Carbon-nitrogen-bond-forming reductive elimination of arylamines from palladium(II) phosphine complexes

Article abstract of DOI:10.1021/ja971057x

A series of monomeric palladium amido complexes of the form trans-(PPh₃)₂Pd(Ar)(NAr'₂) and (DPPF)Pd(Ar)NAr'₂) (DPPF = 1,1'-bis(diphenylphosphino)ferrocene) and dimeric palladium amido complexes of the form {(PPH₃)Pd(Ar)μ-NHR)}₂ (R = Ph, t-Bu) have been prepared by the reaction of lithium and potassium amides with palladium aryl halide complexes. An X-ray crystal structure of (DPPF)Pd(p-NMe₂C₆H₄)[N(p-CH₃C₆H₄)₂] was obtained. Upon thermolysis in the presence of PPh₃, serving as a trapping agent, both the monomeric and dimeric palladium amido complexes underwent C-N-bond-forming reductive elimination to form arylamines in high yields along with a Pd(0) species. Reductive elimination was also observed from azametallacycle (PPh₃)Pd(η²-C₆H₄C₆H₄-NH), to form carbazole and Pd(PPh₃)₄ at room temperature. Mechanistic studies on the reductive elimination reactions of the monomeric PPh₃-ligated amido complexes indicated the presence of two competing pathways for the formation of amine. At low [PPh₃], reductive elimination occurs via phosphine dissociation to form a three-coordinate intermediate; however, as [PPh₃] is increased, a pathway of reductive elimination from a four-coordinate complex becomes dominant. The DPPF-ligated palladium amido complexes directly eliminated amine from the four-coordinate complex. The mechanism of the reductive elimination from dimeric palladium amido complexes was also studied. These complexes undergo reductive elimination of amine via dimer dissociation to generate a three-coordinate intermediate analogous to those formed by the PPh₃-ligated monomeric amido complexes. The C-N-bond forming reductive elimination reactions were accelerated by electron-withdrawing groups on the Pd bound aryl group and by electron-donating groups on the amido ligand, suggesting that the aryl group acts as an electrophile and the amido ligand acts as a nucleophile.

Full text of DOI:10.1021/ja971057x

8232
J. Am. Chem. Soc. 1997, 119, 8232-8245
Carbon-Nitrogen-Bond-Forming Reductive Elimination of
Arylamines from Palladium(II) Phosphine Complexes
Michael S. Driver and John F. Hartwig*
Contribution from the Department of Chemistry, Yale UniVersity, P.O. Box 208107
New HaVen, Connecticut 06520-8107
ReceiVed April 3, 1997X
Abstract: A series of monomeric palladium amido complexes of the form trans-(PPh3)2Pd(Ar)(NAr′2) and (DPPF)-
Pd(Ar)(NAr′2) (DPPF ) 1,1′-bis(diphenylphosphino)ferrocene) and dimeric palladium amido complexes of the form
{(PPh3)Pd(Ar)(µ-NHR)}2 (R ) Ph, t-Bu) have been prepared by the reaction of lithium and potassium amides with
palladium aryl halide complexes. An X-ray crystal structure of (DPPF)Pd(p-NMe2C6H4)[N(p-CH3C6H4)2] was
obtained. Upon thermolysis in the presence of PPh3, serving as a trapping agent, both the monomeric and dimeric
palladium amido complexes underwent C-N-bond-forming reductive elimination to form arylamines in high yields
along with a Pd(0) species. Reductive elimination was also observed from azametallacycle (PPh3)Pd(η2-C6H4C6H4-
NH), to form carbazole and Pd(PPh3)4 at room temperature. Mechanistic studies on the reductive elimination reactions
of the monomeric PPh3-ligated amido complexes indicated the presence of two competing pathways for the formation
of amine. At low [PPh3], reductive elimination occurs via phosphine dissociation to form a three-coordinate
intermediate; however, as [PPh3] is increased, a pathway of reductive elimination from a four-coordinate complex
becomes dominant. The DPPF-ligated palladium amido complexes directly eliminated amine from the four-coordinate
complex. The mechanism of the reductive elimination from dimeric palladium amido complexes was also studied.
These complexes undergo reductive elimination of amine via dimer dissociation to generate a three-coordinate
intermediate analogous to those formed by the PPh3-ligated monomeric amido complexes. The C-N-bond forming
reductive elimination reactions were accelerated by electron-withdrawing groups on the Pd bound aryl group and by
electron-donating groups on the amido ligand, suggesting that the aryl group acts as an electrophile and the amido
ligand acts as a nucleophile.
Introduction
mediates. More recently, a catalytic process which leads to the
addition of aniline across an olefin has been reported to form a
metal-amido intermediate through oxidative addition of aniline
to an iridium center.24 Metal-amido complexes have also been
implicated as key intermediates present in the palladium-
catalyzed amination of aryl halides and aryl triflates.25-34 Yet,
synthetic routes to the reactive intermediates proposed in these
Low-valent late-transition-metal amido complexes are highly
reactive molecules involved in catalysis, but for which little
fundamental reactivity is known. These complexes are believed
to be reactive because of a mismatch between the soft transition-
metal center with the hard amido ligand.1,2 Metal-amido
complexes have been proposed as intermediates in a number
of catalytic processes. The catalytic hydrogenation of imines3-21
and nitriles22,23 is thought to proceed via metal-amido inter-
(15) Ng, Y.; Chan, C.; Meyer, D.; Osborm, J. A. J. Chem. Soc., Chem.
Commun. 1990, 869.
(16) Spindler, F.; Pugin, B.; Blaser, H. U. Angew. Chem., Int. Ed. Engl.
1990, 29, 558.
(17) Vastag, S.; Bakos, J.; Toros, S.; Takach, N. E.; King, R. B.; Heil,
B.; Marko, L. J. Mol. Catal. 1984, 22, 283.
(18) Willoughby, C. A.; Buchwald, S. L. J. Am. Chem. Soc. 1992, 114,
7562.
X Abstract published in AdVance ACS Abstracts, August 15, 1997.
(1) Pearson, R. G. J. Am. Chem. Soc. 1963, 85, 3533.
(2) Pearson, R. G. J. Chem. Educ. 1968, 45, 643.
(3) Amrani, Y.; Lecomte, L.; Simou, D.; Bakos, J.; Toth, I.; Heil, B.
Organometallics 1989, 8, 542.
(4) Bakos, J.; Toty, I.; Heil, B.; Marko, L. J. Organomet. Chem. 1985,
279, 23.
(5) Bakos, J.; Toth, I.; Heil, B.; Szalontai, G.; Parkanyi, L.; Fulop, V. J.
Organomet. Chem. 1989, 370, 263.
(6) Bakos, J.; Orosz, A.; Heil, B.; Laghmari, M.; Lhoste, P.; Sinou, D.
J. Chem. Soc., Chem. Commun. 1991, 1684.
(7) Becalski, A. G.; Cullen, W. R.; Fryzuk, M. D.; James, B. R.; Kang,
G. J.; Rettig, S. J. Inorg. Chem. 1991, 30, 5002-5008.
(8) Burk, M. J.; Feaster, J. E. J. Am. Chem. Soc. 1992, 114, 6266.
(9) Chan, Y. N. C.; Osborn, J. A. J. Am. Chem. Soc. 1990, 112, 9400.
(10) Cullen, W. R.; Fryzuk, M. D.; James, B. R.; Kutney, J. P.; Kang,
G. J.; Herb, G.; Thorburn, I. S.; Spogliarich, R. J. Mol. Catal. 1990, 62,
243.
(11) Imine hydrogenation catalyzed by rhodium and iridium complexes
would be expected to occur through an amide complex. However,
mechanistic data is scarce. A direct observation of imine insertion that
produces an amide complex potentially similar to those in catalytic systems
is the following: Fryzuk, M. D.; Piers, W. E. Organometallics 1990, 9,
986-98.
(19) Willoughby, C. A.; Buchwald, S. L. J. Org. Chem. 1993, 58, 7627.
(20) Willoughby, C. A.; Buchwald, S. L. J. Am. Chem. Soc. 1994, 116,
11703-11714.
(21) Willoughby, C. A.; Buchwald, S. L. J. Am. Chem. Soc. 1994, 116,
8952.
(22) Armor, J. N. Inorg. Chem. 1978, 17, 203-213.
(23) Yoshida, T.; Okano, T.; Otsuka, S. J. Chem. Soc., Chem. Commun.
1979, 870.
(24) Casalnuovo, A. L.; Calabrese, J. C.; Milstein, D. J. Am. Chem. Soc.
1988, 110, 6738-6744.
(25) Kosugi, M.; Kameyama, M.; Migita, T. Chem. Lett. 1983, 927-
928.
(26) Kosugi, M.; Kameyama, M.; Sano, H.; Migita, T. Nippon Kagaku
Kaishi 1985, 3, 547-551.
(27) Paul, F.; Patt, J.; Hartwig, J. F. J. Am. Chem. Soc. 1994, 116, 5969-
5970.
(28) Guram, A. S.; Buchwald, S. L. J. Am. Chem. Soc. 1994, 116, 7901-
7902.
(29) Louie, J.; Paul, F.; Hartwig, J. F. Organometallics 1996, 15, 2794-
2805.
(30) Louie, J.; Hartwig, J. F. Tetrahedron Lett. 1995, 36, 3609.
(31) Guram, A. S.; Rennels, R. A.; Buchwald, S. L. Angew. Chem., Int.
Ed. Engl. 1995, 34, 1348.
(32) Driver, M. S.; Hartwig, J. F. J. Am. Chem. Soc. 1996, 118, 7217-
7218.
(12) For a recent review see: James, B. R. Chem. Ind. 1995, 62, 167-
180.
(13) Kang, G. J.; Cullen, W. R.; Fryzuk, M. D.; James, B. R.; Kutney,
J. P. J. Chem. Soc., Chem. Commun. 1988, 1466.
(14) Longley, C. J.; Goodwin, T. J.; Wilkinson, G. Polyhedron 1986, 5,
1625.
S0002-7863(97)01057-3 CCC: $14.00 © 1997 American Chemical Society

C-N-Bond-Forming ReductiVe Elimination
J. Am. Chem. Soc., Vol. 119, No. 35, 1997 8233
catalytic processes are not well developed and have attracted
significant attention only recently. The high reactivity of
metal-amido complexes as well as their facile decomposition
through low-energy pathways such as â-hydrogen elimina-
tion35,36 has limited the number of isolated amido complexes.
As a consequence, the fundamental reaction chemistry of metal-
amido complexes has remained undeveloped.
Scheme 1
Presently, metal complexes containing arylamido,35-47 sily-
lamido,35,36,48 and parent amido35,36,49-51 (-NH2) ligands have
been characterized. Alkylamido complexes have remained
rare.46,52-56 Arylamido compounds have displayed insertion
chemistry,35-47 exchanges with even mildly acidic H-X
bonds,35-47,57 and in some cases N-H-bond-forming reac-
tions.35,36,43,45,46,58 Yet, amido complexes which undergo reduc-
tive elimination to form the carbon-nitrogen bond of amines
have been observed in only a few cases.54,59-61 Although
reductive eliminations to form C-C and C-H bonds are among
the fundamental reactions of organometallic chemistry, little is
known about analogous reactions forming C-N bonds. A better
understanding of C-N-bond-forming reductive elimination
reactions is essential for the further development of catalytic
processes that construct amines and nitrogen heterocycles.
Previously, we have communicated our preliminary mechanistic
results on C-N-bond-forming reductive elimination from some
isolated palladium(II) amido complexes.54 We now report the
results of an extensive synthetic and mechanistic study of C-N-
bond-forming reductive eliminations from a series of monomeric
and dimeric palladium(II) amido complexes.
Monomeric palladium complexes of the form L2PdAr(NAr′2)
have been prepared, and they reductively eliminate arylamine,
ArNAr′2. The reductive elimination chemistry of these com-
plexes involves two concurrent pathways, and it is the existence
of these two pathways that allows for aryl halide amination
reactions to occur with complexes containing chelating ligands.
Monomeric complexes containing chelating phosphine ligands
and a variety of aryl groups and amido groups have revealed
the electronic properties of the transition state for C-N-bond-
forming reductive elimination. We have also prepared some
dimeric palladium(II) amido complexes, one of which is a rare
example of an alkylamido complex. The dimeric amido
complexes also reductively eliminate arylamine. In this case,
kinetic studies show that the bridging interaction is cleaved
either reversibly or irreversibly depending on the amido group
and that the resulting three-coordinate intermediate undergoes
elimination of amine.
(33) Wolfe, J. P.; Wagaw, S.; Buchwald, S. L. J. Am. Chem. Soc. 1996,
118, 7215-7216.
(34) Louie, J.; Driver, M. S.; Hamann, B. C.; Hartwig, J. F. J. Org. Chem.
1997, in press.
(35) Bryndza, H. E.; Tam, W. Chem. ReV. 1988, 88, 1163 and references
therein.
(36) Fryzuk, M. D.; Montgomery, C. D. Coord. Chem. ReV. 1989, 95,
1-40 and references therein.
(37) Glueck, D. S.; Winslow, L. J.; Bergman, R. G. Organometallics
1991, 10, 1462-1479.
(38) Glueck, D. S.; Bergman, R. G. Organometallics 1991, 10, 1479-
1486.
(39) Hartwig, J. F.; Andersen, R. A.; Bergman, R. G. J. Am. Chem. Soc.
1989, 111, 2717.
(40) Hartwig, J. F.; Bergman, R. G.; Andersen, R. A. J. Am. Chem. Soc.
1991, 113, 6499.
Results
(41) Hartwig, J. F.; Andersen, R. A.; Bergman, R. G. Organometallics
1991, 10, 1875.
(42) Martin, G. C.; Boncella, J. M.; Wurcherer, E. J. Organometallics
1991, 10, 2804.
(43) Seligson, A. L.; Cowan, R. L.; Trogler, W. C. Inorg. Chem. 1991,
30, 3371.
A. Synthesis and Characterization of Palladium(II)
Amido Complexes. 1. Monomeric Palladium Amido Com-
plexes. The syntheses of trans-(PPh3)2Pd(C6H5)[N(C6H5)2] (1)
and trans-(PPh3)2Pd(C6H5)[N(p-CH3C6H4)2] (2) are summarized
in Scheme 1. Addition of KN(C6H5)2 or KN(C6H4p-Me)2 to a
THF solution of trans-(PPh3)2Pd(C6H5)(I) led to the clean
formations of 1 or 2, which were isolated in 86% and 78%
yields, respectively, after crystallization from THF/Et2O. Com-
plexes 1 and 2 were characterized by standard spectroscopic
and microanalytical techniques. The formation of complexes
1 and 2 was signified by a shift of the 31P{1H} NMR resonance
(1 δ 21.4, s; and 2 δ 21.0, s) upfield from that of the starting
palladium aryl halide complex (δ 24.3, s), the observation of
tolyl methyl groups for 2, and the appropriate aryl 13C{1H}
signals. Both complexes 1 and 2 are red, and the UV-vis
spectrum of 1 showed an absorbance at λmax ) 489.6 nm with
(44) For a well-characterized example and related references, see: Ruiz,
J.; Martinez, M. T.; Vicente, C.; Garcia, G.; Lopez, G.; Chaloner, P.;
Hitchcock, P. B. Organometallics 1993, 12, 4321.
(45) Rahim, M.; White, C.; Rheingold, A. L.; Ahmed, K. J. Organo-
metallics 1993, 12, 2401-2403.
(46) Rahim, M.; Bushweller, C. H.; Ahmed, K. J. Organometallics 1994,
13, 4952-4958.
(47) Boncella, J. M.; Villanueva, L. A. J. Organomet. Chem. 1994, 465,
297-304.
(48) Fryzuk, M. D.; Montgomery, C. D.; Rettig, S. J. Organometallics
1991, 10, 467-473.
(49) Joslin, F. L.; Johnson, M. P.; Mague, J. T.; Roundhill, D. M.
Organometallics 1991, 10, 2781.
(50) Park, S.; Rheingold, A. L.; Roundhill, D. M. Organometallics 1991,
10,
(51) Koelliker, R.; Milstein, D. J. Am. Chem. Soc. 1991, 113, 8524.
(52) Klein, D. P.; Hayes, J. C.; Bergman, R. G. J. Am. Chem. Soc. 1988,
110, 3704.
an extinction coefficient of ꢀ ) 800 cm-1 M-1
.
2. Monomeric Palladium Amido Complexes Containing
Cis-Chelating Phosphines. The observation that reductive
elimination of arylamine can occur from a four-coordinate
palladium amido complex (Vide infra) led us to seek the
synthesis of palladium amido complexes possessing cis-chelating
phosphine ligands. Such cis-chelated complexes would allow
us to observe directly the C-N-bond-forming reductive elimina-
tion in the absence of competing processes such as phosphine
dissociation and cis/trans isomerization. Despite kinetic results
described below that demonstrate elimination can occur from
four-coordinate complexes, the general chemistry of palladium
amido aryl complexes with chelating phosphine ligands was
dominated by reactions other than C-N-bond-forming reductive
(53) Dewey, M. A.; Knight, A.; Arif, A.; Gladysz, J. A. Chem. Ber.
1992, 125, 815-824.
(54) Driver, M. S.; Hartwig, J. F. J. Am. Chem. Soc. 1995, 117, 4708-
4709.
(55) Driver, M. S.; Hartwig, J. F. J. Am. Chem. Soc. 1996, 118, 4206-
4207.
(56) Hartwig, J. F. J. Am. Chem. Soc. 1996, 118, 7010.
(57) Bryndza, H. E.; Fong, L. K.; Paciello, R. A.; Tam, W.; Bercaw, J.
E. J. Am. Chem. Soc. 1987, 109, 1444.
(58) Rahim, M.; Ahmed, K. J. Organometallics 1994, 13, 1751-1756.
(59) Villanueva, L. A.; Abboud, K. A.; Boncella, J. M. Organometallics
1994, 13, 3921-3931.
(60) Koo, K.; Hillhouse, G. L. Organometallics 1995, 14, 4421-4423.
(61) Matsunaga, P. T.; Hess, C. R.; Hillhouse, G. L. J. Am. Chem. Soc.
1994, 116, 3665-3666.

8234 J. Am. Chem. Soc., Vol. 119, No. 35, 1997
DriVer and Hartwig
Table 1. Data Collection and Refinement Parameters for X-ray
Structure of (DPPF)Pd[p-N(CH3)2C6H4][N(p-CH3C6H4)2]
empirical formula
formula weight
crystal color/habit
crystal system
C56H52N2P2FePd‚C3H6O0.75
1031.31
red plates
triclinic
lattice parameters:
a ) 11.086(4) Å
b ) 14.901(3) Å
c ) 16.329(3) Å
R ) 73.41(2) °
â ) 78.92(2) °
γ ) 89.16(2) °
2534(2)
volume
space group
Z value
diffractometer
radiation
P-1 (no. 2)
2
Enraf-Nonius CAD-4
Mo KR (λ ) 0.710 69 Å)
-60 °C
T
residuals
goodness of fit indicator
R; Rw 0.051; 0.051
1.49
Table 2. Intramolecular Bond Distances Involving the
Non-Hydrogen Atoms of
(DPPF)Pd[p-N(CH3)2C6H4][N(p-CH3C6H4)2] (3b)a
Figure 1. ORTEP Drawing of (DPPF)Pd[p-N(CH3)2C6H4][N(p-
CH3C6H4)2] (3b). Hydrogen atoms are omitted for clarity.
atom
atom
distance
atom
atom
distance
Pd
Pd
Pd
Pd
P1
P2
N1
C1
C6
C3
C4
C5
2.283(2)
2.391(2)
2.097(7)
2.048(8)
1.39(1)
1.38(1)
1.39(1)
1.40(1)
1.38(1)
1.38(1)
1.41(1)
1.36(1)
1.38(1)
1.40(1)
1.52(1)
C13
C16
C16
C17
C18
C19
C19
C20
C23
N1
N1
N2
N2
N2
C14
C17
C21
C18
C19
C20
C22
C21
C24
C9
C16
C4
C7
C8
C2
1.38(1)
1.41(1)
1.41(1)
1.39(1)
1.40(1)
1.36(1)
1.53(1)
1.39(1)
1.42(1)
1.39(1)
1.374(9)
1.40(1)
1.45(1)
1.45(1)
1.40(1)
elimination. The addition of DPPE (1,2-bis(diphenylphosphi-
no)ethane) to a solution of 2 led to the decomposition of the
DPPE backbone by P-C bond cleavage and formation of
vinyldiphenylphosphine62 in 93% yield, as shown in Scheme
1. Addition of other chelating phosphine ligands such as DPPP
(1,3-bis(diphenylphosphino)propane), DPPBz (1,2-bis(diphen-
ylphosphino)benzene), and DPPEn (1,2-bis(diphenylphosphino)-
ethylene) did not produce stable amido complexes, and no
triarylamine products were observed.
C1
C2
C3
C4
C5
C9
C9
C10
C11
C12
C12
C6
C10
C14
C11
C12
C13
C15
Nevertheless, we found that amido complexes containing the
chelating phosphine ligand DPPF (DPPF ) 1,1′-bis(diphen-
ylphosphino)ferrocene) could be isolated or observed spectro-
scopically and that these complexes produced arylamines in high
yields by reductive elimination. As shown in Scheme 1,
addition of DPPF to a THF solution of 2 generated the DPPF-
ligated product 3a in 54% yield. A series of monomeric
palladium ditolylamido complexes 3a-f possessing the cis-
chelating ligand DPPF were prepared by addition of KN(p-
CH3C6H4)2 to (DPPF)Pd(Ar)(X) complexes (Ar ) p-NMe2C6H4,
p-CH3OC6H4, p-CH3C6H4, p-CH3SC6H4, p-ClC6H4)63 as shown
in Scheme 1 and were isolated in yields between 62% and 78%
by recrystallization from THF/Et2O. These complexes displayed
two sharp doublets in the 31P{1H} NMR spectrum, demonstrat-
ing the cis, four-coordinate geometry shown in Scheme 1.
3. X-ray Crystal Structure of Complex 3b. During our
studies on the DPPF-ligated palladium amido complexes, we
were able to obtain X-ray quality crystals of complex 3b. The
ORTEP drawing of 3b is shown in Figure 1; data collection
and refinement parameters, bond distances, and bond angles are
provided in Tables 1-3. Complex 3b crystallized in the triclinic
space group P-1 (no. 2). The geometry at the Pd is square-
planar; the sum of the four angles at palladium is 359.97°. The
P-Pd-P bite angle is believed to be important in controlling
the rates for reductive elimination. The 100.47° angle in 3b is
larger than other neutral palladium DPPF complexes.64-67 The
C1
a Distances are in angstroms. Estimated standard deviations in the
least significant figure are given in parentheses.
N atom of the amido ligand is planar; the sum of the three angles
at nitrogen is 358.60°. The substituents at nitrogen are nearly
orthogonal to the Pd square plane; the plane of the nitrogen
forms a dihedral angle of 77.56° with the Pd square plane. The
Pd-N distance is 2.10Å and falls in the range of M-N distances
for other late metal arylamido complexes, despite the larger size
of the diarylamido group.44,45,59,68-70 The Pd-bound aryl group
is also nearly orthogonal to the Pd square plane. In this case
the dihedral angle between the aryl group and the square plane
is 75.26°. The two Pd-P distances are different by over 0.1
Å. The Pd-P1 distance involving the phosphine trans to the
amido ligand is 2.28 Å, while the Pd-P2 distance involving
the phosphine trans to the aryl group is longer, 2.39 Å. The
phosphorus-Cp(centroid)-Cp(centriod)-phosphorus tortional
angle is 37.84°, which is large for ferrocene-based phosphine
ligands and is presumably related to the large bite angle. The
two Fe-centroid distances are both 1.64 Å.
4. Synthesis of a Palladium Azametallacycle. The met-
allacyclic amine complex trans-(PPh3)Pd(C6H4C6H4NH2)(I) (4)
was prepared by oxidative addition of 2′-amino-2-iodobiphenyl
(66) Hayashi, T.; Kumada, M.; Higuchi, T.; Hirotsu, K. J. Organomet.
Chem. 1987, 334, 195.
(67) Clemente, D. A.; Pilloni, G.; Coran, B.; Longato, B.; Tripicchio-
Camellini, M. Inorg. Chim. Acta 1986, 115, L9.
(62) This phosphine is commercially available. The 1H NMR, GC
retention time, and GC/MS were compared to an authentic commercial
sample. In pp .
(63) Brown, J. M.; Guiry, P. J. Inorg. Chim. Acta 1994, 220, 249.
(64) Hayashi, T.; Konishi, M.; Kobori, Y.; Kumada, M.; Higuchi, T.;
Hirotsu, K. J. Am. Chem. Soc. 1984, 106, 158.
(65) Butler, I. R.; Cullen, W. R.; Kim, T. J.; Retting, S. J.; Trotter, J.
Organometallics 1985, 4, 972.
(68) Hope, H.; Olmstead, M. M.; Murray, B. D.; Power, P. P. J. Am.
Chem. Soc. 1985, 107, 712-713.
(69) Kolel-Veetil, M. K.; Rahim, M.; Edwards, J. A.; Rheingold, A. L.;
Ahmed, K. J. Inorg. Chem. 1992, 31, 3877.
(70) Cowan, R. L.; Trogler, W. C. J. Am. Chem. Soc. 1989, 111, 4750.

C-N-Bond-Forming ReductiVe Elimination
J. Am. Chem. Soc., Vol. 119, No. 35, 1997 8235
Table 3. Selected Intramolecular Bond Angles Involving the
Non-Hydrogen Atoms of
Scheme 3
(DPPF)Pd[p-N(CH3)2C6H4][N(p-CH3C6H4)2] (3b)a
atom atom atom
angle
atom atom atom
angle
P1
P1
P1
P2
P2
N1
Pd
Pd
Pd
Pd
Pd
Pd
Pd
P1
P1
P1
P1
P1
P2
P2
P2
P2
P2
P2
N1
N1
N1
N2
N2
N2
C1
C1
C1
C2
P2
100.47(9) C25
Fe
Fe
Fe
Fe
Fe
Fe
C6
C9
C41 177.9(4)
C42 140.8(4)
C43 113.9(4)
C44 111.7(4)
N1
C1
N1
C1
C1
171.9(2)
84.9(2)
87.6(2)
174.4(2)
87.0(3)
C25
C25
C25
C26
C26
C1
5. Dimeric Palladium Amido Complexes. The preparation
of dimeric complexes [(PPh3)Pd(p-CH3C6H4)(µ-NH-t-Bu)]2 (6)
and [(PPh3)Pd(p-CH3C6H4)(µ-NHC6H5)]2 (7) are summarized
in Scheme 3. These complexes were prepared by addition of
LiNH-t-Bu or KNHPh to trans-(PPh3)2Pd(p-CH3C6H4)(I). On
a small scale, addition of LiNH-t-Bu in THF led to formation
of 6 in 60-80% yields by 31P NMR spectroscopy (P(o-tolyl)3
as internal standard) and the reaction to form 7 was similarly
high yielding. However, isolation of 6 and 7 was hampered
by similar solubility of the side product Pd(PPh3)4. Thus, pure
samples of 6 and 7 were obtained by addition of 4-iodotoluene
to crude reaction solutions to convert the soluble Pd(PPh3)4 to
the sparingly soluble (PPh3)2Pd(p-CH3C6H4)(I) and to allow for
isolation of the amido complexes by crystallization. Analytically
pure 6 and 7 were obtained by this procedure in 20% and 28%
yields.
C27
C40 178.2(4)
C5 121.9(8)
40.3(3)
C45 113.7(3)
C51 113.6(3)
C45 102.5(4)
Pd
N1
C10 120.8(8)
C14 123.1(8)
C14 116.0(8)
C11 121.6(8)
C12 123.6(9)
C13 115.7(8)
C15 124(1)
C40
C40
C45
Pd
Pd
Pd
C23
C23
C28
Pd
N1
C10
C9
C9
C9
C51
97.8(4)
C51 105.4(4)
C23 116.9(3)
C28 115.8(3)
C34 113.1(3)
C28 102.1(4)
C34 104.9(4)
C34 102.3(4)
C10
C11
C12
C12
C12
C13
C14
C16
C16
C16
C17
C18
C19
C19
C19
C20
C21
C10
C11
C11
C13
C12
C9
C15 120(1)
C14 121.5(9)
C13 121.6(9)
C17 121.3(7)
C21 121.7(8)
C21 116.9(8)
C18 120.5(8)
C19 121.8(8)
C20 117.5(9)
C22 119.9(9)
C22 123(1)
C9
118.4(5)
N1
Pd
C16 117.8(5)
C16 122.4(7)
N1
C9
C4
C4
C7
Pd
Pd
C2
C1
C17
C16
C17
C18
C18
C20
C19
C16
C7
C8
C8
C2
C6
C6
C3
118.3(8)
118.7(8)
113.0(8)
123.1(6)
121.2(6)
115.6(7)
123.0(8)
The NMR analysis of 6 and 7 indicated the rigid puckered
structures in Scheme 3 for the two complexes. The 31P{1H}
NMR spectrum of 6 contained one singlet at δ 25.7. Addition-
C21 122.4(9)
C20 120.9(8)
1
ally, for the two amido ligands of 6 appeared in the H NMR
spectrum as a single resonance for the t-Bu protons (δ 1.3, s)
and one doublet for the NH protons (δ -0.1, d, J ) 5.3 Hz)
coupled to a single phosphine ligand. The splitting of the NH
resonance by a single phosphine ligand indicated that the
phosphine ligands are oriented anti to one another. The presence
of a single phosphine 31P NMR signal and a single set of amido
1H NMR signals for 6 indicates the geometry in Scheme 3. A
ring-puckered geometry in which the alkyl groups of the amido
ligands are oriented syn to each other or a geometry containing
a planar Pd-N ring, with either syn or anti amido groups, would
produce equivalent phosphine and amido groups.
a Angles are in degrees. Estimated standard deviations in the least
significant figure are given in parentheses.
Scheme 2
The 31P NMR spectrum of 7 contained two singlets at δ 29.69
and 29.73 for inequivalent phosphine ligands. The phosphine
resonances for 7 coalesced upon heating a sample in the NMR
probe at 95 °C. The 1H NMR spectrum of 7, however,
contained two inequivalent NH resonances (δ 1.21, d, 4.3 Hz;
1.42, d, 5.3 Hz), each coupled to a single phosphine ligand.
Again, the coupling of the NH resonances to a single phosphine
required the phosphine ligands to possess an anti geometry. The
inequivalence of the phosphine ligands and the doublet N-H
signals reveal the geometry in Scheme 3 for 7. The ring
containing the two Pd and two N atoms is markedly puckered
giving the molecule a “butterfly” configuration, and the NH
protons are oriented anti to one another. This type of ring-
puckered geometry has been previously observed for other
dimeric bridging amido compounds.44,50,55,71-73
to Pd(PPh3)4 in benzene at 50 °C as shown in Scheme 2. The
product precipitated out of solution as a beige solid during the
course of the reaction and was isolated in 84% yield. The 31P-
{1H} NMR spectrum of 4 consisted of a singlet at δ 38.9 while
1
the H NMR spectrum at 50 °C contained a broad resonance
centered at δ 5.13 for the two NH protons. Two NH resonances
1
at δ 5.26 and δ 5.00 were observed in the H NMR spectra
obtained below 10 °C. This dynamic behavior is attributed to
slow inversion of the ring-puckered structure in Scheme 2.
Infrared vibrations at 3304 and 3233 cm-1 were observed for
the two νNH
.
B. C-N-Bond-Forming Reductive Elimination of Amines.
1. C-N-Bond-Forming Reductive Elimination as a Cycliza-
Addition of KN(TMS)2 to a THF or toluene solution of 4
generated the azametallacycle 5 as shown in Scheme 2.
Complex 5 could not be isolated due to its reactivity at room
temperature, but it was formed as the only transition-metal
product that could be observed by NMR and was clearly
identified in solution by NMR techniques. The formation of 5
was demonstrated by the appearance of a new resonance at δ
31.9 in the 31P{1H} NMR spectrum, an upfield shift of the NH
resonance to δ 2.00 (d, J ) 7.8 Hz), and a new NH band in the
(71) O’Mahoney, C. A.; Parkin, I. P.; Williams, D. J.; Woollins, J. D.
Polyhedron 1989, 8, 1979.
(72) Alcock, N. W.; Bergamini, P.; Kemp, T. J.; Pringle, P. G. J. Chem.
Soc., Chem. Commun. 1987, 235.
(73) Park, S.; Roundhill, D. M.; Rheingold, A. L. Inorg. Chem. 1987,
26, 3972.
(74) Kong, K.; Cheng, C. J. Am. Chem. Soc. 1991, 113, 6313-6315.
(75) Reductive elimination and aryl group exchange are competing, but
separate processes. Their dependence on ligand concentration identity is
markedly different. Please see the following three references.
(76) Okeefe, D. F.; Dannock, M. C.; Marcuccio, S. M. Tetrahedron Lett.
1992, 33, 6679.
(77) Baranano, D.; Hartwig, J. F. J. Am. Chem. Soc. 1995, 117, 2937.
(78) Morita, D. K.; Stille, J. K.; Norton, J. R. J. Am. Chem. Soc. 1995,
117, 8576-8581.
IR spectrum at 3298 cm-1
. The similarity of the N-H
stretching frequency to that of a bridging palladium arylamido
(Vide supra)59 and the reduced frequency from terminal aryla-
mides,37,40,41,70 along with the scarcity of three-coordinate
Pd(II) species, suggests that the structure of 5 is dimeric.

8236 J. Am. Chem. Soc., Vol. 119, No. 35, 1997
DriVer and Hartwig
Scheme 4
Scheme 7
Scheme 5
Scheme 6
was assigned as the cis-ligated isobutylamido aryl complex 8a
shown in Scheme 7. Consistent with this proposal, PhNH-i-
Bu was formed in 64% yield based on (DPPF)Pd(Ph)(I) upon
warming a sample of 8a to room temperature.
A new palladium amido complex 8b was also formed upon
addition of KNHPh to a THF solution of (DPPF)Pd(Ph)(I). The
31P{1H} NMR spectrum of this complex again contained two
doublets at δ 25.1 and 6.3 (J ) 38 Hz). To confirm the
formation of a monomeric N-bound anilido complex, (DPPF)-
Pd(Ph)(I) was reacted with [15N]KNHPh to observe the 31P-
15N couplings in 8b. Coupling between the anilido ligand and
each inequivalent phosphine was observed. The trans JNP
coupling was 43 Hz, while the cis coupling was 4 Hz. All of
the coupling constants for the labeled compound are included
in the Experimental Section. Anilido complex 8b underwent
reductive elimination of amine at or below room temperature.
Ph-NHPh formed from 8b in 80% yield after standing at room
temperature for 1.5 h in the presence of added PPh3.
During the course of our studies with KNHPh, we found that
the addition of excess KNHPh reacted further with 8b to form
the complex 9. The 31P{1H} NMR spectrum of 9 contained
two singlets at δ 22.3 and -16.6, the upfield shift being similar
to that of free DPPF. We had originally assigned 9 a dimeric
structure containing bridging anilido groups.32 However,
complex 9-15N that was prepared from excess [15N]KNHPh
displayed a 31P{1H} NMR spectrum in which the downfield
resonance exhibited 31P-15N coupling (δ 22.3, d, JNP ) 46 Hz)
and displayed a 15N{1H} NMR spectrum in which two
resonances (δ 86.5, d, JNP ) 46 Hz; 89.2, s) other than that for
free [15N]KNHPh were observed for inequivalent anilido ligands.
The observation of inequivalent anilido ligands, only one of
which is coupled to phosphorous, is inconsistent with a dimeric
structure and is consistent with the structure for 9 in Scheme 7.
Recently, related studies conducted on the reactivity of PPh3-
and DPPF-ligated palladium aryl halide complexes with KNC4H4
have also resulted in the isolation of some unusual anionic
palladium bispyrrolyl complexes X[(L)Pd(Ar)(NC4H4)2] (X )
K, NMe4; L ) PPh3, η1-DPPF), the chemistry of which will be
reported elsewhere.81
tion. Intramolecular reductive elimination reactions from the
strained metallacycle 5 in Scheme 2 to form carbazole occurred
at room temperature as shown in Scheme 4. Upon standing at
room temperature for 1 h, 5 underwent C-N-bond-forming
reductive elimination to give carbazole in 92% yield.
2. Reductive Elimination of Arylamine from Monomeric
Palladium Amido Complexes 1 and 2. Heating of monomeric
complexes 1 and 2, in the presence of PPh3, led to the reductive
elimination of triarylamines (Scheme 5). The only palladium
species observed by 31P NMR spectroscopy at the end of the
reaction was the palladium(0) triphenylphosphine complexes,
which exist predominantly as a rapidly exchanging mixture of
Pd(PPh3)3 and PPh3.79,80 Complex 1 formed NPh3 in 84% yield
upon heating in THF solvent at 80 °C for 5-12 h. Complex 2
produced C6H5N(p-CH3C6H4)2 in 77% yield after heating at 110
°C for 3-6 h in toluene-d8 solution. In the absence of added
phosphine, the yields of arylamine from 1 and 2 were less than
20%. Although the reaction was not fully investigated under
these conditions, diarylamine and biphenyl were the major
organic products. These reactions are among the first examples
of C-N-bond-forming reductive elimination to form amine59,60
and the first reductive elimination of a triarylamine.
3. Reductive Elimination of Amine from DPPF-Ligated
Palladium Amido Complexes. The DPPF-ligated palladium
aryl amido complexes 3a-f underwent high-yielding (82-94%)
reductive elimination reactions to form the corresponding
triarylamine after 2-8 h at 75 °C in the presence of free PPh3
as shown in Scheme 6.
We also observed reductive elimination reactions from DPPF-
ligated complexes containing primary amido ligands as shown
in Scheme 7. These primary amido complexes were not
isolated, but were cleanly generated in solution. They under-
went facile reductive eliminatoin of mixed secondary amines
at room temperature or below. The addition of LiNH-i-Bu to
a THF solution of (DPPF)Pd(Ph)(I) at 0 °C resulted in the
formation of a new species which displayed two doublets in
the 31P{1H} NMR spectrum (δ 26.5, 18.3; J ) 22.0 Hz) and a
new isobutyl group by 1H NMR spectrometry. The 31P chemical
shifts of this product appeared downfield of those for (DPPF)-
Pd(Ph)(I), in a similar manner to the downfield shifts of the
resonances for 3a. Given the clean reactivity of potassium
amides with (DPPF)Pd(Ph)I to give 3a and the similarity in
chemical shifts, the reaction product formed at low temperature
4. Reductive Elimination Reactions of Dimeric Palladium
Amido Complexes 6 and 7. In a similar fashion to the reactions
of monomeric palladium amido complexes, the dimeric tert-
butylamido complex 6 and the dimeric anilido complex 7
underwent C-N-bond-forming reductive elimination to form
amine when heated in the presence of PPh3 (Scheme 8). A
toluene solution containing 6 and added PPh3 formed (p-
CH3C6H4)NH-t-Bu in 90% yield upon heating for 8 h at 95 °C,
while complex 7 formed NHPh2 in 87% yield in the presence
of PPh3 after heating for 24 h at 95 °C. The palladium(0)
triphenylphoshine complexes that rapidly equilibrate with PPh3
were the only palladium species observed at the end of the
(79) (Triphenylphosphine)palladium, although (PPh3)4Pd when isolated,
is predominantly dissociated to (PPh3)3Pd and free phosphine in solution:
Amatore, C.; Pfluger, F. Organometallics 1990, 9, 2276-2282.
(80) Tolman C. A.; Seidel, W. C.; Gerlach, D. H. J. Am. Chem. Soc.
1972, 94, 2669.
(81) Driver, M. S.; Mann, G.; Hartwig, J. F. Manuscript in preparation.

C-N-Bond-Forming ReductiVe Elimination
J. Am. Chem. Soc., Vol. 119, No. 35, 1997 8237
Scheme 8
1
reaction by H and 31P{1H} NMR spectroscopy. Again, the
addition of PPh3 to trap the Pd(0) species formed was essential
for obtaining high yields of coupled amine product.
Figure 2. Plot of kobs vs 1/[PPh3] for the reductive elimination of di-
C. Mechanistic Investigation of the Reductive Elimination
of Amines. 1. Mechanism of Reductive Elimination of
Arylamine from 1 and 2. The formation of the C-N bond of
amines from reactive metal-amido complexes is a key step in
the catalytic formation of arylamines.25-31 A clear description
of how the reductive elimination of amines occurs will help
the understanding of the chemistry of the active catalyst in the
aryl halide amination process. This information has aided in
the design of new catalysts with improved selectivity and
activity.32-34 We have studied in detail the mechanism of the
C-N-bond-forming reductive elimination of arylamines from
the palladium amido complexes described above.
Kinetic data for the reductive elimination of triphenylamine
from 1 were obtained by monitoring the disappearance of 1 in
THF solution by UV-vis spectroscopy. Complex 1 is red and
has an absorption at λmax ) 489.6 nm (ꢀ ) 800 cm-1 M-1),
which was used to monitor the concentration of 1 throughout
the reaction. The products of the reaction did not show any
absorption bands above 300 nm. The reactions were heated at
80 °C in sealed quartz cuvettes. UV-vis spectra were taken
every 30 min. The reactions were conducted with [1] of 2.3
mM and [PPh3] ranging from 22 to 68 mM to create pseudo-
first-order conditions. Plots of ln (absorbance) vs time were
linear over three half-lives, indicating a first-order dependence
on [1]. By monitoring the reductive elimination at varied
phosphine concentrations and graphing ln (kobs) vs ln [PPh3],
an inverse first-order dependence of rate on [PPh3] (slope )
-1.1 ( 0.2) was deduced. Unfortunately, this complex proved
difficult to isolate reproducably in pure form, since it was
typically obtained as powders. Further, the lack of well-resolved
resonances in the 1H NMR spectrum precluded simple analysis
of its purity. Thus, we conducted a thorough study of the
mechanism of the reductive elimination from diarylamido
complexes using the p-tolyl analog 2.
p-tolylphenylamine from 2. R ) 0.97 for y ) (1.1 (( 0.2) × 10-3
+
[(2.1 (( 0.2) × 10-4]x.
number of rate constants were measured in that case and they
were measured within a smaller range of concentrations, most
likely precluding our detection of a second pathway rather than
indicating different chemistry between the two compounds.
It seemed likely that the noninteger reaction order revealed
in the studies with 2 resulted from competing processes that
were zero and first order in [PPh3]. A plot of kobs vs 1/[PPh3],
which is given in Figure 2, is more informative. It revealed
the two competing reaction pathways. This plot was linear with
a positive slope and a positive, nonzero y intercept. The yields
of triarylamine in these kinetic runs were not dependent on the
changing [PPh3]. Yields between 70% and 80% were observed
for all reactions. These data show that two competing pathways
exist for reductive elimination, one of which involves a
monophosphine complex formed by phosphine dissociation and
one of which involves a bis-phosphine complex. The detailed
arguments for this conclusion are presented in the Discussion
section.
2. Mechanism of Reductive Elimination from DPPF-
Ligated Complexes 3a-f. According to the results in the
previous section, cis-chelated complexes should eliminate
triarylamine directly from the four-coordinate complex allowing
us to obtain mechanistic data on the C-N-bond-forming
reductive elimination in the absence of any competing processes.
Kinetic data for the elimination of triarylamine from complexes
3a-f were obtained by monitoring the disappearance of the
palladium amido complex by 1H NMR spectroscopy in toluene-
d8. The reactions were conducted in the NMR probe heated at
75 °C with spectra being obtained periodically. The reactions
were conducted with a 13 mM concentration of palladium amido
complex, and concentrations of PPh3 ranging from 53 to 160
mM. The results of the kinetic analysis are summarized in Table
4. Plots of ln [3a-f] vs time were linear over three half-lives
for all the complexes 3a-f, demonstrating a first-order depen-
dence, and the observed rate constants were independent of
[PPh3].
3. Electronic Effects on Reductive Elimination Reactions
of DPPF-Ligated Palladium Amido Complexes 3a-f. Previ-
ous mechanistic studies conducted on the palladium-catalyzed
amination of aryl halides have indicated the presence of a
reactive palladium-amido intermediate and that the rates of
catalysis were accelerated by aryl halides with electron-
withdrawing substituents.29 Also, electron-poor aryl halides
often give higher yields of arylamine products in the catalytic
amination, particularly for reactions of alkyl amines.32,34 This
enhanced yield is presumably due to an acceleration of the
reductive elimination step of the catalytic cycle relative to
competitive decomposition pathways, such as â-hydrogen
elimination, which ultimately generates arene. A better under-
standing of the electronic influences on the reductive elimination
reaction of amido complexes will help design of new catalysts
for improved turnover numbers, substrate compatibility, and
product selectivity, while providing fundamental information
on an important class of reaction.
Complex 2 contains a tolyl methyl group that allowed for
NMR spectroscopic monitoring of the reactions and allowed
for assessment of the purity of different batches of amido
complex without microanalytical data. Further, we found that
these diarylamido complexes decomposed upon irradiation with
a Hg arc lamp. Thus, a full study of the effect of phosphine
1
concentration on the reaction rate was conducted by H NMR
spectroscopy using complex 2.
Kinetic data for the reductive elimination of PhN(p-
CH3C6H4)2 from 2 were obtained in toluene-d8 solvent by
1
monitoring the disappearance of the amidotolyl group by H
NMR spectroscopy. The reactions were conducted in the NMR
probe at 110 °C; spectra were obtained at 1 min intervals. The
reactions were conducted with [2] of 6.7 mM and [PPh3] ranging
from 52 to 470 mM to create pseudo-first-order conditions. Plots
of ln [2] vs time were linear over three half-lives for all reaction
conditions, again demonstrating a first-order dependence on [2].
The rate of reductive elimination was again inhibited by an
increase in [PPh3]. However, graphs of ln (kobs) vs ln [PPh3],
1
which involved the higher reagent concentrations used in H
NMR spectroscopy as well as a wide range of [PPh3], showed
noninteger reaction orders between zero and one. This nonin-
teger order was not observed with 1. However, a smaller

8238 J. Am. Chem. Soc., Vol. 119, No. 35, 1997
Table 4. kobs for Reductive Elimination Reactions of 3a-f
DriVer and Hartwig
[(PPh3)(Ph)Pd(µ-NH-t-Bu)2Pd(Tol)(PPh3] (6c) would form dur-
ing the reaction. Instead, no evidence for the formation of the
mixed dimer 6c was obtained over the course of the reaction.
The tert-butyl resonances for 6, 6b, and 6c can be distinguished
complex
conc (M)
[PPh3] (M)
kobs (s-1
)
3a
3b
3c
3d
3e
3f
0.013
0.013
0.013
0.013
0.013
0.013
0.11
0.11
0.11
0.11
0.11
0.11
29.4 × 10-4
2.5 × 10-4
2.1 × 10-4
13.8 × 10-4
35.2 × 10-4
43.3 × 10-4
1
by H NMR spectroscopy at 50 °C.82
Kinetic data for the reductive elimination of diphenylamine
from 7 in toluene-d8 were obtained by monitoring the dis-
1
appearance of an aromatic or NH resonance of 7 by H NMR
spectroscopy. The reactions were heated at 95 °C, and 1H NMR
spectra were obtained at 2 h intervals. The reactions were
conducted with [7] between 5.9 and 22.4 mM and [PPh3]
between 47.7 mM and 143 mM to create pseudo-first-order
conditions. The results of the kinetic analysis are summarized
in Table 5. The observed rate constants for 7 were independent
of [PPh3]. The plots of 2 × [7]1/2 vs time were linear over
three-half lives. Again, since half and first order reactions are
difficult to distinguish by the linearity of a 2 × [7]1/2 or ln [7]
plot, we measured rate constants for reactions with different
initial concentrations. Plots of 2 × [7]1/2 gave comparable rate
constants, kobs, when initial concentrations of 7 were 5.9 mM
and 22.4 mM, confirming that the reactions are half order in
[7]. Furthermore, first-order plots for the decay of [7] (ln [7]
vs time) produced rate constants that were proportional to the
square root of the initial concentration of 7, again indicating
that the reaction has a half, rather than first, order dependence
on [7].
Table 5. kobs for Reductive Elimination Reactions of 5 and 6
conc
(M)
first order
kobs (s-1
complex
[PPh3]
)
half order kobs
6
6
6
7
7
7
7
0.0066 0.040
0.0066 0.12
7.2 × 10-5 ( .3
7.2 × 10-5 ( .2
7.4 × 10-5 ( .2
-
-
-
0.015
0.12
0.0058 0.092 12.0 × 10-6 ( .1 1.2 × 10-6 ( .1 M0.5
0.023
0.092
5.7 × 10-6 ( .2 1.0 × 10-6 ( .3 M0.5
0.0095 0.048
0.0095 0.14
-
-
1.4 × 10-6 ( .2 M0.5
1.6 × 10-6 ( .3 M0.5
Kinetic data for the reductive elimination of triarylamine from
complexes 3a-f were collected as described above and are
summarized in Table 4. Indeed, we did observe an acceleration
of C-N-bond-forming reductive elimination with complexes
containing aryl groups with electron-withdrawing substituents.
Complex 3f with a p-Cl substituent reacted approximately 20
times faster than complex 3b containing a p-OMe substituent,
and the overall order of rate constants was p-Cl > p-SMe >
p-H > p-Me > p-NMe2 > p-OMe.
Discussion
Reductive Elimination of Arylamine from Palladium
Amido Complexes. Both the monomeric and dimeric pal-
ladium amido complexes produce arylamine by C-N-bond-
forming reductive elimination. These isolable palladium amido
complexes provide a means to model the reactivity of the
palladium amido complexes proposed27,29 in the catalytic
formation of arylamine with discrete, well-characterized com-
pounds. Although our complexes are the first examples of
reductive elimination of triarylamines, a few other systems have
been reported to undergo C-N-bond-forming reductive elimina-
tion to give amines.
A qualitative comparison of the rates of reductive elimination
for amido complexes with electronically different amido ligands
was obtained by monitoring the elimination reactions of
diarylamido complex 3a, alkylamido 8a, and anilide 8b. The
reductive elimination reaction was accelerated by electron-
donating substituents on the nitrogen. Complex 8a, containing
an alkyl amido group, formed coupled amine rapidly at room
temperature while 8b took a couple of hours at room temperature
to react fully. Complex 3a, with a diaryl amido ligand, was
stable at room temperature and reductively eliminated amine
over several hours at 75 °C.
The reactivity of some monomeric PMe3-ligated complexes
of the form trans-(PMe3)2Pd(R)(NHAr) (R ) Me, Ph) have been
studied.59 When the complexes were heated at moderate
temperatures (80 °C), phosphine dissociation occurred and
dimeric amido complexes formed. Heating for longer times at
higher temperature (110 °C) resulted in the slow reductive
elimination of arylamine. In these studies, it was undetermined
whether the reductive elimination reaction occurred from the
monomeric or dimeric species. Some amido and azametalla-
cycle complexes of Ni(II) have also been seen to eliminate
arylamines.60 In this case, reductive elimination was initiated
by the addition of an oxidant to convert the Ni(II) to Ni(III).
The formation of amine from Ni(III) was proposed to proceed
through radical processes. There have also been reports of
reductive elimination that form organic amides. In the Fe-
facilitated synthesis of â-lactams, the final step in the synthesis
involves the oxidation of the Fe center to promote C-N-bond-
forming reductive elimination of the product.83 Ni(II) amido
complexes have also been seen to reductively eliminate organic
amides upon reaction with CO,61 although it is uncertain in this
4. Mechanism of Arylamine Elimination from Dimeric
6 and 7. To compare the reactivities of the monomeric and
dimeric palladium amido complexes, the mechanisms for the
reductive elimination of amine from 6 and 7 were investigated.
Kinetic data for the reductive elimination of (p-CH3C6H4)NH-
t-Bu from 6 in toluene-d8 solution were obtained by monitoring
the disappearance of the NH or the tolyl group Me resonance
1
of 6 by H NMR spectroscopy. The reactions were heated at
1
95 °C, and H NMR spectra were obtained periodically. The
reactions were conducted with [6] of either 4.3 mM or 10.3
mM and [PPh3] of either 40.0 mM or 119 mM to create pseudo-
first-order conditions. The results of the kinetic analysis are
presented in Table 5. Plots of the ln [6] vs time were linear
over three half-lives. Since it is difficult to distinguish half-
and first-order reactions by the linearity of a 2 × [6]1/2 or ln
[6] plot, we measured rate constants for reactions with different
initial concentrations. The values of kobs from the first-order
plot were independent of the initial concentrations of 6,
indicating that the reaction is first order in palladium dimer 6.
The observed rate constants were also independent of phosphine
concentration. Even at high [PPh3], no evidence for monomeric,
bis-phosphine palladium amido complexes was obtained during
the reactions.
(82) A sample of the mixed dimer was prepared by the reaction of
(PPh3)2Pd(tol)I and (PPh3)2Pd(Ph)I with LiNH-t-Bu in THF, and it was
analyzed by 31P and 1H NMR spectroscopy in C6D6. The 31P NMR signals
1
for the three compounds were not resolvable, but the H NMR signals for
To probe for reversible dimer dissociation, reactions were
conducted with a mixture of 6 and its phenyl analog [(PPh3)-
Pd(Ph)(µ-NH-t-Bu)]2 (6b). If reversible cleavage of the dimeric
complexes occurred during the reactions, then the mixed dimer
the t-Bu peaks were distinguishable at 50 °C: 6a δ 1.237 ppm; 6b δ 1.218
ppm; 6c δ 1.228 ppm and 1.226 ppm.
(83) Berryhill, S. R.; Price, T.; Rosenblum, M. J. Org. Chem. 1983, 48,
158-162.

C-N-Bond-Forming ReductiVe Elimination
J. Am. Chem. Soc., Vol. 119, No. 35, 1997 8239
Scheme 9
whether the isomerization occurs by phosphine association or
dissociation, there is no net loss or gain of ligand before the
rate determining step. Therefore, this pathway will show no
rate dependence on [PPh3]. The rate equation for this pathway
is presented in eq 3.
-d[2]
dt
) kobs[2]
(3)
where
kobs ) K5k6
Trans-to-cis isomerization of related dialkyl complexes has been
shown to occur at temperatures lower than those of the reductive
elimination reactions, and reductive elimination reactions that
do not occur by ionic pathways occur from cis complexes.
Pathway D shows an unusual dissociative mechanism in
which a phosphine ligand is lost from 2 to generate a three-
coordinate intermediate which then dimerizes. The dimeric
complex then undergoes reductive elimination to produce
triphenylamine. Although this mechanism seems unlikely, many
isolated late-metal amido complexes have been dimeric(Vide
supra),35,36,44,50,55,59,87 and these dimeric complexes do ultimately
give amines by reductive elimination. In addition, previous
observations of C-N-bond-forming reductive elimination in-
dicated that amine formation occurred competitively with
dimerization.59 Pathway D would give rise to the rate expres-
sion in eq 4, which predicts a second-order dependence on [2]
and an inverse second order dependence on [PPh3].
case whether the CO inserts into the Ni-R or Ni-NR2 bond
prior to reductive elimination.
Mechanism of Reductive Elimination from Monomeric
Palladium Amido Complexes 1 and 2. Some of the possible
mechanisms considered for the elimination of amine from 1
and 2 are provided in Scheme 9. Our kinetic data indicate two
parallel pathways for reductive elimination from the PPh3-ligated
complexes, one of which dominates in the case of the PPh3
complexes at low [PPh3]. To begin the discussion on the
mechanism of reductive elimination, we will first consider the
kinetic data obtained for the monomeric PPh3 complexes 1 and
2. Our discussion will focus on the results for 2, since the
kinetic data for the reactions of this complex are most complete.
Pathway A consists of an associative mechanism in which
PPh3 coordinates to 2 to create a five-coordinate intermediate,
which then reductively eliminates arylamine. This mechanism
would result in the rate expression in eq 1.
-d[2]
dt
) kobs[2]2
(4)
where
-d[2]
dt
k4k7k8
) kobs[2]
(1)
kobs
)
k-4k-7[PPh3]2 + k-4k8[PPh3]2 + k7k8
k1k2[PPh3]
k-1 + k2
kobs
)
Pathway E consists of a dissociative mechanism in which 2
reversibly loses a phosphine ligand to generate a three-coordinate
intermediate that reductively eliminates arylamine. This mech-
anism would result in the rate expression in eq 5 where k9 <<
k-4[PPh3]. Under these conditions, the rate expression predicts
a first-order dependence on [2] and an inverse first-order
dependence on [PPh3].
Should two mechanisms occur simultaneously, then the
overall rate equation will be the sum of the rate expressions for
the two competing pathways. For example, should pathway C
and E compete with each other, then the overall rate equation
would be the one shown in eq 6.
This expression predicts a first order dependence on both [2]
and [PPh3]. Our kinetic data show an inverse first-order
dependence on phosphine concentration ruling out pathway A.
Pathway B shows the mechanism for direct elimination of
arylamine from the four-coordinate trans complex. This mech-
anism seems unlikely because earlier studies conducted on the
reductive elimination of C-C and C-H bonds have determined
that the ligands involved in bond formation need to be oriented
cis to one another.84-86 Yet, previous observations of C-N-
bond-forming reductive elimination have suggested radical
pathways60 for the formation of amine, making pathway B worth
consideration. The rate equation for pathway B is shown in eq
2 and would provide a first order dependence on [2] and no
dependence on [PPh3].
-d[2]
dt
) kobs[2]
(5)
where
-d[2]
dt
) k3[2]
(2)
k4k9
kobs
)
k-4[PPh3] + k9
Pathway C consists of a mechanism in which 2 isomerizes
to a cis, four-coordinate complex, which then undergoes
reductive elimination to form triphenylamine. Regardless of
This rate equation can be simplified with the assumption that
phosphine association is much faster than the actual C-N-bond-
forming step. In this case, k9 << k-4. With this assumption,
eq 6 simplifies to the expression in eq 7.
(84) Gillie, A.; Stille, J. K. J. Am. Chem. Soc. 1980, 102, 4933-4941.
(85) Sostero, S.; Traverso, O. J. Organometal. Chem. 1983, 246, 325-
329.
(86) Ozawa, F.; Iro, T.; Nakamura, Y.; Yamamoto, A. Bull. Chem. Soc.
Jpn. 1981, 54, 1868-1880.
(87) Casalnuovo, A. L.; Calabrese, J. C.; Milstein, D. Inorg. Chem. 1987,
26, 971-973.

8240 J. Am. Chem. Soc., Vol. 119, No. 35, 1997
DriVer and Hartwig
for C-C-bond-forming reductive elimination reactions. The
series of studies conducted on the reductive elimination of ethane
from both cis- and trans-(PPh2R)2PdR′2 (R ) Ph, Me; R′ )
Me, Et) have shown that initial phosphine dissociation is
required for reductive elimination.84,86 The Au(III) complex
(PPh3)AuMe3 must also lose its phosphine ligand before the
reductive elimination of ethane.90 The reductive elimination
of 1,1-dimethylcyclopropane is thought to proceed via ligand
dissociation from the platinocyclobutane (P-i-Pr3)2Pt(CH2C(CH3)2-
CH2).89
-d[2]
) kobs[2]
dt
(6)
(7)
where
k4k9
kobs ) K5k6 +
k-4[PPh3] + k9
-d[2]
) kobs[2]
dt
Nevertheless, some reductive elimination reactions to form
C-C bonds have been observed directly from four-coordinate
complexes. In Stille’s studies,84 reductive elimination of ethane
occurred from complexes containing chelating phosphines, but
with rates that were approximately 100 times slower than those
with monodentate ligands. In studying palladium-catalyzed C-
cross-coupling reactions, Hayashi found that the presence of
chelating ligands inhibits decompostion pathways such as
â-hydrogen elimination and leads to high yields of coupled
product via reductive elimination.82 Of course, it is difficult to
determine in the reactions of complexes with chelating ligands
whether the bis-phosphine undergoes reductive elimination or
whether dissociation of an arm of the chelate occurs before
reductive elimination. Stang and Young have reported rates
for C-C-bond-forming reductive elimination from platinum
vinyl alkynyl and diaryl complexes that are either independent
of [PPh3] or slightly accelerated by added PPh3. These results
could indicate either elimination without phoshine dissociation
or irreversible phosphine dissociation. Our kinetic results for
complexes with monodentate ligands explicitly show that two
reaction pathways are followed for the reductive elimination
of arylamines, one through a three-coordinate monophosphine
complex and one through a four-coordinate bisphosphine
complex.
The reductive elimination to form a carbon-heteroatom bond
from four-coordinate complexes has previously been observed
for the formation of sulfides from (DPPE)Pd(Ar)(SR).77 To
test whether dechelation was occurring, analogous thiolate
complexes with more rigid chelating ligands were studied. The
DPPE, bis(diphenylphosphino)benzene, and bis(diphenylphos-
phino)ethene complexes all eliminated at similar rates, and these
rates were slower than the extremely rapid rates for reductive
elimination from three-coordinate 14-electron monophoshine
thiolates generated from tin thiolates.92 Thus, a rapid elimination
from the unsaturated intermediate and slower elimination from
a four-coordinate intermediate, again, appears to occur. C-O-
bond-forming reductive eliminations of esters has been observed
by Yamamoto, and these reactions occur by reductive elimina-
tion from both three- and four-coordinate complexes.
where
k4k9
kobs ) K5k6 +
k-4[PPh3]
In the case of both complexes 1 and 2, the reductive
elimination reactions were first order in [2] and inhibited by
increasing [PPh3]. Our results for 2 involve high [PPh3] and a
wide range of [PPh3]. In this case, the graph of ln (kobs) vs ln
[PPh3] was curved at high [PPh3], indicating a change in the
reaction order in phosphine as [PPh3] was increased. However,
the plot of kobs vs 1/[PPh3] was linear and displayed a positive
slope and positive, nonzero intercept. The positive slope of
this plot reveals one reaction pathway that is inverse first order
in [PPh3]. The nonzero intercept reveals a competing pathway
that is independent of [PPh3]. In other words, this plot
corresponds to eq 7, the equation for two competing pathways,
one of which is independent of [PPh3] and one of which is
inversely dependent on [PPh3]. The pathway that shows inverse
first-order dependence on [PPh3] is path E, the pathway in which
C-N bond formation occurs within a three-coordinate 14-
electron intermediate. The pathway that is independent of
[PPh3] could be path B, the direct elimination from the isolated
trans isomer, or path C, elimination from the four-coordinate,
16-electron cis isomer. It is most reasonable that C-N bond
formation occurs from the cis isomer. We have obtained no
evidence for radical pathways and amide dissociation is unlikely
considering the nonpolar solvent and the absence of a substantial
solvent effect. Previous concerted reductive eliminations are
well known and require a cis arrangement of ligands. Thus, at
low [PPh3], reductiVe elimination from the three-coordinate
intermediate is the primary pathway for formation of triaryl-
amine (pathway E); howeVer, as [PPh3] is increased, pathway
C through the cis, four-coordinate intermediate contributes
substantially to the total rate. This information on PPh3-ligated
complexes clearly implies that the pathway for reductiVe
elimination of amine from an amido complex containing a
chelating ligand will inVolVe a four-coordinate intermediate.
Comparison of C-N Bond Formation to C-H-, C-C-,
C-S-, and C-O-Bond-Forming Reductive Eliminations.
Similar kinetic results were obtained for the reductive elimina-
tion of carborane from the complex trans-(PPh3)2(CO)Ir(σ-carb)-
(H)Cl.88,89 For this system, the reductive elimination occurred
primarily via phosphine dissociation to give a five-coordinate
intermediate which eliminated carborane. However, at high
[PPh3], the five-coordinate intermediate competitively underwent
isomerization and reassociation of phosphine ligand to give the
cis-oriented six-coordinate complex which eliminated carborane.
For square-planar complexes, C-H-bond-forming reductive
eliminations are believed to occur predominantly from four-
coordinate species.
Mechanism of Reductive Elimination of Arylamine from
DPPF-Ligated Complexes. The kinetic data for reductive
elimination from complexes 3a-f demonstrate that the reductive
elimination reaction is first order in the palladium amido
complex. The observed rate constants were also independent
of [PPh3], again ruling out an associative mechanism for the
reductive elimination. The added phosphine serves only as a
trap for the Pd(0) fragment formed after reductive elimination
of amine. These data are consistent with our hypothesis that
reductive elimination of triarylamine from the DPPF-ligated
palladium amido complexes occurs directly from the four-
coordinate complex.
Electronic Effects on the Reductive Elimination of Aryl-
Previous studies conducted on square-planar, d8 metal
complexes have shown the presence of dissociative mechanisms
(90) Komiya, S.; Albright, T. A.; Hoffman, R.; Kochi, J. K. J. Am. Chem.
Soc. 1976, 98, 7255.
(88) Basato, M.; Morandini, F.; Longato, B.; Bresadola, S. Inorg. Chem.
1984, 23, 649-653.
(89) carb ) 7-C6H5-1,7-C2B10H10.
(91) DiCosimo, R.; Whitesides, G. M. J. Am. Chem. Soc. 1982, 104,
3601.
(92) Louie, J.; Hartwig, J. F. J. Am. Chem. Soc. 1995, 117, 11598.

C-N-Bond-Forming ReductiVe Elimination
J. Am. Chem. Soc., Vol. 119, No. 35, 1997 8241
Scheme 10
Figure 3. Free energy relationships for the C-N-bond-forming
reductive elimination reactions of 3a-f. The left plot shows log(kobs
/
nisms considered for the elimination of amine from 6 are shown
in Scheme 10. Pathway A consists of a mechanism involving
phosphine association which acts to cleave the palladium dimer
and form one three- and one four-coordinate monomeric species,
one or the other of which reductively eliminates amine product.
This mechanism would give the rate expression shown in eq 8
which predicts a half-order dependence on [6] and a first order
dependence on [PPh3].
kobs(H)) vs σ; the right plot shows log(kobs/kobs(H)) vs σj.
amine. The relative rates for reductive elimination of 3a-f
directly probe the electronic properties of the transition state
because reductive elimination occurs from the observed four-
coordinate complex without any preceding ligand dissociation
or isomerization. The plot on the left of Figure 3 shows that
the substituent effects do not follow a simple linear free-energy
relationship with the σ values derived from the acidity of
substituted benzoic acids. Similar results were also observed
in the case of sulfide reductive elimination from (DPPE)Pd-
(Ar)(SR). The separation of the resonance and inductive σ
effects has been previously described for reactions of organic
compounds.93,94 In this treatment, a synthetic σ value (σj) is
constructed from separate resonance and inductive σ values. A
σ generated from weighting the resonance effect more than the
inductive effect (σj ) 2σR + σI) provides the best fit to our data
for elimination of amines from Pd(II). The plot of log (kobs) vs
σ is shown on the right of Figure 3. Although a small number
of points are used in the analysis of this paper, an identical
effect was observed in the reductive elimination of sulfides that
involved 11 different substituents. Thus, it appears to be a
general result for C-X-bond-forming reductive elimination from
palladium complexes that resonance effects dominate over
inductive effects, although we draw no quantitative conclusions
from the data of the amido complexes. An analysis of
Yamamoto’s previous data95 for formation of sp2-sp3 C-C-
bonds from Pd(II) also reveals that resonance effects are more
important than inductive effects in this reaction. In the case of
C-C- and C-H-bond-forming reductive eliminations, theoreti-
cal studies have indicated that the reactions are accelerated by
electron-donating substituents.96 In the case of bond formation
involving sp2-hybridized carbon atoms this predicted trend is
clearly not observed. More recently, Calhorda95 has provided
a theoretical study on bond formation involving vinyl groups
that may be more relevant.
-d[6]
dt
) kobs[6]1/2
(8)
where
k10k11[PPh3]
k-10 + k11
kobs
)
Our kinetic data show a first-order dependence on [6] and no
dependence on [PPh3], ruling out the associative mechanism of
pathway A.
Pathway B shows a mechanism consisting of reversible dimer
cleavage to generate a three-coordinate intermediate, which
eliminates amine product. This mechanism would result in the
rate expression in eq 9 giving a half-order dependence on [6]
and no dependence on [PPh3].
-d[6]
dt
) kobs[6]1/2
(9)
where
k12k13
kobs
)
k-12 + k13
This mechanism would also result in formation of the mixed
dimer 6c from the thermolysis of the mixture of 6 and 6b.
Reversible dimer cleavage is inconsistent with the observed first-
order dependence on [6] and absence of the cross-over product
6c.
Pathway C involves partial cleavage of the dimeric 6 to form
a complex containing one three- and one four-coordinate
palladium. This intermediate complex could be formed revers-
ibly or irreversibly from 6. In either case, the rate of the
reductive elimination process would be first order in [6] and
zero order in [PPh3] as shown in the rate expression in eq 10.
Our qualitative studies conducted with varied amido ligands
is consistent with the results of the electronic effects of the aryl
substituents. Since electron-withdrawing substituents on the aryl
ligand accelerate reductive elimination, it is reasonable that a
complementary effect from the amido groupselectron-donating
substituents accelerating reductive eliminationswould be ob-
served. The overall reactivity of the amido groups was
diarylamido < arylamido < alkylamido. This trend shows that
electron-donating groups accelerate reductive elimination in this
case. In other words, the amido group acts as a nucleophile,
and the aryl group acts as an electrophile.
-d[6]
dt
) kobs[6]
(10)
Mechanism of Reductive Elimination from Dimeric Pal-
ladium Amido Complexes 6 and 7. Some possible mecha-
where
(93) Wells, P. R.; Ehrenson, S.; Taft, R. W. Prog. Phys. Org. Chem.
1969, 6, 147.
(94) Ehrenson, S.; Brownlee, R. T. C.; Taft, R. W. Prog. Phys. Org.
Chem. 1973, 10, 1.
k14k15
kobs
)
k-14 + k15
(95) Ozawa, F.; Kurihara, K.; Fujimori, M.; Hidaka, T.; Toyoshima, T.;
Yamamoto, A. Organometallics 1989, 8, 180-188.
(96) Tatsumi, K.; Hoffman, R.; Yamamoto, A.; Stille, J. K. Bull. Chem.
Soc. Jpn. 1981, 54, 1857-1867.
This pathway would also predict the absence of the mixed dimer
6c during the course of the reductive elimination. Although
the intermediate in this pathway is perhaps counterintuitive, it

8242 J. Am. Chem. Soc., Vol. 119, No. 35, 1997
DriVer and Hartwig
is likely to be an intermediate in the cleavage of dimeric amido
complexes and its intermediacy in the reductive elimination
process is consistent with the first-order behavior in [6], zero-
order behavior in [PPh3], and the absence of mixed dimer 6c
during the reaction.
However, pathway D is also consistent with these data.
Pathway D consists of a mechanism involving irreversible dimer
cleavage to give a three-coordinate intermediate. The three-
coordinate complex then undergoes rapid reductive elimination
of amine. This pathway would result in the rate expression
shown in eq 11 that contains a first-order dependence on [6]
and no dependence on [PPh3].
DPPF-ligated palladium primary alkylamido complex. The
dimeric palladium amido complexes 6 and 7 also underwent
C-N-bond-forming reductive elimination. The mechanism of
reductive elimination of the dimeric complexes involved dimer
dissociation to generate two three-coordinate intermediates that
were analogous to those formed by the PPh3-ligated monomeric
amido complexes. In the case of the alkyl amido dimer 6, dimer
cleavage was irreversible while for the anilido dimer 7, dimer
dissociation was reversible due to the presence of the electron-
withdrawing anilido group.
Experimental Section
General Methods. Unless otherwise noted, all reactions and
manipulations were performed in an inert atmosphere glovebox or by
using standard Schlenk techniques. All 31P{1H} NMR chemical shifts
are reported in parts per million relative to an 85% H3PO4 external
standard. Shifts downfield of the standard are reported as positive.
Benzene, toluene, THF, ether, and pentane solvents were distilled from
sodium/benzophenone prior to use. Triphenylphosphine was sublimed
prior to use in kinetic reactions. trans-(PPh3)2Pd(Ar)I98 was prepared
according to literature procedures. All other chemicals were used as
received from commercial suppliers.
Preparation of trans-[P(C6H5)3]2Pd(C6H5)[N(C6H5)2] (1). Into a
20 mL vial was weighed 105 mg of [P(C6H5)3]2Pd(C6H5)(I) (0.126
mmol). This material was suspended in 12 mL of THF and 27 mg of
KN(C6H5)2 (0.13 mmol) was added to the suspension as a solid. The
reaction mixture was stirred for 30 min at room temperature and turned
a deep red color. The reaction mixture was filtered through a medium
fritted funnel. The resulting THF solution was concentrated under
vacuum. Crystalline material was obtained by layering with Et2O and
cooling at -35 °C. The yield of red crystalline product was 93 mg
(84%). 1H NMR: (C6D6) δ 6.26 (d, 7.6 Hz, 1H), 6.62 (t, 7.1 Hz, 1H),
6.80-6.92 (m, 21H), 7.06 (t, 7.7 Hz, 4H), 7.43 (d, 7.2 Hz, 4H), 7.47-
7.56 (m, 12H), 7.71-7.75 (m, 2H). 31P{1H} NMR: (THF) δ 21.4. IR
(cm-1): (KBr) 3052 (m), 2991 (w), 2921 (m), 2849 (m), 1574 (s), 1468
(s), 1433 (s) 1313 (s) 1218 (m), 1170 (m), 1095 (s), 987 (m), 791 (s),
741 (s), 691 (s). UV-vis: (THF) λmax ) 489.6 nm (ꢀ ) 880 cm-1
M-1). Anal. Calcd for C54H45NP2Pd: C, 74.01; H, 5.18; N, 1.60.
Found: C, 73.46; H, 5.55; N, 1.50.
Preparation of trans-[P(C6H5)3]2Pd(C6H5)[N(p-CH3C6H4)2] (2).
Into a 20 mL vial was weighed 218 mg (0.261 mmol) of trans-(PPh3)2-
Pd(Ph)(I). The material was suspended in 15 mL of THF and 68 mg
(0.289 mmol) of KN(p-CH3C6H4)2 was added as a solid. The reaction
mixture was stirred for 1 h at room temperature and turned a deep red
color. The reaction mixture was filtered through a medium fritted
funnel. The resulting THF solution was concentrated by vacuum.
Crystalline material was obtained by addition of Et2O and cooling at
-35 °C for 12 h. The yield of red crystalline product was 184 mg
(78%). 1H NMR: (C6D6) δ 2.26 (s, 6H), 6.40 (t, 8.3 Hz, 2H), 6.52 (t,
7.1 Hz, 1H), 6.86 (d, 8.3 Hz, 4H), 6.91-7.01 (m, 20 H), 7.39 (d, 8.3
Hz, 4H), 7.50-7.56 (m, 12H). 13C{1H} NMR: (C6D6) δ 20.7 (s), 120.0
(s), 121.5 (s), 122.4 (s), 126.2 (s), 127.7 (s), 128.5 (s), 129.4 (s), 131.4
(t, 21.9 Hz), 134.5 (t, 6.0 Hz), 140.0 (t, 3.8 Hz), 149.2 (t, 7.0 Hz),
156.6 (s). 31P{1H} NMR: (toluene) δ 21.0 (s). Anal. Calcd for
C56H49N1P2Pd: C, 74.37; H, 5.46; N, 1.55. Found: C, 74.18; H, 5.66;
N, 1.38.
General Procedure for the Preparation of (DPPF)Pd(Ar)X.
(DPPF)Pd(Ph)(I).63 trans-(PPh3)2Pd(Ph)(I) (302 mg, 0.362 mmol) was
suspended in 4 mL of THF. DPPF (221 mg, 0.399 mmol) was added
to the stirred suspension as a solid. The reaction was allowed to stir
at room temperature for 30 min. The product was precipitated by the
addition of 15 mL of pentane. The product was collected as a solid
by filtration on a fritted glass funnel to yield 272 mg (87%). The
product was used without further purification.
-d[6]
dt
) k16[6]
(11)
This mechanism also predicts the absence of the mixed dimer
6c. Although it seems more likely that full cleavage of 6
preceded elimination, we sought firm data that the dimeric
complexes fully cleaved to monomers before reductive elimina-
tion.
The PPh3-ligated anilido complexes underwent reductive
elimination more slowly than the alkyl amido complexes.
Qualitatively, reductive elimination of amine from 6 was much
faster than reaction of 7 (8 h vs 24 h at 95 °C). This slower
reactivity of 7 is most likely due to slower reductive elimination
since 7 contains the less electron-rich anilido ligand. Thus, the
dimer cleavage is more likely to be reversible in the case of
anilido complex 7. The kinetic data for the reductive elimination
from 7 did reveal reversible dimer cleavage to form two
monomeric, three-coordinate 14-electron intermediates. The rate
of the reductive elimination was half order in [7] and indepen-
dent of [PPh3]. These kinetic data are inconsistent with pathway
C inVolVing incomplete dimer cleaVage, and are consistent with
only pathway B inVolVing reVersible dimer dissociation to form
a three-coordinate intermediate which then undergoes C-N-
bond-forming reductiVe elimination to giVe amine product. As
a result, reaction of the tert-butylamido dimer by pathway D
involving full dimer cleavage, rather than pathway C involving
partial dimer cleavage, is almost certainly followed for the
elimination from the alkylamido dimer 6. Thus, the dimeric
complexes undergo C-N-bond-forming reductive eliminations
through the same three-coordinate intermediates that dominate
the chemistry of the PPh3-ligated monomeric diarylamido
complexes at low [PPh3].
Conclusions
A series of monomeric and dimeric palladium amido com-
plexes have been prepared which undergo C-N-bond-forming
reductive elimination to form amines when heated. Using PPh3-
ligated palladium amido complexes 1 and 2, the presence of
two concurrent mechanisms for the reductive elimination of
amine were detected. At low [PPh3], a mechanism involving
phosphine dissociation to generate a three-coordinate intermedi-
ate which then eliminates amine is the primary pathway for
reductive elimination. As [PPh3] is increased, this dissociative
pathway is inhibited, and a mechanism involving reductive
elimination from a four-coordinate complex dominates. These
results led to the preparation of a series of DPPF-ligated
palladium amido complexes which reductively eliminated amine
directly from the cis, four-coordinate palladium amido complex.
Using these DPPF-ligated complexes, we observed that the
reductive elimination reaction is accelerated by electron-
withdrawing substituents of the palladium-bound aryl ring and
that the resonance effects (σR) were more substantial than the
inductive effects (σI). We were also able to observe high-
yielding reductive elimination of mixed arylamines from a
Preparation of (DPPF)Pd(p-CH3OC6H4)(I). Using the general
procedure with 250 mg (0.289 mmol) of trans-(PPh3)2Pd(p-CH3OC6H4)-
(I) and 176 mg (0.318 mmol) of DPPF gave 201 mg (78%) of product.
(97) Calhorda, M. J.; Brown, F. M.; Cooley, N. A. Organometallics 1991,
10, 1431-1438.
(98) Fitton, P.; Johnson, M. P.; McKeon, J. E. J. Chem. Soc., Chem.
Commun. 1968, 6-7.

C-N-Bond-Forming ReductiVe Elimination
J. Am. Chem. Soc., Vol. 119, No. 35, 1997 8243
1H NMR: (C6D6) δ 3.28 (s, 3H), 3.66 (m, 2H), 3.71 (m, 2H), 3.93 (m,
2H), 4.53 (m, 2H), 6.44 (d, 6.7 Hz, 2H), 6.80-7.09 (m, 14H), 7.40-
7.46 (m, 4H), 8.34-8.39 (m, 4H). 31P{1H} NMR: (C6D6) δ 29.4 (d,
32.4 Hz), 8.9 (d, 32.4 Hz).
Preparation of (DPPF)Pd(p-CH3C6H4)(I). Using the general
procedure with 248 mg (0.292 mmol) of trans-(PPh3)2Pd(p-CH3C6H4)-
(I) and 178 mg (0.321 mmol) of DPPF gave 208 mg (81%) of product.
1H NMR: (C6D6) δ 2.04 (s, 3H), 3.65 (m, 2H), 3.72 (m, 2H), 3.93 (m,
2H), 4.50 (m, 2H), 6.52 (d, 6.6 Hz, 2H), 6.81-7.12 (m, 14H), 7.39-
7.45 (m, 4H), 8.28-8.34 (m, 4H). 31P{1H} NMR: (C6D6) δ 25.1 (d,
34.8 Hz), 8.0 (d, 34.8 Hz).
Preparation of (DPPF)Pd(p-CH3SC6H4)(Br). Using the general
procedure with 260 mg (0.309 mmol) of trans-(PPh3)2Pd(p-CH3SC6H4)-
(I) and 188 mg (0.340 mmol) of DPPF gave 218 mg (81%) of product.
1H NMR: (CDCl3) δ 3.72 (m, 2H), 4.16 (m, 2H), 4.50 (m, 2H), 4.66
(m, 2H), 6.55 (d, 6.7 Hz, 2H), 6.84-6.89 (m, 2H), 7.15-7.26 (m, 4H),
7.34-7.50 (m, 12H), 8.01-8.05 (m, 4H). 31P{1H} NMR: (C6D6) δ
29.4 (d, 33.0 Hz), 9.2 (d, 33.0 Hz).
Preparation of (DPPF)Pd(p-ClC6H4)(I). Using the general pro-
cedure with 240 mg (0.276 mmol) of trans-(PPh3)2Pd(p-ClC6H4)(I) and
168 mg (0.304 mmol) of DPPF gave 226 mg (91%) of product. 1H
NMR: (CDCl3) δ 2.28 (s, 3H), 3.64 (m, 2H), 4.16 (m, 2H), 4.50 (m,
2H), 4.68 (m, 2H), 6.60 (d, 7.3 Hz, 2H), 6.86-9.82 (m, 2H), 7.13-
7.18 (m, 4H), 7.29-7.50 (m, 12H), 8.03-8.10 (m, 4H). 31P{1H}
NMR: (C6D6) δ 25.2 (d, 33.0 Hz), 8.7 (d, 33.0 Hz).
5.3 Hz), 73.97 (d, 32.3 H), 74.97 (d, 8.8 Hz) 75.48 (d, 10.6 Hz), 79.34
(dd, 40.4 Hz, 6.3 Hz), 113.61 (d, 8.8 Hz), 122.17 (s), 123.18 (s), 127.72
(d, 9.3 Hz), 127.93 (d,9.3 Hz), 128.26 (s), 128.81 (s), 129.99 (d, 17.9
Hz), 133.33 (d, 50.3 Hz), 134.86 (d, 11.5), 135.41 (d, 13.3), 135.44
(d, 32.4 Hz), 136.62 (d, 2.8 Hz), 142.74 (dd, 124.3 Hz, 9.0 Hz), 147.41
(s), 153.45 (s). 31P{1H} NMR: (toluene) δ 14.1 (d, 33.4 Hz), 23.1 (d,
33.4 Hz). Anal. Calcd for C56H52FeN2P2Pd: C, 68.83; H, 5.36; N,
2.87. Found: C, 69.22; H, 5.48; N, 2.61.
Preparation of (DPPF)Pd(p-CH3OC6H4)[N(p-CH3C6H4)2] (3c).
Following the procedure for the preparation of 3a, (DPPF)Pd(p-CH3-
OC6H4)(I) (180 mg. 0.201 mmol) was reacted with KN(p-CH3C6H4)2
(52 mg, 0.221 mmol) in THF. Recrystallization at -35 °C from a
concentrated THF solution layered with ether gave 143 mg of pure 3c
(74%). 1H NMR: (C6D6) δ 2.23 (s, 6H), 3.17 (s, 3H), 3.72 (m, 2H),
3.95 (m, 2H), 4.15 (m, 2H), 4.50 (m, 2H), 6.26 (d, 8.1 Hz, 2H), 6.85
(d, 8.1 Hz, 2H), 6.90 (d, 8.2 Hz, 4H), 6.99-7.10 (m, 12H), 7.35 (d,
8.2 Hz, 4H), 7.62-7.69 (m, 4H), 7.80-7.86 (m, 4H). 13C{1H} NMR:
(C6D6) δ 20.98 (s), 54.40 (s), 72.18 (d, 5.6 Hz), 73.35 (d, 5.2 Hz),
73.91 (d, 33.1 Hz), 74.94 (d, 9.2 Hz), 75.54 (d, 12.9 Hz), 79.08 (dd,
41.4 Hz, 7.0 Hz), 112.90 (d, 8.8 Hz), 122.13 (s), 123.38 (s), 127.72 (d,
9.4 Hz), 127.99 (d, 11.4 Hz), 128.24 (s), 128.83 (s), 130.12 (d, 3.8
Hz), 133.01 (d, 51.3 Hz), 134.75 (d, 12.5 Hz), 135.27 (d, 34.3 Hz),
135.36 (d, 12.5 Hz), 136.68 (d, 3.8 Hz), 146.68 (dd, 124.2 Hz, 9.6
Hz), 153.30 (s), 156.54 (s). 31P{1H} NMR: (toluene) δ 23.2 (d, 33.5
Hz), 14.6 (d, 33.5 Hz). Anal. Calcd for C55H49FeNOP2Pd: C, 68.51;
H, 5.12; N, 1.45. Found: C, 68.22; H, 4.98; N, 1.29.
Preparation of (DPPF)Pd(p-CH3C6H4)[N(p-CH3C6H4)2] (3d).
Following the procedure for the preparation of 3a, (DPPF)Pd(p-
CH3C6H4)(I) (158 mg. 0.180 mmol) was reacted with KN(p-CH3C6H4)2
(52 mg, 0.221 mmol) in THF. Recrystallization at -35 °C from a
concentrated THF solution layered with ether gave 116 mg of pure 3d
(68%). 1H NMR: (C6D6) δ 1.94 (s, 3H), 2.23 (s, 6H), 3.73, (m, 2H),
3.95 (m, 2H), 4.16 (m, 2H), 4.48 (m, 2H), 6.40 (d, 6.3 Hz, 2H), 6.89-
6.93 (m, 6H), 7.00-7.10 (m, 12H), 7.34 (d, 8.3 Hz, 4H), 7.62-7.66
(m, 4H), 7.81-7.85 (m, 4H); 31P{1H} NMR: (toluene) δ 23.6 (d, 34.2
Hz), 14.6 (d, 34.2 Hz). Anal. Calcd for C55H49FeNP2Pd: C, 69.67;
H, 5.21; N, 1.48. Found: C, 69.88; H, 5.42; N, 1.21.
Preparation of (DPPF)Pd(p-CH3SC6H4)[N(p-CH3C6H4)2] (3e).
Following the procedure for the preparation of 3a, (DPPF)Pd(p-CH3-
SC6H4)Br (210 mg. 0.241 mmol) was reacted with KN(p-CH3C6H4)2
(52 mg, 0.221 mmol) in THF. Recrystallization at -35 °C from a
concentrated THF solution layered with ether gave 158 mg of pure 3e
(67%). 1H NMR: (C6D6) δ 1.92 (s, 3H), 2.28 (s, 6H), 3.72 (m, 2H),
3.94 (m, 2H), 4.16 (m, 2H), 4.47 (m, 2H), 6.58 (d, 6.0 Hz, 2H), 6.83-
7.10 (m, 18H), 7.31 (d, 8.2 Hz, 4H), 7.60-7.74 (m, 4H), 7.76-7.84
(m, 4H). 13C{1H} NMR: (C6D6) δ 20.96 (s), 21.37 (s), 72.29 (d, 5.4
Hz), 73.42 (d, 5.4 Hz), 73.78 (d, 35.7 Hz), 74.96 (d, 9.0 Hz), 75.55 (d,
13.0 Hz), 78.61 (dd, 41.6 Hz, 7.1 Hz), 122.16 (s), 123.56 (s), 125.63
(s), 126.43 (d 7.8 Hz), 127.82 (d 8.7 Hz), 128.34 (d, 11.6 Hz), 128.63
(s), 128.86 (s), 129.24 (s), 130.21 (s), 132.77 (d, 50.4 Hz), 134.62 (d,
11.0 Hz), 135.10 (d, 37.2 Hz), 136.88 (s), 153.19 (s), 156.02 (dd, 127.6
Hz, 7.3 Hz). 31P{1H} NMR: (toluene) δ 23.4 (d, 34.8 Hz), 14.6 (d,
34,8 Hz). Anal. Calcd for C55H49FeNP2PdS: C, 67.39; H, 5.04; N,
1.43. Found: C, 67.02; H, 4.91; N, 1.74.
Preparation of (DPPF)Pd(p-NMe2C6H4)(Br). {[P(o-tolyl)3](p-
27
NMe2C6H4)Pd(µ-Br)}2 (240 mg, 0.197 mmol) was suspended in 4
mL of THF. DPPF (240 mg, 0.433 mmol) was added to the stirred
suspension as a solid. The reaction was allowed to stir at room
temperature for 10 min. The product was precipitated by the addition
of 15 mL of pentane. The product was collected as a solid by filtration
on a fritted glass funnel to yield 281 mg (83%). The product was
used without further purification. 1H NMR: (C6D6) δ 2.46 (s, 6H),
3.67 (m, 2H), 3.74 (m, 2H), 3.93 (m, 2H), 4.53 (m, 2H), 6.30 (d, 7.0
Hz, 2H), 6.85-7.06 (m, 14H), 7.44-7.50 (m, 4H), 8.35-8.41 (m, 4H).
31P{1H} NMR: (C6D6) δ 29.2 (d, 33.6 Hz), 8.5 (d, 33.6 Hz).
Preparation of (DPPF)Pd(C6H5)[N(p-CH3C6H4)2] (3a). (DPPF)-
Pd(C6H5)(I) (174 mg, 0.201 mmol) was suspended in 15 mL of toluene.
KN(p-CH3C6H4)2 (52 mg, 0.221 mmol) was added to the solution as a
solid. The reaction was stirred at room temperature for 30 min, and it
turned a deep red color. The reaction mixture was filtered through
Celite, and the toluene was removed under vacuum. The resulting solid
was dissolved in 8 mL of THF. The THF solution was concentrated
under vacuum and layered with Et2O. Pure crystalline material was
obtained by cooling the layered solution at -35 °C for 12 h. The yield
of 3a was 129 mg (69%). 1H NMR: (C6D6) δ 2.23 (s, 6H), 3.72 (m,
2H), 3.94 (m, 2H), 4.15 (m, 2H), 4.48 (m, 2H), 6.55 (m, 3H), 6.89 (d,
8.2 Hz, 4H), 7.00-7.06 (m, 14H), 7.32 (d, 8.2 Hz), 4H, 7.65 (m, 4H),
7.83 (m, 4H). 13C{1H} NMR: (C6D6) δ 20.98 (s), 72.22 (d, 6.1 Hz),
73.38 (d, 5.6 Hz), 73.85 (d, 34.1 Hz), 74.98 (d, 8.9 Hz), 75.56 (d, 11.3
Hz), 78.97 (dd, 41.3 Hz, 8.9 Hz), 121.98 (s), 123.45 (s), 126.25 (d, 8.8
Hz), 127.51 (d, 9.1 Hz), 127.81 (d, 9.6 Hz), 128.05 (s), 128.62 (s),
129.91 (s), 129.91 (s), 132.89 (d, 49.9 Hz), 134.45 (d, 11.8 Hz), 135.15
(d, 12.9 Hz), 135.26 (d, 34.2 Hz), 136.40 (s), 153.30 (s), 158.80 (dd,
121.2 Hz, 8.1 Hz). 31P{1H} NMR: (THF) δ 23.5 (d, 35.4 Hz), 14.9
(d, 35.4 Hz). IR (cm-1): (KBr) 3051 (m), 3010 (m), 2908 (m), 2854
(m), 1598 (s), 1566 (s), 1549 (s), 1490 (s), 1432 (m), 1389 (s), 1314
(s), 1292 (m), 1259 (s), 1219 (s), 1200 (w), 1170 (w), 1098 (m), 1063
(s), 1016 (s), 995 (m), 899 (w), 886 (w), 866 (m), 839 (m), 829 (m),
809 (s), 799 (s), 772 (s), 746 (s), 728 (s), 691 (m), 638 (w), 631 (w),
587 (w), 565 (m), 540 (s), 514 (s), 488 (m), 159 (m). Anal. Calcd for
C54H47FeNP2Pd: C 69.43, H 5.07, N 1.50. Found: C 69.18, H 4.93,
N 1.25.
Preparation of (DPPF)Pd(p-ClC6H4)[N(p-CH3C6H4)2] (3f). Fol-
lowing the procedure for the preparation of 3a, (DPPF)Pd(p-ClC6H4)-
(I) (176 mg. 0.196 mmol) was reacted with KN(p-CH3C6H4)2 (51 mg,
0.217 mmol) in THF. Recrystallization at -35 °C from a concentrated
THF solution layered with ether gave 118 mg of pure 3f (62%). 1H
NMR: (C6D6) δ 2.22 (s, 6H), 3.70 (m, 2H) 3.94 (m, 2H), 4.13, (m,
2H), 4.45 (m, 2H), 6.57 (d, 6.7 Hz, 2H), 6.83-6.88 (m, 6H), 6.98-
7.10 (m, 12H), 7.27 (d, 8.2 Hz, 4H), 7.55-7.62 (m, 4H), 7.75-7.82
(m, 4H). 13C{1H} NMR: (C6D6) δ 20.92 (s) 72.29 (d, 5.5 Hz), 73.42
(d, 7.9 Hz), 73.78 (d, 34.3 Hz), 74.91 (d, 8.6 Hz), 75.61 (d, 11.1 Hz),
78.38 (dd. 43.4 Hz, 8.2 Hz), 122.11 (s), 123.77 (s), 126.12 (d, 8,5 Hz),
127.73 (d, 8.5 Hz), 127.89 (s), 128.13 (d, 11.8 Hz), 128.25 (s), 128.88
(s), 130.29 (d, 11.8 Hz), 132.50 (d, 50.5 Hz), 134.51 (d, 11.4 Hz),
134.91 (d, 36.4 Hz), 135.31 (d, 12.8 Hz), 137.46 (s), 153.03 (s), 157.05
(dd, 125.2 Hz, 8.5 Hz). 31P{1H} NMR: (toluene) δ 23.5 (d, 34.2 Hz),
15.2 (d, 34.2 Hz); Anal. Calcd for C54H46ClFeNP2Pd: C, 66.96; H,
4.79; N, 1.45. Found: C, 66.74; H, 4.95; N, 1.30.
Preparation of (DPPF)Pd(p-NMe2C6H4)[N(p-CH3C6H4)2] (3b).
Following the procedure for the preparation of 3a, (DPPF)Pd(p-
NMe2C6H4)(Br) (180 mg, 0.209 mmol) was reacted with KN(p-
CH3C6H4)2 (52 mg, 0.221 mmol) in THF. Recrystallization at -35
°C from a concentrated THF solution layered with ether gave 159 mg
of pure 3b (78%). 1H NMR: (C6D6) δ 2.24(s, 6H), 2.37 (s, 6H), 3.74
(m, 2H), 3.95 (m, 2H), 4.17 (m, 2H), 4.49 (m, 2H), 6.10 (d, 7.1 Hz,
2H), 6.84 (t, 7.9 Hz, 2H), 6.90 (d, 8.2 Hz, 4H), 7.00-7.07 (m, 12H),
7.36 (d, 8.2 Hz, 4H), 7.66-7.70 (m, 4H), 7.83-7.86 (m, 4H). 13C-
{1H} NMR: (C6D6) δ 21.00 (s), 41.03 (s), 72.13 (d, 5.6 Hz), 73.28 (d,
Preparation of (PPh3)Pd(C6H4-C6H4NH2)(I) (4). Into a 50 mL
screw-capped test tube, 602 mg (0.522 mmol) of Pd[P(C6H5)3]4 was

8244 J. Am. Chem. Soc., Vol. 119, No. 35, 1997
DriVer and Hartwig
suspended in 25 mL of toluene. To this suspension, a solution of 231
mg (0.783 mmol) of 2′-amino-2-iodobiphenyl in 10 mL of toluene was
added. The test tube was sealed and removed from the drybox. The
reaction mixture was heated at 50 °C for 12 h during which time the
product precipitated out of solution as a beige solid. The tube was
taken back into the drybox, and the product was collected by filtration
on a medium fritted funnel. The product was washed with 2 × 3 mL
of Et2O to yield 292 mg (84%) of 4. 1H NMR: (CDCl3, 65 °C) δ
5.13 (broad, 2H), 6.51 (d, 4.0 Hz, 2H), 6.90 (m, 1H), 7.3-7.4 (m,
11H), 7.45-7.6 (m, 9H). 31P NMR: (THF) δ 38.9. IR (cm-1): (KBr)
3304 (m), 3233 (m), 3064 (w), 3034 (w), 1565 (m), 1492 (m), 1436
(m), 1100 (m), 1033 (s), 867 (m), 743 (s), 732 (s), 705 (s). Anal.
Calcd for C30H25NIPPd: C, 54.28; H, 3.80; N, 2.11. Found: C, 54.22;
H, 3.91; N, 2.14.
solution was transferred to an NMR tube. The NMR tube was cooled
at -35 °C in the drybox freezer. LiNH-i-Bu (1.6 mg, 0.020 mmol)
was added to the NMR tube as a solid. The tube was quickly placed
in the NMR: probe at -80 °C. The probe was warmed until reaction
occurred at 0 °C. 31P{1H} NMR: (toluene, 0 °C) δ 26.5 (d, 22.0 Hz),
1
18.3 (d, 22.0 Hz). Selected resolved H NMR resonances: (C7D8) δ
2.55 (m, CH2), 1.96 (m, NH), 0.53, d, 7.5 Hz, CH3).
Generation of (DPPF)Pd(Ph)(NHPh) (8b). (DPPF)Pd(Ph)(I) (12.6
mg, 0.0146 mmol) was suspended in 0.6 mL of THF, and the solution
was transferred to an NMR tube. The NMR tube was cooled at -35
°C in the drybox freezer. KNHPh (2.1 mg, 0.016 mmol) was added to
the NMR tube as a solid. The tube was quickly placed in the NMR
probe cooled at -80 °C. The probe was incrementally warmed until
reaction occurred at 0 °C. 31P{1H} NMR: (THF, 0 °C) δ 25.1 (d, 38
Hz), 6.3 (d, 38 Hz).
Preparation of anti-{[P(C6H5)3]Pd(p-CH3C6H4)(µ-NH-t-Bu)}2 (6).
Into a 20 mL vial was weighed 352 mg (0.415 mmol) of [P(C6H5)3]2-
Pd(p-CH3C6H4)(I). This material was suspended in 10 mL of Et2O
and to the resulting suspension was added 36 mg (0.46 mmol) of LiNH-
t-Bu as a solid. The reaction was stirred for 24 h, over which time the
white suspension became pale yellow. The Et2O was decanted from
the solid. Fresh Et2O (10 mL) was then added and 15 mg (0.069 mmol)
of p-CH3C6H4I was added to react with the Pd(PPh3)4 byproduct. After
15 min, the solid was isolated by filtration using a medium fritted
funnel. The solid was then extracted with 3 × 4 mL of toluene, leaving
behind the palladium halide complex. The toluene was then concen-
trated. The product began to crystallize upon concentration and was
isolated by pipetting off the supernatant. Pentane (10 mL) was then
added to the supernatant to isolate the remaining product. The solvent
mixture was cooled at -35 °C for 5 h. The total yield was 46 mg
(20%). 1H NMR: (C6D6) δ -0.10 (d, 5.3 Hz, 2H), 1.27 (s, 18H),
2.20 (s, 6H), 6.65 (broad s, 2H), 6.90 (broad s, 2H) 6.95-7.06 (m,
18H), 7.55 (broad s, 2H), 7.59 (broad s, 2H), 7.61(t, 8.4 Hz, 12H). 1H
NMR: (C6D6, 75 °C) δ -0.07 (d, 5.3 Hz, 2H), 1.26 (s, 18H), 2.16 (s,
6H), 6.71 (d, 6.0 Hz, 4H), 6.91-7.09 (m, 18H), 7.26 (d, 6.0 Hz, 4H),
7.5(t, 8.4 Hz, 12H). 31P NMR: (toluene) δ 25.7. IR (cm-1): (KBr)
3301 (w), 3070 (w), 3050 (m), 2957 (m), 2916 (m), 2861 (m), 1582
(m), 1481 (s), 1432 (s), 1384 (m), 1358 (m), 1350 (m), 1187 (s), 1096
(s), 1010 (s), 917 (s), 905 (s) 791 (s), 746 (s), 698 (s). Anal. Calcd
for C58H64N2P2Pd2: C, 65.48; H, 6.06; N, 2.63. Found: C, 65.49; H,
6.09; N, 2.55.
Generation of 8b with [15N]KNHPh. (DPPF)Pd(Ph)(I) (15.0 mg,
0.0174 mmol) was dissolved in 0.4 mL of THF. [15N]KNHPh (2.4
mg, 0.018 mmol) was dissolved in 0.2 mL of THF, and the two
solutions were combined. The reaction mixture was transferred to an
NMR tube and quickly placed in the NMR probe cooled to 0 °C. 31P-
{1H} NMR: (THF, 0 °C) δ 25.1 (dd, JPP ) 38 Hz, JNP ) 43 Hz) 6.3
(dd, JPP ) 38 Hz, JNP ) 4 Hz). 15N{1H} NMR: (THF, 0 °C) δ 69.5
(d, JNP ) 43 Hz).
Generation of 9 with [15N]KNHPh. (DPPF)Pd(Ph)(I) (15.4 mg,
0.0178 mmol) was dissolved in 0.6 mL of THF. [15N]KNHPh (22.1
mg, 0.167 mmol) was added as a solid. The reaction was transferred
to an NMR tube and quickly placed into the NMR probe, which was
cooled to 0 °C. 31P{1H} NMR: (THF, 0 °C) δ 22.3 (d, JNP ) 46 Hz),
-16.6 (s); 15N{1H} NMR: (THF, 0 °C) δ 89.2 (s), 86.5 (d, JNP ) 46
Hz).
Preparation of anti-{[P(C6H5)3]Pd(C6H5)(µ-NH-t-Bu)}2 (6b). Into
a 20 mL vial was weighed 357 mg of [P(C6H5)3]2Pd(C6H5)(I) (0.428
mmol). This material was suspended in 15 mL of Et2O, and 27 mg
(.34 mmol) of LiNH-t-Bu was added to the suspension as a solid. The
reaction mixture was stirred for 2 h. To the reaction mixture was then
added 30 mg (0.147 mmol) of C6H5I, and the resulting solution was
stirred for an additional 30 min. The reaction was filtered through a
medium fritted funnel to remove the solid palladium aryl halide
complex. The ether was removed under vacuum to give a yellow oily
residue. The oily residue was layered with pentane and cooled at -35
°C for 12 h to give 62 mg (28%) of product. 1H NMR: (C6D6, 55 °C)
δ -0.11 (d, 4.8 Hz, 2H), 1.22 (s, 18H), 6.86 (t, 5.3 Hz, 4H), 6.90-
7.13 (m, 24H), 7.26 (t, 8.2 Hz, 12H). 31P NMR: (toluene) δ 26.0. IR
(cm-1): (KBr) 3317 (w), 3303 (w), 3049 (s), 2955 (s), 2861 (m), 1563
(s), 1479 (m), 1468 (m), 1434 (s), 1354 (w), 1185 (s), 1094 (s) 1058
(w), 1021 (m), 996 (w), 922 (w), 906 (w), 888 (w), 804 (w) 745 (m),
729 (s), 697 (s).
Kinetic Analysis of the Reductive Elimination Reaction of 1. Into
a vial was weighed 20 mg (0.023 mmol) of 1. The solid was dissolved
in approximately 5 mL of anhydrous THF and transferred to a 10 mL
volumetric flask. The volumetric flask was then filled with enough
THF to make 10 mL of a 0.0023 M solution. Into another vial was
weighed the appropriate amount of PPh3. A 4.5 mL aliquot of the
palladium solution was syringed from the volumetric flask into the vial
containing the phosphine. Once all of the phosphine was dissolved,
the solution was transferred into a quartz UV-vis cell equipped with
a stopcock. The reaction was heated at 80 °C in a constant temperature
water bath and a UV-vis spectrum was taken every 30 min. The
concentration of palladium amide was monitored at λ ) 489.6 nm
throughout the course of the reaction. The following pseudo-first-order
rate constants were obtained at different concentrations of PPh3 (k,
[PPh3]): 1.7 × 10-4 s-1, 0.023 M; 1.3 × 10-4 s-1, 0.034 M; 8.0 ×
10-5 s-1, 0.045 M; 7.3 × 10-5 s-1, 0.057 M; 4.8 × 10-5 s-1, 0.068 M.
Kinetic Analysis of the Reductive Elimination Reaction of 2. Into
a vial was weighed 30.2 mg (0.0334 mmol) of 2. The solid was
dissolved in approximately 3 mL of toluene-d8 and transferred into a
5 mL volumetric flask. The volumetric flask was filled with toluene-
d8 to make 5 mL of a solution that had [2] ) 0.0067 M. Triphen-
ylphosphine-d15 was weighed into a separate vial. A 0.5 mL aliquot
of the volumetric solution was transferred by syringe to the vial
containing the phosphine. Upon dissolution of the phosphine, the
solution was transferred to an NMR tube. The tube was cooled in a
dry ice/acetone bath and flame sealed. This procedure results in a tube
Preparation of anti-{[P(C6H5)3]Pd(C6H5)(µ-NHC6H5)}2 (7). In
a 20 mL vial, 283 mg (0.339 mmol) of (PPh3)2Pd(Ph)I was stirred in
15 mL of toluene. To this suspension was added 66.8 mg (0.509 mmol)
of KNHPh as a solid. The reaction was stirred at room temperature
for 2 h. A 31P{1H} NMR spectrum of the reaction mixture indicated
complete consumption of (PPh3)2Pd(Ph)I and the formation of 7 and
Pd(PPh3)4. The reaction mixture was filtered through Celite, and the
toluene solvent was removed by vacuum to obtain a dry solid. To the
resulting solids was added 6 mL of fresh toluene. This suspension
was stirred for 2 h to extract Pd(PPh3)4 from the product 7. The
suspension was filtered through a medium fritted funnel, and 7 was
collected as a solid. Recrystallization of the solid at -35 °C from a
concentrated THF solution layered with Et2O gave pure crystalline 7
in 28% yield (50.2 mg). 1H NMR: (toluene-d8) δ 1.22 (d, 4.3 Hz,
1H), 1.42 (d, 5.3 Hz, 1H), 6.58-6.69 (m, 4H), 6.74-6.99 (m, 16H),
7.11-7.16 (m, 18H), 7.24-7.44 (m, 12H); 31P NMR (THF) δ 29.73
(s), 29.69 (s). IR (cm-1): (KBr) 3307 (w), 3044 (m), 3028 (m), 2998
(w), 2966 (w), 2922 (w), 1592 (s), 1564 (s), 1485 (s), 1471 (m), 1437
(s), 1340 (w), 1310 (w), 1226 (s), 1205 (m), 1100 (s), 1063 (m), 1023
(s), 998 (m), 816 (s), 742 (s). Anal. Calcd for C60H52N2P2Pd2: C,
66.99; H, 4.87; N, 2.60. Found: C, 66.81; H, 4.79; N, 2.54.
Generation of [(PPh3)Pd(C6H4-C6H4NH)]2 (5). A solution of 10
mg (0.015 mmol) of (PPh3)Pd(C6H4-C6H4NH2)(I) in 0.5 mL of toluene-
d8 was placed in an NMR tube and cooled at -40 °C. To the cooled
solution was added a solution of 3.0 mg (0.015 mmol) of KN(TMS)2
in 0.2 mL of toluene-d8. The tube was placed into an NMR probe
which had been cooled to -40 °C. The NMR probe was slowly heated
until reaction occurred at -15 °C. 1H NMR: (toluene-d8, 20 °C) δ
1.96 (d, 7.6 Hz, 1H), 6.54 (t, 7.2 Hz, 1H), 6.62 (m, 1H), 6.73 (t, 7.4
Hz, 1H), 6.80-6.97 (m, 12H), 7.43-7.50 (m, 8H). 31P NMR: (THF)
δ 31.9. IR (cm-1) (C7D8): 3289 (w, NH).
Generation of (DPPF)Pd(Ph)(NH-i-Bu) (8a). (DPPF)Pd(Ph)(I)
(12.1 mg, 0.0140 mmol) was suspended in 0.6 mL of toluene, and the

C-N-Bond-Forming ReductiVe Elimination
J. Am. Chem. Soc., Vol. 119, No. 35, 1997 8245
sealed at atmospheric pressure at room temperature. Reaction rates
were measured by H NMR spectroscopy at 110 °C. Samples were
vial along with a small amount (<1 mg) of trimethoxybenzene.
Toluene-d8 (0.4 mL) was syringed into the vials to dissolve the solids.
The solutions were transferred to NMR tubes and sealed as described
above. Kinetic data was obtained as described above. The results of
the kinetic analysis are summarized in Table 3.
X-ray Structure of 3b. A red plate crystal of C56H52N2P2-
FePd‚C3H6O0.75 having approximate dimensions of 0.08 × 0.10 × 0.24
mm was obtained by layering a THF solution of 3b with ether at -35
°C and was mounted on a glass fiber. Data was collected at -60 °C
on an Enraf-Nonius CAD-4 diffractometer with graphite monochro-
mated Mo KR radiation.
Cell constants and an orientation matrix for data collection obtained
from a least-squares refinement using the setting angles of 18 carefully
centered reflections in the range 5.24 < 2θ < 13.47° corresponded to
a triclinic cell with dimensions given in Table 2. On the basis of
packing considerations, a statistical analysis of intensity distribution,
and the successful solution and refinement of the structure the space
group was determined to be: P-1 (no. 2).
1
shimmed at room temperature and removed. The NMR probe was then
heated to 110 °C. The sample was placed into the probe, quickly
reshimmed, and an automated program was initiated that collected single
pulse experiments with at least 1 min between data acquisitions. The
resonance for the tolyl CH3 was used to monitor the concentration of
palladium amido complex throughout the course of the reaction. The
following pseudo-first-order rate constants were obtained at different
concentrations of PPh3 (k, [PPh3]): 5.0 × 10-3 s-1, 0.052 M; 4.0 ×
10-3 s-1, 0.061 M; 4.4 × 10-3 s-1, 0.079 M; 3.3 × 10-3 s-1, 0.087 M;
3.2 × 10-3 s-1, 0.10 M; 2.6 × 10-3 s-1, 0.14 M; 2.4 × 10-3 s-1, 0.16
M; 2.2 × 10-3 s-1, 0.19 M; 2.0 × 10-3 s-1, 0.19 M; 1.2 × 10-3 s-1
,
0.47 M.
Kinetic Analysis of the Reductive Elimination Reaction of 3a-
f. A stock solution was made by dissolving 101 mg (0.385 mmol) of
PPh3 in 3.5 mL of toluene-d8. The palladium amido complex was
weighed into a small vial: 5.9 mg (0.0063 mmol) 3a, 6.2 mg (0.0064
mmol) 3b, 6.1 mg (0.0063 mmol) 3c, 6.0 mg (0.0063 mmol) 3d, 6.2
mg (0.0063 mmol) 3e, 6.1 mg (0.0063 mmol) 3f. A 0.5 mL aliquot of
the PPh3 stock solution was transferred by syringe into each of the six
vials. Upon dissolution, each sample was transferred to an NMR
sample tube. The tubes were frozen in N2(l) and flame sealed, resulting
in a tube sealed at atmospheric pressure at room temperature. Reaction
rates were measured by 1H NMR spectroscopy at 75 °C using a
procedure analogous to that for monitoring reaction of 2. The results
of the kinetic analysis are summarized in Table 2.
Of the 10 832 reflections which were collected 10 281 were unique
(Rint ) 0.109). The intensities of three representative reflections, which
were measured after every 60 min of X-ray exposure time, remained
constant throughout data collection indicating crystal and electronic
stability (no decay correction was applied). The linear absorption
coefficient for Mo KR is 7.4 cm-1
. Azimuthal scans of several
reflections indicated no need for an absorption correction. The data
were corrected for Lorentz and polarization effects.
The structure was solved by the Patterson method. The non-
hydrogen atoms were refined anisotropically. The hydrogen atoms were
included in calculated positions. In the case of the methyl group
hydrogens, one hydrogen atom was located in the difference map and
included at an idealized distance to set the orientation of the other
hydrogen atoms. The THF solvent molecule appeared disordered and
the multiplicity of the THF molecule was approximately 0.75%.
The standard deviation of an observation of unit weight was 1.49.
The weighting scheme was based on counting statistics and included a
factor (p ) 0.02) to downweight the intense reflections. Plots of ∑w(|Fo
- |Fc|)2 versus |Fo|, reflection order in data collection, sin θ/λ, and
various classes of indices showed no unusual trends. The maximum
and minimum peaks on the final difference Fourier map corresponded
to 0.67 and -0.51 e/Å3 respectively.
Phosphine Dependence on the Reductive Elimination Reaction
of 3f. A stock solution was prepared by dissolving 12.1 mg (0.0126
mmol) of 3f in 1.2 mL of toluene-d8. PPh3 (6.9 mg, 0.026 mmol or
20.6 mg, 0.079 mmol) was weighed into a small vial. A 0.5 mL aliquot
of the stock solution was transferred to each of the two vials. Upon
dissolution, each sample was transferred to an NMR tube. The tubes
were frozen in N2(l) and flame sealed, resulting in a tube sealed at
atmospheric pressure at room temperature. Reaction rates were
1
measured by H NMR spectroscopy at 75 °C as described above.
Kinetic Analysis of the Reductive Elimination Reaction of 6.
Triphenylphosphine (6.8 mg, 0.026 mmol or 20.2 mmol 0.0771 mmol)
and 6 (3.0 mg, 0.0043 mmol or 7.2 mg, 0.010 mmol) and a small
amount (<1 mg) of trimethoxybenzene were weighed into the same
vial. Toluene-d8 (0.65 mL) was syringed into each vial to dissolve the
solids. Once the solids were dissolved, the solutions were transferred
to NMR tubes. The tubes were frozen in N2(l) to create a vacuum and
were flame sealed. This procedure results in a tube sealed at
atmospheric pressure. The samples were heated at 95 °C in a constant
temperature water bath, and 1H NMR spectra were taken periodically.
The concentration of palladium amide complex was measured by the
integration of the NH resonance or tolyl resonance relative to the
trimethoxybenzene internal standard. The results of the kinetic analysis
are summarized in Table 3.
Neutral atom scattering factors were taken from Cromer and Waber.99
Anomalous dispersion effects were included in Fcalc; the values for ∆f ′
and ∆f ′′ were those of Cromer. All calculations were performed using
the TEXSAN crystallographic software package of Molecular Structure
Corporation.
Acknowledgment. This work was generously supported by
the Department of Energy (DE-RG02-96ER14678). We also
gratefully acknowledge support from a DuPont Young Professor
Award, a Union Carbide Innovative Recognition Award, a
National Science Foundation Young Investigator Award, and a
Dreyfus Foundation New Faculty Award, and a Camille Dreyfus
Teacher-Scholar award for support for this work. J.F.H. is a
fellow of the Alfred P. Sloan Foundation. We thank Johnson-
Matthey Alpha/Aesar for donation of palladium chloride.
Kinetic Analysis of the Reductive Elimination Reaction of 7. A
stock solution was prepared by dissolving 30 mg of triphenylphosphine-
d
15 (0.11 mmol) in 1.2 mL of toluene-d8. Compound 7 (3.1 mg, 0.0029
mmol or 12.1 mg, 0.0113 mmol) was weighed into a small vial with
a small amount (<1 mg) of trimethoxybenzene. A 0.5 mL aliquot of
the stock solution was transferred by syringe to each of the vials to
dissolve the solids. The solutions were transferred to NMR tubes. The
tubes were frozen in N2(l) to create a vacuum and were flame sealed.
This procedure results in a tube sealed at atmospheric pressure. The
samples were heated at 95 °C in a constant temperature water bath,
Supporting Information Available: Tables of positional
parameters and B(eq), U values, intramolecular distances,
intramolecular bond angles, Cartesian coordinates, and torsion
or conformational angles for hydrogen and non-hydrogen atoms
and a fully labeled PLUTO diagram (21 pages). See any current
masthead page for ordering and Internet access instructions. (3b).
1
and H NMR spectra were taken periodically. The concentration of
palladium amide complex was measured by the integration of an
aromatic or NH resonance relative to the trimethoxybenzene internal
standard. The results of the kinetic analysis are summarized in Table
3.
JA971057X
Phosphine Dependence of the Reductive Elimination Reaction
of 7. Triphenylphosphine-d15 (5.3 mg, 0.019 mmol or 15.9 mg, 0.0573
mmol) and 7 (4.1 mg, 0.0038 mmol) were weighed into the same small
(99) Cromer, D. T.; Waber, J. T. International Tables for X-ray
Crystallography; The Kynoch Press: Birmingham England, 1974; Vol. IV,
Table 2.3.1.