New insights into xantphos/Pd-catalyzed C-N bond forming reactions: A structural and kinetic study

Article abstract of DOI:10.1021/om050715g

Both (Xantphos)Pd(dba) and Pd(Xantphos)₂ were identified in mixtures of Pd₂(dba)₃ and Xantphos by use of ³¹P NMR and independent syntheses. At high ligand concentrations, Pd(Xantphos)₂ was found to be the predominant species. Reaction calorimetry was employed to determine whether the formation of Pd(Xantphos) ₂ affects the rate at which the Pd-catalyzed C-N bond formation occurs between 4-tert-butylbromobenzene and morpholine. Indeed, the concentration of Xantphos dramatically influences the activity of the catalyst, with high concentrations of Xantphos inhibiting the reaction rate due to the formation of Pd(Xantphos)₂. Two plausible hypotheses for the low activity of Pd(Xantphos)₂ as a precatalyst are (1) a slow rate of Xantphos dissociation from Pd(Xantphos)₂ inhibits the formation of an active (Xantphos)Pd⁰ species and (2) the high degree of insolubility of Pd(Xantphos)₂ results in the precipitation of a significant fraction of the precatalyst from the reaction mixture. Although the equilibrium constant for ligand dissociation could not be determined by either magnetization transfer or variable-temperature experiments due to its slow rate, the more soluble Pd(4,7-di-tert-butylXantphos)₂ complex demonstrated different activity relative to Pd(Xantphos)₂. A comprehensive study of these two complexes indicates that the lower activity of the bis-ligated Pd species is a result of a combination of aforementioned processes, i.e., the slow rate of ligand dissociation and the insolubility of Pd(Xantphos)₂, with the rate of ligand dissociation serving as the primary turnover-limiting factor.

Full text of DOI:10.1021/om050715g

82
Organometallics 2006, 25, 82-91
New Insights into Xantphos/Pd-Catalyzed C-N Bond Forming
Reactions: A Structural and Kinetic Study
Liane M. Klingensmith, Eric R. Strieter, Timothy E. Barder, and Stephen L. Buchwald*
Department of Chemistry, Massachusetts Institute of Technology, Cambridge, Massachusetts 02139
ReceiVed August 18, 2005
Both (Xantphos)Pd(dba) and Pd(Xantphos)2 were identified in mixtures of Pd2(dba)3 and Xantphos by
use of 31P NMR and independent syntheses. At high ligand concentrations, Pd(Xantphos)2 was found to
be the predominant species. Reaction calorimetry was employed to determine whether the formation of
Pd(Xantphos)2 affects the rate at which the Pd-catalyzed C-N bond formation occurs between 4-tert-
butylbromobenzene and morpholine. Indeed, the concentration of Xantphos dramatically influences the
activity of the catalyst, with high concentrations of Xantphos inhibiting the reaction rate due to the
formation of Pd(Xantphos)2. Two plausible hypotheses for the low activity of Pd(Xantphos)2 as a precatalyst
are (1) a slow rate of Xantphos dissociation from Pd(Xantphos)2 inhibits the formation of an active
(Xantphos)Pd0 species and (2) the high degree of insolubility of Pd(Xantphos)2 results in the precipitation
of a significant fraction of the precatalyst from the reaction mixture. Although the equilibrium constant
for ligand dissociation could not be determined by either magnetization transfer or variable-temperature
experiments due to its slow rate, the more soluble Pd(4,7-di-tert-butylXantphos)2 complex demonstrated
different activity relative to Pd(Xantphos)2. A comprehensive study of these two complexes indicates
that the lower activity of the bis-ligated Pd species is a result of a combination of aforementioned processes,
i.e., the slow rate of ligand dissociation and the insolubility of Pd(Xantphos)2, with the rate of ligand
dissociation serving as the primary turnover-limiting factor.
Introduction
Palladium-catalyzed C-N bond forming reactions have
become one of the most important cross-coupling reactions in
synthetic organic chemistry. Recently, advances have been made
with the development of catalyst systems that exhibit increased
selectivity and wide substrate scope.1 One catalyst system, based
on the ligand Xantphos2 (Figure 1), has been particularly
successful in broadening the substrate scope of amination
reactions.3 By utilizing a Pd/Xantphos catalyst system, diffi-
cult reactions such as the N-arylation of heteroarylamines
and amides,3e-g,5c the coupling of amines with ortho-function-
Figure 1. Xantphos (1).
alized base-sensitive aryl halides,4 the amination of aryl
nonaflates,3i and the N-arylation of 2-oxazolidinones3j can be
accomplished.
Although Pd/Xantphos catalyst systems have been employed
in various types of reactions, the reason behind their efficiency
is largely unknown.5 Van Leeuwen and co-workers initially
employed Xantphos in Pd-catalyzed amination reactions3a after
the observation that other large bite-angle ligands such as
DPEphos6 could function as excellent supporting ligands for
Pd. Both Van Leeuwen5b and our group5c have demonstrated
that Xantphos can serve as a trans-chelating ligand when bound
to Pd(II). The bite-angle of Xantphos in these complexes is
around 153°, which is much larger than the calculated flexible
bite-angle (97-135°).7 The cause of this large bite-angle may
be the interaction between the oxygen atom on the Xantphos
ligand and palladium center, which is present in the crystal
structure of a oxidative addition product, i.e., (Xantphos)Pd-
(Ar)Br (where Ar ) 4-cyanophenyl) complex.3f Additionally,
it is possible that the Pd-O interaction facilitates key steps in
the catalytic process by promoting the dissociation of one
phosphine moiety of Xantphos from the metal center. More
specifically, reductive elimination from a trans-chelating com-
plex must be preceded by a trans to cis isomerization, an event
(1) For two recent reviews see: (a) Jiang, L.; Buchwald, S. L. Metal-
Catalyzed Cross-Coupling Reactions, 2nd ed.; Wiley-Interscience: Wein-
heim. (b) Hartwig, J. F. Handbook of Organopalladium Chemistry for
Organic Synthesis; Negishi, E., Ed.; Wiley-Interscience: Weinheim, 2002.
(2) Kranenburg, M.; van der Burgt, Y. E. M.; Kamer, P. C. J.; van
Leeuwen, P. W. N. M. Organometallics 1995, 14, 3081.
(3) (a) Guari, Y.; van Es, D. S.; Reek, J. N. H.; Kamer, P. C. J.; van
Leeuwen, P. W. N. M. Tetrahedron Lett. 1999, 40, 3789. (b) Harris, M.
C.; Geis, O.; Buchwald, S. L. J. Org. Chem. 1999, 64, 6019. (c) Yang, B.
H.; Buchwald, S. L. Org. Lett. 1999, 1, 35. (d) Wagaw, S.; Yang, B. H.;
Buchwald, S. L. J. Am. Chem. Soc. 1999, 121, 10251. (e) Yin, J.; Buchwald,
S. L. Org. Lett. 2000, 2, 1101. (f) Yin, J.; Buchwald, S. L. J. Am. Chem.
Soc. 2002, 124, 6043. (g) Yin, J.; Zhao, M. M.; Huffman, M. A.; McNamara,
J. M. Org. Lett. 2002, 4, 3481. (h) Ji, J.; Li, T.; Bunnelle, W. H. Org. Lett.
2003, 5, 4611. (i) Anderson, K. A.; Menez-Perez, M.; Priego, J.; Buchwald,
S. L. J. Org. Chem. 2003, 68, 9563. (j) Cacchi, S.; Fabrizi, G.; Goggiamani,
A.; Zappia, G. Org. Lett. 2001, 3, 2539. (k) Artamkina, G. A.; Sergeev, A.
G.; Beletskaya, I. P. Tetrahedron Lett. 2001, 42, 4381. (l) Sergeev, A. G.;
Artamkina, G. A.; Beletskaya, I. P. Tetrahedron Lett. 2003, 44, 4719. (m)
Browning, R. G.; Mahmud, H.; Badarinarayana, V.; Lovely, C. J.
Tetrahedron Lett. 2001, 42, 7155. (n) Anbazhagan, M.; Stephens, C. E.;
Boykin, D. W. Tetrahedron Lett. 2002, 43, 4221. (o) Jean, L.; Rouden, J.;
Maddaluno, J.; Lasne, M.-C. J. Org. Chem. 2004, 69, 8893.
(4) Anderson, K. A.; Buchwald, S. L. Unpublished results.
(5) For Xantphos studies see: (a) Guari, Y.; van Strijdonck, G. P. F.;
Boele, M. D. K.; Reek, J. N. H.; Kamer, P. C. J.; van Leeuwen, P. W. N.
M. Chem. Eur. J. 2001, 7, 475. (b) Kamer, P. C. J.; Van Leeuwen, P. W.
N. M.; Reek, J. N. H. Acc. Chem. Res. 2001, 34, 895. (c) Yin, J.; Buchwald,
S. L. J. Am. Chem. Soc. 2002, 124, 6043.
(6) Sadighi, J. P.; Harris, M. C.; Buchwald, S. L. Tetrahedron Lett. 1998,
39, 5327.
(7) Van der Veen, L. A.; Keeven, P. K.; Schoemaker, G. C.; Reek, J. N.
H.; Kamer, P. C. J.; van Leeuwen, P. W. N. M.; Lutz, M.; Spek, A. L.
Organometallics 2000, 19, 872.
10.1021/om050715g CCC: $33.50 © 2006 American Chemical Society
Publication on Web 12/08/2005

Xantphos/Pd-Catalyzed C-N Bond Forming Reactions
Figure 2. 4,7-Di-tert-butylXantphos (2).
Organometallics, Vol. 25, No. 1, 2006 83
catalyzed coupling of 4-tert-butylbromobenzene and morpholine
using the ligand Xantphos at varying palladium-to-ligand ratios.
The activity of the catalyst is dramatically dependent on the
concentration of ligand relative to palladium due to formation
of the much less reactive Pd(Xantphos)2 species. Magnetization
transfer experiments were used to probe the hypothesis that the
bis-ligated Pd species has low reactivity due to a large binding
constant for ligand on the bis-ligated species, and reaction
calorimetry was performed using Pd(Xantphos)2 and Pd(4,7-
di-tert-butylXantphos)2 as precatalysts to identify whether the
solubility of the bis-ligated species affects reaction rate. It was
discovered that the main cause for the low reactivity of Pd-
(Xantphos)2 is a large binding constant for Xantphos from the
bis-ligated species, causing the generation of the active (Xant-
phos)Pd0 species to be relatively slow. Additionally, the
insoluble nature of Pd(Xantphos)2 in THF and toluene is
apparent by examining the differences in reactivity between Pd-
(Xantphos)2 and Pd(4,7-di-tert-butylXantphos)2.
that may occur through dissociation of one phosphine group of
the ligand. 5b
Van Leeuwen and co-workers have performed a detailed
kinetic analysis of these reactions employing both a (Xantphos)-
Pd(Ar)Br complex and a cationic (Xantphos)Pd(Ar)OTf
complex.5a However, these studies, which are based only on
initial rates, may not capture the entire kinetic picture of the
reaction, as recent studies have shown anomalous rate behavior
while measuring reaction kinetics under these conditions.8a
Our goal was to obtain insight into the composition of the
catalyst based upon Xantphos and Pd in C-N bond forming
reactions. These findings not only would yield important
mechanistic insight but would also be advantageous in designing
new ligands that could induce better selectivity and substrate
scope than is currently possible. To accomplish this goal, the
more soluble 4,7-di-tert-butylXantphos (2) (Figure 2)9 was used
in NMR studies and in separate syntheses of possible precatalytic
species to aid in the identification of in-situ-generated pre-
catalytic species. Both (4,7-di-tert-butylXantphos)Pd(dba) and
Pd(4,7-di-tert-butylXantphos)2 were identified as precatalytic
species when mixtures of Pd2(dba)3 and 4,7-di-tert-butylXant-
phos were analyzed by 31P NMR. The relative reactivity of the
two precatalysts follows the order (4,7-di-tert-butylXantphos)-
Pd(dba) . Pd(4,7-di-tert-butylXantphos)2 based on 31P NMR
studies of the coupling of 4-tert-butylbromobenzene and mor-
pholine using the Pd/4,7-di-tert-butylXantphos system. Reaction
calorimetry was also used to explore the rate of the palladium-
Results and Discussion
Initial studies focused on determining the structure(s) of the
Xantphos/Pd precatalyst. We wanted to ascertain the nature and
distribution of species present in a mixture of Xantphos and
Pd2(dba)3 by 31P NMR. However, due to the low solubility of
Xantphos and Pd2(dba)3 in common organic solvents, experi-
ments utilizing standard NMR techniques could not be con-
ducted. To circumvent this issue, initial studies were conducted
using the more soluble 4,7-di-tert-butylXantphos, 2. The 31P
NMR spectrum of a mixture of Pd2(dba)3 and 4,7-di-tert-
butylXantphos in a 1:1 L:Pd ratio exhibits two separate
resonances, corresponding to two distinct species (Figure 3).
The 31P NMR spectrum of complex B has an intricate splitting
pattern at low temperature that begins to coalesce to one peak
Figure 3. 31P NMR spectra of the precatalyst mixture (1:1 Pd:L) at 0 and 50 °C.

84 Organometallics, Vol. 25, No. 1, 2006
Klingensmith et al.
Figure 4. 31P NMR spectra of the precatalyst mixture (1:1 and 1:3 Pd:L) at RT.
Scheme 1. Synthesis of Various (4,7-Di-tert-butylXantphos)Pd Complexes
at higher temperatures. The other resonances, ascribed to
atoms bound to one palladium center. The 31P NMR spectrum
complex A, are consistent with two inequivalent phosphorus
of a mixture similar to that above, but with an excess of 2 (i.e.,

Xantphos/Pd-Catalyzed C-N Bond Forming Reactions
Organometallics, Vol. 25, No. 1, 2006 85
Figure 5. 31P NMR spectra of palladium complexes derived from (4,7-di-tert-butylXantphos)Pd(COT). The spectra of Pd(4,7-di-tert-
butylXantphos)2 and (4,7-di-tert-butylXantphos)Pd(dba) were obtained in THF-d8, while the spectrum of the precatalyst mixture was obtained
in PhMe-d8.
3:1 L:Pd), indicated the presence of only one species, aside from
unbound ligand, that corresponds to complex B (Figure 4). This
is suggestive of the conversion of complex A into complex B
at high ligand concentrations. The same experiment was at-
tempted with Xantphos; however, an insoluble yellow-green
precipitate formed, rendering a solution state NMR analysis im-
possible. Fortunately, Pd(Xantphos)2 could be isolated from a
mixture of Xantphos and Pd2(dba)3 as a yellow solid10 and was
characterized by both MALDI-TOF-MS and elemental analysis.
To identify the resonances observed in the 31P NMR of the
precatalyst mixture, complexes 3, 5, and 6 were independently
prepared (Scheme 1). (4,7-Di-tert-butylXantphos)PdCl2 (3) was
synthesized using a procedure analogous to that used in the
preparation of PdCl2(dppf).11 By utilizing the straightforward
route to bisphosphine-cyclooctatetraene Pd(0) complexes de-
veloped by Brown and Cooley,12 4 was assembled from 3. As
cyclooctatetraene is an extremely labile ligand, both (4,7-di-
tert-butylXantphos)Pd(dba)13 (5) and Pd(4,7-di-tert-butylXant-
phos)2 (6) can be easily obtained from 3; the 31P NMR spectra
are illustrated in Figure 5. (4,7-Di-tert-butylXantphos)Pd(dba)
corresponds to the resonances at 9.8 and 12.4 ppm. The two
phosphorus atoms on palladium are inequivalent due to the
nonsymmetric bonding of dba ligand on the palladium center,
which accounts for the observed P-P coupling. The 31P NMR
resonances for the phosphorus atoms in Pd(tert-butylXantphos)2
appear at 0.8 and 3.4 ppm (in THF). As illustrated in Figure 4,
(8) (a) Singh, U. K.; Strieter, E. R.; Blackmond, D. G.; Buchwald, S. L.
J. Am. Chem. Soc. 2002, 124, 14104. (b) Nielsen, L. P. C.; Stevenson, C.
P.; Blackmond, D. G.; Jacobsen, E. N. J. Am. Chem. Soc. 2004, 126, 1360.
(c) Streiter, E. R.; Blackmond, D. G.; Buchwald, S. L. J. Am. Chem. Soc.
2005, 127, 4120.
(11) Hayashi, T.; Konishi, M.; Kobori, Y.; Kumada, M.; Higuchi, T.;
Hirotsu, K. J. Am. Chem. Soc. 1984, 106, 158.
(9) van der Veen, L. A.; Kamer, P. C. J.; van Leeuwen, P. W. N. M.
Angew. Chem., Int. Ed. 1999, 38, 336.
(10) Once isolated, Pd(Xantphos)2 appears green if there is a small
amount of palladium black present.
(12) (a) Brown, J. M.; Cooley, N. A. Organometallics 1990, 9, 353. (b)
Katz, T. J.; Garratt, P. J. J. Am. Chem. Soc. 1964, 86, 4876.
(13) This is most likely the complex formed; however, attempts to isolate
and purify this complex have not been successful.

86 Organometallics, Vol. 25, No. 1, 2006
Klingensmith et al.
Figure 7. Space-filling model of Pd(4,7-di-tert-butylXantphos)2.
Gray ) carbon, red ) oxygen, orange ) phosphorus, and pink )
palladium.
Table 1. Selected Bond Lengths (Å) and Bond Angles (deg)
for Pd(4,7-di-tert-butylXantphos)2
bond
length
bond
angle
Pd(1)-P(1)
Pd(1)-P(2)
Pd(2)-P(3)
Pd(2)-P(4)
Pd(2)-P(5)
Pd(2)-P(6)
2.3884(12)
2.3454(11)
2.3857(11)
2.3798(11)
2.3870(10)
2.3998(12)
P(1)-Pd(1)-P(2)
P(1)-Pd(1)-P(1A)
P(3)-Pd(2)-P(4)
P(3)-Pd(2)-P(5)
P(3)-Pd(2)-P(6)
P(4)-Pd(2)-P(5)
P(4)-Pd(2)-P(6)
P(5)-Pd(2)-P(6)
108.73(4)
106.49(4)
108.79(4)
113.47(4)
108.94(4)
108.08(4)
108.05(4)
109.38(4)
7). The distortions from a tetrahedral geometry most likely exist
due to the highly crowded and geometrically strained Pd center
constantly attempting to reside in a minimum energy structure.
To test the relative reactivity of (4,7-di-tert-butylXantphos)-
Pd(dba) and Pd(tert-butylXantphos)2 in an amination reaction,
a slurry of NaOt-Am, morpholine, and 4-tert-butylbromobenzene
was injected into a NMR tube, sealed with a Teflon septum,
containing a precatalyst mixture as prepared above. The NMR
tube was quickly placed in the spectrometer preheated to 60
°C. It was found that Pd(tert-butylXantphos)2 remains un-
changed, while a large quantity of (4,7-di-tert-butylXantphos)-
Pd(dba) is converted into other unidentified species.15 Presum-
ably, these species stem from amine and/or NaOt-Am addition
to the dibenzylideneacetone ligand bound to the Pd center.
Importantly, these data suggest that Pd(tert-butylXantphos)2 is
relatively unreactive compared to (4,7-di-tert-butylXantphos)-
Pd(dba).
Reaction calorimetry experiments were conducted to verify
that the rate of reaction was affected by the ratio of (Xantphos)-
Pd(dba) to Pd(Xantphos)2 in the precatalyst mixture. The
palladium-catalyzed coupling of 4-tert-butylbromobenzene and
morpholine at 60 °C (Figure 8) was used in this investigation
due to its rapid reaction rate and high conversion of aryl bromide
to product.16 Our initial studies revealed that the Pd:Xantphos
ratio has a dramatic effect on reaction rate, as shown in Figure
8. Interestingly, the initial addition of Xantphos increases the
rate of C-N bond formation; however, at higher Xantphos
concentrations the reaction rate is suppressed. Ultimately, a
Xantphos:Pd ratio of 1:1 is optimal for maximum turnover
frequency, suggesting that the active catalyst contains only one
Figure 6. (a) ORTEP diagram of Pd(4,7-di-tert-butylXantPhos)2
with hydrogen atoms, benzene molecule, and ether molecule
removed for clarity. Thermal ellipsoids are at 30% probability. (b)
ORTEP diagram of Pd(4,7-di-tert-butylXantPhos)2 with hydrogens,
phenyl groups, benzene molecule, and ether molecule removed for
clarity. Thermal ellipsoids are at 30% probability.
the 31P NMR spectrum of this complex at 0 °C possesses a
complex splitting pattern, suggesting that 6 does not possess
tetrahedral geometry.
An X-ray-quality crystal of 614 was grown by preparing the
complex by the route shown in Scheme 1 and subsequently
layering a solution of 6 in benzene with ether. The ORTEP
diagram is shown in Figure 6a. To allow for better interpretation,
the phenyl groups have been removed for clarity, as shown in
Figure 6b. The asymmetric unit contains a molecule of Pd(4,7-
di-tert-butylXantphos)2 with slightly distorted tetrahedral ge-
ometry (Pd-P bond distances and angles in Table 1) as well as
a half-molecule of Pd(tert-butylXantphos)2 possessing C2 sym-
metry. It is apparent from both solution state NMR studies,
conducted at RT and 60 °C, and X-ray crystallography,
conducted at -173 °C, that this complex is fluxional in both
states likely due to the crowding around the Pd center (Figure
(14) The only other palladium(0) bis-xanthene crystal structure reported
is Pd(ethylXantphos)2 (in this case, it is diethylphosphines instead of
diphenylphosphines on the xanthene backbone): Raebiger, J. W.; Miedaner,
A.; Curtis, C. J.; Miller, S. M.; Anderson, O. P.; DuBois, D. L. J. Am.
Chem. Soc. 2004, 126, 5502.
(15) See Supporting Information for NMR spectra.
(16) Toluene was used to make a stock solution of Xantphos so that a
more accurate amount of Xantphos could be injected into the reaction vessel.
Toluene was used rather than 1,4-dioxane due to solubility problems.

Xantphos/Pd-Catalyzed C-N Bond Forming Reactions
Organometallics, Vol. 25, No. 1, 2006 87
Scheme 2. Equilibria of Palladium/Xantphos Species
Xantphos ligand. Indeed, both Sigman and Stahl have recently
observed similar behavior while examining the rate dependence
on ligand concentration during the palladium-catalyzed aerobic
oxidation of alcohols.17 Thus, on the basis of both the 31P NMR
and calorimetric experiments, it is believed that at higher
concentrations of Xantphos formation of Pd(Xantphos)2 occurs,
which dominates the Pd speciation, thereby leading to a decrease
in the reaction rate. Others have reported trends that can account
for the formation of this species during Pd-catalyzed C-N bond
forming processes. For example, our group has reported that
“an unusual dependence on catalyst loading”, where the
reactions being studied were more efficient at lower catalyst
concentrations, was observed.5c Moreover, Yin has reported that
difficult reactions employing Xantphos are often run at low
concentrations (0.25-0.13 M), which would minimize the
formation of Pd(Xantphos)2.3g
We found that in this case exchange is too slow to measure
before relaxation occurs, thus preventing the quantification of
the dissociation constant. These data suggest that the binding
constant of 2 is very large in Pd(4,7-di-tert-butylXantphos)2,
which is consistent with its poor activity as a precatalyst.
To test the hypothesis that the insolubility of Pd(Xantphos)2
restricts its activity as a precatalyst, experiments were performed
at varying ligand-to-palladium ratios with both Xantphos and
4,7-di-tert-butylXantphos (Figure 9). At high ligand-to-pal-
ladium ratios (>1.6), the catalyst based upon 2 continues to
promote the reaction. However, performing the reaction under
identical conditions, except for substitution of Xantphos for 4,7-
di-tert-butylXantphos, did not yield product, due to the formation
of the inactive Pd(Xantphos)2 species. However, at L:Pd ratios
between 1 and 1.6, the kinetic rate profiles are essentially
identical for Xantphos and 4,7-di-tert-butylXantphos.
There exist as least two possible hypotheses that could
account for the low activity of Pd(Xantphos)2 in Pd-catalyzed
C-N bond forming reactions. The first premise is that an
equilibrium process exists between the bis-ligated palladium
species and the mono-ligated palladium species in solution, and
this equilibrium favors the bis-ligated species (Scheme 2). The
second possibility is that a solubility equilibrium process exists
that favors the unsolvated Pd(Xantphos)2 and therefore causes
the majority of the Pd to reside as a highly insoluble precipitate,
unable to enter the catalytic cycle.
Attempts were made using a 31P NMR magnetization transfer
experiment to measure the rate of dissociation of a ligand from
Pd(Xantphos)2, similar to a procedure that has been previously
reported by Grubbs and co-workers.18 A solution containing
Pd(4,7-di-tert-butylXantphos)2 and 4,7-di-tert-butylXantphos
(1.0:1.5) in benzene-d6 was allowed to reach equilibrium in an
NMR probe. Once this occurred, the resonance corresponding
to the free ligand was selectively inverted by a 180° pulse. After
variable mixing times between 0.01 and 5.12 s, a nonselective
90° pulse was applied. Using this method, if free ligand were
to exchange rapidly with ligand bound to the Pd center, the
peak area for the complex would decrease following exchange
with the inverted signal of the free ligand. However, even at
60 °C no change in the value of the peak area was observed,
implying no appreciable exchange (or less exchange than is
capable of being detected by NMR) was occurring between the
complexed and free ligand. This method is useful only for rates
of exchange that are large enough relative to (1) the relaxation
rate of the complexed ligand and (2) the relaxation rate of the
free ligand. That is, the rate of magnetization loss due to
exchange of the complexed ligand with free ligand must be large
relative to the rate at which the NMR signal of the complexed
ligand returns to equilibrium and/or the rate at which the free
ligand relaxes back to equilibrium following selective inversion.
To further examine the hypothesis that the insolubility of Pd-
(Xantphos)2 can restrict its activity as a precatalyst, reactions
were catalyzed by employing both Pd(Xantphos)2 and Pd(tert-
butylXantphos)2 as precatalysts. Rate versus time plots are
depicted in Figure 10a, and a plot of rate versus fractional
conversion is depicted in Figure 10b for the coupling of 4-tert-
butylbromobenzene and morpholine.
The reaction catalyzed by Pd(Xantphos)2 exhibits a very
interesting kinetic profile in that the rate is slowly increasing
throughout the reaction. This is in contrast to the kinetic profile
of the corresponding reaction with Pd(4,7-di-tert-butylXant-
phos)2, where the rate is relatively constant throughout the
reaction, corresponding to zero-order kinetics. These kinetic data
are very similar to an observation of zero-order kinetics made
(17) (a) Schultz, M. J.; Adler, R. S.; Zierkiewicz, W.; Privalov, T.;
Sigman, M. S. J. Am. Chem. Soc. 2005, 127, 8499. (b) Steinhoff, B. A.;
Guzei, I. A.; Stahl, S. S. J. Am. Chem. Soc. 2004, 126, 11268. (c) Steinhoff,
B. A.; Stahl, S. S. Org. Lett. 2002, 4, 4179.
(18) Sanford, M. S.; Love, Jennifer, A.; Grubbs, R. H. J. Am. Chem.
Soc. 2001, 123, 6543.
Figure 8. Reaction rate vs [Xantphos]:[Pd] ratio. [ArBr]0 ) 0.25
M; [Amine]0 ) 0.30 M (1.2 equiv); [NaOtAm]0 ) 0.35 M (1.4
equiv); Pd2(dba)3 2.5 mol % Pd based on ArBr; Xantphos 2-4
mol % based on ArBr; 3 mL of 1,4-dioxane and 1 mL of toluene.
Reaction rate is at 10% conversion of ArBr.

88 Organometallics, Vol. 25, No. 1, 2006
Klingensmith et al.
Figure 9. Plot of rate vs [Ligand]:[Pd] for both Xantphos (blue
line) and 4,7-di-tert-butylXantphos (red line). [ArBr]0 ) 0.25 M;
[Amine]0 ) 0.30 M (1.2 equiv); [NaOtAm]0 ) 0.35 M (1.4 equiv);
Pd2(dba)3 2.5 mol % Pd based on ArBr; ligand 2.0-4.0 mol %
based on ArBr; 3 mL of 1,4-dioxane and 1 mL of toluene.
both by van Leeuwen5a and by Hartwig19 while studying
palladium-catalyzed C-N bond formation using a Pd/Xantphos
and a Pd/BINAP catalyst system, respectively. Blackmond and
Buchwald8a later demonstrated that the zero-order dependence
on substrate may arise from the slow rate of active catalyst
formation. This would be the case if a slow dissociation of 2
from Pd(4,7-di-tert-butylXantphos)2 was occurring. A slow
dissociation process is also occurring from Pd(Xantphos)2 in
addition to a slow equilibrium allowing more precatalyst into
solution, a process that accounts for the increasing rate of
reaction.
Conclusions
Products from mixing Xantphos and Pd2(dba)3 were identified
as Pd(Xantphos)2 and (Xantphos)Pd(dba). This was accom-
plished by separately synthesizing the analogous 4,7-di-tert-
butylXantphos species and comparing 31P NMR spectra. (Xant-
phos)Pd(dba) serves as the precatalyst, and Pd(Xantphos)2
demonstrates extremely low activity as a precatalyst.
Figure 10. (a) Reaction rate vs time for reactions catalyzed by
Pd(4,7-di-tert-butylXantphos)2 (red line) and Pd(Xantphos)2 (blue
line). [ArBr]0 ) 0.25 M; [Amine]0 ) 0.30 M (1.2 equiv);
[NaOtAm]0 ) 0.35 M (1.4 equiv); 2.5% Pd(L)2 based on [ArBr];
4 mL of 1,4-dioxane. (b) Reaction rate vs fractional conversion
for reactions catalyzed by Pd(4,7-di-tert-butylXantphos)2 (red line)
and Pd(Xantphos)2 (blue line). [ArBr]0 ) 0.25 M; [Amine]0 ) 0.30
M (1.2 equiv); [NaOtAm]0 ) 0.35 M (1.4 equiv); 2.5% Pd(L)2
based on [ArBr]; 4 mL of 1,4-dioxane. ∆Hrxn ) 203.6 kJ/mol for
Pd(4,7-di-tert-butylXantphos)2 and ∆Hrxn ) 214.1 kJ/mol for Pd-
(Xantphos)2.
Reaction rates were essentially identical for reactions cata-
lyzed by Xantphos and 4,7-di-tert-butylXantphos at L:Pd ratios
between 1.0 and 1.5, suggesting that the insolubility of Pd-
(Xantphos)2 is not the cause of its low activity, since Pd(tert-
butylXantphos)2 is entirely soluble. Furthermore, it was dem-
onstrated through reaction calorimetry that Pd(Xantphos)2 is in
equilibrium with Pd(Xantphos) by use of Pd(Xantphos)2 as the
precatalyst. This equilibrium was probed by use of a magnetiza-
tion transfer experiment, and it was found that formation of
Pd(Xantphos)2 in the palladium-catalyzed amination of 4-tert-
butylbromobenzene significantly decreases the rate of reaction,
not due to solubility, but due to a very large binding constant
for the ligand on the bis-ligated species, which causes a slow
generation of mono-ligated active catalyst. This is the first time
the equilibrium of phosphine dissociation from a bis(xanthane-
based phosphine)Pd(0) complex has been examined and sub-
sequently applied to determining the detrimental effect of using
a ratio of g 1.5:1 L:Pd, where L ) Xantphos, in C-N bond
forming reactions.
Materials and Methods
Reagents. Toluene and THF were purchased from J. T. Baker
in CYCLE-TAINER solvent delivery kegs and vigorously purged
with argon for 2 h. The solvent was further purified by passing it
under argon pressure through two packed columns of neutral
alumina (THF) or through neutral alumina and copper(II) oxide
(toluene). 1,4-Dioxane, benzene, and morpholine were purchased
from Aldrich Chemical Co. in SureSeal containers and taken into
a glovebox before use. Xantphos, dichlorobis(acetonitrile)palladium-
(II), tris(dibenzylideneacetone)dipalladium(0), and lithium granules
(19) Alcazar-Roman, L. M.; Hartwig, J. F.; Rheingold, A. L.; Liable-
Sands, L. M.; Guzei, I. A. J. Am. Chem. Soc. 2000, 122, 4618.

Xantphos/Pd-Catalyzed C-N Bond Forming Reactions
Organometallics, Vol. 25, No. 1, 2006 89
were acquired from Strem Chemicals, Inc. and used without further
purification. 1-Bromo-4-tert-butylbenzene was purchased from
Aldrich Chemical Co. and distilled from CaH2 prior to use. Sodium
tert-amylate (NaOtAm), purchased from Aldrich, was stored and
used inside of the glovebox. 4,5-Dibromo-2,7-di-tert-butyl-9,9-
dimethylxanthene, n-BuLi (2.5 M in SureSeal bottle), and cyclooc-
tatetraene were purchased from Aldrich Chemical Co. and used
without further purification. Chlorodiphenylphosphine (98%) was
purchased from Strem Chemical Co. and distilled over CaH2 under
reduced pressure prior to use. All reagents used in reaction
calorimetry experiments were handled and stored in a nitrogen-
filled glovebox, except for tris(dibenzylideneacetone)dipalladium-
(0), which was weighed in air into a septum-sealed vial. This vial
was then evacuated/backfilled with argon three times before it was
taken into a glovebox.
1
Analytical Methods. H NMR spectra were obtained on either
a Bruker 400 MHz or a Varian Mercury 300 MHz spectrometer,
with chemical shifts reported with respect to residual solvent peaks.
31P{1H} NMR spectra were obtained on either a Varian 500 MHz
or a Varian Mercury 300 MHz, with chemical shifts reported with
respect to calibration with an external standard of phosphoric acid
(0 ppm). MALDI-TOF was performed on a Bruker Omniflex
calibrating externally with a ProteoMass Peptide MALDI-MS
calibration kit. Melting points (uncorrected) were obtained on a
Mel-Temp capillary melting point apparatus. Gas chromatographic
analyses were performed on a Hewlett-Packard 6890 gas chroma-
tography instrument with an FID detector using a 25 m × 0.20
mm capillary column with cross-linked methyl siloxane as a
stationary phase. Elemental analyses were obtained from Atlantic
Microlab, Inc. (Norcross, GA).
Figure 11. Fraction conversion vs time for calorimetric and GC
data. [ArBr]0 ) 0.25 M; [Amine]0 ) 0.30 M (1.2 equiv);
[NaOtAm]0 ) 0.35 M (1.4 equiv; 1:1 Xantphos:Pd from Pd2(dba)3
2.5 mol % Pd based on ArBr; 3 mL of 1,4-dioxane and 1 mL of
toluene.
produced by measuring the heat flow from the sample vessel every
6 s during the reaction. Due to the delay between the instantaneous
heat flow being evolved from the reaction vessel and the time the
thermophile sensor detects the heat flow, the raw data curve must
be calibrated. To accomplish this calibration, a constant amount of
current was passed through a resistor in the sample chamber of the
calorimeter, thereby producing a known quantity of heat. This
process results in a response curve, which is then transformed into
a square wave, allowing for the response time of the instrument to
be calculated using the WinCRC software. Application of the
response time to the raw data results in a “tau-corrected data curve”.
The tau-corrected data curve is a plot of heat flow (mJ s-1) versus
time. The reaction rate, which is directly proportional to the heat
flow (eq 1), fractional conversion (eq 2), and instantaneous
concentrations of reactants/products can all be calculated from this
tau-corrected data curve.
Reaction Calorimetry Experimental Details. Reactions were
performed in either an Omnical SuperCRC or an Omnical Reactmax
reaction calorimeter. The instrument contains an internal magnetic
stirrer and a differential scanning calorimeter (DSC), which
compares the heat released or consumed in a sample vessel to an
empty reference vessel. The reaction vessels were 16 mL borosili-
cate screw-thread vials fit with open-top black phenolic screw caps
and white PTFE septa (KimbleBrand) charged with Teflon-coated
magnetic stir bars. Sample volumes did not exceed 4.2 mL. A stock
solution of Xantphos or tert-butylXantphos was made by dissolving
0.250 mmol of ligand (145 mg of Xantphos, 173 mg of tert-
butylXantphos) in 5.0 mL of toluene in a volumetric flask to make
a 0.050 M solution. Pd2(dba)3 was weighed in air and brought into
the glovebox by evacuating and backfilling a small vial with argon
three times. Dioxane (1 mL) was then added and stirred to form a
slurry. This slurry was then added to the previously weighed
NaOtAm in the reaction vessel. The desired amount of ligand was
added by taking a portion of the stock solution and diluting to a
total volume of 1.0 mL in toluene, so that each reaction contained
a constant amount of toluene with varying amounts of ligand (i.e.,
if 0.025 mmol of ligand was desired, 0.50 mL of the stock solution
was delivered to a vial and 0.50 mL of toluene was added; if 0.03
mmol of ligand was desired, 0.60 mL of the stock solution was
delivered to a vial, and 0.40 mL of toluene was added). This solution
of ligand was added to the calorimeter vial containing the NaOtAm
and Pd2(dba)3, and finally 2.0 mL of dioxane was added and the
reaction vessel was sealed. This vessel was then removed from the
glovebox and placed in the calorimeter and stirred for 1 h, allowing
the contents of the vessel to reach thermal equilibrium. Simulta-
neously, a syringe containing 1-bromo-4-tert-butylbenzene and
morpholine was placed in the sample injection port of the
calorimeter and was allowed to thermally equilibrate. The reaction
was initiated by injecting the mixture of aryl bromide and amine
into the stirred catalyst/NaOtAm solution. The temperature of the
DSC was held constant at 333 K using the internal temperature
controller in the calorimeter, ensuring that the reaction would
proceed under isothermal conditions. A raw data curve was
q ) ∆HrxnVr
(1)
tq(t) dt
∫
∫
t0
fractional conversion )
(2)
tq(t) dt
t0
As a control, conversion measured by GC analysis was compared
to conversion measured by heat flow (Figure 11). Agreement
between the two curves suggests that calorimetric analysis was a
valid method for studying rates of this type of reaction.
Crystal Structure Determination of Pd(4,7-di-tert-butylXant-
phos)2. Crystals suitable for X-ray diffraction were obtained by
layering ether on a saturated solution of Pd(4,7-di-tert-butylXant-
Phos)2 in benzene in a glovebox. A single crystal (0.17 × 0.12 ×
0.09 mm3) was mounted on a magnetic glass pin and placed on
the goniometer head under a stream of N2 delivered from a
Cyrostream 700 at 100 K. A Siemens Platform three-circle
diffractometer equipped with an APEX CCD detetctor was used
to obtain the data. The crystal was exposed to Mo KR radiation (λ
) 0.71073 Å), collecting 10 s frames, of which 230 195 measured
and 26 495 independent reflections were observed, with Rint
)
0.1009 in C2/c (space group #15), to d ) 0.80 (2θ ) 52.78°). Data
were processed using SAINT supplied by Siemens Industrial
Automation, Inc., and structure determination was completed by

90 Organometallics, Vol. 25, No. 1, 2006
Klingensmith et al.
1
direct methods using SHELXTL, V6.10, G. M. Sheldrick, Univer-
sity of Go¨ttingen. The structure was refined on F2 by full-matrix
least-squares methods, and an absorption correction was applied
with SADABS. All non-hydrogen atoms were refined anistropically,
except for the extremely disorded ether molecule. All hydrogens
were placed in calculated positions and left to ride on their parent
atoms. The benzene molecule was disordered and each carbon was
set at half-occupancy. The refinement of 1465 parameters using
26 495 reflections and 0 restraints gave R1 ) 0.0637, wR2 ) 0.1639
[I > 2σ(I)], goodness of fit on F2 ) 1.070, ∆Fmax/min ) 1.398/-
1255, 1234, 1190, 1094, 741, 706, 692 cm-1. H NMR (CD2Cl2,
4
300 MHz): δ 7.70 (d, J(H,H) ) 1.70 Hz, 2H), 7.23 (m, 22H),
1.87 (s, 6H), 1.26 (s, 18H). 13C NMR (CD2Cl2, 500 MHz): δ 153.3
(m), 148.4 (m), 135.7 (m), 135.0 (m), 130.6 (s), 130.2 (s), 128.8
(s), 128.6 (s), 128.5 (s), 128.5 (s), 127.7 (s), 119.4 (m), 118.9 (m),
38.0 (m), 35.5 (s), 31.7 (s), 26.8 (bs) (complexity of spectrum due
to 31P-13C coupling). 31P NMR{1H} (CD2Cl2, 300 MHz): δ 23.3.
MALDI-MS: observed C47H48O2P2PdCl (complex-Cl); theoretical
829.1907 (26.6%), 830.1922 (64.9%), 831.1915 (100.0%), 832.1927
(60.2%), 833.1907 (99.7%), 834.1935(45.4%), 835.1913 (57.5%),
836.1938 (25.7%), 837.1923 (14.9%); found 829.2071 (35.8%),
830.2068 (76.1%), 831.2242 (100.0%), 832.1912 (70.5%), 833.2227
(97.3%), 834.2245 (51.2%), 835.1700 (59.4%), 836.2003 (25.6%),
837.1684 (5.5%).
0.967 e Å-3
.
Magnetization Transfer Experiment. Pd(4,7-di-tert-butylX-
antphos)2 (34.2 mg, 0.0230 mmol) and 4,7-di-tert-butylXantphos
(24.2 mg, 0.0350 mmol) were weighed inside of the glovebox into
a small vial. They were dissolved in benzene-d6 (0.70 mL) and
transferred to a screw-cap septum-sealed NMR tube. The tube was
equilibrated in the NMR probe at either 20 or 60 °C. The free tert-
butylXantphos was selectively inverted using a 180° pulse. After
variable mixing times between 0.0100 and 5.12 s, a nonselective
90° pulse was applied. Sixteen transients with a relaxation delay
of 35 s (T1 of Pd(4,7-di-tert-butylXantphos)2 is 1.17 s; T1 of 4,7-
di-tert-butylXantphos is 6.46 s) were needed to obtain a spectrum
with an acceptable signal-to-noise ratio. 1H decoupling was applied
during the 90° pulse. Integration values at the variable mixing times
for the complex were determined and found to remain constant,
meaning that no detectable exchange occurred between the com-
plexed and free ligand.
Sample Analysis of Precatalyst by 31P NMR. 4,7-Di-tert-
butylXantphos (69.9 mg, 0.100 mmol) and Pd2(dba)3 (45.9 mg,
0.0500 mmol) were dissolved in toluene (2.0 mL) and allowed to
stir for 2 h inside of a glovebox. This solution was filtered over a
glass frit to remove insoluble matter and concentrated to a volume
of 0.7 mL. This solution was then transferred to a septum-sealed
NMR tube.
Preparation of Cyclooctatetradienide Solution (0.30 M). This
preparation was the same as that used by Katz and co-workers.12
THF (16 mL) was added to a flame-dried three-neck 25 mL round-
bottom flask under argon and cooled to -78 °C. Lithium granules
(76 mg, 11 mmol, washed with hexanes to remove mineral oil)
were added under a positive flow of argon. Cyclooctatetraene (0.54
mL, 4.8 mmol) was then added via syringe. The mixture was stirred
overnight while warming to room temperature to form a green-
blue solution that could be stirred at room temperature until use.
Best results were obtained by using the solution the same day, but
it can be stored for up to 4 days with minimal decomposition.
Preparation of (4,7-Di-tert-butylXantphos)Pd(cyclooctatet-
raene) (4). Inside of a glovebox under nitrogen atmosphere, PdCl2-
(4,7-di-tert-butylXantphos) (170 mg, 0.20 mmol) was weighed into
a 25 mL round-bottom flask equipped with a stirbar. The flask was
sealed with a rubber septum and further sealed with black electrical
tape. The flask was removed from the glovebox, and THF (8.0
mL) was added to form a yellow slurry. The slurry was degassed
by three freeze/pump/thaw cycles and finally cooled to -78 °C.
Cyclooctatetradienide (0.70 mL of the 0.30 M solution in THF,
0.20 mmol) was added dropwise over 5 min and then allowed to
stir for 30 min to form a green slurry. Cannula transferring a portion
of this solution to a flame-dried septum-sealed NMR tube under
argon allowed for 31P NMR analysis. Decomposition will begin to
Material Preparation. Preparation of 4,7-Di-tert-butylXant-
phos (2). To a flame-dried 500 mL round-bottom flask was added
4,5-dibromo-2,7-di-tert-butyl-9,9-dimethylxanthene (5.0 g, 10.4
mmol), which was dissolved in THF (150 mL) under a positive
flow of argon. This solution was cooled to -78 °C, and n-BuLi
(8.8 mL of a 2.5 M solution in hexanes, 22 mmol) was added
dropwise over 20 min. The solution was allowed to stir at -78 °C
for 2 h, then chlorodiphenylphosphine (4.5 mL, 24 mmol) was
added dropwise over 45 min. With stirring, the solution was allowed
to warm to room temperature overnight. The solution was washed
with water (3 × 100 mL), dried over MgSO4, and concentrated
with the aid of a rotary evaporator to give a light yellow oil. With
vigorous stirring, EtOH (50 mL) was slowly added to the yellow
oil to form a slurry of crude 4,7-di-tert-butylXantphos, which was
filtered and recrystallized from toluene/EtOH to afford 6.13 g (85%)
of a white solid. 1H NMR (CD2Cl2, 300 MHz): δ 7.42 (d, 4J(H,H)
) 2.40 Hz, 2H), 7.24 (m, 20H), 6.55 (m, 2H), 1.67 (s, 6H), 1.11
(s, 18H). 31P NMR{1H} (CD2Cl2, 300 MHz): δ -16.3. Lit. mp
194-195 °C,10 experimental mp 194-195 °C.
Preparation of PdCl2(4,7-di-tert-butylXantphos) (3). This
procedure was adapted from Hayashi’s procedure to make PdCl2-
(dppf).11 A slurry of benzene (20 mL) and dichlorobis(acetonitrile)-
palladium(II) (518 mg, 2.00 mmol) was allowed to stir in a septum-
sealed 100 mL round-bottom flask in a glovebox. 4,7-Di-tert-
butylXantphos (1.38 g, 2.00 mmol) was dissolved in benzene (20
mL) and added slowly with stirring to the dichlorobis(acetonitrile)-
palladium(II) slurry. This mixture was allowed to stir for 12 h,
during which time a yellow precipitate, as well as an orange
solution, formed. The yellow solid was filtered over a glass frit in
a glovebox and washed with benzene (10 mL) and ether (10 mL)
until the supernatant was clear and finally dried under high vacuum
to afford 457 mg (29%) of a yellow solid. Mp: 171 °C dec. IR
(KBr): 3057, 2964, 2906, 2869, 1479, 1436, 1426, 1395, 1364,
occur at room temperature, and this complex was not isolable. 31
P
NMR{1H} (THF, 300 MHz): δ 11.384 (s).
Preparation of (4,7-Di-tert-butylXantphos)Pd(dba) (5). Diben-
zylideneacetone (280 mg, 1.2 mmol) was weighed into a 25 mL
round-bottom flask and evacuated/backfilled with argon three times.
The solid was dissolved in THF (5.0 mL), and the solution was
then subjected to three freeze/pump/thaw cycles. While still cold,
it was slowly cannula transferred to the (4,7-di-tert-butylXantphos)-
Pd(cyclooctatetraene) solution prepared above. This mixture was
stirred for 30 min while warming to room temperature to form a
red-yellow solution. All attempts to isolate this complex led to
decomposition. However, 31P NMR analysis prior to isolation
attempts revealed the species previously identified in precatalyst
solutions (see Figure 8). In solution before attempted isolation: 31P
NMR{1H} (THF, 300 MHz): δ 12.3 (d, 2J(P,P) ) 29.1 Hz), 9.82
2
(d, J(P,P) ) 29.4 Hz).
Preparation of Pd(4,7-di-tert-butylXantphos)2 (6). 4,7-Di-tert-
butylXantphos (830 mg, 1.2 mmol) was weighed into a 25 mL
round-bottom flask and evacuated/backfilled with argon three times.
The solid was dissolved in THF (5.0 mL) and then subjected to
three freeze/pump/thaw cycles. While still cold, it was slowly
cannula transferred to the (4,7-di-tert-butylXantphos)Pd(cyclooc-
tatetraene) solution prepared above. This mixture was allowed to
stir for 30 min while warming to room temperature to form a yellow
solution. This yellow solution was taken into the glovebox and
filtered. The resulting solution was concentrated, dissolved in
benzene, filtered, and finally layered with ether. Bright yellow
crystals formed, which were suitable for X-ray analysis (140 mg,
48%). Mp: 162 °C dec. IR (KBr): 3053, 2964, 2905, 2867, 2280,

Xantphos/Pd-Catalyzed C-N Bond Forming Reactions
Organometallics, Vol. 25, No. 1, 2006 91
1585, 1477, 1426, 1398, 1361, 1284, 1256, 1240, 742, 696 cm-1
.
use of MALDI-TOF-MS analysis. IR (KBr): 2924, 2854, 1461,
1398, 1377, 1222 cm-1. MALDI-MS: calcd for C78H64O2P4Pd:
theoretical 1260.2894 (22.9%), 1261.2909 (63.4%), 1262.2911
(100.0%), 1263.2907 (64.6%), 1264.2914 (77.2%), 1265.2933
(47.5%), 1266.2931 (35.5%), 1267.2949 (25.1%); Found 1260.3405
(24.0%), 1261.3285 (67.4%), 1262.3166 (100.0%), 1263.3162
(73.2%), 1264.3300 (79.3%), 1265.3424 (47.5%), 1266.3491
(35.5%), 1267.3104 (25.1%). Anal. Calcd for C78H64O2P4Pd: C,
74.14; H, 5.10. Found: C, 74.44; H, 4.97. Mp: 164 °C dec.
1H NMR (C6D6, 300 MHz): δ 7.05 (m, 48H), 1.78 (m, 12), 1.20
(m, 36H). 13C NMR (C6D6, 500 MHz): δ 154.4 (bs), 153.7 (bs),
145.3 (m), 141.7 (m), 139.7 (bs), 137.4 (bs), 135.0 (s), 133.9 (s),
132.7 (m), 131.2 (bs), 124.6 (m), 123.6 (m), 121.0 (m), 36.8 (s),
35.3, (m), 33.0 (s), 32.2 (m), 22.9 (bs). 31P NMR {1H} (C6D6, 300
MHz): δ 3.70 (m), 1.58 (m). Anal. Calcd for C94H96O2P4Pd: C,
75.87; H, 6.50. Found: C, 75.97; H, 6.48.
Preparation of Pd(Xantphos)2. Xantphos (579 mg, 1.00 mmol)
and Pd2(dba)3 (229 mg, 0.250 mmol) were weighed into a flame-
dried 500 mL round-bottom flask and evacuated/backfilled with
argon three times. Toluene (300 mL) was added, and the solution
was stirred for 4 h. The solution was then filtered with a cannula
filter into another flame-dried round-bottom flask under argon to
remove insoluble matter. This solution was concentrated slightly
and allowed to rest overnight so that any extra palladium black
would settle. The resulting solution was filtered again and finally
concentrated to dryness. At this point, the yellow solid was stirred
in toluene (100 mL) overnight to remove dibenzylidene acetone
and excess Xantphos. The remaining yellow solid was isolated by
filtration and is sparingly soluble in all common organic solvents.
The identity of the species was confirmed to be Pd(Xantphos)2 by
Acknowledgment. Support has been provided by the
National Institutes of Health (GM 58160), the American
Chemical Society (Organic Division Fellowship to E.R.S.,
sponsored by Albany Molecular Research), Merck, and Novartis.
Supporting Information Available: Tables of X-ray crystal-
lographic data, atomic coordinates, bond lengths and bond angles,
and anisotropic thermal parameters for Pd(4,7-di-tert-butylXant-
phos)2 (PDF and CIF files). This material is available free of charge
via the Internet at http://pubs.acs.org.
OM050715G