were used to plot Hammett correlation.
The competitive reactions were also performed with an excess of aryl
halides (1.4 mmol of bromobenzene and 1.4 mmol of substituted
bromoarene) relative to n-butyl acrylate (0.14 mmol) (see Supporting
Information).
a Reaction conditions: DMA (5 mL), NaOAc (1.4 mmol), n-butyl
acrylate (1.2 mmol) PhI (1 mmol), 1 (1.0 × 10-3 mmol), and NBu4Br (0.2
mmol). Hg(0) (300:1 Hg:Pd) addition after 2 h.
for 2 h. The cleavage solution was dried under vacuum and the residual
amount of free halobenzoic acid was employed to obtain the resin
loading.
Typical Experiment for the Heck Reaction of n-Butyl Acrylate
with the Substrate Attached to the Wang Resin. A 10-mL resealable
Schlenk flask was evacuated and back-filled with argon and charged
with sodium acetate (1.4 mmol, 112 mg), tetrabutylammonium bromide
(0.2 mmol, 64 mg), and the Wang resin containing the attached
4-iodobenzoic acid (500 mg; I loading: 0.68 mmol g-1, 0.35 mmol of
I), DMA (5 mL), n-butyl acrylate (0.5 mmol, 72 µL), and palladacycle
1 (147 µL of a 2.4 × 10-3 mol L-1 DMA solution, 3.5 × 10-4 mmol).
The reaction mixture was stirred at 30 °C for 24 h. The Wang resin
was collected on a filter funnel, washed with DMF (3×), H2O (3×),
DMF (3×), CH2Cl2 (3×), and MeOH (3×), and dried under vacuum.
The Wang resin was suspended in the cleavage solution (3 mL,
trifluoroacetic acid:CH2Cl2 1:1 (v/v)) and stirred for 2 h. The cleavage
solution was dried under vacuum and 1,3,5-tri-hydroxibenzene utilized
1
as internal standard for H NMR quantification of the Heck product.
For Heck reactions with the acrylic resin, the cleavage of the Heck
product was performed with dry methanol (4 mL) and LiBH4 (200 mg)
for 24 h. The reaction was quenched with the addition of wet methanol
1
and the Heck product was quantified by H NMR.
DCT Inhibition Test. Two identical Heck experiments consisting
of n-butyl acrylate (1.2 mmol, 172 µL), iodobenzene (1 mmol, 112
µL), 5 mL of N,N-dimethylacetamide, sodium acetate (1.4 mmol, 112
mg), tetrabutylammonium bromide (0.2 mmol, 64 mg), and palladacycle
1 in dimethylacetamide (3.2 mg, 10-2 mmol) were stirred at 30 °C.
After 15 min, DCT (2 × 10-2 mmol, 4 mg) was added to one reaction
vessel. Aliquots were taken from both reactions and analyzed by GC.
Results and Discussion
Hg(0) Poisoning Test. Two identical Heck experiments consisting
of n-butyl acrylate (1.2 mmol, 172 µL), iodobenzene (1 mmol, 112
µL), 5 mL of N,N-dimethylacetamide, sodium acetate (1.4 mmol, 112
mg), tetrabutylammonium bromide (0.2 mmol, 64 mg), and palladacycle
1 in N,N-dimethylacetamide (42 µL of a 2.4 × 10-3 mol L-1 solution,
10-3 mmol) were stirred at 80 °C. Aliquots were taken from both
reactions and analyzed by GC. After 2 h, Hg (0.3 mmol, 60 mg) was
added to one reaction vessel.
Attachment of Halobenzoic Acids to Wang Resin. Wang resin (2
g; OH loading: 1 mmol g-1, 2 mmol of OH) was suspended in 9:1
(v/v) CH2Cl2/DMF in a Schlenk flask. The halobenzoic acid (6 mmol),
dicyclohexylcarbodiimide (1.24 g, 6 mmol), and 4-dimethylamino-
pyridine (25 mg, 0.2 mmol) were added to the resin suspension, and
the reaction mixture was refluxed for 4 h. The resin was collected in
a filter funnel, washed with DMF (3×), H2O (3×), DMF (3×), CH2Cl2
(3×), and MeOH (3×), and dried under vacuum. For an estimation of
resin loading, 300 mg of Wang resin modified with the halobenzoic
acid was suspended in trifluoroacetic acid:CH2Cl2 1:1 (v/v) and stirred
Inhibition Tests. As already pointed out, several known
selective poisoning tests were applied to the catalytic systems
in order to characterize homogeneous or heterogeneous com-
ponents. One of the most popular is a Hg poisoning test that
relies on the formation of amalgam with colloidal palladium,
although poisoning of molecular Pd(0) cannot be excluded. Of
note is that, as pointed by Whitesides,26 a Hg poisoning test
can confirm a homogeneous catalytic system but not a hetero-
geneous one. Mercury addition can even improve the catalytic
activity of a heterogeneous hydrogenation reaction.27 The Hg(0)
poisoning test was performed in the Heck reaction of n-butyl
acrylate and iodobenzene promoted by palladacycle 1 (Scheme
1). When compared to control reactions, a complete inhibition
of the catalytic activity was observed after Hg (300 equiv)
addition (Scheme 2).
This positive response to mercury poisoning test unfortunately
cannot be taken as a confirmation of the presence of heteroge-
neous catalytically active palladium species in the Heck reaction
but does confirm the intervention of Pd(0) species.
(21) Herrmann, W. A.; Brossmer, C.; Ofele, K.; Reisinger, C. P.; Priermeier,
T.; Beller, M.; Fischer, H. Angew. Chem., Int. Ed. Engl. 1995, 34, 1844-
1848.
(22) Gruber, A. S.; Zim, D.; Ebeling, G.; Monteiro, A. L.; Dupont, J. Org. Lett.
2000, 2, 1287-1290.
(23) Consorti, C. S.; Ebeling, G.; Rodembusch, F.; Stefani, V.; Livotto, P. R.;
Rominger, F.; Quina, F. H.; Yihwa, C.; Dupont, J. Inorg. Chem. 2004, 43,
530-536.
(26) Whitesides, G. M.; Hackett, M.; Brainard, R. L.; Lavalleye, J.-P. P. M.;
Sowinski, A. F.; Izumi, A. N.; Moore, S. S.; Brown, D. W.; Staudt, E. M.
Organometallics 1985, 4, 1819-1830.
(27) Georgiades, G. C.; Sermon, P. A. J. Chem. Soc., Chem. Commun. 1985,
975-976.
(24) Wong, H. N. C.; Sondheimer, F. Tetrahedron 1981, 37, 99-109.
(25) Rabideau, P. W. OPPI Briefs 1986, 18, 113-116.
9
12056 J. AM. CHEM. SOC. VOL. 127, NO. 34, 2005

Heck Reaction Promoted by a CN−Palladacycle
A R T I C L E S
Table 1. Catalytic Activity of Palladacycle 1 with Polymer-Supported Substratesa
b
a Reaction conditions: 24 h, DMA (3 mL), resin (0.35 mmol), base (0.5 mmol), NBu4Br (0.1 mmol), n-butyl acrylate (0.5 mmol). 1H NMR yield after
resin cleavage. c Cleavage with TFA/CH2Cl2. d Cleavage with LiBH4 (internal standard 1,3,5-trihydroxibenzene).
The Collman test28 is based on the ability of substrates
attached to cross-linked polymers in distinguishing homoge-
neous or heterogeneous catalytically active species. Only soluble
or highly solvated species are able to diffuse through the
Figure 1. Dibenzo[a,e]cyclooctatetraene (DCT).
polymeric matrix, and as a consequence, only homogeneous
catalytically active species are expected to show catalytic
activity. The Collman test was successfully utilized with slight
obtained ([ArX]/[Pd] ) 1.0 × 105 (entries 3 and 6, Table 1),
even without low concentrations of free substrate (three-phase
test). These results strongly suggest that homogeneous catalyti-
cally active species are responsible for the catalytic activity
observed.
modifications recently in the Heck reaction of n-butyl acrylate
and halobenzenes attached to poly(ethylene glycol) (three-phase
test),29 and no catalytic activity was observed with heterogeneous
sources of palladium (Pd/alumina). However, the addition of a
small quantity of a free substrate in the above catalytic system
causes the insoluble substrate to be completely converted in
the Heck product. This results strongly suggests the involvement
of homogeneous catalytically active species in the Heck reaction
even with traditional heterogeneous catalytic precursors: the
oxidative addition of the free haloarene to the Pd(0) surface
causes the release of Pd(II) species into the reaction solution,
where the reaction proceeds according to the classical homo-
geneous mechanism and the substrate attached to the insoluble
polymer can be converted.
In addition to showing that homogeneous palladium species
were present with the Collmann test, we decided to run the test
with the homogeneous species selective inhibitor dibenzo[a,e]-
cyclooctatetraene (DCT, Figure 1).30 This poisoning test, known
as Crabtree’s test, is based on the ability of DCT to form stable
and inert homogeneous complexes with platinum metals. The
coordination of DCT to a heterogeneous surface is highly
unfavorable, due to his rigid conformation. The irreversibility
of the DCT-Pd complex formation and, as a consequence, the
test validity depend on the reaction temperature. Since the Heck
reaction promoted by palladacycle 1 can performed at room
temperature, the Crabtree test could be applied for the first time
to the Heck reaction.
The catalytic activity of the Heck reaction of iodo- and
bromoarenes attached to cross-linked polystyrene promoted by
palladacycle 1 is almost similar to those conducted with free
substrates, and quantitative yields of the coupling products can
be obtained after resin cleavage. Huge catalytic activities were
The general experimental protocol proposed by Crabtree
includes the prior generation of the catalytically active species
(28) Collman, J. P.; Kosydar, K. M.; Bressan, M.; Lamanna, W.; Garrett, T. J.
Am. Chem. Soc. 1984, 106, 2569-2579.
(29) Davies, I. W.; Matty, L.; Hughes, D. L.; Reider, P. J. J. Am. Chem. Soc.
2001, 123, 10139-10140.
(30) Anton, D. R.; Crabtree, R. H. Organometallics 1983, 2, 855-859.
9
J. AM. CHEM. SOC. VOL. 127, NO. 34, 2005 12057

A R T I C L E S
Consorti et al.
Table 2. DCT Inhibition Testsa
a Reaction conditions: DMA (5 mL), NaOAc (1.4 mmol), olefin (1.2
mmol), ArX (1 mmol), 1 (1.0 × 10-3 mmol), and NBu4Br (0.2 mmol).
Reaction run with the Wang resin: DMA (3 mL), resin (0.35 mmol), base
(0.5 mmol), NBu4Br (0.1 mmol), olefin (0.5 mmol). DCT (2:1 DCT:Pd)
was added after 15 min. Relative reaction rate was obtained by comparison
of conversions at 15 min and 24 h. b GC yield (internal standard methyl
benzoate). 1H NMR yield (internal standard 1,3,5-trihydroxybenzene) after
c
Figure 2. Heck reaction of iodobenzene and n-butyl acrylate. Reaction
conditions: DMA (25 mL), NaOAc (0.250 mol L-1), n-butyl acrylate (0.213
mol L-1), PhI (0.179 mol L-1), NBu4Br (0.035 mol L-1), 1 (0.179 mmol
L-1), 80 °C. (0) Reaction performed without pretreatment of the pallada-
cycle. ([) Reaction performed with pretreatment of the palladacycle 1.
d
resin cleavage with TFA/CH2Cl2. 1H NMR yield (internal standard 1,3,5-
trihydroxybenzene) after resin cleavage with LiBH4.
and the subsequent contact of the active species with DCT for
at least 2 h prior to the addition of substrates. With minor
adaptation, the Crabtree test was applied to the Heck reaction
of iodobenzene with n-butyl acrylate performed at room
temperature. After 15 min, aliquots were taken from the
reactions for quantification, and DCT (2 equiv) was added to
one reaction vessel. After 24 h, the reactions were stopped for
conversion quantification and comparison. The same experi-
mental protocol for the inhibition test was performed with the
iodobenzene substrate attached to the Wang resin. It may be
argued that the catalytic activity of homogeneous species will
be suppressed by DCT and heterogeneous active species will
not present any activity in the polymeric matrix. In both cases
a partial inhibition of the catalytic activity was observed (Table
2), with only 40% of the original catalytic activity maintained
for the free substrate and 50% for the insoluble substrate. The
fact that the Heck reaction catalytic activity was in part inhibited
by DCT is further strong experimental evidence of the presence
of homogeneous catalytically active Pd species.
Figure 3. Time-resolved UV profile of the palladacycle 1 absorption band
with pretreatment time. Reaction conditions: 1 (3.17 × 10-4 mol L-1),
n-butyl acrylate (9.5 × 10-3 mol L-1), NaOAc (1.2 × 10-2 mol L-1), and
NBu4Br (1 × 10-3 mol L-1), RT.
At this point, the information gathered with poisoning tests
cannot be regarded as an unambiguous proof of a homogeneous
Heck reaction, and we cannot exclude the participation of
palladium colloids or clusters as catalytically active species, as
the test range is not known. To gain insight into the reaction
mechanism, detailed TEM and UV-vis studies were performed.
UV-Vis and TEM Studies. A preliminary study of the Heck
reaction of iodobenzene and n-butyl acrylate promoted by
palladacycle 1 reveals sigmoidal kinetics for the iodobenzene
conversion (curve A, Figure 2). It has been found that pretreat-
ment of the palladacycle 1 with n-butyl acrylate, NaOAc, and
NBu4Br for 10 min before the iodobenzene addition cause a
pronounced change in the kinetics of the iodobenzene conversion
(curve B, Figure 2). The exponential decay observed for the
iodobenzene conversion suggests that the palladacycle has been
transformed into the catalytically active species during the
pretreatment time.
Figure 4. Time-resolved UV-vis spectra of the reaction of iodobenzene
with n-butyl acrylate after a pretreatment time of 40 min. Reaction
conditions: 1 (3.17 × 10-4 mol L-1), n-butyl acrylate (9.5 × 10-3 mol
L-1), NaOAc (1.2 × 10-2 mol L-1), and NBu4Br (1.6 × 10-3 mol L-1),
PhI (1.2 × 10-1 mol L-1), room temperature.
A clearer picture of the hypothesis that catalytically active
species are formed in the pretreatment of palladacycle 1 with
n-butyl acrylate, NaOAc, and NBu4Br was provided by UV-
vis spectrometry studies. Time-resolved UV-vis was utilized
be seen in Figure 3, the gradual decrease of the absorption band
at 360 nm is accompanied by a growth in the spectra baseline,
possibly related to Pd(0) colloid formation.
After a total pretreatment time of 40 min, iodobenzene (1.2
× 10-1 mol L-1) was added to the reaction mixture. Figure 4
shows the evolution of a new set of bands with absorption
maxima at 540 and 420 nm and a shoulder at 320 nm over a
to follow the reaction of palladacycle 1 (3.17 × 10-4 mol L-1
)
with n-butyl acrylate (9.5 × 10-3 mol L-1), NaOAc (1.2 ×
10-2 mol L-1), and NBu4Br (1.6 × 10-3 mol L-1).
Figure 3 shows the evolution of the absorption band of
palladacycle 1 at 360 nm in a reaction time of 30 min. As can
9
12058 J. AM. CHEM. SOC. VOL. 127, NO. 34, 2005

Heck Reaction Promoted by a CN−Palladacycle
A R T I C L E S
Kinetic Modeling. The progress of the Heck reaction of
iodobenzene and n-butyl acrylate was successfully monitored
by time-resolved IR using an ATR probe. The iodobenzene and
trans-butyl cinnamate concentrations as a function of time were
determined by integration of the Ph-I stretching vibration (735
cm-1) and the out-of-plane deformation at 772 cm-1 of the
trans-butyl cinnamate vinylic hydrogen atoms (Figure 7)
employing a response factor determined experimentally by
concentration calibration curves. This methodology provides an
accurate and noninvasive integral method for the determination
of the concentration profile for iodobenzene and the Heck
reaction product trans-butyl cinnamate during the Heck reaction,
since no aliquots are taken from the reaction mixture. Chro-
matographic analysis of iodobenzene conversion differs from
the ATR data no more that 1%.
In our preliminary studies, we tested the ATR methodology
to determine the dependence of reaction rates with the concen-
tration of the reaction components at 80 °C using the initial
rate method for experimental curves obtained after pretreatment
of the catalytic precursor (see Supporting Information). We have
found a first-order dependence of reaction rate in palladacycle
1 concentration and a zero-order dependence in NaOAc and
NBu4Br concentrations. Interestingly, saturation kinetics was
found for both iodobenzene and n-butyl acrylate substrates. The
existence of an association equilibrium involving olefin coor-
dination to palladium has been reported previously31 as a highly
unfavorable step, and in this case, both iodobenzene and n-butyl
acrylate can follow saturation kinetics.
Figure 5. TEM micrograph showing the palladium colloids found in the
pretreatment solution of palladacycle 1. Reaction conditions: 1 (3.17 ×
10-4 mol L-1), n-butyl acrylate (9.5 × 10-3 mol L-1), NaOAc (1.2 × 10-2
mol L-1), and NBu4Br (1.6 × 10-3 mol L-1), room temperature.
On the basis of these data, we propose the Heck reaction
mechanism depicted in Figure 8. We assume that, after a
pretreatment period of 10 min with NaOAc, NBu4Br, and
n-butyl acrylate, the palladacycle 1 is completely converted into
the catalytically active Pd(0) species [1]. The active species [1]
can consist of palladium colloids identified by TEM analysis
or even highly active forms of low-ligated Pd(0) species
stabilized by anions (acetate or halides) and/or solvent mol-
ecules, associated with the quaternary ammonium salt. The
proposed Heck catalytic cycle is initiated by oxidative addition
of iodobenzene to [1], followed by the reversible coordination
of the olefin ([Acr]) to the oxidative addition product. All the
unimolecular subsequent steps (olefin insertion, â-elimination,
reductive elimination, and Heck product dissociation) are
indistinguishable kinetically and for this reason are combined
in a single step characterized by the kinetic constant k3.
Using the steady-state approximation for the species [1], [2],
and [3], the reaction rate eq 1 can be easily found (for details,
see the Supporting Information):
Figure 6. EDS analysis of the palladium colloids.
reaction time of 3 h. Although the new species formed cannot
be unambiguously assigned, they provide strong evidence for
the formation of homogeneous Pd(II) species, regarding the
similarity of the bands with those of simple PdXn- salts. Thus,
these results are in agreement with the process of formation of
palladium colloids starting from the palladacycle and the
subsequent oxidative addition of the iodoarene on the colloid
surface and detachment of the homogeneous Pd(II) species, as
pointed out by Reetz.17b
The same decrease in the absorption band of palladacycle 1
and appearance of the homogeneous Pd(II) bands were observed
in the Heck reaction performed without the pretreatment of the
catalytic precursor.
Detailed TEM analysis was performed by withdrawing
samples of the reaction before and after iodobenzene addition
to the pretreatment solution by simply depositing the solution
on carbon-coated copper grids. Palladium colloids were found
in the pretreatment solution, as can be seen in the TEM
micrograph presented in Figure 5. Palladium colloids were
identified by the characteristic palladium lines at 2.83 keV (LR1)
and 3.03 keV (Lâ1) in the EDS analysis (Figure 6). After
iodobenzene addition, no attempts to identify palladium colloids
were successful, indicating that homogeneous species were
formed starting from the palladium colloids.
The experimental evidence gathered from the inhibition tests,
UV-vis, and TEM studies strongly establishes that palladium
colloids are generated starting from palladacycle 1 and they act
merely as reservoirs of homogeneous palladium species and the
Heck reaction catalytic cycle occurs in the homogeneous phase.
k1k2k3[PhI][Acr][Pd]
-d[PhI]/dt )
k2k3[Acr] + k1k2[PhI][Acr] + k1k3[PhI]
The fitting of the reaction rate equation (eq 1) to the
experimental curves was performed with the kinetic simulator
software Dynafit.32 The kinetic model depicted in Figure 8 was
simultaneously fitted to a set of experimental curves obtained
after pretreatment of the catalytic precursor showing good
agreement with experimental curves. The simulation of the
(31) (a) Rosner, T.; Pfaltz, A.; Blackmond, D. G. J. Am. Chem. Soc. 2001, 123,
4621. (b) Rosner, T.; Le Bars, J.; Pfaltz, A.; Blackmond, D. G. J. Am.
Chem. Soc. 2001, 123, 1848-1855.
(32) Kuzmic, P. Anal. Biochem. 1996, 237, 260-273.
9
J. AM. CHEM. SOC. VOL. 127, NO. 34, 2005 12059

A R T I C L E S
Consorti et al.
Figure 7. Time-resolved ATR-IR showing the C-I stretching vibration (735 cm-1) and trans-n-butyl cinnamate deformation (772 cm-1) bands versus
time during the Heck reaction of iodobenzene and n-butyl acrylate promoted by palladacycle 1. Reaction conditions: DMA (25 mL), NaOAc (0.250 mol
L-1), n-butyl acrylate (0.213 mol L-1), PhI (0.179 mol L-1), NBu4Br (0.035 mol L-1), 1 (0.179 mmol L-1), 80 °C.
simplified (k2k3 ) 9375 L mol-1 s-2; k1k3 ) 6578 L mol-1 s-2
;
k1k-2 ) 12 400 L mol-1 s-2). This kinetic model also predicts
that relative concentrations of substrate will change the distribu-
tion of intermediates involved in the catalytic cycle. A slight
excess of alkene relative to iodobenzene leads to a rapid rise in
the Pd(0) concentration while when using a slight excess of
iodobenzene, relative to alkene, the oxidative addition product
is the resting state of the catalytic cycle. This behavior is clearly
demonstrated in Figure 10, which shows the concentrations of
the palladium intermediates predicted by the rate equation (1)
during the Heck reaction starting from different relative substrate
concentrations.
The important conclusion that arises from analysis of this
kinetic model is that the initial rate measurements can be highly
distorted due to decomposition pathways of the intermediates.
In our case we notice palladium black deposition in reactions
performed with a deficiency of iodobenzene and reaction
deactivation by formation of PdXnm- salts via reductive coupling
of the oxidative addition product in reactions performed with
an excess of iodobenzene. The possible deactivation-regenera-
tion pathways involved in the catalytic process are summarized
in Figure 11. The first decomposition pathway (deactivation I),
i.e., the formation of Pd nanoparticles and eventually Pd metal
through agglomeration of Pd(0), has been observed several
times.33,34,35 The agglomeration process is autocatalytic and
thermodynamically downhill. In the cases where ligands and/
or stabilizing agents are present (exerting a kinetic control on
the nucleation step),36 the regeneration process occurs through
oxidative addition of the haloarene to form soluble Pd(II) species
(regeneration I). This regeneration process is dependent on
halide concentration, and the Pd nucleation rate increases with
a decrease in the halide concentration.37
Figure 8. Postulated Heck reaction mechanism.
proposed mechanism was performed for reactions starting with
three different initial concentrations of n-butyl acrylate, iodo-
benzene, and palladacycle 1. In this study we avoided artificially
low concentrations of n-butyl acrylate or iodobenzene, since in
the initial rate method we observed a high degree of reaction
deactivation. The initial concentrations of the catalytically active
species [1] after the preactiVation period were considered to
be the same as the palladacycle 1. The fitting of the model
mechanism to experimental data is presented in Figure 9, and
the kinetic constants are in Table 3.
It can be seen from the rate constants estimated by the kinetic
model that, even with an observed zero-order in iodobenzene
concentration over a wide range of initial concentrations
employed in this study, the bimolecular oxidative addition step
has the highest energy barrier. It is also noteworthy that no
simple order in substrate concentration can be derived from eq
1, as the denominator terms in the rate equation cannot be
(33) Rocaboy, C.; Gladysz, J. A. New J. Chem. 2003, 27, 39-49.
(34) de Vries, A. H. M.; Mulders, J. M. C. A.; Mommers, J. H. M.; Henderickx,
H. J. W.; de Vries, J. G. Org. Lett. 2003, 5, 3285-3288.
(35) Rocaboy, C.; Gladysz, J. A. Org. Lett. 2002, 4, 1993-1996.
9
12060 J. AM. CHEM. SOC. VOL. 127, NO. 34, 2005

Heck Reaction Promoted by a CN−Palladacycle
A R T I C L E S
Figure 9. Simultaneous fit of eq 1 to experimental curves obtained with different initial concentrations (mmol L-1) of palladacycle 1, iodobenzene, and
n-butyl acrylate.
Table 3. Rate Constants Determined by Fitting Eq 1 to
Experimental Data
NBu4Br used in the experiments possibly contains some quantity
of NBu3 that is responsible for the reduction of Pd(II) salts
constant
fitting
error
resulting from the deactivation pathway II (Figure 11). In this
respect, it is well-known that at temperatures above 100 °C
NBu4Br undergoes partial decomposition with the liberation of
NBu3.39 To check these hypotheses, some experiments were
performed using different types of additives (Table 4). It is clear
that there are differences between the use of dried and wet NBu4-
Br (compare entries 1 and 2, Table 4) on the catalytic
performance of palladacycle 1. This behavior was already
observed in Heck coupling reactions promoted by other pal-
ladium catalyst precursors.40 It is also quite fitting that the dried
salt additive can be substituted by equal amounts (0.2 equiv)
of NBu3 (compare entries 1 and 3, Table 4), and in the absence
of both salt additive and NBu3 only marginal catalytic activities
were observed (see entries 2 and 4, Table 4). Note that in the
reactions performed with dried NBu4Br, no deposition of
metallic palladium was observed in opposition to those per-
formed in the presence of NBu3, where metallic palladium is
formed.
The synergic effect between NaOAc and NBu3 combinations
can be rationalized in terms of an ideal equilibrium between
the deactivation I/regeneration I pathways (Figure 11) involving
the catalytically active species.41 The use of NaOAc alone at
lower temperatures promotes the complete deactivation of the
system, since the palladium halide species formed during the
reaction are not reduced by NaOAc at temperatures below 130
°C. In opposition, the use of sole tertiary amines as base induces
the formation of metallic palladium due to their strong and faster
reducing properties toward palladium halides species when
compared to other bases, such as NaOAc.41 Therefore for these
particular patterns, there are specific relative NaOAc/NBu3
concentrations that can generate an ideal equilibrium between
the rates of deactivation/regeneration of the Pd catalytically
active species, resulting in a higher catalytically active process.
Aromatic Substituent Effect on Reaction Rate. The linear
free-energy relationship through a Hammett correlation42 has
k1
k2
k-2
k3
22.3 L mol-1 s-1
33 L mol-1 s-1
556 s-1
0.3 L mol-1 s-1
2 L mol-1 s-1
10 s-1
295 s-1
53 s-1
The second decomposition pathway consists of the formation
of Pd(II) halide species34,37,38 (such as PdX2, PdX3-, PdX4
2-
,
and Pd2X62-) together with the formation of biaryls and
haloarene reduction product (deactivation II). These Pd(II) salts
and homocoupling and hydrogenolysis products are generated
from the oxidative addition intermediate, probably through a
bimolecular process.37 The regeneration (regeneration II) of the
catalytically active species occurs through the reduction Pd(II)
halides promoted by the base.
The same kinetic model also led to a good fit on reactions
performed without the initial pretreatment of the palladacycle
1. By simply adding a reduction step involving the palladacycle
1 and the additive NBu4Br, a good agreement with the
experimental curve is obtained, as can be seen in Figure 12
with a kinetic constant for precursor reduction of ka ) 3.5 ×
10-4 L mmol-1 s-1
.
All attempts to fit the proposed kinetic model into the Heck
reaction experimental curves obtained without the pretreatment
of the palladacycle 1 failed due to oscillation in the initial
induction period with the initial concentrations of NaOAc,
n-butyl acrylate, and iodobenzene. The only significant effect
in reaction rate and induction period was found with different
initial concentrations of NBu4Br, as can be seen in Figure 13.
The Role of NBu4Br. The reaction performed in the absence
of the additive NBu4Br is very slow and occurs with sedimenta-
tion of metallic palladium. It is worthwhile noting that for the
reactions performed with pretreatment of palladacycle 1 the rates
are independent of the salt additive concentration.
This salt effect is much probably related to the role of NBu4-
Br as stabilizing agent for the colloidal palladium, which hinders
its agglomeration to metallic palladium. Moreover, the “dried”
(39) (a) Amirnasr, M.; Nazeeruddin, M. K.; Gra¨tzel, M. Thermochimica Acta
2000, 348, 105-114. (b) Prasad, M. R. R.; Krishnamurthy, V. N.
Thermochim. Acta 1991, 185, 1-10.
(40) Bo¨hm, V. P. W.; Herrmann, W. A. Chem. Eur. J. 2000, 6, 1017-1025.
(41) Shmidt, A. F.; Khalaika, A.; Bylkova, V. G. Kinet. Catal. 1998, 39, 194-
199.
(42) Fuchs, R.; Lewis, E. S. Linear Free-Energy Relations. In InVestigations of
Rate and Mechanisms of Reactions, 4th ed.; Bernasconi, C. F., Ed.;
Wiley: New York, 1986, p 777.
(36) Calo, V.; Nacci, A.; Monopoli, A.; Laera, S.; Cioffi, N. J. Org. Chem.
2003, 68, 2929-2933.
(37) Shmidt, A. F.; Khalaika, A. Kinet. Catal. 1998, 39, 803-807.
(38) de Vries, A. H. M.; Parlevliet, F. J.; van de Vondervoort, L. S.; Mommers,
J. H. M.; Henderickx, H. J. W.; Walet, M. A. M.; de Vries, J. G. AdV.
Synth. Catal. 2002, 344, 996-1002.
9
J. AM. CHEM. SOC. VOL. 127, NO. 34, 2005 12061

A R T I C L E S
Consorti et al.
Figure 10. Distribution of palladium species within the Heck reaction catalytic cycle using different [PhI]/[Acr] initial concentration relations.
Figure 11. Possible deactivation/regeneration of the Pd species involved
in the Heck catalytic process.
Figure 12. Fitting of the kinetic model to the Heck reaction of iodobenzene
with n-butyl acrylate performed without pretreatment of the palladacycle
1. Reaction conditions: DMA (25 mL), NaOAc (0.250 mol L-1), n-butyl
acrylate (0.213 mol L-1), PhI (0.179 mol L-1), NBu4Br (0.035 mol L-1),
1 (0.179 mmol L-1), 80 °C.
been used several times to compare the substituent effects of
aryl halides in the Heck reaction.43-47 In most of the cases
investigated so far, positive F values were observed, and
electron-withdrawing groups are known to increase the rate of
the Heck reaction. This electronic effect is more pronounced
for chloroarenes than for bromoarenes, followed by iodoarenes.
Interestingly, the observed F are on the same order of magnitude,
whatever the nature of the Pd catalyst precursor, except in the
cases where σ- was used. However, it is relatively difficult to
extract relevant mechanistic information in a multistep reaction
sequence, since all of the individual steps involved may affect
the final value. Moreover, it is also very difficult to separate
these contributions from each other. Nevertheless, the Hammett
correlation is of enormous significance if the differences between
the constants obtained for the different catalyst precursors are
used for comparison purposes and not their absolute values.
To get more insight into the type of influence that the different
substrates may have in a competitive Heck reaction involving
differently substituted aryl halides, the kinetic model was
extended for two aryl halides (Figure 14). The mechanism is
the same as used in the kinetic modeling (Figure 8), except that
9
12062 J. AM. CHEM. SOC. VOL. 127, NO. 34, 2005

Heck Reaction Promoted by a CN−Palladacycle
A R T I C L E S
Figure 13. Heck reaction of iodobenzene and n-butyl acrylate employing
different initial NBu4Br concentrations. Reaction conditions: DMA (25 mL),
NaOAc (0.250 mol L-1), n-butyl acrylate (0.213 mol L-1), PhI (0.179 mol
L-1), 1 (0.179 mmol L-1), NBu4Br (0, 0.017, 0.070, and 0.13 mol L-1), 80
°C.
Table 4. Base and Additive Effect in the Heck Reaction between
Iodobenzene and n-Butyl Acrylate at 30 °C Promoted by
Palladacycle 1a
entry
additive
conv (%)b
yield (%)b
Figure 14. Proposed mechanism of a competitive Heck reaction involving
two different aryl halides.
1
2
3
4
dried NBu4Br
wet NBu4Br
NBu3
100
6
100
10
100
6
100
10
none
a Reaction conditions: [PhI]/[Pd] ) 1.0 × 102, 24 h, DMA (5 mL),
NaOAc (1.4 mmol), n-butyl acrylate (1.2 mmol), PhI (1 mmol), and additive
(0.2 mmol); 30 °C. b Yield and conversions determined by GC (methyl
benzoate as internal standard).
Figure 15. Palladacycles used in the Heck competitive experiments.
in this case the reaction starts with the two aryl halides
competing for the same catalytically active species [1] for the
formation of the respective oxidative addition products and the
catalytic cycle proceeds with their corresponding rate constants.
The treatment of this mechanism with the steady-state
approximation furnishes the following rate equations for the
consumption of competing haloarenes:
in the rate constant for the oxidative addition step. According
to eq 4 the competitive reactions should be performed under
pseudo-zero-order conditions relative to the aryl halide (large
excess of the aryl halides) or by determining the substrate
consumption at very low conversions where the reaction rate
does not depend on the aryl halide concentration. Indeed, both
assumptions have been corroborated experimentally (see the
Supporting Information).
A series of competitive experiments of 4- and 3-substituted
iodo- and bromoarenes with n-butyl acrylate (iodo- or bro-
mobenzene were used as standards) were performed using
palladacycle 1, and for comparative purposes, the same experi-
ments have been repeated with CP (2), CS (3), and NCN (4)
palladacycles (Figure 15).
k1k2k3k′k′[PhI][Acr][Pd]
2
3
-d[PhI]/dt )
-d[PhI]/dt )
(2)
(3)
a
k1k2k3k′k′[PhI][Acr][Pd]
2
3
a
a ) k2k3k′k′[Acr] + k1k′k′3[PhI](k-2 + k3 + k2[Acr]) +
2
3
2
k′k2k3[RPhI](k-′ 2 + k3′+ k′[Acr])
The reaction was stopped at very low substrate conversions,
and the relative rate constants were determined directly using
the relative yield obtained by GC. The Hammett parameter was
obtained through the slope defined by the straight line that
includes the meta-substituents, and the constant (σ0, σ, or σ-)
that best fits the para position was chosen, as recommended
for this kind of experiment.42 The competitive experiments
performed between bromo- or iodoarenes with only one
substituted aryl halide or with a group of aryl halides that possess
similar reactivity (all containing electron-withdrawing or all
containing electron-donating groups) gave the same relative
reactivity values. Similarly, the same relative reactivity values
2
2
The division of eqs 2 and 3 gives eq 4, which expresses the
relative reactivity of the two substrates that in its turn is equal
to the relative rate of formation of the two different products.
k′[RPhI]
d[RPhI]/dt
d[PhI]/dt
1
)
(4)
k1[RPhI]
It is now clear that, whatever the rate-limiting step, in a
competitive Heck reaction the changes in the relative reactivity
involving different aryl halides is strictly bound to the changes
9
J. AM. CHEM. SOC. VOL. 127, NO. 34, 2005 12063

A R T I C L E S
Consorti et al.
Table 5. R and r Values Obtained in Competitive Heck
Experiments Using 3- and 4-Aryl Bromides and Iodides with
n-Butyl Acrylate Promoted by Palladacycles 1-4 Using σ0
palladacycle
X
R
r
palladacycle
X
R
r
1
2
3
4
Br
Br
Br
Br
0.98
0.98
0.98
0.97
2.5
2.4
2.5
2.5
1
2
3
4
I
I
I
I
0.99
0.98
0.98
0.97
1.8
1.7
1.8
1.8
Figure 18. Eyring plot for the Heck reaction of iodobenzene and n-butyl
acrylate promoted by palladacycle 1. Reaction conditions: NaOAc (0.250
mol L-1), NBu4Br (0.035 mol L-1), n-butyl acrylate (21.3 mol L-1), PhI
(1.78 mol L-1), 1 (0.178 mmol L-1). Y ) (18.5 ( 1.0) + (-8287 ( 370)X
(R ) 0.99).
increases with the presence of electron-withdrawing substituents
on the aryl halide, and consequently, the transition state for the
oxidative addition is stabilized for groups that remove electronic
density from the Cipso of the aryl halide. The excellent fit of the
slope with σ0 (using the meta-substituents parameters of the
aryl halide as standard) shows that there is no formal charge
on the transition state of the oxidative step, as suggested by
theoretical calculations.48
Figure 16. Hammett correlation for the competitive Heck reaction between
bromoarenes and n-butyl acrylate at 150 °C in DMA promoted by different
palladacycles using the σ0 constants [palladacycles: 1 (O), 2 (1), 3 (4), 4
(9)].
Activation Parameters. The activation parameters for the
reaction between iodobenzene and n-butyl acrylate promoted
by palladacycle 1 after a pretreatment time of 10 min were
determined using the initial reaction rates at different temper-
atures employing the Eyring equation. The reactions were
performed with a large excess of n-butyl acrylate to ensure first-
order kinetics relative to iodobenzene, and consequently the rate
law (Equation 1) can be reduced to ri ) k1[PhI]0[Pd]0. The
Eyring plot is presented in Figure 18, and the activation
parameters ∆Hq ) 69 ( 3 kJ mol-1 and ∆Sq ) -43 ( 8 J
mol-1 K-1 could be determined, corresponding to the oxidative
addition step if the kinetic model is correct. Indeed, the ∆Hq of
69 ( 3 kJ mol-1 is identical to those found for the oxidative
addition of iodobenzene with Pd(PPh3)4.45,46 Moreover, the
entropy of activation of -43 ( 8 J mol-1 K-1 is consistent
with an associative process involving the development of a
partial charge in the transition state.
Conclusions. The palladacycle {Pd[κ1-C, κ1-N-Cd(C6H5)C(Cl)-
CH2NMe2](µ-Cl)}2, 1, is merely a reservoir of catalytically
active Pd(0) species [1] (Pd colloids or even highly active forms
of low-ligated Pd(0) species) that undergoes oxidative addition
of the aryl halide on the surface with subsequent detachment
of the homogeneous Pd(II) species. The reaction of iodobenzene
and n-butyl acrylate presents a first-order dependence of reaction
rate with palladacycle 1 concentration and a zero-order depen-
dence on NaOAc and NBu4Br concentrations. Saturation kinetics
was found for both iodobenzene and n-butyl acrylate substrates.
Figure 17. Hammett correlation for the Heck competitive reaction between
iodoarenes and n-butyl acrylate at 80 °C in DMA promoted by different
palladacycles using the σ0 constants [palladacycles: 1 (O), 2 (1), 3(4),
4(9)].
were obtained whether the competitive experiments were
performed under pseudo-first-order conditions relative to the
n-butyl acrylate or relative to the aryl halides (see Supporting
Information).
The use of the σ0 Hammett parameter42 results in excellent
linear correlation (Table 5) for bromoarenes (F ) 2.5, Figure
16) and iodoarenes (F ) 1.8, Figure 17) for the Heck reaction,
indicating that in both cases there is an electronic influence on
the reaction rate.
(43) Portnoy, M.; Milstein, D. Organometallics 1993, 12, 1655-1664.
(44) Ohff, M.; Ohff, A.; Boom, M. E. V. D.; Milstein, D. J. Am. Chem. Soc.
1997, 119, 11687-11688.
(45) Amatore, C., Pfluger, F. Organometallics 1990, 9, 2276-2282.
(46) Fauvarque, J.-F.; Pfluger, F. J. Organomet. Chem. 1981, 208, 419-427.
(47) Bo¨hm, V. P. W.; Herrmann, W. A. Chem. Eur. J. 2001, 7, 4191-4197.
(48) (a) Sundermann, A.; Uzan, O.; Martin, J. M. L. Chem. Eur. J. 2001, 7,
1703-1711. (b) Senn, H. M.; Ziegler, T. Organometallics 2004, 23, 2980-
2988.
It is also quite impressive that the same F value was obtained
(2.4-2.5 for Ar-Br and 1.7-1.8 for Ar-I) for all the four
palladacycles, indicating that they form the same active species
in the oxidative addition step. The rate of the oxidative addition
9
12064 J. AM. CHEM. SOC. VOL. 127, NO. 34, 2005

Heck Reaction Promoted by a CN−Palladacycle
A R T I C L E S
The plausible mechanism involves a preactivation step that
consists of the complete conversion of 1 into the catalytically
active Pd(0) species [1]. The main catalytic cycle is initiated
by oxidative addition of iodobenzene to [1], followed by the
reversible coordination of the olefin to the oxidative addition
product. All unimolecular subsequent steps (olefin insertion,
â-elimination, reductive elimination, and Heck product dis-
sociation) are indistinguishable kinetically and can be combined
in a single step. This kinetic model also predicts that a slight
excess of alkene relative to iodobenzene leads to a rapid rise in
the Pd(0) concentration while when using a slight excess of
iodobenzene, relative to alkene, the oxidative addition product
is the resting state of the catalytic cycle. The kinetic model
extended for a competitive Heck reaction involving two-different
aryl halides indicates that, whatever the rate-limiting step, in a
competitive Heck reaction, the changes in the relative reactivity
involving different aryl halides is strictly bound to the changes
in the rate constant of the oxidative addition step. The same F
value (2.4-2.5 for Ar-Br and 1.7-1.8 for Ar-I) observed in
competitive experiments catalyzed by 1 and CS, CP, and NCN
palladacycles suggests that the same active species is involved
in the oxidative addition step. Moreover, the activation enthalpy
of ∆Hq ) 69 ( 3 kJ mol-1 obtained for iodobenzene is identical
to those reported for the oxidative addition of iodobenzene with
Pd(PPh3)4. The excellent fit of the slope with the σ0 Hammett
parameter and entropy of activation of -43 ( 8 J mol-1 K-1 is
consistent with an associative process involving the development
of a partial charge in the transition state (no formal charge) for
the oxidative step.
Acknowledgment. We thank CNPq, CAPES, and CT-PETRO
for partial financial support and a scholarship to C.S.C. We also
thank Dr. G. Machado for the TEM analysis and Prof. Dr. G.
Eberlin for help in the preparation of DCT.
Supporting Information Available: Reaction orders deter-
mined by the initial rate method, deduction of rate equations,
script used in Dinafit kinetic simulations, typical ATR calibra-
tion curve, and Hammett plots obtained with different substrates
relative concentrations. This material is available free of charge
via the Internet at http://pubs.acs.org.
JA051834Q
9
J. AM. CHEM. SOC. VOL. 127, NO. 34, 2005 12065