Experimental and theoretical investigations of new dinuclear palladium complexes as precatalysts for the amination of aryl chlorides

Article abstract of DOI:10.1021/ja057825z

A series of new palladium dinuclear species with general formula [Pd ₂X(μ-X){μ- P^tBu₂(Bph-R)}] (X = Cl, Br; Bph = biphenyl; R = H, Me, NMe₂) have been prepared. The two palladium centers in these species are bridged by one of the aromatic rings of the biphenyl group present in the corresponding phosphine. The X-ray crystal structure of one of these complexes has been obtained, providing a clear picture of the bonding pattern. The stability of these dimers in solution is shown to be highly dependent on the nature of the phosphine R group and also on the bridging halide. When R = NMe₂, the dimers dissociate, yielding the palladium(II) compounds PdX₂{P^tBu₂(BPh-NMe ₂)} (X = Cl, Br), and the X-ray crystal structure of one of them (X = Br) has shown that the biphenyl group from the phosphine interacts directly with the metal center. This interaction seems to play an important role in stabilizing the otherwise coordinatively unsaturated palladium(II) complex. In contrast, when R = H or Me, the analogous monomeric palladium(II) complexes are unstable and undergo cyclometalation to generate a palladium(II) dinuclear species in which each of the two phosphines cyclometalates with the palladium centers forming a strained four-membered ring. In addition to their unusual structures, these aryl-bridged dimers have also proven to be excellent precatalysts for the amination of aryl chlorides. To rationalize some of the experimental results, a detailed DFT computational study has been carried out and is presented herein.

Full text of DOI:10.1021/ja057825z

Published on Web 04/15/2006
Experimental and Theoretical Investigations of New Dinuclear
Palladium Complexes as Precatalysts for the Amination of
Aryl Chlorides
Ute Christmann,†,§ Dimitrios A. Pantazis,‡ Jordi Benet-Buchholz,†
John E. McGrady,# Feliu Maseras,†,£ and Ramo´n Vilar*,§
Contribution from the Department of Chemistry, Imperial College London,
South Kensington, London SW7 2AZ, U.K., Institute of Chemical Research of Catalonia (ICIQ),
43007 Tarragona, Spain, Unitat de Qu´ımica F´ısica, Edifici Cn, UniVersitat Auto`noma de
Barcelona, 08193 Bellaterra, Spain, Department of Chemistry, UniVersity of York,
Heslington, York YO10 5DD, U.K., and WestCHEM, Department of Chemistry,
UniVersity of Glasgow, G12 8QQ, U.K.
Received November 17, 2005; E-mail: r.vilar@imperial.ac.uk
Abstract: A series of new palladium dinuclear species with general formula [Pd2X(µ-X){µ- PtBu2(Bph-R)}]
(X ) Cl, Br; Bph ) biphenyl; R ) H, Me, NMe2) have been prepared. The two palladium centers in these
species are bridged by one of the aromatic rings of the biphenyl group present in the corresponding
phosphine. The X-ray crystal structure of one of these complexes has been obtained, providing a clear
picture of the bonding pattern. The stability of these dimers in solution is shown to be highly dependent on
the nature of the phosphine R group and also on the bridging halide. When R ) NMe2, the dimers dissociate,
yielding the palladium(II) compounds PdX2{PtBu2(BPh-NMe2)} (X ) Cl, Br), and the X-ray crystal structure
of one of them (X ) Br) has shown that the biphenyl group from the phosphine interacts directly with the
metal center. This interaction seems to play an important role in stabilizing the otherwise coordinatively
unsaturated palladium(II) complex. In contrast, when R ) H or Me, the analogous monomeric palladium(II)
complexes are unstable and undergo cyclometalation to generate a palladium(II) dinuclear species in which
each of the two phosphines cyclometalates with the palladium centers forming a strained four-membered
ring. In addition to their unusual structures, these aryl-bridged dimers have also proven to be excellent
precatalysts for the amination of aryl chlorides. To rationalize some of the experimental results, a detailed
DFT computational study has been carried out and is presented herein.
to a palladium(0) center (usually stabilized by bulky ligands)12-16
is followed by a transmetalation reaction to yield a palladium-
(II) complex that contains the two moieties to be coupled. The
final step is then the reductive elimination of the product with
the concomitant regeneration of the active palladium(0) catalyst.
Introduction
Palladium-mediated cross-coupling reactions to form C-C,
C-N, C-O, and C-S bonds are important organometallic
transformations in organic synthesis.1-11 These reactions are
generally proposed to proceed through a mechanism that
involves three distinct steps. Oxidative addition of an aryl halide
Although the original palladium catalysts for cross-coupling
reactions required reactive aryl iodides and bromides as
substrates, the past few years have seen important advances in
the development of more efficient catalytic systems that can
also activate aryl chlorides17-25 (which are more readily
§ Imperial College London.
† Institute of Chemical Research of Catalonia.
£ Universitat Auto`noma de Barcelona.
‡ University of York.
# University of Glasgow.
(1) Hartwig, J. F. Angew. Chem., Int. Ed. 1998, 37, 2046-2067.
(2) Tykwinski, R. R. Angew. Chem., Int. Ed. 2003, 42, 1566-1568.
(3) Wolfe, J. P.; Wagaw, S.; Marcoux, J.-F.; Buchwald, S. L. Acc. Chem. Res.
1998, 31, 805-818.
(12) Hartwig, J. F. Synlett 1997, 329-340.
(13) Hartwig, J. F. Acc. Chem. Res. 1998, 31, 852-860.
(14) Stambuli, J. P.; Incarvito, C. D.; Buehl, M.; Hartwig, J. F. J. Am. Chem.
Soc. 2004, 126, 1184-1194.
(4) Miyaura, N.; Suzuki, A. Chem. ReV. 1995, 95, 2457-2483.
(5) Cardenas, D. J. Angew. Chem., Int. Ed. 2003, 42, 384-387.
(6) Diederich, F., Stang, P. J., Eds. Metal-Catalyzed Cross-Coupling Reactions;
Wiley: New York, 1998; Vol. p, 517 pp.
(15) Stambuli, J. P.; Buehl, M.; Hartwig, J. F. J. Am. Chem. Soc. 2002, 124,
9346-9347.
(16) Aranyos, A.; Old, D. W.; Kiyomori, A.; Wolfe, J. P.; Sadighi, J. P.;
Buchwald, S. L. J. Am. Chem. Soc. 1999, 121, 4369-4378.
(17) Navarro, O.; Kelly, R. A., III; Nolan, S. P. J. Am. Chem. Soc. 2003, 125,
16194-16195.
(7) Hiyama, T.; Shirakawa, E. Handbook of Organopalladium Chemistry for
Organic Synthesis; Wiley: New York, 2002; Vol. 1, pp 285-309.
(8) Echavarren, A. M.; Cardenas, D. J. Metal-Catalyzed Cross-Coupling
Reactions, 2nd ed.; Wiley-VCH: Weinheim, Germany, 2004; Vol. 1, pp
1-40.
(18) Littke, A. F.; Fu, G. C. Angew. Chem., Int. Ed. 2002, 41, 4176-4211.
(19) Bedford, R. B.; Cazin, C. S. J.; Hazelwood, S. L. Angew. Chem., Int. Ed.
2002, 41, 4120-4122.
(9) Espinet, P.; Echavarren, A. M. Angew. Chem., Int. Ed. 2004, 43, 4704-
4734.
(20) Bedford, R. B.; Hazelwood, S. L.; Limmert, M. E. Chem. Commun. 2002,
2610-2611.
(10) Beller, M.; Zapf, A. Handbook of Organopalladium Chemistry for Organic
Synthesis; Wiley: New York, 2002; Vol. 1, pp 1209-1222.
(11) Frisch, A. C.; Beller, M. Angew. Chem., Int. Ed. 2005, 44, 674-688.
(21) Bedford, R. B.; Welch, S. L. Chem. Commun. 2001, 129-130.
(22) Muci, A. R.; Buchwald, S. L. Top. Curr. Chem. 2002, 219, 131-209.
9
6376
J. AM. CHEM. SOC. 2006, 128, 6376-6390
10.1021/ja057825z CCC: $33.50 © 2006 American Chemical Society

Dinuclear Pd Precatalysts for Amination of Aryl Chlorides
A R T I C L E S
Chart 1
available and usually cheaper than their bromide or iodide
counterparts). For the less reactive aryl chlorides, the oxidative
addition of the substrate is usually the rate-limiting step in the
catalytic cycle.26 Consequently, the most recent generation of
catalysts is based on palladium complexes containing electron-
rich ligands (phosphines or N-heterocyclic carbenes) which
facilitate the oxidative addition step. It has been suggested that
highly unsaturated [Pd0L] species (especially when L is a bulky
ligand) are generated in the reaction medium, and that these
12-electron species are responsible for the enhanced activity of
these catalytic systems.12,13,27-30 In most cases, standard pal-
ladium sources, such as Pd2(dba)3 and Pd(OAc)2, have been
employed to generate the catalytic species in the presence of
the corresponding phosphine ligands. More recently, Hartwig31,32
and others33 have shown that the palladium(I) dimer [Pd2(µ-
Br)2(PtBu3)2], previously synthesized by us,34,35 can be used as
a novel single-source precatalyst for cross-coupling reactions.
High rates for the amination and Suzuki-Miyaura cross-
coupling of aryl chlorides and bromides31 were observed when
this complex was used as precatalyst. The high reactivity of
this compound has been attributed to the in situ formation of
the monoligated complex [Pd(PtBu3)] either by a dispropor-
tionation reaction or by direct reduction of the palladium dimer
in the presence of substrate and base.
of these systems has been attributed to a combination of
electronic and steric properties of the ligands that favor both
the oxidative addition and reductive elimination steps in the
catalytic cycle.
In a preliminary communication,49 we reported the formation
of the monophosphine palladium(0) complex Pd(dba){PtBu2-
(Bph-H)} (1) (where Bph-H ) biphenyl; see Chart 1) which,
upon addition of Pd(COD)X2 (X ) Cl, Br), reacts to form the
novel dinuclear palladium complexes Pd2(µ-X)X{µ-PtBu2-
(Bph-H)} (2, X ) Cl; 3, X ) Br) (see Scheme 1).
Scheme 1
Since the reactivity of palladium(I) dimers is highly dependent
on the nature of the coordinated phosphines, the synthesis of a
series of dipalladium complexes with phosphines (or other
ligands) with a range of different electronic and steric properties
is an attractive target. As part of our systematic efforts to prepare
novel dinuclear palladium compounds with better catalytic
properties for cross-coupling reactions, we have recently turned
our attention to sterically demanding biaryl phosphines. Buch-
wald has shown that several of these phosphines are excellent
supporting ligands for the palladium-catalyzed formation of
C-C,36-39 C-N,40-45 and C-O22,46-48 bonds using aryl chlo-
rides, bromides, and sulfonates as substrates. The effectiveness
Besides their unusual structural features, these complexes
generate catalytically active species for the amination of aryl
chlorides and bromides at room temperature. To investigate the
synthesis, reactivity, and catalytic properties of this new type
of dipalladium compounds in depth, we have carried out the
systematic study presented herein. More specifically, we
describe the products resulting from the reaction between Pd2-
(dba)3, the biphenyl phosphines PtBu2(Bph-R) (R ) H, Me,
NMe2; see Chart 1), and Pd(COD)X2 (X ) Cl, Br) and use
density functional theory to probe the nature of the metal-ligand
interaction. Critically, we show that the nature of the phosphine
has an important effect on the stability and reactivity of the
products and also on their effectiveness as precatalysts for the
room temperature amination of aryl chlorides.
(23) Zapf, A.; Ehrentraut, A.; Beller, M. Angew. Chem., Int. Ed. 2000, 39, 4153-
4155.
(24) Zapf, A.; Beller, M. Chem. Commun. 2005, 431-440.
(25) Grasa, G. A.; Viciu, M. S.; Huang, J.; Nolan, S. P. J. Org. Chem. 2001,
66, 7729-7737.
(26) Galardon, E.; Ramdeehul, S.; Brown, J. M.; Cowley, A.; Hii, K. K.; Jutand,
A. Angew. Chem., Int. Ed. 2002, 41, 1760-1763.
(27) Widenhoefer, R. A.; Buchwald, S. L. Organometallics 1996, 15, 3534-
3542.
(28) Widenhoefer, R. A.; Buchwald, S. L. Organometallics 1996, 15, 2755-
2763.
Results and Discussion
(29) Christmann, U.; Vilar, R. Angew. Chem., Int. Ed. 2005, 44, 366-374.
(30) Paul, F.; Patt, J.; Hartwig, J. F. J. Am. Chem. Soc. 1994, 116, 5969-5670.
(31) Stambuli, J. P.; Kuwano, R.; Hartwig, J. F. Angew. Chem., Int. Ed. 2002,
41, 4746-4748.
Synthesis of the Dinuclear Palladium Complexes Pd2(µ-
X)X{µ-PtBu2(Bph-R)} (R ) H, Me, NMe2; X ) Cl, Br). On
the basis of the recently reported49 synthesis of complexes 2
and 3 (see Scheme 1), the analogous methyl-substituted di-
nuclear compounds Pd2(µ-X)X{µ-PtBu2(Bph-Me)} (X ) Cl, 4;
Br, 5) were successfully prepared from the reaction between
Pd2(dba)3 and 2 equiv of PtBu2(Bph-Me) followed by addition
of Pd(COD)X2 (X ) Cl, Br) (see Scheme 2).
(32) Hama, T.; Liu, X.; Culkin, D. A.; Hartwig, J. F. J. Am. Chem. Soc. 2003,
125, 11176-11177.
(33) Prashad, M.; Mak, X. Y.; Liu, Y.; Repic, O. J. Org. Chem. 2003, 68, 1163-
1164.
(34) Vilar, R.; Mingos, D. M. P.; Cardin, C. J. J. Chem. Soc., Dalton Trans.
1996, 4313-4314.
(35) Dura-Vila, V.; Mingos, D. M. P.; Vilar, R.; White, A. J. P.; Williams, D.
J. J. Organomet. Chem. 2000, 600, 198-205.
(36) Milne, J. E.; Buchwald, S. L. J. Am. Chem. Soc. 2004, 126, 13028-13032.
(37) Barder, T. E.; Buchwald, S. L. Org. Lett. 2004, 6, 2649-2652.
(38) Walker, S. D.; Barder, T. E.; Martinelli, J. R.; Buchwald, S. L. Angew.
Chem., Int. Ed. 2004, 43, 1871-1876.
(44) Strieter, E. R.; Blackmond, D. G.; Buchwald, S. L. J. Am. Chem. Soc.
2003, 125, 13978-13980.
(39) Nguyen, H. N.; Huang, X.; Buchwald, S. L. J. Am. Chem. Soc. 2003, 125,
11818-11819.
(45) Zim, D.; Buchwald, S. L. Org. Lett. 2003, 5, 2413-2415.
(46) Vorogushin, A. V.; Huang, X.; Buchwald, S. L. J. Am. Chem. Soc. 2005,
127, 8146-8149.
(40) Charles, M. D.; Schultz, P.; Buchwald, S. L. Org. Lett. 2005, 7, 3965-
3968.
(41) Jiang, L.; Buchwald, S. L. Metal-Catalyzed Cross-Coupling Reactions, 2nd
ed.; Wiley-VCH: Weinheim, Germany, 2004; Vol. 2, pp 699-760.
(42) Anderson, K. W.; Mendez-Perez, M.; Priego, J.; Buchwald, S. L. J. Org.
Chem. 2003, 68, 9563-9573.
(47) Hamada, T.; Chieffi, A.; Ahman, J.; Buchwald, S. L. J. Am. Chem. Soc.
2002, 124, 1261-1268.
(48) Kuwabe, S.-i.; Torraca, K. E.; Buchwald, S. L. J. Am. Chem. Soc. 2001,
123, 12202-12206.
(43) Huang, X.; Anderson, K. W.; Zim, D.; Jiang, L.; Klapars, A.; Buchwald,
S. L. J. Am. Chem. Soc. 2003, 125, 6653-6655.
(49) Christmann, U.; Vilar, R.; White, A. J. P.; Williams, D. J. Chem. Commun.
2004, 1294-1295.
9
J. AM. CHEM. SOC. VOL. 128, NO. 19, 2006 6377

A R T I C L E S
Christmann et al.
1
Figure 1. A section of the 13C-1H correlation spectrum for 2 showing the cross-peaks for the 13C and H resonances of the biphenyl moiety; the three
cross-peaks of the complexed phenyl ring are all shifted to higher field (500 MHz, CD2Cl2).
Scheme 2
All the hydrogen and carbon atoms associated with the
coordinated aromatic ring (C7-C12) are considerably shifted
relative to the free phosphine and also to the shifts of the atoms
in the aromatic ring (C1-C6) that is not bound directly to the
palladium centers. Clearly, the strong interaction between the
ring (C7-C12) and the Pd-Pd unit shown previously by an
X-ray crystal structure is retained in solution in all four dimers,
2-5.
The dipalladium complex Pd2(µ-Cl)Cl{PtBu2(Bph-NMe2)} (6)
The products were isolated and characterized on the basis of
spectroscopic and analytical techniques. The dinuclear structure
of 4 and 5 was supported by elemental analyses and also by
ESI mass spectroscopy, which show peaks corresponding to
[M - X]+ (X ) Cl, Br). The 31P{1H} NMR spectra of these
complexes show only one singlet each (at 62.1 ppm for 4 and
67.9 ppm for 5) which is consistent with the proposed
was prepared using an analogous procedure to the one described
above. The reaction mixture containing Pd2(dba)3 and PtBu2-
(Bph-NMe2) was stirred for 48 h (until no further free phosphine
was observed in the 31P{1H} NMR spectrum). At this point,
solid Pd(COD)Cl2 was added, causing a darkening of the
solution, and after approximately 1 h a dark precipitate started
to form. The reaction mixture was stirred for 2 further hours,
after which time it was filtered and crystals of 6 suitable for
X-ray crystallography (and spectroscopic characterization) were
obtained from the filtrate. The synthesis of the bromide analogue
of 6, Pd(µ-Br)Br{µ-PtBu2(Bph-NMe2)} (7), by a similar protocol
was, in contrast, unsuccessful, the reaction instead leading to a
monometallic species that will be discussed in a subsequent
section.
1
formulation. The H and 13C{1H} NMR spectra of the two
complexes are particularly indicative of the presence of a
bridging phenyl ring from the biphenyl group since all the
aromatic signals corresponding to the ring are shifted to higher
field. Analogous shifts have been previously observed in the
spectra of the dimeric species 2 and 3. To assign unambiguously
all the resonances, 2D NMR studies were carried out for
compounds 2-5. In Figures 1 and 2, the HSQC (13C-1H
correlation) spectra of 2 (the synthesis and X-ray crystal structure
of which we have previously reported49) and 4 are shown as
examples.
The X-ray crystallographic study of 6 confirmed it to be a
dipalladium complex where the metal centers are bridged by a
chloride and by the biphenyl group of PtBu2(Bph-NMe2). The
9
6378 J. AM. CHEM. SOC. VOL. 128, NO. 19, 2006

Dinuclear Pd Precatalysts for Amination of Aryl Chlorides
A R T I C L E S
1
Figure 2. A section of the 13C-1H correlation spectrum for 4 showing the cross-peaks for the 13C and H resonances of the biphenyl moiety; the four
cross-peaks of the complexed phenyl ring are all shifted to higher field (500 MHz, CD2Cl2).
Pd1-C10, Pd1-C11, Pd2-C8, Pd2-C7, and Pd2-C12) are
relatively short, but the plane of the coordinated aromatic ring
is not orthogonal to the Pd2Cl plane. In fact, the contacts from
the palladium atoms to the carbon atoms at the aromatic ring
are shorter on the side of the NMe2 group (Pd1-C11 ) 2.471(3)
Å vs Pd1-C9 ) 2.617(5) Å and Pd2-C12 ) 2.451(3) Å vs
Pd2-C8 ) 2.613(4) Å). Moreover, the aromatic ring directly
linked to the phosphorus atom does not lie in the Pd2Cl plane,
but is tilted by 10.5° away from the NMe2 substituent, possibly
due to the steric effects of the methyl group.
The solid-state structure of 6 is consistent with the charac-
terization of the bulk sample in solution. Its 31P{1H} NMR
spectrum shows a singlet at 63.6 ppm consistent with the single
phosphorus environment present in the complex. The 1H NMR
spectrum of a freshly prepared sample of 6 in CD2Cl2 shows
again that four of the aromatic protons are considerably shifted
Figure 3. Ortep plot (thermal ellipsoids shown at 50% probability level)
of 6. Hydrogen atoms have been omitted for the sake of clarity. Selected
to higher field (δH ) 6.50-5.87 ppm) in comparison to the
resonances observed for the free phosphine. As was the case
for 2-5, this is a clear indication of the coordination of one of
the biphenyl aromatic rings to the Pd-Pd unit. Although the
compound was fully characterized both in solution and in the
solid state, it should be pointed out that if it is left in solution
for more than ca. 2 h a second singlet in the 31P{1H} NMR
spectrum appears (at 76.9 ppm), indicating the conversion of
the dinuclear compound into a new species (see below for
discussion).
distances (Å): Pd1-Pd2 2.5329(3); Pd1-Cl1 2.3859(8); Pd1-Cl2 2.4391(8);
Pd2-Cl2 2.3488(3); Pd2-P1 2.2663(7); Pd1-C9 2.617(5); Pd1-C10
2.099(3); Pd1-C11 2.471(3); Pd2-C8 2.613(4); Pd2-C7 2.136(2); Pd2-
C(12) 2.451(3).
molecular structure of 6 (Figure 3) is similar to those of 2 and
3 reported in our preliminary communication, where we
described the coordination of the aromatic ring as η3:η3.49 The
coordinated aromatic ring contains a NMe2 group which lowers
the symmetry of the system, making the molecule chiral. Atom
C7 has the characteristic pyramidalization detected in the
structures of the analogous compounds 2 and 3, with an angle
of 19.5° between the plane of the aromatic ring and the exocyclic
bond at C7. All six contacts to the aromatic ring (Pd1-C9,
In our preliminary communication, we described the coor-
dination of the arene ring in 2 and 3 as η3:η3, a relatively
uncommon but not unprecedented structural motif.50-54 How-
9
J. AM. CHEM. SOC. VOL. 128, NO. 19, 2006 6379

A R T I C L E S
Christmann et al.
Figure 4. Computed structure for complex 3m (left) and qualitative description of the bonding of the bridging phenyl to the palladium atoms (right).
ever, the subtle variations in Pd-C bond lengths described above
have prompted us to explore the bonding in these systems in
greater detail using density functional theory.55 The model
system chosen for the computational studies was Pd2(µ-X)X-
{µ-PH2(Bph-R)}, where the two tert-butyl groups on each
phosphine were replaced by hydrogen atoms. The structures of
all six possible combinations of X (Br, Cl) and R (H, Me, NMe2)
groups (2m-7m) were optimized in the gas phase. All opti-
mized geometries share the same principal features, and only a
representative example, 3m, Pd2(µ-Br)Br{µ-PH2(Bph-H)}, is
shown in Figure 4, which also introduces the atom numbering
that will be used in the subsequent discussion.
suggests significant double bond character and a partial breaking
of the aromaticity of the phenyl ring as a result of its interaction
with the palladium centers. In isolation, the loss of aromaticity
would, of course, also lead to a contraction of C7-C12 and
C11-C10, but these two are the bonds interacting most strongly
with the palladium atoms, and back-donation from the metals
masks the expected lengthening. We therefore suggest that the
most appropriate description for the bonding in compound 3m,
[Pd2Br(µ-Br){µ-PH2(Bph-H)}], is the µ-η2:η2 mode presented
in the right-hand side of Figure 4, with η2 interactions between
Pd2, C7 and C12 and Pd1, C11 and C10, giving the formal
16-electron count expected for an approximately square planar
coordination sphere.
The main features of the computed geometry of 3m, most
notably the Pd-Pd (2.578 Å cf. experimental value of 2.563
Å) and Pd-(µ-Br) distances (2.485 and 2.508 Å cf. experiment
values of 2.462, 2.563 Å), are in reasonable agreement with
the X-ray structure of 3. There are, however, some differences
related to the arrangement of the bridging phenyl group which
we consider significant. As in the X-ray crystal structure, the
plane of the phenyl group is almost perfectly parallel to the
Pd-Pd axis, with computed Pd2-C7 and Pd1-C10 distances
of 2.165 and 2.148 Å, respectively. This phenyl group plane is
not, however, perpendicular to the Pd-Br-Pd plane, but instead
is quite tilted, with C12 and C11 much closer to their respective
palladium centers (Pd2-C12 ) 2.371 Å; Pd1-C11 ) 2.252
Å) than C9 and C8 (Pd2-C8 ) 2.744 Å; Pd1-C9 ) 2.790 Å).
The differences in Pd-C distances of 0.373 and 0.538 Å are
significantly larger than those in the X-ray structure (0.297 and
0.102 Å), suggesting that the degree of tilt is overestimated in
the calculation. The long computed Pd1-C9 and Pd2-C8
distances (>2.7 Å) have led us to re-examine the η3:η3
description proposed previously. The trends in C-C distances
are also interesting in this regard: C9-C8 is only 1.371 Å,
which is the shortest C-C distance in the phenyl ring by more
than 0.04 Å, the other five distances being 1.428, 1.416, 1.421,
1.436, and 1.450 Å for C7-C12, C12-C11, C11-C10, C10-
C9, and C7-C8, respectively. The short C9-C8 distance
It is possible that the discrepancy between crystallographic
and computed structures could be an artifact associated with
the replacement of the phosphine tert-butyl groups by less bulky
and less electron-donating hydrogen atoms in our computational
model. We have checked this possibility by carrying out two
additional calculations designed to probe the electronic and steric
influence of the phosphine substituents. To introduce the
electronic effects of the substituents, we carried out a geometry
optimization on the model system 3m′, Pd2(µ-Br)Br{µ-PMe2-
(Bph-H}, where the tert-butyl groups have been replaced by
methyl groups instead of hydrogens. The resulting geometry
was very similar to that obtained for 3m; in particular, the tilt
of the phenyl ring was conserved, with two short (Pd2-C12,
Pd1-C11, 2.359 and 2.261 Å, respectively) and two long (Pd2-
C8, Pd1-C9, 2.738 and 2.787 Å, respectively) Pd-C distances.
The steric influence of the phosphine substituent was then
modeled through an ONIOM calculation [ONIOM(mPW1PW91:
UFF)] on the real system 3, with the tert-butyl substituents
included in the MM region. Results were again similar to those
of the calculation on 3m, with Pd2-C12, Pd1-C11 distances
shorter than 2.4 Å and Pd2-C8, Pd1-C9 distances longer than
2.75 Å. These additional calculations convince us that the tilting
of the phenyl group is not a computational artifact, and the fact
that the tilting is much less exaggerated in the X-ray structure
(leading to our previous η3:η3 description) is most probably due
to disorder between the structure shown in Figure 4 and its
enantiomer, where the C8-C9 bond is closer to the metal. The
barrier separating the enantiomers is very small, and rapid
interconversion between the two in solution would mean that
the asymmetric coordination of the arene ring is not apparent
in the NMR data.
(50) Allegra, G.; Tettamanti Casagrande, G.; Immirzi, A.; Porri, L.; Vitulli, G.
J. Am. Chem. Soc. 1970, 92, 289-293.
(51) Kannan, S.; James, A. J.; Sharp, P. R. J. Am. Chem. Soc. 1998, 120, 215-
216.
(52) Dupont, J.; Pfeffer, M.; Rotteveel, M. A.; De Cian, A.; Fischer, J.
Organometallics 1989, 8, 1116-1118.
(53) Sommovigo, M.; Pasquali, M.; Leoni, P.; Braga, D.; Sabatino, P. Chem.
Ber. 1991, 124, 97-99.
(54) Retboll, M.; Edwards, A. J.; Rae, A. D.; Willis, A. C.; Bennett, M. A.;
Wenger, E. J. Am. Chem. Soc. 2002, 124, 8348-8360.
(55) Computational studies on related systems have been recently reported:
Barder, T. E.; Walker, S. D.; Martinelli, J. R.; Buchwald, S. L. J. Am.
Chem. Soc. 2005, 127, 4685-4696.
As noted above, the computed structures for 2m, 4m, 5m,
6m, and 7m are very similar to those of 3m, and their structures
9
6380 J. AM. CHEM. SOC. VOL. 128, NO. 19, 2006

Dinuclear Pd Precatalysts for Amination of Aryl Chlorides
A R T I C L E S
and energies are collected in the Supporting Information. The
only qualitative difference with the discussion above arises from
the presence of the Me or NMe2 substituents which leads to
two distinct minima, one where the uncoordinated double bond
carries the substituent, the other where it does not. For 5m
(X ) Br and R ) Me), these two local minima, 5m-a and 5m-
b, are only 0.8 kcal/mol apart. The situation with R ) NMe2 is
even more complex because different orientations of the lone
pair on the nitrogen are possible. In fact, for system 6m (X )
Cl and R ) NMe2), three distinct local minima were located,
all of which lie within 3.5 kcal/mol of each other. The most
stable of the three is not the one that most closely resembles
the X-ray structure for 6, but given the small energy differences
involved, the solid-state structure is probably strongly influenced
by crystal packing effects. The conformation of the absolute
minimum is, however, probably of little chemical relevance
because the different isomers are likely to be involved in fast
equilibria.
Dissociation and Rearrangement of the Dinuclear Pal-
ladium Complexes Pd2(µ-X)X{µ-PtBu2(Bph-R)} (R ) H, Me,
NMe2; X ) Cl, Br) in Solution. We noted in the previous
section that attempted synthesis of the bromide compound Pd-
(µ-Br)Br{µ-PtBu2(Bph-NMe2)} (7) was unsuccessful. In fact,
when a mixture of Pd2(dba)3 and PtBu2(Bph-NMe2) (which had
been stirred for 48 h) was reacted with Pd(COD)Br2, the
formation of a dark solid was observed within 1 h (along with
some palladium metal formed as a mirror in the reaction vessel).
The crude solid was isolated by filtration, and a dark blue
compound was extracted using dichloromethane. After filtering
this solution, the dichloromethane was evaporated under reduced
pressure to give a dark blue solid, which was further purified
by column chromatography to yield a product that was
formulated, based on spectroscopic and structural characteriza-
tion, as the new palladium(II) compound PdBr2{PtBu2(BPh-
NMe2)} (8) (see Scheme 3).
Figure 5. Ortep plot (thermal ellipsoids shown at 50% probability level)
of 8. Hydrogen atoms have been omitted for the sake of clarity. Selected
distances (Å) and angles (deg): Pd1-P1 2.2886(6); Pd1-C7 2.191(3);
Pd1-Br1 2.4870(3); Pd1-Br2 2.4728(4); C1-C2 1.393(4); C2-C7
1.531(4); C7-C8 1.469(6); C7-C12 1.435(4); P1-Pd1-C7 85.52(9); Pd1-
C7-C2 109.39(19); C1-C2-C7 120.8(2); P1-C1-C2 116.31(17); Br1-
Pd1-Br2 86.606(12); C2-C7-C10 134.
also an elongation of the aromatic bonds C7-C8 and C7-C12
to 1.469(6) and 1.435(4) Å, respectively. The single bond
C7(sp2)-C2(sp2) is also significantly elongated to 1.531(4) Å.
As a consequence of the strain in the system, C7 shows a
deviation of approximately 0.11 Å from the mirror plane that
defines the Cs pseudo-symmetry. The spectroscopic character-
ization in solution of a bulk sample of 8 is consistent with the
structure found by X-ray crystallography.
The structure of 8 was confirmed by X-ray single-crystal
structural analysis (see Figure 5). The compound crystallizes
in the Pnma space group with a Cs pseudo-symmetry plane
defined by the approximately square planar coordination sphere
around the palladium(II) center. The tert-butyl groups and the
aromatic ring containing the NMe2 group are disordered on both
sides of this plane. A structure refinement at lower symmetries
was also tested (Pna21), leading to similar R values and a strong
correlation at the refinement of the atoms. The most significant
feature of this structure is the interaction between the palladium
center and the biphenyl ipso carbon (C7). The presence of this
type of Pd-Cipso bond in palladium(II) complexes is unusual
but not unknown (see for example the work by Kocovsky,56
and by Campora and Carmona57,58). The Pd-Cipso bond length
of 2.191(3) Å is similar to that reported by Kocovsky (2.187
Å)56 but shorter than those described previously by Carmona
and Campora (2.343(6) and 2.39(1) Å).57,58 It is likely that this
short Pd-Cipso distance is playing an important role in stabilizing
the otherwise coordinatively unsaturated complex. As expected,
the Pd-Cipso interaction causes a pyramidalization at C7 and
We also noted in the previous section that the dipalladium
compound 6 is prone to decomposition in solution, and the
elucidation of the structure of 8 led us to speculate that the
decomposition product of 6 could be its chloride analogue.
Consequently, a sample of 6 was dissolved in CD2Cl2, and the
solution was monitored by 31P{1H} NMR spectroscopy for 24
h. After approximately 2 h, a new resonance appeared in the
spectrum of this solution (at δP ) 76.9 ppm), while the
resonance associated with the original dipalladium complex 6
(δP ) 63.6 ppm) slowly decreased in intensity. After 12 h, the
31P{1H} NMR spectrum indicated that 6 had completely
converted into the new product with δP ) 76.9 ppm. No further
changes were observed after an additional 24 h of stirring. The
compound associated with the δP ) 76.9 ppm signal was
isolated and, on the basis of spectroscopic characterization,
formulated as the palladium(II) compound PdCl2{PtBu2(Bph-
NMe2)} (9) (see below for synthesis of the same complex via
a different route). Although no structural characterization of this
compound was carried out, we anticipate that a Pd-Cipso
interaction similar to that in 8 stabilizes the otherwise coordi-
natively unsaturated palladium(II) complex. The 13C{1H} NMR
spectrum of this compound is certainly consistent with the
presence of such an interaction since the chemical shift of the
(56) Kocovsky, P.; Vyskocil, S.; Cisarova, I.; Sejbal, J.; Tislerova, I.; Smrcina,
M.; Lloyd-Jones, G. C.; Stephen, S. C.; Butts, C. P.; Murray, M.; Langer,
V. J. Am. Chem. Soc. 1999, 121, 7714-7715.
(57) Campora, J.; Gutierrez-Puebla, E.; Lopez, J. A.; Monge, A.; Palma, P.;
Del Rio, D.; Carmona, E. Angew. Chem., Int. Ed. 2001, 40, 3641-3644.
(58) Campora, J.; Lopez, J. A.; Palma, P.; Valerga, P.; Spillner, E.; Carmona,
E. Angew. Chem., Int. Ed. 1999, 38, 147-151.
Cipso appears considerably shifted at δc ) 86.5 ppm (compared
to 117.4 ppm for the free phosphine).
9
J. AM. CHEM. SOC. VOL. 128, NO. 19, 2006 6381

A R T I C L E S
Christmann et al.
Figure 6. Computed structures for complexes 12m-a (left) and 12m-b (right).
Scheme 3
Once the decomposition path of 6 in solution was established,
it was of considerable interest to know whether the other
dinuclear compounds, 2-5, would behave in an analogous way
if left in solution over extended periods of time. When a sample
of 4 was dissolved in CD2Cl2, a new signal at 75.6 ppm started
to appear after approximately 10 h, while the resonance
associated with 4 (δP ) 62.1 ppm) decreased in intensity, and
palladium black was formed in the NMR tube. The chemical
shift of the new species (75.6 ppm) is in the same region of the
spectrum as those observed for 8 and 9, leading us to formulate
it as the palladium(II) complex PdCl2{PtBu2(Bph-Me)} (10).
To establish whether the palladium(II) compounds 8 and 9
could be prepared directly from a palladium(II) source and the
corresponding phosphine, the reactions between Pd(COD)X2
(X ) Cl, Br) and PtBu2(Bph-NMe2) were carried out. The
reactions were monitored by 31P{1H} NMR spectroscopy which
indicated that, in each case, the only product formed was indeed
the corresponding PdX2{µ-PtBu2(Bph-NMe2)} compound (X )
Br, 8; X ) Cl, 9). The products were isolated in high yields
and fully characterized, confirming that they could be obtained
either through decomposition of the dinuclear species or directly
from Pd(COD)X2 (see Scheme 3). The reactions between Pd-
(COD)Cl2 and PtBu2(Bph-Me) were also carried out, yielding
the expected PdCl2{µ-PtBu2(Bph-Me)} compound (10). How-
ever, as will be discussed in the following section, this complex
converted to yet another product when left in solution for longer
periods of time (ca. 12 h) (see following section for detailed
discussion).
The relatively unusual η1 coordination mode of the arene in
8 (and, presumably, also 9) raises questions concerning the
nature of the metal-arene interaction and the role of these
interactions in controlling the stability of potential intermediates
in the catalytic cycle. To compare the relative stability of this
type of compound with different combinations of halide and
phosphine, computational studies were carried out on the
compounds with all possible combinations of X and R groups
(although only the first three have been isolated experimentally).
The following model systems, PdX2(PH2Bph-R) 8m (X ) Br,
R ) NMe2), 9m (X ) Cl, R ) NMe2), 10m (X ) Br, R )
Me), 11m (X ) Cl, R ) Me), 12m (X ) Br, R ) H), and 13m
(X ) Cl, R ) H), were studied.
In the less stable of the two isomers, 12m-b, the coordination
of the arene ring is rather similar to that in the dinuclear species
3m, with two short Pd-C distances, Pd-C7 and Pd-C8, of
2.614 and 2.425 Å, respectively, consistent with an η2 coordina-
tion mode, albeit a highly distorted one. In contrast, there is
only one short Pd-C contact, Pd-C8 (2.415 Å), in the more
stable isomer 12m-a, the next nearest carbon being C12, 2.958
Å away. The highly asymmetric coordination mode in 12m-a
could be described as η1, although the rather short Pd-H
distance indicates some degree of C-H agostic character.
Consistent with this, the C8-H bond length of 1.098 Å is
significantly longer than the other four C-H bonds in the phenyl
ring, which all fall between 1.083 and 1.085 Å. Our previous
experience suggests that a lengthening of more than 0.01 Å in
such a strong bond is indicative of an agostic bond. Calculations
on 8m, 9m, 10m, 11m, and 13m all follow the same pattern,
with two distinct local minima present in each case, one C-H
agostic, the other featuring an η2-CdC coordination mode. It
is rather unusual for a C-H agostic bond to compete with an
η2-CdC interaction, but in this case, the chelating nature of
the ligand prevents the arene ring from adopting an orientation
that optimizes the latter. This is most clearly illustrated by a
further calculation on the model system PdBr2(PH3)(C6H6),
where the phenyl group linking the phosphorus center and the
coordinating arene ring has been removed. In this case, where
the benzene ring is completely unconstrained, it rotates by
approximately 30°, such that the C7-C8 vector is perpendicular
to the P-Pd-Br plane. The overall effect of the chelating ligand
is therefore to destabilize the η2-CdC interaction and cause a
distinct slippage toward a highly asymmetric η2, or even η1,
We first present the results for 12m (with X ) Br and R )
H), where the absence of a Me or NMe2 substituent simplifies
the discussion. Two different isomers of 12m were optimized,
12m-a and 12m-b, both of which are presented in Figure 6.
9
6382 J. AM. CHEM. SOC. VOL. 128, NO. 19, 2006

Dinuclear Pd Precatalysts for Amination of Aryl Chlorides
A R T I C L E S
Figure 7. Ortep plot (thermal ellipsoids shown at 50% probability level) of 14. Hydrogen atoms have been omitted for the sake of clarity. Selected distances
(Å) and angles (deg): Pd1-P1 2.2392(3); Pd1-C2 1.9868(12); Pd1-Cl1 2.4668(1); Pd1-Cl1A 2.4261(3); C1-C2 1.4047(15); P1-Pd1-C2 68.69(3);
P1-C1-C2 95.07(8); Pd1-P1-C1 86.29(4); Pd1-C2-C1 109.57(9); P1-Pd1-Cl1 106.792(11); C2-Pd1-Cl1A 99.02(3); Cl1-Pd1-Cl1A 85.463(11).
mode. Mealli59 and Barder60 have recently discussed similar
features in the bonding in constrained Pd-arene complexes.
of the aromatic group noted in the X-ray study. A comparison
of the computed structures with (8) and without (8m) the tert-
butyl groups clearly indicates that the unusual coordination mode
is again a result of a compromise between the steric bulk of the
phosphine substituents, favoring a symmetric orientation of the
arene ring and hence η1 coordination, and the intrinsic electronic
driving force favoring the η2 alternative. The fact that the
chelating nature of the ligand prevents the system from
optimizing the η2-CdC interaction makes the potential energy
surface for ring slippage very soft and hence allows steric factors
to play a dominant role.
The relatively weak interaction between the metal and the
arene ring in both 12m-a and 12m-b prompted us to reintroduce
the tert-butyl groups on phosphorus into our computational
model, as their steric influence may be of similar magnitude to
the strength of the metal-arene bond. This possibility is most
apparent in the structure of 12m-a, where the interaction
between Pd and the C8-H bond introduces a torsion about the
P-C bond and forces one of the hydrogen atoms on phosphorus
to point directly toward the hydrogen bound to the ortho carbon
of the phenyl group. We have therefore carried out ONIOM
calculations on the complete complex 12, PdBr2(PtBu2Bph-H),
incorporating the tert-butyl groups in the MM partition. Two
different isomers could again be located, 12-a and 12-b,
separated by only 0.4 kcal/mol, but both now feature relatively
symmetric η2-CdC coordination and differ only in the confor-
mational arrangement of the tert-butyl groups. The local minima
corresponding to agostic C-H species therefore disappear when
the bulky phosphine substituents are included, and ONIOM
calculations on systems 8-13 confirm this to be the case in all
the systems. As anticipated above, the disappearance of the
agostic minima can be traced to the steric clash between the
tert-butyl groups and the ortho C-H of the ring bound to
phosphorus. The bulky substituents on the phosphorus force the
ortho C-H group to bisect the C-P-C angle, rather than
pointing directly at one of the substituents on P, as was the
case in 12m-a. This, in turn, forces the remote arene ring to
adopt a more symmetric position with respect to the palladium
center and hence favors the distorted η2-CdC coordination
mode.
Synthesis of the Cyclopalladated 14. As noted in the
previous section, when a sample of 10 was left in solution for
more than 12 h, the 31P{1H} NMR spectrum indicated that it
was fully converted to yet another product, 14, with δP ) -13.1
ppm, which was isolated from the reaction mixture and fully
characterized via spectroscopic, structural, and analytical tech-
niques. It has previously been reported that cyclometalated
phosphines featuring four-membered rings show very different
31P NMR chemical shifts compared to their analogous noncy-
clometalated counterparts.61 On the basis of the unusually high
field resonance for 14, the complex was therefore formulated
as a cyclometalated dipalladium(II) compound (see Scheme 4)
in which intramolecular activation of an ortho C-H of the
PtBu2(Bph-Me) phosphine generates a strained four-member
ring.
The proposed structure of 14 was unambiguously established
by a single-crystal X-ray structural analysis, which reveals a
dipalladium(II) species bridged by two chlorides, each palladium
center being part of a cyclometalated four-membered ring
(Figure 7). The compound crystallizes as a chloroform solvate
with Ci molecular symmetry, with the phosphorus atoms located
in a trans position with respect to each other. The Pd1-C2
distance (1.9868(12) Å) is within the range of values found in
similar four-membered complexes with a sp2-hybridized carbon.
As a result of the strain produced by the unusual four-membered
ring (atoms Pd1-P1-C1-C2), the angle P1-Pd1-C2 is
reduced to 68.69(3)° (compared to average values of 67.8° in
similar monomeric systems with a four-membered azapallada-
cycle).62,63 The other endocyclic angles in the four-membered
Perhaps the most striking aspect of the ONIOM calculation
on complex 8 (X ) Br, R ) NMe2) is that the most stable
isomer, 8-a, features a highly asymmetric coordination mode,
which could be described either as η1 or as distorted η2, precisely
as is observed in the X-ray structure. The computed Pd-C7
distance is 2.439 Å, significantly shorter than the Pd-C12 and
Pd-C8 distances of 2.777 and the 2.974 Å. The calculated Pd-
C7 distance is rather longer than the X-ray value of 2.191 Å,
but the origin of the discrepancy may again lie in the disorder
(59) Catellani, M.; Mealli, C.; Motti, E.; Paoli, P.; Perez-Carreno, E.; Pregosin,
P. S. J. Am. Chem. Soc. 2002, 124, 4336-4346.
(60) Barder, T. E. J. Am. Chem. Soc. 2006, 128, 898-904.
(61) Garrou, P. E. Chem. ReV. 1981, 81, 229-266.
(62) Sole, D.; Vallverdu, L.; Solans, X.; Font-Bardia, M.; Bonjoch, J. Orga-
nometallics 2004, 23, 1438-1447.
9
J. AM. CHEM. SOC. VOL. 128, NO. 19, 2006 6383

A R T I C L E S
Scheme 4
Christmann et al.
ring are also smaller than expected (Pd1-P1-C1 ) 86.29(4)°;
P1-C1-C2 ) 95.07(8)°; and Pd1-C2-C1 ) 109.57(9)°). The
difference between the P1-Pd1-Cl1 (106.792(11)°) and C2-
Pd1-Cl1A (99.02(3)°) angles is probably due to the repulsive
steric effects of the bulky tert-butyl group. As the computational
studies reported below demonstrate, the steric bulk of these
groups plays a key role in driving the formation of the four-
membered ring rather than its six-membered analogue.
Once the conversion of 4 to the cyclometalated product 14
(via 10) was confirmed, the analogous process for 2, 3, and 5
was investigated by 31P{1H} NMR spectroscopy. In all three
cases, the biphenyl-bridged dipalladium compounds convert to
the corresponding cyclometalated products (all of which appear
at a chemical shift between -14.0 and -18.7 ppm) via a process
that involves initial dissociation of the dimer into species
analogous to the palladium(II) compound 8. The conversion
times are highly dependent on the coordinated halide and
phosphine and on the solvent used; all compounds dissociate
and cyclometalate more rapidly in CH2Cl2 than in THF. The
least stable of the four compounds proved to be Pd(µ-Br)Br-
{µ-PtBu2(Bph-H)} (3), which starts to dissociate in THF after
ca. 3.5 h and eventually converts to the corresponding four-
membered cyclometalated product. The other three compounds
remain stable in THF solution for at least 3 h.
which we might anticipate to be more stable, was also accessible
through activation of a C-H bond on the remote arene ring
(see Scheme 5). To probe the origin of the regioselectivity of
the cyclometalation reaction, we have surveyed the potential
energy surface for the model complex 14m, [Pd(µ-Cl)(PH2-
Bph)]2. In this dimeric species, the tert-butyl groups have again
been replaced by hydrogens, as has the methyl group on the
arene ring in 14. Geometry optimizations confirmed the presence
of two distinct isomers, one with four-membered rings, labeled
as 14m-4r, the other with two six-membered rings, labeled as
14m-6r. The geometry of 14m-4r is reasonably similar to the
X-ray structure of 14, with a diamond-shaped Pd(µ-Cl)2Pd core
and Pd-Cl distances of 2.463 and 2.394 Å (cf. the experimental
values of 2.4668(1) and 2.4261(3) Å). The alternative structure,
14m-6r, also features a diamond-shaped core, albeit rather more
puckered than that in 14m-4r. The most conspicuous feature
of the comparison of the two isomers is, however, their
energies: 14m-6r is 23.2 kcal/mol more stable than 14m-4r,
the experimentally observed isomer. In light of our previous
results, it seemed likely that the steric bulk of the tert-butyl
groups may play a significant role in determining the relative
stabilities of the products; in particular, the introduction of the
remote arene group into the metal coordination sphere in 14-6r
necessarily increases the steric crowding in this case. Steric
effects will be sensitive to the nature of the phosphine
substituents, so a new set of calculations with the ONIOM
method were carried out on system 14, where the four tert-
butyl substituents on the phosphines are introduced at the
molecular mechanics (UFF) level. We have located a number
of different local minima for these systems; the ones we have
selected for analysis are 14-4r and 14-6r, shown in Figure 8.
The geometries are similar to those from the corresponding 14m
model systems, but their relative energies are quite different;
the inclusion of the tert-butyl groups destabilizes 14-6r relative
to 14-4r, and the two are now almost isoenergetic, 14-4r lying
only 0.3 kcal/mol above 14-6r. It is thus evident that steric
effects strongly favor the structure with the four-membered
rings. The reason for this can be clearly seen in Figure 8, where
Scheme 5
The formation of the four-membered metallacycle in 14 was
rather unexpected because an alternative six-membered ring,
(63) Sole, D.; Vallverdu, L.; Solans, X.; Font-Bardia, M.; Bonjoch, J. J. Am.
Chem. Soc. 2003, 125, 1587-1594.
9
6384 J. AM. CHEM. SOC. VOL. 128, NO. 19, 2006

Dinuclear Pd Precatalysts for Amination of Aryl Chlorides
A R T I C L E S
Figure 8. Computed structures for complexes 14-4r (above) and 14-6r (below). Nonsterically active hydrogen atoms are omitted for clarity.
Table 1. Catalytic Amination of 4-Chlorotoluene with p-Toluidine
the remote arene ring in 14-4r is not constrained by coordination
to the metal. As a result, it lies perpendicular to the P-C vector
and avoids a destabilizing steric clash with the tert-butyl groups.
(see Scheme 6) Using Different Palladium Sourcesa
Conversion (%)b
at Different Times (h)
In contrast, in 14-6r, the remote arene ring is coordinated to
entry
precatalyst
[mol % Pd]
T
0.5
1
2
3
24
the metal and, as a result, the ortho C-H group of the ring
bonded to the phosphorus is forced to bisect the two tert-butyl
groups. The result is a much more dramatic steric clash, causing
the P-C bonds in 14-6r to elongatesthe average value is 1.895
Å compared to 1.872 Å in 14-4r, and a lengthening of 0.023 Å
is energetically significant for a strong single bond such as P-C.
The repulsive Cl1-H2 contact is also closer in 14-6r (ca. 2.80
Å) than in 14-4r (ca. 3.06 Å). Entropic effects further tilt the
balance in favor of 14-4r because, unlike 14-6r, it retains a
freely rotating C-C single bond. Explicit calculation of the
entropies of the two species suggests that this contributes a
further 4 kcal/mol in favor of 14-4r, bringing it below 14-6r
on the free-energy surface.
Catalytic Amination of Aryl Halides Using Dipalladium
Compounds 2-5 as Precatalysts. As stated in the Introduction,
we reported in a preliminary communication that 2 and 3 are
effective precatalysts for the amination of aryl bromides and
chlorides at room temperature. To study the catalytic properties
of these compounds in more detail, the model reaction shown
in Scheme 6 was carried out using different dipalladium
compounds as precatalysts. Since the dinuclear complex 6 is
not particularly stable in solution, we decided to study only
complexes 2-5 and to compare their catalytic properties with
the species generated by mixing Pd2(dba)3 with the correspond-
ing phosphine. As can be seen in Table 1, the four complexes
accelerate the reaction at room temperature and, particularly,
in the case of 4 and 5, in very short periods of time. To probe
the rate of catalysis, the reaction with the dinuclear dimer 4 (at
0.5 mol %) was monitored over time (Figure 9). More than
1
2
3
4
5
6
7
8
9
2
2
3
3
4
4
5
5
1
2
1
2
1
2
1
2
rt
rt
rt
rt
rt
rt
rt
rt
rt
21 38 45
50 66 86
29 48 63
59 70 86
82 92 96
75 100
93 100
75 99
91 100
99 100
91 96 99 100
92 93 97
95 98 100
99 100
Pd2dba3/PtBu2Bph
1
0.5
1
4c
7
4
10
6
10 Pd2dba3/PtBu2Bph
80 °C
90d
4e
11 Pd2dba3/PtBu2Bph(Me)
12 Pd2dba3/PtBu2Bph(Me)
rt
1
80 °C 58 88 99e
a Conditions: 1.0 equiv of aryl chloride, 1.2 equiv amine, 1.4 equiv of
NaOtBu, THF (1.2 mL/mmol aryl chloride. b GC conversion of aryl chloride
(see Scheme 4), average of at least two runs. The conversion of this aryl
chloride to the corresponding coupled product is very clean with no other
products being formed; consequently, the GC % conversion (instead of GC
% yield) is a good measure of the catalytic activity of the complexes. c Ratio
L/Pd of 1:1, in toluene (2 mL/mmol aryl chloride). d Ratio L/Pd of 1:1, in
toluene (2 mL/mmol aryl chloride), 2.5 h, isolated yield.64 e Ratio L/Pd of
2:1, in toluene (2 mL/mmol aryl chloride).
Scheme 6
60% conversion was reached within the first 5 min, and after
20 min, more than 80% of the product was obtained. Although
these are preliminary investigations, the results presented in
Figure 9 show that the activation of the catalyst occurs readily
since high yields of the cross-coupling product are produced
within the first few minutes of reaction. Interestingly, when the
9
J. AM. CHEM. SOC. VOL. 128, NO. 19, 2006 6385

A R T I C L E S
Christmann et al.
Figure 9. Plot of % conversion versus time for the reaction in toluene of 4-chlorotoluene with p-toluidine at room temperature in the presence of 4 and a
Pd2(dba)3/PtBu2(Bph-Me) mixture. The amount of compound added in each case was such that a 1 mol % of palladium was present in the solution.
reaction between Pd2(dba)3 and PtBu2(Bph-Me) (the same
phosphine present in 4) was monitored at room temperature,
the cross-coupling yields were very low even after 2 h (see Table
1 and Figure 9). Only when this reaction was warmed to 80 °C
in toluene were the yields comparable to the systems catalyzed
by the dinuclear compound. This apparent slower rate of reaction
for the mixture Pd2(dba)3/PtBu2(Bph-Me) in comparison to the
corresponding dimer 4 may simply be a consequence of the
slow reaction of Pd2(dba)3 with bulky biphenyl phosphines at
room temperature. Hence, the rate-limiting step of the reaction
with Pd2(dba)3/PtBu2(Bph-Me) is most probably the formation
of the catalytically active species.
products are above 93% for the coupling of this amine with
para-substituted aryl chlorides (entries 2 and 5), even when the
substituent is an electron-donating group, such as methoxy. The
usually difficult coupling of morpholine with an ortho-
substituted aryl chloride (entry 3) gave the corresponding
product in 85% yield using 2 mol % of palladium. Excellent
yields (95%) are obtained when the ortho-substituted 2-chloro-
toluene is coupled with benzylamine at room temperature within
1 h.
The coupling of 4-chlorotoluene with benzylamine was also
investigated, giving 76% conversion of the aryl chloride within
1 h (using 1 mol % of Pd). However, the reaction gave both
the mono- and bisarylated products in a 5:1 ratio. Due to the
relatively low conversion (in comparison to the other reactions)
and the formation of more than one product, only the GC
conversion was recorded and no further treatment of the mixture
was carried out (and hence this entry is not shown in Table 2).
This preliminary screening indicates that 4 is an effective
precatalyst for the room temperature amination of both electron-
deficient and electron-rich aryl chlorides. Furthermore, we have
shown that the system is compatible with both aliphatic and
aromatic amines (primary and secondary). To the best of our
knowledge, the rates at which these reactions take place are
considerably higher than those reported when a mixture of a
palladium source and the corresponding phosphine is employed
to generate the catalyst in situ.64 We anticipate that even better
yields and reaction rates will be achieved by optimizing the
reaction conditions at which these dimers catalyze the amination
of aryl chlorides.
To confirm that the catalytic activity of these systems was
not due to one of the “decomposition” products (e.g., the
cyclometalated compound), the above reaction was investigated
in the presence of compound 14 as a potential catalyst. Using
analogous experimental conditions to the ones described above
(1% mol catalyst, i.e., 2% palladium loading, in THF and at
room temperature), no coupling product was detected with this
palladium(II) compound after 4 h. This indicates that the high
activity of the aryl-bridged dimers 2-5 is not due to the
corresponding cyclometalated compound.
Once it was established that the biaryl-bridged dimers
(specially 4 and 5) are effective precatalysts for the coupling
of 4-chlorotoluene with p-toluidine, it was of interest to
investigate the scope of these compounds for the catalysis of
othersmore challengingssubstrates. Thus, the coupling of a
range of substrate combinations (see Table 2) was investigated
using compound 4 as a precatalyst. This dimer was chosen as
it can be synthesized in high yields and is not air-sensitive (hence
it can be weighted out and handled in air). Table 2 summarizes
the results of these amination reactions. As can be seen from
entries 2, 3, and 5, it is possible to use a cyclic secondary
aliphatic amine (namely, morpholine) as substrate in the
amination reaction. The isolated yields of the corresponding
Conclusions
A series of new dinuclear palladium species with a bridging
aryl group have been prepared. These compounds have proved
(64) Wolfe, J. P.; Tomori, H.; Sadighi, J. P.; Yin, J.; Buchwald, S. L. J. Org.
Chem. 2000, 65, 1158-1174.
9
6386 J. AM. CHEM. SOC. VOL. 128, NO. 19, 2006

Dinuclear Pd Precatalysts for Amination of Aryl Chlorides
A R T I C L E S
Table 2. Room Temperature Amination of Aryl Chloridesa
a Reaction conditions: 1.0 equiv of aryl chloride, 1.2 equiv of amine, 1.4 equiv of NaOtBu, 0.5-1 mol % of Pd2(µ-Cl)Cl(PtBu2Bph-Me) (4), THF (1.1
mL/mmol aryl chloride, rt). b Isolated yields (average of two runs). c The spectroscopic data for the isolated products were identical to those reported before.65,66
Experimental Section
to be excellent precatalysts for the room temperature amination
of aryl chlorides (including ortho-substituted and electron-rich).
Crystal structures of the dinuclear and mononuclear species
initially suggested highly unusual and symmetric η3:η3 and η1
arene coordination modes, respectively. Density functional
theory has, however, failed to locate minima corresponding to
these structures and suggests instead that the coordination modes
are better described as distorted η2:η2 and η2, respectively. The
observed crystal structures are therefore most likely to be a result
of crystallographic disorder between isomers with very similar
energies. The high degree of coordinative flexibility shown by
the arene ring of the biphenyl group may have important
implications in the catalytic cycle, where it could act as an
electronic buffer. The aryl-bridged dinuclear palladium com-
pounds dissociate in solution, yielding palladium(II) monophos-
phine complexes which, in some cases, decompose to form
dinuclear cyclometalated products containing unusual four-
membered metallacycles. In the absence of the tert-Bu groups,
calculations indicate that the thermodynamic product should
feature six- rather than four-membered rings. The inclusion of
the phosphine substituents in the computational model, however,
reverses this preference by destabilizing the more crowded six-
membered product. This offers clear evidence for the important
role of steric bulk of the phosphine. The overall reaction
chemistry, and presumably catalytic efficiency, of these systems
therefore depends critically on both the electronic and steric
properties of the phosphine ligand, as well as solvent and
temperature.
General. All experimental procedures were carried out using
standard high vacuum and Schlenk line techniques under an atmosphere
of dry argon. Glassware was dried in an oven at 150 °C prior to use.
CH2Cl2, hexane, THF, methanol, diethyl ether, and toluene were dried
using a solvent purification system (SPS, Innovative Technology) and
degassed with argon prior to use. H, 13C, and 31P NMR spectra were
recorded on a Bruker Avance 400 or Bruker Avance 500 spectrometer
and referenced to residual H and 13C signals of the solvents or 85%
H3PO4 as an external standard (31P). Elemental analyses were carried
out at the Universidad de Santiago de Compostela. ESI-MS and MALDI
spectra were obtained by J. Barr at the Institute of Chemical Research
of Catalonia (ICIQ). GC-MS were recorded on an Agilent Technology
6890N gas chromatographer equipped with an Agilent-5973 Mass
Selective Detector using a HP-5 methyl silicone column. The following
compounds, Pd(COD)Br2,67 Pd(COD)Cl2,67 [Pd2(dba)3]‚C6H6,68 Pd2(µ-
Br)Br{µ-PtBu2(Bph-H)},49 and Pd2(µ-Cl)Cl{µ-PtBu2(Bph-H)},49 were
synthesized following literature procedures. The phosphines, PtBu2(Bph-
H), PtBu2(Bph-Me), and PtBu2(Bph-NMe2), were purchased from Strem
chemicals. The NMR assignments were confirmed by H COSY, H
NOESY, HSQC (1H-13C), and HMBC (1H-13C) experiments. The
numbering scheme for NMR assignments is shown in Figures 10 and
11.
Synthesis of Pd2(µ-Cl)Cl{µ-PtBu2(Bph-H)} (2). This compound
was prepared according to the method we previously reported in a
1
1
1
1
(65) Wolfe, J. P.; Buchwald, S. L. J. Am. Chem. Soc. 1997, 119, 6054-6058.
(66) Kataoka, N.; Shelby, Q.; Stambuli, J. P.; Hartwig, J. F. J. Org. Chem. 2002,
67, 5553-5566.
(67) Drew, D.; Doyle, J. R. Inorg. Synth. 1990, 28, 346-349.
(68) Ukai, T.; Kawazura, H.; Ishii, Y.; Bonnet, J. J.; Ibers, J. A. J. Organomet.
Chem. 1974, 65, 253-266.
9
J. AM. CHEM. SOC. VOL. 128, NO. 19, 2006 6387

A R T I C L E S
Christmann et al.
stirred for further 5 h after which time a dark solid precipitated. The
reaction mixture was filtered to isolate the solid product, and from the
filtrate, crystals grew overnight. The solvent was removed via cannula,
and the crystals were washed with diethyl ether (3 × 3 mL) and dried
under vacuum. The crystalline material was analyzed by X-ray
1
1
crystallography and by H and 31P{1H} spectroscopy. H NMR (CD2-
Figure 10. Numbering scheme for the NMR assignments of compounds
2-6.
Cl2, 500 MHz): δ 7.80-7.74 (m, 1H), 7.65-7.59 (m, 1H), 7.56-7.46
3
(m, 2H), 6.50-6.45 (m, 1H), 6.09-5.98 (m, 2H), 5.87 (d, JHH ) 7.6
3
Hz, 1H), 2.34 (s, 3H, Me), 1.44 (d, 9H, JPH ) 15.0 Hz, C(CH3)3),
preliminary communication. The complete and detailed NMR assign-
3
1.24 (d, 9H, JPH ) 14.9 Hz, C(CH3)3). 31P{1H} NMR (CD2Cl2, 202
1
ments are herein reported. H NMR (400 MHz, CD2Cl2): δ 7.78-
MHz): δ 63.6 (s).
7.73 (m, H-6), 7.63-7.58 (m, H-4), 7.56-7.50 (m, H-5), 7.34-7.30
(m, H-3), 6.63-6.58 (m, H-10), 6.47-6.41 (m, H-9, H-11), 6.05-
6.01 (m, H-8, H-12), 1.36 (d, 18H, 3JPH ) 14.9 Hz, C(CH3)3). 13C{1H}
NMR (126 MHz, CD2Cl2): δ 146.9 (d, JPC ) 19.8 Hz, C-2), 136.9 (d,
JPC ) 32.8 Hz, C-1), 133.7 (d, JPC ) 1.7 Hz, C-6), 132.6 (d, JPC
)
19.0 Hz, C-3), 132.5 (d, JPC ) 4.3 Hz, C-4), 129.1 (d, JPC ) 4.7 Hz,
C-5), 110.0 (d, JPC ) 3.9 Hz, C-8, C-12), 107.8 (d, JPC ) 3.5 Hz, C-9,
C-11), 105.0 (d, JPC ) 6.9 Hz, C-7), 69.9 (d, JPC ) 1.3 Hz, C-10),
36.3 (d, JPC ) 11.5 Hz, C(CH3)3), 30.0 (d, JPC ) 5.4 Hz, C(CH3)3).
Synthesis of Pd2(µ-Cl)Cl{µ-PtBu2(Bph-Me)} (4). A toluene solu-
tion (10 mL) of Pd2dba3‚C6H6 (128 mg, 0.128 mmol) and PtBu2(Bph-
Me) (80 mg, 0.257 mmol) was stirred for 18 h. To the resulting orange-
red solution was added Pd(COD)Cl2 (67 mg, 0.257 mmol), and the
reaction mixture was stirred for further 15 h after which time a dark
solid precipitated. The reaction mixture was filtered and the solid residue
washed with diethyl ether (3 × 5 mL). The crude product was purified
by flash column chromatography (CH2Cl2/CH3OH 9:1) on silica gel to
give 4 as dark-green solid. Yield: 114 mg (74%). 1H NMR (400 MHz,
CD2Cl2): δ 7.80-7.75 (m, H-6), 7.65-7.60 (m, H-4), 7.56-7.50 (m,
Figure 11. Numbering scheme for the NMR assignments of compounds
8-10.
Synthesis of Pd(Br)2{η1-[PtBu2(Bph-NMe2)]} (8). Method A: A
Schlenk flask was charged with Pd2dba3‚C6H6 (148 mg, 0.148 mmol)
and PtBu2(Bph-NMe2) (101 mg, 0.298 mmol), evacuated, and backfilled
with argon. Toluene (10 mL) was added, and the resulting reaction
mixture was stirred for 48 h. After this time, Pd(COD)Br2 (111 mg,
0.297 mmol) was added as a solid. After 1 h, a dark solid precipitated
(together with the formation of palladium metal on the walls of the
reaction vessel). The solution was filtered and the solid washed with
diethyl ether (3 × 3 mL). The residue was extracted with CH2Cl2, the
combined washings were concentrated, and hexane was added to
precipitate a dark blue solid. The solution was filtered, and the crude
solid was purified by column chromatography on silica gel (CH2Cl2
f CH2Cl2/MeOH 10:1). Yield: 75 mg (42%). Single crystals suitable
for X-ray crystallography were obtained by slow evaporation of a
3
H-5), 7.30 (d, JHH ) 7.92 Hz, H-3), 6.62-6.57 (m, H-10), 6.44 (d,
3JHH ) 6.20 Hz, H-11), 6.32-6.27 (m, H-9), 5.86 (d, 3JHH ) 7.60 Hz,
3
H-8), 1.71 (s, 3H, Me), 1.40 (d, 9H, JPH ) 14.9 Hz, C(CH3)3), 1.33
(d, 9H, 3JPH ) 14.5 Hz, C(CH3)3). 13C{1H} NMR (126 MHz, CD2Cl2):
δ 146.9 (C-7), 138.1 (d, JPC ) 30.27 Hz, C-1), 134.4 (C-6), 132.9 (d,
JPC ) 1.89 Hz, C-4), 132.8 (d, JPC ) 16.62 Hz, C-3), 129.1 (d, JPC
)
1
saturated dichloromethane solution of 8. H NMR (400 MHz, CD2-
4.92 Hz, C-5), 128.8 (d, JPC ) 3.99 Hz, C-12), 112.7 (d, JPC ) 2.72
Hz, C-11), 107.1 (d, JPC ) 3.99 Hz, C-2), 106.2 (d, JPC ) 3.83 Hz,
C-8), 105.2 (d, JPC ) 2.35 Hz, C-9), 71.2 (C-10), 37.1 (d, JPC ) 12.0
Hz, C(CH3)3), 36.7 (d, JPC ) 12.0 Hz, C(CH3)3), 30.8 (d, JPC ) 5.6
Hz, C(CH3)3), 30.7 (d, JPC ) 5.5 Hz, C(CH3)3), 21.0 (Me). 31P{1H}
NMR (202 MHz, CD2Cl2): δ 62.1 (s). HRMS-ESI calcd for C23H32-
ClNPPd2 (M+ + MeCN - Cl): 602.0035. Found: 602.0068. Anal.
Calcd for C21H29Cl2PPd2: C, 42.31; H, 4.90. Found: C, 42.34; H, 5.22.
Synthesis of Pd2(µ-Br)Br{µ-PtBu2(Bph-Me)} (5). A toluene solu-
tion (8 mL) of Pd2dba3‚C6H6 (103 mg, 0.103 mmol) and PtBu2(Bph-
Me) (64 mg, 0.207 mmol) was stirred for 18 h. To the resulting orange-
red solution was added Pd(COD)Br2 (62 mg, 0.165 mmol), and the
reaction mixture was stirred for further 5 h after which time a dark
solid precipitated. The reaction mixture was filtered and the brown solid
residue washed with diethyl ether (3 × 3 mL). The crude solid was
extracted with THF (3 × 3 mL), and the combined THF fractions were
concentrated. After the addition of hexane, a dark solid precipitated;
this was isolated by filtration and dried under reduced pressure. Yield:
Cl2): δ 7.84-7.78 (m, H-6), 7.68-7.57 (m, H-10), 7.38-7.20 (m, H-4,
3
3
H-5, H-8), 7.02 (d, JHH ) 8.80 Hz, H-11), 6.71 (d, JHH ) 7.45 Hz,
3
H-3), 6.55 (dd, JHH ) ca. 7.44, ca. 7.44 Hz, H-9), 3.25 (s, NMe2),
3
3
1.74 (d, 9H, JPH ) 14.3 Hz, C(CH3)3), 1.65 (d, 9H, JPH ) 14.9 Hz,
C(CH3)3). 31P{1H} NMR (CD2Cl2, 202 MHz): δ 77.8 (s). HRMS-ESI
calcd for C22H32BrNPPd (M+ - Br): 526.0494. Found: 526.0429.
Method B: The phosphine PtBu2Bph(NMe2) (93 mg, 0.275 mmol) and
Pd(COD)Br2 (103 mg, 0.2751 mmol) were dissolved in CH2Cl2 (6 mL).
The reaction mixture was stirred for 7 h, and after this time, the solvent
was removed under reduced pressure. The residue was washed with
diethyl ether (3 × 6 mL). The crude product was purified on a silica
gel column (CH2Cl2 f CH2Cl2/MeOH 10:1) to obtain a black-blue
1
solid. Yield: 85 mg (51%). H NMR (400 MHz, CD2Cl2): δ 7.84-
7.79 (m, H-6) 7.65-7.59 (m, H-10), 7.34-7.24 (m, H-4, H-5, H-8),
7.02 (d, 3JHH ) 8.76 Hz, H-11), 6.73-6.69 (m, H-3), 6.55 (dd, 3JHH
)
7.80, 7.02 Hz), 3.25 (s, NMe2), 1.75 (d, 9H, 3JPH ) 14.3 Hz, C(CH3)3),
1.64 (d, 9H, 3JPH ) 15.0 Hz, C(CH3)3). 31P{1H} NMR (202 MHz, CD2-
Cl2): δ 77.9 (s). Anal. Calcd for C22H32Br2NPPd: C, 43.48; H, 5.31;
N, 2.30. Found C, 43.22; H, 5.12; N, 2.24.
1
30 mg (26%). H NMR (500 MHz, THF-d8): δ 7.96-7.90 (m, H-6),
3
7.68-7.65 (m, H-4), 7.59-7.56 (m, H-5), 7.43 (d, JHH ) 7. 90 Hz,
3
H-3), 6.91-6.87 (m, H-10), 6.50 (d, JHH ) 5.70 Hz, H-11), 6.29-
Synthesis of Pd(Cl)2{η1-[PtBu2(Bph-NMe2)]} (9). Method A: A
Schlenk flask was charged with Pd2dba3‚C6H6 (84 mg, 0.084 mmol)
and PtBu2(Bph-NMe2) (57 mg, 0.169 mmol), evacuated, and backfilled
with argon. Then toluene (6 mL) was added. The reaction mixture was
stirred for 48 h. To the red-orange solution was added Pd(COD)Cl2
(36 mg, 0.127 mmol). The reaction solution was stirred for further 8
h. The solution was filtered and the residue washed with diethyl ether
(3 × 3 mL). The crude product was purified by flash column
chromatography on silica gel (CH2Cl2 f CH2Cl2/MeOH 10:1). Yield:
49 mg (73%, related to used amounts of Pd(COD)Cl2). 1H NMR (400
3
6.24 (m, H-9), 6.01 (d, JHH ) 7.60 Hz, H-8), 1.70 (s, 3H, Me), 1.41
3
3
(d, 9H, JPH ) 15.0 Hz, C(CH3)3), 1.35 (d, 9H, JPH ) 14.5 Hz,
C(CH3)3). 31P{1H} NMR (202 MHz, THF-d8): δ 67.9 (s). HRMS-ESI
calcd for C23H32BrNPPd2 (M+ + MeCN - Br): 645.9529. Found:
645.9541. Anal. Calcd for C21H29Br2PPd2: C, 36.82; H, 4.27. Found:
C, 36.41; H, 3.97.
Synthesis of Pd2(µ-Cl)Cl{µ-PtBu2(Bph-NMe2)} (6). A toluene
solution (10 mL) of Pd2dba3‚C6H6 (0.150 g, 0.150 mmol) and PtBu2-
Bph-NMe2 (0.103 g, 0.301 mmol) was stirred for 48 h. To the resulting
red solution was added Pd(COD)Cl2, and the reaction mixture was
3
MHz, CD2Cl2): δ 7.81-7.76 (m, H-6), 7.72-7.66 (ddd, JHH ) 9.2,
9
6388 J. AM. CHEM. SOC. VOL. 128, NO. 19, 2006

Dinuclear Pd Precatalysts for Amination of Aryl Chlorides
A R T I C L E S
4
6.56 Hz, JHH ) 1.28 Hz, H-10), 7.35-7.26 (m, H-4, H-5, H-8), 6.93
p-toluidine (1.2 mmol) were added. This was followed by the addition
of the aryl chloride (1 mmol) and THF (1.2 mL/mmol aryl chloride).
The reaction mixture was stirred (see Table 1 for details of reaction
times), and the samples were taken with a syringe at different times,
quenched with water, passed trough a small alumina column, and
analyzed by GC-MS. For data points below 30 min, a stock solution
of aryl chloride in THF was used instead of adding the neat aryl chloride
into the reaction. The formulation of the reaction product (i.e., di-p-
tolylamine) was confirmed by spectroscopic analysis.65
(d, 3JHH ) 8.72 Hz, H-11), 6.76-6.69 (m, H-3), 6.54 (dd, 3JHH ) 7.36,
7.36 Hz, H-9), 3.28 (s, NMe2), 1.71 (d, 9H, 3JPH ) 14.5 Hz, C(CH3)3),
1.66 (d, 9H, 3JPH ) 15.0 Hz, C(CH3)3). 31P{1H} NMR (162 MHz, CD2-
Cl2): δ 76.9 (s). MALDI(+) MS: m/z 484 [M+ - Cl]. Method B:
The phosphine PtBu2Bph(NMe2) (47 mg, 0.136 mmol) and Pd(COD)-
Cl2 (39 mg, 0.136 mmol) were dissolved in CH2Cl2 (5 mL). The solution
turned purple after 30 min. After 6 h, all phosphine has been consumed
(as judged by in situ 31P{1H} NMR), and the solvent was removed
under reduced pressure. The residue was washed with diethyl ether
(2 × 5 mL) and dried under vacuum to obtain 9 as a purple powder.
Yield: 63 mg (89%). 1H NMR (400 MHz, CD2Cl2): δ 7.81-7.76 (m,
H-6), 7.69 (ddd, 3JHH ) 8.75, 6.98 Hz, 4JHH ) 1.50 Hz, H-10), 7.34-
General Procedure for the Room Temperature Amination of
Aryl Chlorides (Table 2). The precatalyst 4 (1-2 mol % of Pd) was
weighed out in air and transferred to an oven-dried Schlenk flask, which
was evacuated and backfilled with argon. Then, NaOtBu (1.4 mmol)
was added, and the flask was evacuated and backfilled with argon again.
The amine (1.2 mmol) and the aryl chloride (1 mmol) were added
through a septum followed by the addition of THF (1 mL/mmol aryl
chloride) (p-toluidine as a solid was added following the addition of
NaOtBu). The reaction was monitored by GC-MS, and after complete
consumption of the aryl chloride or after a certain period of time (see
Table 2 for details), the reaction mixture was quenched with water.
The mixture was extracted with CH2Cl2, the organic layer was dried
over MgSO4 and concentrated. The residue was purified by column
chromatography on silica gel with ethyl acetate/hexane.
3
7.25 (m, H-4, H-5, H-8), 6.94 (d, JHH ) 8.75 Hz, H-11), 6.74-6.70
3
(m, H-3), 6.54 (dd, JHH ) 7.50, 7.05 Hz, H-9), 3.28 (s, NMe2), 1.71
3
3
(d, 9H, JPH ) 15.2 Hz, C(CH3)3), 1.66 (d, 9H, JPH ) 14.4 Hz,
C(CH3)3). 13C{1H} NMR (100 MHz, CDCl3): δ 167.9 (C-12), 152.3
(d, JPC ) 15.4 Hz, C-2), 142.5 (C-10), 135.9 (C-6), 132.3 (d, JPC
)
1.97 Hz, C-arom.), 130.6 (d, JPC ) 14.82 Hz, C-3), 130.3 (C-1), 130.1
(d, JPC ) 1.48 Hz, C-arom.), 126.4 (d, JPC ) 5.71 Hz, C-arom.), 116.4
(C-11), 115.0 (C-9), 86.5 (C-7), 45.5 (N(CH3)2), 41.8 (C(CH3)3), 41.3
(d, JPC ) 7.32 Hz, C(CH3)3), 32.0 (d, JPC ) 2.25 Hz, C(CH3)3), 31.90
(d, JPC ) 2.90 Hz, C(CH3)3). 31P{1H} NMR (162 MHz, CD2Cl2): δ
76.9 (s). MALDI(+) MS: m/z 484 [M+ - Cl]. Anal. Calcd for C22H32-
Cl2NPPd: C, 50.93; H, 6.22; N, 2.70. Found: C, 50.94; H, 6.08; N,
2.63.
Synthesis of Pd(Cl)2{η1-[PtBu2(Bph-Me)]} (10). In a glovebox, the
phosphine (96 mg, 0.3082 mmol) and Pd(COD)Cl2 (80 mg, 0.280
mmol) were weighed and transferred to Schlenk. The Schlenk was
removed from the glovebox, and CH2Cl2 (3 mL) was added. After the
reaction was stirred for 2.5 h, the solvent was removed under reduced
pressure. The obtained orange oil was broken by the addition of hexane
and filtered. This procedure was repeated three times in total to yield
a red-brown powder. The crude product was purified by column
chromatography on silica gel (CH2Cl2/MeOH 20:1). Yield: 112 mg
X-ray Structure Determination: Crystals of 6, 8, and 14 were
obtained from toluene, CH2Cl2 (slow evaporation), and a mixture of
CHCl3/hexane, respectively. The measured crystals were prepared under
inert conditions immersed in perfluoropolyether as protecting oil for
manipulation. Data Collection: Measurements were made on a Bruker-
Nonius diffractometer equipped with a APPEX 2 4K CCD area detector,
a FR591 rotating anode with MoKR radiation, Montel mirrors as
monochromator, and a Kryoflex low-temperature device (T ) -173
°C). Full-sphere data collection was used with ω and æ scans. Programs
used: Data collection Apex2 V. 1.0-22 (Bruker-Nonius 2004), data
reduction Saint + Version 6.22 (Bruker-Nonius 2001), and absorption
correction SADABS V. 2.10 (2003). Structure Solution and Refine-
ment: SHELXTL Version 6.10 (Sheldrick, 2000) was used.
1
(81%). H NMR (400 MHz, CD2Cl2): δ 7.89-7.82 (m, 2H), 7.53-
3
3
7.33 (m, 4H), 6.95 (d, JHH ) 7.36 Hz, 1H), 6.81 (d, JHH ) 7.20 Hz,
1H), 2.50 (s, Me), 1.70 (d, 9H, 3JPH ) 15.0 Hz, C(CH3)3), 1.65 (d, 9H,
3JPH ) 15.0 Hz, C(CH3)3). 31P{1H} NMR (162 MHz, CD2Cl2): δ 75.6
(s). ESI-MS: m/z+ 455 (M+ - Cl). Elemental analyses for this
compound are not provided due to the presence of traces of the
cyclometalated product 14 (see below) detected by 31P{1H} NMR
spectroscopy.
Computational Details. All calculations were carried out with the
Gaussian03 package.69 Pure DFT calculations used the mPW1PW91
functional, a hybrid functional70 with modified Perdew-Wang exchange
and Perdew-Wang correlation.71 An effective core potential was used
to represent the 28 innermost electrons of palladium, and the associated
SDD basis set was used for its valence electrons.72 A valence double-ú
plus polarization 6-31G(d,p) basis set was used for all other atoms.73-75
Hybrid QM/MM calculations with the ONIOM method76,77 used the
same DFT description for the QM region and the UFF force field for
the MM region.78 The nonbonded radius for chlorine was increased by
0.4 Å to account for its anionic character.79 All geometry optimizations
were full, with no restriction.
Synthesis of Palladacycle 14. Pd(COD)Cl2 (59 mg, 0.206 mmol)
and PtBu2Bph(Me) (65 mg, 0.208 mmol) were dissolved in CH2Cl2 (3
mL), and the reaction solution was stirred for 3 days at room
temperature after which time a white precipitate had formed. The solvent
was removed under reduced pressure. The crude product was purified
by flash column chromatography on silica gel (CH2Cl2/hexane 6:1).
The obtained product was washed once with cold hexane and dried
under vacuum to give an off-white solid. Yield: 40 mg (42%). Crystals
suitable for X-ray diffraction studies were grown from a chloroform/
hexane mixture. 1H NMR (500 MHz, THF-d8): δ 7.67-7.01 (m, 12H),
6.89 (dd, J ) 7.6, 3.1 Hz, 2H), 2.11 (s, 6H, Me), 1.32 (br s, 36H,
C(CH3)3). 13C{1H} NMR (126 MHz, THF-d8): δ 143.0 (C-arom.), 142.2
(d, JPC ) 4.3 Hz, C-arom.), 140.2 (d, JPC ) 34.2 Hz, C-arom.), 137.3
(C-arom.), 136.3 (d, JPC ) 24.9 Hz, C-arom.), 131.1 (C-arom.), 130.5
(C-arom.), 130.1 (d, JPC ) 4.7 Hz, C-arom.), 129.5 (C-arom.), 129.0
(C-arom.), 128.6 (d, JPC ) 9.1 Hz, C-arom.), 125.8 (C-arom.), 36.7 (d,
JPC ) 12.6 Hz, C(CH3)3), 30.8 (br s, C(CH3)3), 20.8 (s, Me, 6H). 31P{1H}
NMR (162 MHz, THF-d8): δ -13.1 (s). Anal. Calcd for C42H56Cl2P2-
Pd2‚H2O: C, 54.56; H, 6.32. Found: C, 54.45; H, 6.45.
Acknowledgment. We thank EPSRC (U.K.) (GR/R74970/
01 and GR/R74963/01) and the ICIQ Foundation (Spain) for
financial support, and Johnson Matthey for a loan of PdCl2.
Dr. Gabriel Gonzalez and Mr. Kerman Go´mez are thanked for
(69) Frisch, M. J.; et al. Gaussian 03, revision c.02; Wallingford, CT, 2004.
(70) Adamo, C.; Barone, V. J. Chem. Phys. 1998, 108, 664-675.
(71) Perdew, J. P.; Burke, K.; Wang, Y. Phys. ReV. B: Condens. Matter 1996,
54, 16533-16539.
(72) Andrae, D.; Haeussermann, U.; Dolg, M.; Stoll, H.; Preuss, H. Theor. Chim.
Acta 1990, 77, 123-141.
(73) Hehre, W. J.; Lathan, W. A. J. Chem. Phys. 1972, 56, 5255-5257.
(74) Hariharan, P. C.; Pople, J. A. Theor. Chim. Acta 1973, 28, 213-222.
(75) Francl, M. M.; Pietro, W. J.; Hehre, W. J.; Binkley, J. S.; Gordon, M. S.;
DeFrees, D. J.; Pople, J. A. J. Chem. Phys. 1982, 77, 3654-3665.
(76) Maseras, F.; Morokuma, K. J. Comput. Chem. 1995, 16, 1170-1179.
(77) Vreven, T.; Morokuma, K. J. Comput. Chem. 2000, 21, 1419-1432.
(78) Rappe, A. K.; Casewit, C. J.; Colwell, K. S.; Goddard, W. A., III; Skiff,
W. M. J. Am. Chem. Soc. 1992, 114, 10024-10035.
General Procedure for the Amination of 4-Chlorotoluene with
p-Toluidine. The corresponding precatalyst (2-5) was weighed out
in air and transferred to an oven-dried Schlenk. The Schlenk flask was
evacuated and backfilled with argon, then NaOtBu (1.4 mmol) and
(79) Ujaque, G.; Maseras, F.; Eisenstein, O. Theor. Chem. Acc. 1997, 96, 146-
150.
9
J. AM. CHEM. SOC. VOL. 128, NO. 19, 2006 6389

A R T I C L E S
Christmann et al.
their help with the NMR data. D.A.P.’s visit to ICIQ and his
calculations in CESCA/CEPBA were funded by the HPC-
EUROPA project (RII3-CT-2003-506079), with the support of
the European CommunitysResearch Infrastructure Action under
the FP6 “Structuring the European Research Area” Programme.
data for this paper can also be obtained free of charge via
www.ccdc.cam.ac.uk/conts/retrieving.html (or from the Cam-
bridge Crystallographic Data Centre, 12, Union Road, Cam-
bridge CB2 1EZ, U.K.; fax +44 1223 336033 or e-mail
deposit@ccdc.cam.ac.uk). This material is available free of
charge via the Internet at http://pubs.acs.org.
Supporting Information Available: Cartesian coordinates and
absolute energies for all computed structures. Complete ref 69.
Crystallographic data (CIF). The supplementary crystallographic
JA057825Z
9
6390 J. AM. CHEM. SOC. VOL. 128, NO. 19, 2006