A highly active palladium catalyst for intermolecular hydroamination. Factors that control reactivity and additions of functionalized anilines to dienes and vinylarenes

Article abstract of DOI:10.1021/ja056003z

We report a catalyst for intermolecular hydroamination of vinylarenes that is substantially more active for this process than catalysts published previously. With this more reactive catalyst, we demonstrate that additions of amines to vinylarenes and dienes occur in the presence of potentially reactive functional groups, such as ketones with enolizable hydrogens, free alcohols, free carboxylic acids, free amides, nitriles, and esters. The catalyst for these reactions is generated from [Pd(η³-allyl)CI]₂ (with or without added AgOTf) or [Pd(CH₃CN)₄](BF ₄)₂ and Xantphos (9,9-dimethyl-4,5-bis(diphenylphosphino) xanthene), which generates complexes with large P-Pd-P bite angles. Studies on the rate of the C-N bond-forming step that occurs by attack of amine on an η³-phenethyl and an η³-allyl complex were conducted to determine the effect of the bite angle on the rate of this nucleophilic attack. Studies on model η³-benzyl complexes containing various bisphosphines showed that the nucleophilic attack was faster for complexes containing larger P-Pd-P bite angles. Studies of substituted unsymmetrical and unsubstituted symmetrical model η³-allyl complexes showed that nucleophilic attack on complexes ligated by Xantphos was faster than on complexes bearing ligands with smaller bite angles and that nucleophilic attack on unsymmetrical allyl complexes with larger bite angle ligands was faster than on unsymmetrical allyl complexes with smaller bite angle ligands. However, monitoring of catalytic reactions of dienes by ³¹P NMR spectroscopy showed that the concentration of active catalyst was the major factor that controlled rates for reactions of symmetrical dienes catalyzed by complexes of phosphines with smaller bite angles. The identity of the counterion also affected the rate of attack; reactions of allylpalladium complexes with chloride counterion occurred faster than reactions of allylpalladium complexes with triflate ortetrafluoroborate counterion. As is often observed, the dynamics of the allyl and benzyl complexes also depended on the identity of the counterion.

Full text of DOI:10.1021/ja056003z

Published on Web 01/21/2006
A Highly Active Palladium Catalyst for Intermolecular
Hydroamination. Factors that Control Reactivity and Additions
of Functionalized Anilines to Dienes and Vinylarenes
Adam M. Johns, Masaru Utsunomiya, Christopher D. Incarvito, and
John F. Hartwig*
Contribution from the Department of Chemistry, Yale UniVersity, P.O. Box 208107,
New HaVen, Connecticut 06520-8107
Received August 31, 2005; E-mail: John.Hartwig@yale.edu
Abstract: We report a catalyst for intermolecular hydroamination of vinylarenes that is substantially more
active for this process than catalysts published previously. With this more reactive catalyst, we demonstrate
that additions of amines to vinylarenes and dienes occur in the presence of potentially reactive functional
groups, such as ketones with enolizable hydrogens, free alcohols, free carboxylic acids, free amides, nitriles,
and esters. The catalyst for these reactions is generated from [Pd(η3-allyl)Cl]2 (with or without added AgOTf)
or [Pd(CH3CN)4](BF4)2 and Xantphos (9,9-dimethyl-4,5-bis(diphenylphosphino)xanthene), which generates
complexes with large P-Pd-P bite angles. Studies on the rate of the C-N bond-forming step that occurs
by attack of amine on an η3-phenethyl and an η3-allyl complex were conducted to determine the effect of
the bite angle on the rate of this nucleophilic attack. Studies on model η3-benzyl complexes containing
various bisphosphines showed that the nucleophilic attack was faster for complexes containing larger
P-Pd-P bite angles. Studies of substituted unsymmetrical and unsubstituted symmetrical model η3-allyl
complexes showed that nucleophilic attack on complexes ligated by Xantphos was faster than on complexes
bearing ligands with smaller bite angles and that nucleophilic attack on unsymmetrical allyl complexes
with larger bite angle ligands was faster than on unsymmetrical allyl complexes with smaller bite angle
ligands. However, monitoring of catalytic reactions of dienes by 31P NMR spectroscopy showed that the
concentration of active catalyst was the major factor that controlled rates for reactions of symmetrical dienes
catalyzed by complexes of phosphines with smaller bite angles. The identity of the counterion also affected
the rate of attack: reactions of allylpalladium complexes with chloride counterion occurred faster than
reactions of allylpalladium complexes with triflate or tetrafluoroborate counterion. As is often observed, the
dynamics of the allyl and benzyl complexes also depended on the identity of the counterion.
Introduction
metals, catalysts based on late transition metals are likely to be
more tolerant of polar functional groups in the reagents and to
Catalytic intermolecular hydroamination of olefins remains
a challenging transformation.1-3 To date, complexes of lan-
thanides,4 group IV transition metals,5 and late transition
metals6-11 have been shown to catalyze the intermolecular
addition of amine N-H bonds across vinylarenes and allenes.
Recently, reports of intramolecular hydroaminations of alkenes
have also appeared.1,12-20 Although the lanthanide catalysts have
displayed higher reactivity toward olefins than late transition
be amenable to assembly in air. Thus, if the activities of late
metal catalysts for additions of amines to olefins could be
increased, such that they rival those of early metal catalysts,
then the current advantages of both classes of catalyst could be
gained.
Because an amino group is harder than some functional
groups yet softer than others, it is not clear which functional
groups will be compatible with catalytic hydroamination.
(1) Beller, M.; Breindl, C.; Eichberger, M.; Hartung, C. G.; Seayad, J.; Thiel,
O. R.; Tillack, A.; Trauthwein, H. Synlett 2002, 1579-1594.
(2) Mu¨ller, T. E.; Beller, M. Chem. ReV. 1998, 98, 675-703.
(3) Nobis, M.; Driessen-Holscher, B. Angew. Chem., Int. Ed. 2001, 40, 3983.
(4) Ryu, J. S.; Li, G. Y.; Marks, T. J. J. Am. Chem. Soc. 2003, 125, 12584-
12605.
(12) Bender, C. F.; Widenhoefer, R. A. J. Am. Chem. Soc. 2005, 127, 1070-
1071.
(13) Bexrud, J. A.; Beard, J. D.; Leitch, D. C.; Schafer, L. L. Org. Lett. 2005,
7, 1959-1962.
(14) Collin, J.; Daran, J. C.; Jacquet, O.; Schulz, E.; Trifonov, A. Chem.sEur.
J. 2005, 11, 3455-3462.
(5) Walsh, P. J.; Baranger, A. M.; Bergman, R. G. J. Am. Chem. Soc. 1992,
114, 1708-1719.
(15) Crimmin, M. R.; Casely, I. J.; Hill, M. S. J. Am. Chem. Soc. 2005, 127,
2042-2043.
(6) Al-Masum, M.; Meguro, M.; Yamamoto, Y. Tetrahedron Lett. 1997, 38,
6071-6074.
(16) Hong, S.; Marks, T. J. Acc. Chem. Res. 2004, 37, 673-686.
(17) Hultzsch, K. C.; Hampel, F.; Wagner, T. Organometallics 2004, 23, 2601-
2612.
(7) Besson, L.; Gore, J.; Gazes, B. Tetrahedron Lett. 1995, 36, 3857-3860.
(8) Kawatsura, M.; Hartwig, J. F. J. Am. Chem. Soc. 2000, 122, 9546-9547.
(9) Utsunomiya, M.; Hartwig, J. F. J. Am. Chem. Soc. 2003, 125, 14286-
14287.
(18) Kim, J. Y.; Livinghouse, T. Org. Lett. 2005, 7, 1737-1739.
(19) Lauterwasser, F.; Hayes, P. G.; Brase, S.; Piers, W. E.; Schafer, L. L.
Organometallics 2004, 23, 2234-2237.
(20) Lesueur, W.; Solari, E.; Floriani, C.; ChiesiVilla, A.; Rizzoli, C. Inorg.
Chem. 1997, 36, 3354-3362.
(10) Utsunomiya, M.; Hartwig, J. F. J. Am. Chem. Soc. 2004, 126, 2702-2703.
(11) Utsunomiya, M.; Kuwano, R.; Kawatsura, M.; Hartwig, J. F. J. Am. Chem.
Soc. 2003, 125, 5608-5609.
9
1828
J. AM. CHEM. SOC. 2006, 128, 1828-1839
10.1021/ja056003z CCC: $33.50 © 2006 American Chemical Society

Active Pd Catalyst for Intermolecular Hydroamination
A R T I C L E S
Table 1. Effect of Catalyst Components and Acid on the
Hydroamination of 1,3-Cyclohexadiene with 4-Aminobenzonitrilea
Although scattered hydroaminations in the presence of certain
functional groups have been reported,4,21,22 no study of the
compatibility of catalysts for hydroamination with substrates
bearing functional groups that often interfere with catalytic
reactions, such as alcohols, carboxylic acids, esters, amides,
enolizable keto groups, and nitriles, has been conducted with
any metal. Therefore, we have begun to assess whether late-
metal complexes can catalyze the additions of N-H bonds in
the presence of a variety of accompanying functional groups
(eq 1).
entry
Pd precursor
ligand
additive
yieldb
1
2
3
1% [Pd(η3-allyl)Cl]2
1% [Pd(η3-allyl)Cl]2
2% [(Xantphos)Pd-
(η3-allyl)]OTf
1% [Pd(η3-allyl)Cl]2
1% [Pd(η3-allyl)Cl]2
1% [Pd(η3-allyl)Cl]2
1% [Pd(η3-allyl)Cl]2
1% [Pd(η3-allyl)Cl]2
1% [Pd(η3-allyl)Cl]2
2% Pd(PPh3)4
2% Pd(PPh3)4
-
2% Xantphos
2% Xantphos
-
-
99
3
2
2% AgOTF
-
4
5
6
7
8
9
10
11c
12
13
14
2% Xantphos
2% DPEphos
2% DPPF
2% BINAP
2% DPPPent
4% PPh3
-
-
-
-
-
10% HCl
-
-
-
-
-
10% TFA
10% TFA
10% TfOH
87
7
2
0
0
0
22
11
0
0
0
In combination with this work, we have been conducting
studies to improve the rates of the catalytic hydroamination to
expand reaction scope. Increasing the rate of the desired addition
of the amine across an olefin would likely decrease formation
of products from side reactions and reduce deactivation of the
catalyst. Previous work on the mechanism of the palladium-
catalyzed hydroamination of vinylarenes showed that the C-N
bond in the amine product is formed by nucleophilic attack of
the aryl or aliphatic amine on an η3-phenethyl complex and that
this step can be turnover limiting.9,23 Thus, the activity of the
catalyst is likely to be improved by finding a means to accelerate
nucleophilic attack on the η3-phenethyl complex. Although
nucleophilic attack on allyl groups is a classic organometallic
transformation, the factors that affect the rate of attack on η3-
benzyl groups are not well understood. Even studies that directly
measure the rate of nucleophilic attack on allyl groups as a
function of the ancillary ligand are not well documented.24-26
We report studies that increase the activity of the catalyst
for the hydroamination of vinylarenes and dienes, that assess
the ability of this catalyst to add amines to olefins in the presence
of several types of typically reactive polar functionalities, and
that probe the origin of the increased activity of the catalyst.
We show that the combination of the wide bite-angle bisphos-
phine Xantphos and certain palladium precursors is highly active
for the hydroamination of vinylarenes and 1,3-dienes. This
combination of catalyst precursor and ligand catalyzes hy-
droaminations in the presence of alcohol, carboxylic acid, amide,
ester, enolizable keto, and nitrile functional groups, and our data
imply that the increased activity of this catalyst for the
hydroamination of vinylarenes originates from the effect of bite
angle on the rate of nucleophilic attack on the coordinated η3-
phenethyl ligand.
d
-
-
10% HBF4
10% HCle
a Reaction conditions: 0.5 mmol of 4-aminobenzonitrile, 1.0 mmol of
1,3-cyclohexadiene, 1.0 mL of THF, 20 h at 23 °C. b GC yields, in percent.
c Toluene was used in the place of THF. d 54 wt % in diethyl ether. e 1.0
M in diethyl ether. Ligand abbreviations: Xantphos ) 9,9-dimethyl-4,5-
bis(diphenylphosphino)xanthene, DPEphos ) bis(2-diphenylphosphinophen-
yl)ether, DPPF ) 1,1′-bis(diphenylphosphino)ferrocene, Binap ) 2,2′-
bis(diphenylphosphino)-1,1′-binaphthyl, DPPPent ) 1,5-bis(diphenylphos-
phino)pentane.
Figure 1. Structures of the ligands used in experiments summarized in
Tables 1 and 2.
p-cyanoaniline. Because the palladium-catalyzed addition of
arylamines to vinylarenes and dienes occurs faster with amines
that are more nucleophilic, reactions of this arylamine would
single out the most reactive catalysts. The hydroaminations of
dienes with this amine in the presence of previous catalysts
occurred in low yield: the reaction of p-cyanoaniline with 1,3-
cyclohexadiene catalyzed by Pd(PPh3)4 and 10% added trifluo-
roacetic acid formed only 22% of the addition product after 20
h at 23 °C. The hydroaminations of vinylarenes with this amine
in the presence of previous catalysts also occurred in low
yield: the reaction of p-cyanoaniline with styrene catalyzed by
Pd(PPh3)4 and 5% added triflic acid or by (DPPF)Pd(OTf)2, both
of which catalyze the addition of aniline to styrene,8 formed
less than 15% of the product from addition of p-cyanoaniline
after 20 h at 100 °C.
Results and Discussion
1. Effect of Catalyst Components on Rate. a. Effect of
Ligands and Catalyst Precursor. Our studies toward catalysts
with improved activity and high functional group tolerance were
initiated by conducting reactions with the weakly nucleophilic
(21) Lober, O.; Kawatsura, M.; Hartwig, J. F. J. Am. Chem. Soc. 2001, 123,
4366-4367.
A series of reactions of p-cyanoaniline with 1,3-cyclohexa-
diene catalyzed by palladium complexes of bidentate ligands
are summarized in Table 1. These reactions showed that
complexes of Xantphos (Figure 1), which contain a wide
P-Pd-P bite angle, are substantially more active as catalysts
for the addition of aromatic amines to dienes than are complexes
(22) Qian, H.; Widenhoefer, R. A. Org. Lett. 2005, 7, 2635-2638.
(23) Nettekoven, U.; Hartwig, J. F. J. Am. Chem. Soc. 2002, 124, 1166-1167.
(24) Canovese, L.; Chessa, G.; Santo, C.; Visentin, F.; Uguagliati, P. Organo-
metallics 2002, 21, 4342-4349.
(25) Crociani, B.; Antonaroli, S.; Bandoli, G.; Canovese, L.; Visentin, F.;
Uguagliati, P. Organometallics 1999, 18, 8, 1137-1147.
(26) Crociani, B.; Antonaroli, S.; Canovese, L.; Visentin, F.; Uguagliati, P. Inorg.
Chim. Acta 2001, 315, 172-182.
9
J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006 1829

A R T I C L E S
Johns et al.
Table 2. Effect of Catalyst Components and Acid on the Hydroamination of Styrene with 4-Aminobenzonitrilea
entry
Pd precursor
ligand
acid
yieldb
1
2
3
4
5
6
7
8
2% (Xantphos)Pd(OTf)2
2% (Xantphos)PdCl2
2% (DPPF)Pd(OTf)2
2% [(Xantphos)Pd(CH3CN)2](OTf)2
2% (DPPF)Pd(CH3CN)2(OTf)2
2% Pd(TFA)2
2% Pd(TFA)2
2% Pd(TFA)2
2% Pd(TFA)2
2% Pd(TFA)2
-
-
-
-
-
-
-
-
-
-
81 (93)
0
7 (13)
76 (91)
9 (16)
66
11
2
10
0
3% Xantphos
2% DPPPent
3% BINAP
3% DPEphos
3% DiPrPF
-
2% Xantphos
2% DPPF
2% DPEphos
2% BINAP
2% DPPPent
2% DPPE
-
5% TfOH
20% TfOH
5% TfOH
5% TfOH
20% TfOH
5% TfOH
-
-
-
-
-
-
9
10
11
12
13
14
15
16
17
18
19
2% Pd(PPh3)4
2
2% [Pd(COD)(η3-allyl)]OTf
2% [Pd(COD)(η3-allyl)]OTf
2% [Pd(COD)(η3-allyl)]OTf
2% [Pd(COD)(η3-allyl)]OTf
2% [Pd(COD)(η3-allyl)]OTf
2% [Pd(COD)(η3-allyl)]OTf
-
86 (36)
2 (13)
3 (14)
0 (16)
2 (10)
0 (9)
0
10% TfOH
10% HBF4
c
-
-
0
a Reaction conditions: 0.5 mmol of 4-aminobenzonitrile, 1.0 mmol of styrene, 0.5 mL of dioxane, 20 h at 100 °C. b GC yields, in percent. Values in
parentheses were obtained by conducting the reaction in the presence of 10 mol % TfOH. c 54 wt % in diethyl ether. Ligand abbreviations: Xantphos )
9,9-dimethyl-4,5-bis(diphenylphosphino)xanthene, DPPPent ) 1,5-bis(diphenylphosphino)pentane, Binap ) 2,2′-bis(diphenylphosphino)-1,1′-binaphthyl,
DPEphos ) bis(2-diphenylphosphinophenyl)ether, DiPrPF ) 1,1′-bis(di-i-propylphosphino)ferrocene, DPPF ) 1,1′-bis(diphenylphosphino)ferrocene, DPPPent
) 1,5-bis(diphenylphosphino)pentane, DPPE ) 1,2-bis(diphenylphosphino)ethane.
of related bisphosphines that are less rigid, such as 2,2′-bis-
(diphenylphosphino)biphenyl ether, or that have smaller bite
angles, such as BINAP and DPPF. A comparison of the reactions
with and without AgOTf to abstract chloride from the catalyst
shows that the reactions containing chloride occurred in higher
yields. Reactions initiated with the isolated allylpalladium triflate
complex (Table 1, entry 3) occurred in yields that were similar
to those catalyzed by the species generated in situ from the
combination of allylpalladium chloride and AgOTf.
functionality if they were conducted in the absence of any acid
cocatalyst. However, the reactions catalyzed by several of the
combinations of palladium precursor and ligand occurred in
higher yields in the presence of acid, and others occurred in
higher yields in the absence of added acid. Thus, the precise
effect of acid on the rates and yields is difficult to ascertain
without a determination of the turnover-limiting step in each
case.27
Several trends could be drawn, however. For example,
previously published reactions catalyzed by complexes of DPPF,
BINAP, and PPh3 were slow in the absence of added acid
cocatalyst, unless the isolated bis-triflate complexes were used
as precursor.8 Yet, reactions of complexes of Xantphos generated
in situ occurred at acceptable rates in the absence of added acid.
For example, the complex formed in situ from [Pd(COD)(η3-
allyl)]OTf and Xantphos catalyzed the addition of p-cyanoaniline
to styrene in 86% yield without added TfOH (Table 2, entry
12). The same reaction occurred to lower conversions in the
presence of added triflic acid. Likewise, this slower rate in the
presence of acid contrasts with the effect of acid on the reactions
with most catalyst and substrate combinations and could result,
in this case, from an interaction of the acid cocatalyst with the
free bidentate ligand because displacement of COD from this
precursor is likely to be slower than coordination of bisphos-
phine to Pd(TFA)2. The hydroamination of 1,3-cyclohexadiene
with aniline catalyzed by the combination of Xantphos and
Table 2 summarizes reactions of p-cyanoaniline with styrene
catalyzed by palladium complexes of bidentate ligands in the
presence and absence of triflic acid. The results also show that
complexes of Xantphos are more reactive than those of the other
ligands. This increased reactivity was observed with a series of
palladium precursors. Reactions starting from (chelate)Pd(OTf)2
complexes, from Pd(TFA)2, ligand and added acid, or from [Pd-
(COD)(η3-allyl)]OTf were faster when Xantphos was used as
ligand than when the other bisphosphines were used as ligand.
In contrast to the hydroamination of dienes, the hydroamination
of vinylarenes was much faster when chloride was absent from
the reaction system.
The consistency of the reaction yields with Xantphos as ligand
depended on the catalyst precursor. Reaction of 1,3-cyclohexa-
diene and anilines conducted with isolated (Xantphos)Pd(η3-
allyl)Cl (1) reproducibly occurred in high yield, whereas
reactions catalyzed by the combination of Xantphos and [Pd-
(η3-allyl)Cl]2 were more variable, even with the same substrate
combination. Likewise, reactions of styrene and anilines con-
ducted with (Xantphos)Pd(η3-allyl)OTf (2), generated from 1
and AgOTf, as catalyst occurred in high yields more consistently
than did reactions conducted with (Xantphos)Pd(OTf)2 as
catalyst.
(27) (a) Acid-catalyzed hydroaminations in the absence of metal have been
reported,27b,c but control studies conducted on the addition of p-cyanoaniline
to cyclohexadiene and styrene (Table 1, entries 12-14 and Table 2, entries
18 and 19) showed that acid alone does not catalyze the hydroamination.
Further, the acid-catalyzed reactions of anilines to vinylarenes27b are slower
than the palladium-catalyzed reactions of this work and give a mixture of
products from addition of N-H and C-H bonds across the vinylarene.
The acid-catalyzed reactions of dienes occur to low conversion. (b)
Anderson, L. L. A., J.; Bergman, R. G. J. Am. Chem. Soc. 2005, 127,
14542-14543. (c) Schlummer, B.; Hartwig, J. F. Org. Lett. 2002, 4, 1471-
1474.
b. Effect of Acid. One could envision that the reactions would
be simpler to conduct and could be more compatible with certain
9
1830 J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006

Active Pd Catalyst for Intermolecular Hydroamination
A R T I C L E S
Table 3. Palladium Catalyzed Hydroamination of
Table 4. Palladium Catalyzed Hydroamination of Styrene with
Bifunctional Aromatic Aminesa
1,3-Cyclohexadiene with Bifunctional Aromatic Aminesa
entry
amine
yieldb
entry
amine
yieldb
entry
amine
yieldb
entry
amine
yieldb
1
2
R ) 2-OH
80
89
92
81
71
59
88
8
9
R ) C-CN
71
90
72
99
92
89
91
86
1
2c
3
4
5
6
7
R ) 3-NO2
96
99
81
93
99
98
99
8d
9
10
11
12
13
14d
R ) 2-CN
98
99
97
88
82
95
57
R ) 4-CH2CH2OH
R ) 3-COOH
R ) 3-COCH3
R ) 4-COCH3
R ) 2-SMe
R ) 3-CN
R ) 4-NO2
R ) 3-CN
3
10
11
12
13d
14d
15d
R ) 4-CN
R ) 3-COOH
R ) 3-COCH3
R ) 4-COCH3
R ) 2-SMe
R ) 4-SMe
R ) 4-CN
4c
5
R ) 2-CO2Et
R ) 3-CO2Et
R ) 4-CO2Et
R ) 3-NHCOCH3
R ) 4-CONH2
R ) 2-CO2Et
R ) 3-CO2Et
R ) 4-CO2Et
R ) 2-OH
6
7d
R ) 4-SMe
a Reaction conditions: 1.0 mmol of amine, 2.0 mmol of 1,3-cyclohexa-
diene, 2.0 mL of THF. b Isolated yield (average of two runs), in percent.
c 50 °C. d 4.0 mmol of diene, 1.0 mL of THF, 50 °C.
a Reaction conditions: 1.0 mmol of amine, 2.0 mmol of styrene, 1.0
mL of dioxane. b Isolated yield (average of two runs), in percent. c Reaction
catalyzed by 2 mol % Xantphos, and 2 mol % [Pd(CH3CN)4](BF4)2.
d Reaction run without TfOH cocatalyst.
allylpalladium chloride dimer in the presence of HCl as acid
cocatalyst occurred in a lower yield than the same hydroami-
nation in the absence of the acid cocatalyst (Table 1, entry 4).
In contrast to these reactions that were retarded by acid, the
hydroaminations of styrene with aniline catalyzed by either
(Xantphos)Pd(OTf)2 or [(Xantphos)Pd(CH3CN)2](OTf)2 (Table
2, entries 1 and 4) occurred in higher yields in the presence of
an acid cocatalyst than they did in the absence of an acid
cocatalyst.
Scheme 1. Proposed Mechanisms for the Hydroamination of
Vinylarenes
c. Hydroamination of Dienes and Vinylarenes with Anilines
Bearing Polar and Protic Functional Groups. The reactions
of 1,3-cyclohexadiene with a series of anilines containing diverse
functionalities are summarized in Table 3. In general, reactions
of a 2:1 ratio of diene to amine in the presence of 2 mol %
catalyst afforded good to excellent yields of the desired allylic
amine after 24 h at room temperature. A few of the reactions
conducted with anilines containing electron-withdrawing sub-
stituents (Table 3, entries 2, 8, and 14) required slightly higher
temperatures to occur to completion, but even the reactions of
anilines with the strongly electron-withdrawing nitro group
occurred to completion. Reactions of anilines with functionality
in the ortho position occurred to completion if the ratio of diene
to amine and the overall concentration were increased.
Most striking, the results in Table 3 show that reactions
occurred selectively at the amine N-H bond over many other
types of functional groups and without poisoning of the catalyst.
The reactions occurred selectively at the amine N-H bond over
the carboxylic acid or phenolic O-H bonds; the catalyst was
not poisoned by formation of palladium carboxylates or phe-
noxides. The reaction also occurred with anilines bearing
enolizable hydrogens without the interference of palladium
enolate complexes. Anilines with other groups that could
coordinate to palladium, such as a thioether in the ortho-position
or cyano groups in ortho-, meta-, or para-positions, all reacted
in high yields. Finally, reactions of anilines with pendant esters
occurred without generating the corresponding amide or metal
amidate complex.
shown in Table 3, the reaction of an aromatic amine with a
pendant primary alcohol with styrene occurred in high yield
(Table 4, entry 2). Moreover, addition of the N-H bond of an
aniline derivative containing an N-aryl amide (Table 4, entry
14) or a primary amide (Table 4, entry 15) occurred in high
yield.28 In keeping with the mechanism for these hydroamina-
tions of vinylarenes and dienes (vide infra), hydroaminations
of alkenes under analogous conditions do not occur.
2. Mechanistic Studies of Hydroamination. Origin of the
Difference in Reactivity of Bisphosphine-Ligated Catalysts.
a. Synthesis and Dynamic Studies of η3-Phenethyl and η3-
Benzyl Complexes. Previous mechanistic work on vinylarene
hydroamination suggested that this hydroamination occurs by
one of the two pathways in Scheme 1. Further, these data
indicated that an η3-arylethyl complex was the catalyst resting
state and that nucleophilic attack on this intermediate to form
the C-N bond in the product was turnover limiting.23 The effect
of ligand structure on the rate of additions of amines to η3-
Table 4 summarizes reactions of anilines with styrene. Again,
a 2:1 ratio of olefin to amine with 2 mol % catalyst and 10 mol
% acid cocatalyst led to addition of the arylamine. These
reactions afforded moderate to excellent yields of the Mar-
kovnikov addition product after 24 h at 100 °C. The scope of
these reactions and the compatibility with auxiliary functional
groups was even superior to that of the reactions of anilines
with 1,3-cyclohexadiene. In addition to reactions of the anilines
(28) Results of ongoing studies on the additions of aliphatic amines to dienes
and vinylarenes will be published in due course
9
J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006 1831

A R T I C L E S
Johns et al.
Scheme 2. Synthesis of Benzylpalladium Complexes 4-10
Table 5. Comparison of Pd-C and P-Pd-P Angles of Benzyl Complexes 6, 7, and 9 and Allyl Complexes 1 and 12
phenethyl and η3-allyl complexes and the effect of the differ-
ences in rate of this elementary reaction on the catalytic process
were ambiguous because the rates of palladium-catalyzed allylic
alkylations with malonate nucleophiles, which occur through
analogous allyl complexes, were slower when catalyzed by
palladium complexes of Xantphos than those when catalyzed
by palladium complexes of DPPF or DPEphos.29 To determine
if the increase in rate of hydroamination with complexes of
Xantphos was due to an increase in the rate of nucleophilic
attack on the η3-arylethyl ligand, we prepared η3-benzyl
complexes ligated by Xantphos, DPEphos, DPPF, and (R)-
BINAP, as well as η3-allyl complexes ligated by a Xantphos,
DPEphos, DPPF, (R)-BINAP, and DPPE.
spectra of [(DPPF)Pd(η2-3-trifluoromethylstyrene)] consisted of
a pair of doublets (δ 19.0, 15.6 ppm, JPP ) 17.9 Hz), the spectra
of [(Xantphos)Pd(η2-styrene)] consisted of one broad singlet at
9.86 ppm, and the spectra of [(Xantphos)Pd(η2-3-trifluorom-
ethylstyrene)] consisted of a pair of broad singlets at 11.9 and
8.12 ppm (Figure S3). These data suggest that rotation about
the metal-alkene bond, exchange with free vinylarene, or partial
dissociation of the bisphosphine ligand occur at rates that depend
on the bisphosphine bite angle and vinylarene electronics.
Treatment of the styrene complex ligated by DPPF with
anilinium triflate gave a mixture of the syn and anti η3-phenethyl
complexes (3:1),31 which was recrystallized to yield [(DPPF)-
Pd(η3-phenethyl)]OTf (3) as a single isomer. This observation
suggests that the formation of an η3-phenethyl complex is more
likely to occur by addition of acid to a vinylarene complex than
by the reversed order of addition of a vinylarene to a cationic
palladium hydride generated by protonation of a palladium(0)
complex. The 31P NMR spectra of the major isomer consisted
of a pair of doublets (δ 33.62, 22.90 JPP ) 54.3 Hz). In contrast,
addition of anilinium triflate to the styrene complex ligated by
Xantphos did not form an η3-phenethyl complex that was stable
enough to isolate. Free styrene and zerovalent palladium
complexes were among the products that were generated by
the protonation of the η2-styrene complex ligated by Xantphos.
In lieu of a set of η3-phenethyl complexes with varied
phosphine ligands, we prepared a set of more stable η3-benzyl
complexes ligated by Xantphos, DPEPhos, DPPF, and (R)-
BINAP. [(Xantphos)Pd(η3-benzyl)]Cl (4), [(Xantphos)Pd(η3-
benzyl)]BF4 (5), [(Xantphos)Pd(η3-benzyl)]OTf (6), [(DPEphos)-
Pd(η3-benzyl)]OTf (7), [(DPPF)Pd(benzyl)Cl] (8), [(DPPF)Pd(η3-
benzyl)]OTf (9), and [(R-BINAP)Pd(η3-benzyl)]OTf (10) were
prepared by the addition of the bisphosphine and, when
appropriate, either AgOTf or AgBF4 to (COD)Pd(CH2Ph)Cl
We first sought to prepare η3-phenethyl complexes ligated
by Xantphos and DPPF by combining CpPd(η3-allyl) with the
desired ligand in an excess of styrene30 and adding anilinium
triflate to the resulting styrene complex. The reaction of CpPd-
(η3-allyl) with Xantphos or DPPF in an excess of styrene formed
the corresponding η2-styrene complexes in quantitative yields
if a 50-fold excess of styrene was used. We were unable to
isolate these complexes but were able to characterize them in
solution (see Supporting Information). To obtain spectral data
on these olefin complexes in the absence of the large excess of
styrene, we prepared complexes of electron poor vinylarenes
because they could be prepared with low ratios of olefin to
metal. The Xantphos and DPPF η2-vinylarene complexes of
3-trifluoromethylstryene were generated quantitatively in solu-
tion with only 1 equiv of vinylarene.
The 31P NMR spectra of the series of vinylarene complexes
varied. The 31P NMR spectra of [(DPPF)Pd(η2-styrene)] con-
sisted of a pair of broad singlets at 17.7 and 15.1 ppm, the
(29) The Xantphos complexes, however, showed unusual regioselectivity for
attack at the more substituted carbon of an unsymmetrical allyl substrate.
See: van Haaren, R. J.; Goubitz, K.; Fraanje, J.; van Strijdonck, G. P. F.;
Oevering, H.; Coussens, B.; Reek, J. N. H.; Kamer, P. C. J.; van Leeuwen,
P. W. N. M. Inorg. Chem. 2001, 40, 3363-3372.
(30) Crascall, L. E.; Spencer, J. L. J. Chem. Soc., Dalton Trans. 1992, 3445-
3452.
1
(Scheme 2). These complexes were characterized by H, 31P,
(31) Hii, K. K.; Claridge, T. D. W.; Giernoth, R.; Brown, J. M. AdV. Synth.
Catal. 2004, 346, 983-988.
9
1832 J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006

Active Pd Catalyst for Intermolecular Hydroamination
A R T I C L E S
Scheme 3. Proposed Mechanisms for the Hydroamination of
1,3-Dienes
Figure 2. ORTEP plot of one of the two independent molecules of
[(Xantphos)Pd(η3-benzyl)]OTf (6) at 30% probability level (hydrogen atoms
and triflate counterion omitted for clarity). Selected bond lengths and
angle: Pd1-C40 ) 2.085(6) Å; Pd1-C41 ) 2.238(5) Å; Pd1-C46 )
2.686 Å; Pd1-C42 ) 2.986 Å; Pd1-P1 ) 2.4031(17) Å; Pd1-P2 )
2.3036(16) Å; P2-Pd1-P1 ) 108.11(5)°. Selected bond lengths and angle
for the second independent molecule: Pd1′-C40′ ) 2.095(6) Å; Pd1′-
C41′ ) 2.239(5) Å; Pd1′-C46′ ) 3.091 Å; Pd1′-C42′ ) 2.584(6) Å; Pd1′-
P1′ ) 2.4057(16) Å; Pd1′-P2′ ) 2.3095(15) Å; P2′-Pd1′-P1′ )
107.45(2)°.
These crystallographic data for 6 can be compared to those
for DPEphos-ligated benzyl complex 7 (Figure S6) and DPPF-
ligated benzyl complex 9 (Figure S7, S8). As shown in Table
5, the Pd-C distances in DPEphos-ligated 7 and DPPF-ligated
9 were similar to those in Xantphos complex 6. The bite angle
in 7 was found to be 3.4°-4.1° smaller than that in 6, and the
bite angle in 9 was found to be 5.4°-6.1° smaller than that in
6. The bite angle of 7, 104.08(4)°, was similar to that of the
previously characterized palladium complexes ligated by
DPEphos.29,32-34
13C NMR spectroscopy and elemental analysis. The 31P NMR
spectrum of DPPF-ligated 9 consisted of a typical pair of
doublets. However, the 31P NMR spectrum of Xantphos
complexes 4-6 at room temperature consisted of a single
resonance at 12.9, 11.1, and 13.3 ppm, respectively.
The 1H and 31P NMR spectra of [(Xantphos)Pd(η3-benzyl)]-
OTf (6) were temperature dependent (Figures S1 and S2). The
singlet 31P NMR resonance of triflate 6 was resolved into two
doublets at lower temperatures, with a coalescence temperature
of approximately -25 °C. A single resonance for the two
benzylic protons was observed at room temperature. This signal
broadened, but was not fully resolved, at -80 °C. Therefore, a
rapid η3-to-η1 interconversion occurs at low temperature, but
exchange between T-shaped geometries via Y-shaped intermedi-
ates is slow on the NMR time scale below -25 °C.
X-ray diffraction studies were conducted on the benzylpal-
ladium complexes [(Xantphos)Pd(η3-benzyl)]OTf (6), [(DPEp-
hos)Pd(η3-benzyl)]OTf (7), and [(DPPF)Pd(η3-benzyl)]OTf (9).
Table 5 provides a comparison of distances between the metal
and the three carbons of the benzyl group bound to the metal
and the P-Pd-P bite angles for these three benzyl complexes,
as well as related data on two allyl complexes described in a
later section. The triflate counterion in each of the benzyl
complexes was far from bonding to the metal. Although the
Pd-O distance varied a large amount in the various structures,
the shortest distance was a long 3.28 Å.
[(Xantphos)Pd(η3-benzyl)]OTf (6) (Figure 2) crystallized as
two chemically equivalent but crystallographically independent
molecules with η3-benzyl ligands. The bond distance to the
benzylic carbon (2.085(6) and 2.095(6) Å) was much shorter
than the distance to the ipso carbon (2.238(5) and 2.239(5) Å)
or the distance to the closer of the ortho carbons (2.686 and
2.584(6) Å). The bite angles in these two crystallographically
independent molecules were 108.11(5)° and 107.45(2)°.
b. Synthesis and Dynamic Studies of η3-Allyl Complexes.
Scheme 3 shows how η3-allyl complexes are likely to be
intermediates in the hydroamination of conjugated dienes, just
as the η3-phenethyl complexes appear to be intermediates in
the hydroamination of vinylarenes.23 The allyl-palladium chlo-
ride complexes [(Xantphos)Pd(η3-allyl)]Cl (1), [(DPEphos)Pd-
(η3-allyl)]Cl (11), [(DPPF)Pd(η3-allyl)]Cl35 (12), [((R)-BINAP)-
Pd(η3-allyl)]Cl (13), and [(DPPE)Pd(η3-allyl)]Cl36 (14) were
prepared by the addition of 2 equiv of the appropriate bisphos-
phine to [Pd(η3-allyl)Cl]2. [(Xantphos)Pd(η3-allyl)]BF4 (15),
[(Xantphos)Pd(η3-allyl)]OTf (2), [((R)-BINAP)Pd(η3-allyl)]BF4
(16), and [((R)-BINAP)Pd(η3-allyl)]OTf37 (17) were prepared
by addition of the appropriate bisphosphine to [Pd(η3-allyl)-
Cl]2, followed by 2 equiv of the corresponding silver salt. Each
1
of these complexes was characterized by H, 31P, 13C NMR
spectroscopy and elemental analysis.
(32) Kranenburg, M.; Delis, J. G. P.; Kamer, P. C. J.; van Leeuwen, P. W. N.
M.; Vrieze, K.; Veldman, N.; Spek, A. L.; Goubitz, K.; Fraanje, J. J. Chem.
Soc., Dalton Trans. 1997, 1839-1849.
(33) van Leeuwen, P. W. N. M.; Zuideveld, M. A.; Swennenhuis, B. H. G.;
Freixa, Z.; Kamer, P. C. J.; Goubitz, K.; Fraanje, J.; Lutz, M.; Spek, A. L.
J. Am. Chem. Soc. 2003, 125, 5523-5539.
(34) Zuideveld, M. A.; Swennenhuis, B. H. G.; Boele, M. D. K.; Guari, Y.; van
Strijdonck, G. P. F.; Reek, J. N. H.; Kamer, P. C. J.; Goubitz, K.; Fraanje,
J.; Lutz, M.; Spek, A. L.; van Leeuwen, P. W. N. M. J. Chem. Soc., Dalton
Trans. 2002, 2308-2317.
(35) Cantat, T.; Genin, E.; Giroud, C.; Meyer, G.; Jutand, A. J. Organomet.
Chem. 2003, 687, 365-376.
(36) Paonessa, R. S.; Prignano, A. L.; Trogler, W. C. Organometallics 1985, 4,
647-657.
(37) Jones, M. D.; Paz, F. A. A.; Davies, J. E.; Johnson, B. F. G.; Klinowski,
J. Acta Crystallogr., Sect. E 2003, 59, M538-M540.
9
J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006 1833

A R T I C L E S
Johns et al.
Figure 3. 500 MHz 1H NMR spectra of (Xantphos)Pd(η3-allyl)X at 22 °C
in CD2Cl2 (top, X ) Cl; middle, X)BF4; bottom, X)OTf). # and *
designate the terminal allyl protons and dimethyl backbone of Xantphos,
respectively.
Figure 5. ORTEP plot of (Xantphos)Pd(η3-allyl)Cl (1) at 30% probability
level (hydrogen atoms and cocrystallized benzene omitted for clarity).
Selected distances and angle: Pd1-C40 ) 2.178(4) Å; Pd1-C41 ) 2.160-
(4) Å; Pd1-C42 ) 2.179(4) Å; Pd1-P1 ) 2.3623(10) Å; Pd1-P2 )
2.3502(11) Å; Pd1-Cl1 ) 2.7694(10) Å; P2-Pd1-P1 ) 107.83(4)°.
ligand bearing different counterions is surprisingly large. The
backbone methyl groups can interconvert by reversible dis-
sociation of one-half of the bidentate ligand40 or pseudorotation
via a pentacoordinate palladium complex formed by coordina-
tion of the counterion.41
X-ray diffraction studies were conducted on Xantphos
complex 1 (Figure 5) and DPPF complex 12 (Figure S9). Data
on the M-C distances and P-Pd-P angles are included in
Table 5. These studies showed that the bite angles were 108°
and 101°, respectively. These angles are indistinguishable from
those angles found in the η3-benzyl complexes. In contrast to
most η3-allyl complexes of palladium, Xantphos complex 1
displayed a geometry that can be best described as distorted
square pyramidal with two sites of the equatorial plane occupied
by the η3-allyl ligand and the axial site occupied by a weakly
bound chloride. The metal-chloride distance was long (2.77
Å), but within the sum of the van der Waals radii for palladium
and chloride.42 The structure of DPPF-complex 12 indicated a
weaker Pd-Cl interaction or the lack of an interaction. The
Pd-Cl distance in this complex was 3.16 Å. Although five-
coordinate, 18-electron complexes of palladium are precedented,
they are rare.20,43-49 Only a few examples of (chelate)M(η3-
allyl)X complexes in which X is a coordinated halide have been
structurally characterized,50-55 and only one of these contains
palladium as the central atom.55
1
Figure 4. 400 MHz H NMR spectra of (Xantphos)Pd(η3-allyl)Cl (1) at
various temperatures in CD2Cl2. # and * designate the terminal allyl protons
and dimethyl backbone of Xantphos, respectively.
1
The H NMR spectra of Xantphos-ligated complexes 1, 2,
and 15 with chloride, OTf, and BF4 counterions were distinct
(Figure 3). The terminal protons syn and anti to the central
proton of the allyl group in chloride complex 1 interconverted
on the NMR time scale at room temperature. They were resolved
at -25 °C (Figure 4). The terminal allyl protons of complexes
15 and 2 containing the less-coordinating BF4 and OTf
counterions interconverted more slowly. Signals corresponding
to the syn and anti protons of the allyl termini of both complexes
were resolved at room temperature. The interconversion of two
terminal allyl protons likely occurs by the common η3-η1-η3
mechanism.38 If so, then the more coordinating chloride
counterion appears to promote the η3- η1- η3 process.39
The dynamics of the backbone methyl groups also varied with
the counterion. A single resonance corresponding to the
backbone methyl groups of the Xantphos ligand in 1 and 15
was observed at room temperature, whereas two distinct, broad
resonances corresponding to these methyl groups in complex 2
were observed at room temperature. The methyl groups of the
Xantphos backbone of 1 coalesced at -55 °C (Figure 4), while
the resonances corresponding to the methyl groups of the
Xantphos backbone of 15 coalesced at -18 °C. The difference
between the rates of interconversion of the syn and anti protons
of the allyl group and the two backbone methyl groups of the
(40) Albinati, A.; Kunz, R. W.; Ammann, C. J.; Pregosin, P. S. Organometallics
1991, 10, 1800-1806.
(41) Jackman, L. M.; Cotton, F. A.; Adams, R. D. Dynamic nuclear magnetic
resonance spectroscopy; Academic Press: New York, 1975.
(42) Huheey, J. E.; Keiter, E. A.; Keiter, R. L. Inorganic chemistry: principles
of structure and reactiVity, 4th ed.; HarperCollins College Publishers: New
York, 1993.
(43) Aizawa, S.; Iida, T.; Funahashi, S. Inorg. Chem. 1996, 35, 5163-5167.
(44) Bartsch, R.; Hietkamp, S.; Morton, S.; Peters, H.; Stelzer, O. Inorg. Chem.
1983, 22, 3624-3632.
(45) Brauer, D. J.; Gol, F.; Hietkamp, S.; Peters, H.; Sommer, H.; Stelzer, O.;
Sheldrick, W. S. Chem. Ber.-Recl. 1986, 119, 349-365.
(46) Enzmann, A.; Eckert, M.; Ponikwar, W.; Polborn, K.; Schneiderbauer, S.;
Beller, M.; Beck, W. Eur. J. Inorg. Chem. 2004, 1330-1340.
(47) Fernandez, D.; Garcia-Seijo, M.; Kegl, T.; Petocz, G.; Kollar, L.; Garcia-
Fernandez, M. Inorg. Chem. 2002, 41, 4435-4443.
(48) Louw, W. J.; De Waal, D. J. A.; Kruger, G. J. J. Chem. Soc., Dalton Trans.
1976, 2364-2368.
(49) Ma, J.; Kojima, Y.; Yamamoto, Y. J. Organomet. Chem. 2000, 616, 149-
156.
(50) Brunkan, N. M.; Jones, W. D. J. Organomet. Chem. 2003, 683, 77-82.
(51) Cameron, T. S.; Green, M. L. H.; Munakata, H.; Prout, C. K.; Smith, M.
J. J. Coord. Chem. 2 1972, 43.
(52) Carmona, E.; Palma, P.; Poveda, M. L. Polyhedron 1990, 9, 757-761.
(53) Churchill, M. R.; O’Brien, T. A. J. Chem. Soc. A 1970, 206-212.
(54) Walter, D.; Wilke, G. Angew. Chem., Int. Ed. Engl. 1966, 5, 897-898.
(38) Consiglio, G.; Waymouth, R. M. Chem. ReV. 1989, 89, 257-276.
(39) Oslinger, M.; Powell, J. Can. J. Chem. 1973, 51, 274-287.
9
1834 J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006

Active Pd Catalyst for Intermolecular Hydroamination
A R T I C L E S
Table 6. Nucleophilic Attack of Anilines on [(L2)Pd(η3-CH2Ph)]OTfa
1
entry
complex
X
L2
bite angle (deg)
amine
t1 2 (min)
/
k
obs (s-
)
×
104
1
2
3
4
5
6
7
8
9
6
OTf
Xantphos
108.11(5), 107.45(5)b
R ) H
26.3
41.1
34.6
12.7
54.1
328
610
630
544
4.40
2.81
3.34
9.11
2.13
0.352
0.19
0.18
R ) 3-CN
R ) 4-CN
R ) H
5
7
9
BF4
OTf
OTf
OTf
OTf
OTf
DPEphos
DPPF
104.08(4)b
R ) H
R ) H
R ) 3-CN
R ) 4-CN
R ) H
102.19(5), 101.95(4)b
10
(R)-Binap
95.97(4)c
0.212
a Reaction conditions: 0.014 M Pd complex, 0.070 M amine, 0.014 M PPh3, 9.0 mM 1,3,5-trimethylbenzene (internal standard) in 0.2 mL of CD2Cl2 and
0.6 mL of THF-d8 at 60 °C. b Values taken from crystal structures. For complexes ligated by Xantphos and DPPF, two independent molecules were present
in the unit cell. c The bite angle for [((R)-BINAP)Pd(η3-allyl)]OTf37 was used.
Table 7. Nucleophilic Attack of Anilines on [(L2)Pd(η3-allyl)]Xa
1
entry
complex
L2
X
bite angle (deg)
t1 2 (min)
/
k
obs (s-
)
×
104
1
2
3
4
5
6
7
1
15
2
11
12
13
14
Xantphos
Xantphos
Xantphos
DPEphos
DPPF
Cl
107.83(4)b
108.11(7)c
108.11(7)c
104.08(4)b
100.73(4)b
95.97(4)d
85c
38.8
143
185
55.4
100e
58.4
57.8
2.98
BF4
OTf
Cl
Cl
Cl
0.807
0.625
2.08
0.96
1.98
2.00
(R)-Binap
DPPE
Cl
a Reaction conditions: 0.029 M Pd complex, 2.9 M amine, 0.44 M diphenylacetylene, PPh3 (external standard) in 0.5 mL of CH2Cl2 at 45 °C. b Values
taken from crystal structures. For entry 4 the value from [(DPEphos)Pd(η3-CH2Ph)]OTf was used. c Values obtained from ref 33. d Reference 37. e (20 min.
c. Relative Rates of Reactions of Arylamines with η3-
Benzyl, η3-Allyl, and η3-1,1-Dimethylallyl Complexes Li-
gated by Bisphosphines. The reactions of arylamines with
bisphosphine-ligated, benzyl complexes 4-10 were monitored
by 1H NMR spectroscopy at 60 °C in a mixture of CD2Cl2 and
THF-d8. Triphenylphosphine was added to each reaction to trap
the palladium(0) product. In the presence of PPh3, the observed
palladium(0) products are Pd(bisphosphine)2 and Pd(PPh3)4. The
rate of nucleophilic attack was unaffected by the concentration
of added phosphine; reactions with 5 and 10 equiv of added
PPh3 occurred with indistinguishable rate constants.
roborate counterions are not bound to the metal and would not
be expected to affect the geometry or electrophilicity of the
metal center. These results were used to further improve the
catalyst, as will be subsequently described. More expected, the
chloride complex 4, which would be anticipated to contain an
η1-benzyl group and coordinated chloride,46 did not react with
arylamines to form the N-benzylaniline product.56
Table 7 summarizes the half-lives and rate constants for the
reactions of allylpalladium complexes 1, 2, and 11-15 with
aniline. The rates of reaction of 0.029 M allylpalladium
complexes in CH2Cl2 with 100 equiv of aniline at 45 °C to
form N-allylaniline were measured by 31P NMR spectroscopy.
These reactions were conducted with added diphenylacetylene
to trap the palladium(0) product; reactions of allylpalladium
complexes containing chloride counterions in the presence of
added triphenylphosphine to trap the palladium(0) formed a
mixture of uncharacterized products. The rate of nucleophilic
attack was not affected by the concentration of added diphen-
ylacetylene; reactions with 5 and 10 equiv of added diphenyl-
acetylene occurred with indistinguishable rate constants. Unlike
the correlation of the rate of nucleophilic attack to the bite angle
observed for benzylpalladium complexes, no simple trend in
the rate of nucleophilic attack on the ancillary ligand bite angle
was observed. The Xantphos complex, which contains the
largest bite angle (108°), did undergo nucleophilic attack by
aniline faster than the allyl complexes of the other ligands.
However, nucleophilic attack by aniline on the complexes
Table 6 summarizes the half-lives and rate constants for the
reactions of benzylpalladium complexes 5-7, 9, and 10 with
aniline and the reactions of benzylpalladium complexes 6 and
9 with 3-aminobenzonitrile and 4-aminobenzonitrile, to form
the corresponding N-benzylarylamines. Most striking, these data
show that an increase in bite angle from 102° to 108° leads to
an increase in the rate of nucleophilic attack by an order of
magnitude. In addition, these data show that nucleophilic attack
by aniline is 1.3 to 1.9 times faster than attack by the less
electron-rich 3- and 4-aminobenzonitriles.
The reaction of [(Xantphos)Pd(η3-benzyl)]BF4 (5) with aniline
(Table 6, entry 4) revealed a significant effect of the counterion
on the rate of the nucleophilic attack. The rate of the reaction
of the aniline with BF4 complex 5 was 2 times greater than
that of the analogous reaction with triflate 7 (Table 3, entry 1).
This result was surprising because the triflate and tetrafluo-
(56) Preparation and crystallographic characterization of [(Xantphos)Pd(η3-
benzyl)Br] (18) show an η3 coordination mode for the benzyl ligand with
the bromide coordinated in the apical position. See Supporting Information.
(55) Hansson, S.; Norrby, P. O.; Sjogren, M. P. T.; Akermark, B.; Cucciolito,
M. E.; Giordano, F.; Vitagliano, A. Organometallics 1993, 12, 4940-4948.
9
J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006 1835

A R T I C L E S
Johns et al.
Table 8. Nucleophilic Attack of Aniline on [(L2)Pd(η3-1,1-dimethylallyl)]OTfa
1
entry
complex
equiv PhNH2
T (°
C)
L2
t1 2 (min)
/
k
obs (s-
)
×
104
1
2
3
4
19
5
15
15
200
15
200
200
200
45
45
45
45
45
45
75
75
Xantphos
DPEphos
DPEphos
86.4
32.0
840
1.3
3.6
0.14
6.0
20
21
19.3
5
DPPF
<10% conv @12 h
250
58
6
0.46
2.0
7b
8b
22
(R)-Binap
28.8% conv @1 h
a Reaction conditions: 0.039 M Pd complex, 0.44 M diphenylacetylene, PPh3 (external standard) in 0.5 mL of CH2Cl2 with 5, 15, or 200 equiv of PhNH2
at 45 or 75 °C. b Reaction run in 1,2-dichloroethane.
containing smaller bite angles than the Xantphos complex and
more flexible ligand backbones occurred with similar rates. The
allylpalladium complex ligated by DPPF reacted more slowly
than the other allylpalladium complexes, but the rate constants
were difficult to reproduce, and the value presented for reaction
of this complex (Table 7, entry 5) has significant error.
Like the rates of reactions of the η3-benzyl complexes, the
rates of reactions of the allyl complexes were significantly
affected by the identity of the counterion. One might expect
that the complex with the less coordinating anion would be more
electrophilic, but Xantphos-ligated allyl complexes 2 and 15
containing triflate and tetrafluoroborate counterions reacted
significantly slower than the analogous complex 1 containing
a more coordinating chloride ion. Although unexpected, this
trend for the rates of nucleophilic attack on the allyl complexes
qualitatively paralleled that for the rates of nucleophilic attack
on the Xantphos-ligated benzylpalladium complexes with tet-
rafluoroborate and triflate counterions. Moreover, this trend is
consistent with the yields of the hydroaminations of dienes
catalyzed by complexes with these counterions. The yield of
allylamine from diene hydroamination catalyzed by the com-
bination of Xantphos and [Pd(η3-allyl)Cl]2 was higher than the
yield from the hydroamination conducted in the presence of
AgOTF (Table 1, entries 1 and 2).
The difference between the smooth, positive correlation
between bite angle and rate of nucleophilic attack on the
benzylpalladium complexes and lack of an overarching cor-
relation between bite angle and rate of nucleophilic attack on
the symmetrical allyl complexes led us to conduct experiments
to determine whether the difference resulted from inherent
electronic properties of an η3-benzyl group and an η3-allyl group
or from the symmetries of an η3-benzyl and a symmetric,
unsubstituted η3-allyl ligand. To distinguish between these
possibilities, we prepared a series of 1,1-dimethylallylpalladium
complexes bearing bisphosphines with different bite angles.
[(Xantphos)Pd(η3-1,1-dimethylallyl)]OTf (19),29 [(DPEphos)-
Pd(η3-1,1-dimethylallyl)]OTf (20),29 [(DPPF)Pd(η3-1,1-dim-
ethylallyl)]OTf (21), and [((R)-Binap)Pd(η3-1,1-dimethylallyl)]-
OTf (22) were prepared by the addition of 2 equiv of the
appropriate bisphosphine and 2 equiv of AgOTf to [Pd(η3-1,1-
dimethylallyl)Cl]2.
8). These reactions, like the reaction of the parent allyl
complexes, were conducted with added diphenylacetylene to
trap the palladium(0) product. The reaction of aniline with each
of these complexes formed N-(3-methylbut-2-enyl)aniline from
attack at the unsubstituted terminus of the allyl group. Similar
to the correlation between bite angle and rate of attack on
benzylpalladium complexes, a clear positive correlation was
observed between the ligand bite angle and rate of attack of
aniline on 1,1-dimethylallylpalladium complexes. The effect of
bite angle on the rate of nucleophilic attack on the unsym-
metrical η3-allyl complexes was even greater than that for
nucleophilic attack on the η3-benzyl complexes. Complexes of
Xantphos and DPEphos (Table 8, entries 2 and 3), which have
an approximately 4° difference in bite angle, and complexes of
DPEphos and DPPF (Table 8, entries 4 and 6), which have an
approximately 3° difference in bite angle, displayed an order
of magnitude difference in the rate of nucleophilic attack. This
larger dependence was observed, even though there is no
significant difference in the bite angles of η3-benzyl and η3-
1,1-dimethylallyl complexes of the same bisphosphine (ref 29
and vide supra).
Thus, the correlation between the rate of attack and ligand
bite angle appears to arise when the η3-ligand is substituted
unsymmetrically. The origin of the dependence of the correlation
between rate of attack and bite angle on the substitution pattern
of the η3-ligand may result from differences in the ground-state
structures of the unsymmetrical and symmetrical ligands. X-ray
diffraction showed that the difference between the M-C
distances in the benzyl complex increased with increasing bite
angle, while the two M-C distances were naturally indistin-
guishable in the complexes containing the symmetrical, unsub-
stituted allyl ligand. This greater dissymmetry of the binding
of the unsymmetrical η3-benzyl ligand, and the presumed greater
dissymmetry of the binding of the 1,1-disubstituted allyl ligand,
with increasing size of the bite angle is likely to make one end
of the allyl group more electrophilic than the other and thereby
increase the rate of nucleophilic attack.
d. Identification of the Turnover-Limiting Step.28 To
determine if the effects of ligand and counterion translated to
differences in rates of the overall catalytic process, we conducted
experiments to determine the resting state of the catalyst and
the turnover-limiting step with several catalysts and substrate
combinations. Additions of aniline to styrene catalyzed by
isolated [(DPPF)Pd(η3-benzyl)]OTf (9) and [(Xantphos)Pd(η3-
benzyl)]OTf (6) and additions of aniline to 1,3-cyclohexadiene
The rates of reaction of 0.039 M of each of the 1,1-
dimethylallylpalladium complexes with varying amounts of
aniline in CH2Cl2 or 1,2-dichloroethane at 45 °C or 75 °C,
respectively, were measured by 31P NMR spectroscopy (Table
9
1836 J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006

Active Pd Catalyst for Intermolecular Hydroamination
A R T I C L E S
catalyzed by [(Xantphos)Pd(η3-allyl)Cl] (1) were monitored by
31P NMR spectroscopy. In agreement with previous work,9,23
the [(DPPF)Pd(η3-phenethyl)]OTf (3) complex was the major
palladium complex in the reaction of styrene with aniline
containing HOTf as cocatalyst. These data indicate that attack
of amine on the η3-phenethyl ligand is the turnover-limiting
step of the catalytic cycle when DPPF is used as ligand.
Unfortunately, the 31P NMR spectrum of the reaction of aniline
with styrene catalyzed by [(Xantphos)Pd(η3-benzyl)]OTf (6)
contained many resonances that could not be assigned to specific
complexes at this time. However, 31P NMR spectroscopic
monitoring of the addition of aniline to 1,3-cyclohexadiene
catalyzed by [(Xantphos)Pd(η3-allyl)]Cl (1) showed that [(Xant-
phos)Pd(η3-cyclohexenyl)]Cl (23) was the major intermediate
in solution. This observation implies that nucleophilic attack
of amine on the allyl complex is turnover limiting. Thus, the
faster rate of stoichiometric addition of aniline to the allyl
complexes ligated by Xantphos than that of addition to the allyl
complexes ligated by other bisphosphines accounts for at least
a portion of the faster rate of catalytic hydroamination of dienes
when Xantphos is used as ligand. The identity of the cyclo-
hexenyl complex 23 was confirmed by independent synthesis
Table 9. Effect of Catalyst Components on the Hydroamination of
Styrene with 4-Aminobenzonitrilea
entry
catalyst
yieldb
1
2
3
4
5
6
2% [Pd(CH3CN)4](BF4)2 + 2% Xantphos
2% [Pd(CH3CN)4](BF4)2 + 2% DPPF
2% [Pd(CH3CN)4](BF4)2 + 2% DPEPhos
2% [Pd(CH3CN)4](BF4)2 + 2% BINAP
2% [Pd(CH3CN)4](BF4)2 + 2% DPPPent
2% [Pd(CH3CN)4](BF4)2 + 2% DPPE
96
11
14
3
6
0
a Reaction conditions: 0.5 mmol of 4-aminobenzonitrile, 1.0 mmol of
styrene, 0.5 mL of dioxane. b GC yields, in percent.
Table 10. [Pd(CH3CN)4](BF4)2/Xantphos Catalyzed
Hydroamination of Styrene with Bifunctional Aromatic Aminesa
entry
amine
yieldb
1
2
3
4
5
R ) 3-COCH3
R ) 2-SMe
R ) 4-SMe
R ) 2-CN
83 (81)
56
88 (84)
65
from [Pd(η3-cyclohexenyl)Cl]2 and 2 equiv of Xantphos per
57
R ) 4-CN
65
dimer.
a Reaction conditions: 1.0 mmol of amine, 2.0 mmol of styrene, 1.0
mL of dioxane. b GC yields, in percent. Values in parentheses are isolated
yields (average of two runs), in percent.
e. Relationship between the Rate of Stoichiometric Nu-
cleophilic Attack on η3-Allyl Complexes and the Rate of
Catalytic Hydroamination of 1,3-Dienes. To determine the
origin of the difference in rates of hydroamination of dienes
with the different precatalysts, the reactions of aniline with 1,3-
cyclohexadiene initiated with [(Xantphos)Pd(η3-allyl)]Cl (1),
[(DPPF)Pd(η3-allyl)]Cl (12), and [((R)-BINAP)Pd(η3-allyl)]Cl
(13) as precatalysts were monitored by 31P NMR spectroscopy.
The major palladium complexes in each reaction were precata-
lyst and [(bisphosphine)Pd(η3-cyclohexenyl)]Cl. However, the
concentration of [(bisphosphine)Pd(η3-cyclohexenyl)]Cl was
lower for complexes with smaller bite angles. Complementing
these changes in the concentration of the allyl complex, the
concentration of catalytically inactive [(bisphosphine)PdCl2]58
was greater in reactions catalyzed by complexes containing
bisphosphines with smaller bite angles. The origin of the
relationship between bisphosphine bite angle and concentration
of (bisphosphine)PdCl2 is unclear. However, the decomposition
of bisphosphine allylpalladium complexes bearing halide coun-
terions has previously been shown to form (bisphosphine)-
PdCl2.59 These data provide a second reason that the hydroam-
ination of dienes varies with ligand, besides faster attack of the
amine on the allyl group of Xantphos 1. Considering that the
rate of nucleophilic attack on the symmetric allyl complexes
ligated by DPPF and BINAP was relatively insensitive to the
bite angle (vide supra) but that the concentration of the DPPF-
ligated allyl complex was higher, we propose that the differences
in rates of the hydroamination of dienes by DPPF and BINAP
complexes are due to the differences in concentrations of the
allyl complexes, and not due to a difference in the rate of
nucleophilic attack on the allyl intermediate. Moreover, we
propose that the faster hydroaminations of dienes catalyzed by
Xantphos complexes of palladium result from a sum of the faster
rate of nucleophilic attack on the allyl ligand and a higher
concentration of the allyl intermediate.
3. Further Catalyst Improvements. Reactions Catalyzed
by Bisphosphine and [Pd(CH3CN)4](BF4)2. To determine if
the faster rate of nucleophilic attack of arylamine on [(Xant-
phos)Pd(η3-benzyl)]BF4 (5) vs [(Xantphos)Pd(η3-benzyl)]OTf
(6) translates into a faster catalytic reaction, p-cyanoaniline was
allowed to react with styrene in the presence of a catalyst
generated in situ from [Pd(CH3CN)4](BF4)2 and Xantphos (Table
9). Indeed, the reaction catalyzed by the combination of
Xantphos and [Pd(CH3CN)4](BF4)2 (Table 9, entry 1) occurred
in higher yield than the reaction catalyzed by the analogous
triflate complex [(Xantphos)Pd(CH3CN)2](OTf)2 (Table 2, entry
4).
The combination of Xantphos and [Pd(CH3CN)4](BF4)2 was
also evaluated for the reactions of Table 4 that occurred in
moderate yields with 2 mol % (Xantphos)Pd(η3-allyl)OTf and
10 mol % TfOH as catalyst. Indeed, the reaction of 1-(3-
aminophenyl)ethanone with styrene occurred in higher yield
(Table 10, entry 1) with Xantphos and [Pd(CH3CN)4](BF4)2 as
catalyst, instead of (Xantphos)Pd(η3-allyl)OTf as catalyst.
However, not all reactions catalyzed by the complex generated
from [Pd(CH3CN)4](BF4)2 occurred in higher yield than those
catalyzed by (Xantphos)Pd(η3-allyl)OTf. For example, reactions
of styrene with anilines bearing thiomethyl or nitrile function-
alities occurred in similar yields with the two catalysts (Table
10, entries 2-5), but the rate of the reactions catalyzed by the
combination of Xantphos and [Pd(CH3CN)4](BF4)2 were faster,
and the catalyst can be assembled in situ. Consistent with our
results that showed faster attack of the arylamine on the allyl
(57) Stoichiometric rate data for the nucleophilic addition of arylamines to
benzylpalladium and allylpalladium complexes were obtained at different
temperatures and concentrations than those of the catalytic reaction.
However, a qualitative comparison of the rates of stoichiometric additions
and catalytic transformations shows that they occur on similar time scales.
(58) Trost, B. M.; Strege, P. E.; Weber, L.; Fullerton, T. J.; Dietsche, T. J. J.
Am. Chem. Soc. 1978, 100, 3407-15.
(59) XantphosPdCl2 (24) was independently prepared and characterized by 1H,
13C, 31P NMR spectroscopy and X-ray diffraction.
9
J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006 1837

A R T I C L E S
Johns et al.
Cl]262 were prepared using the reported procedures. (COD)Pd(CH2Ph)-
Cl63 was prepared via an alternate procedure to that found in the
literature.
palladium chloride complexes than on the allyl palladium
complexes with a tetrafluoroborate counterion, reactions of
dienes catalyzed by the combination of Xantphos and [Pd(η3-
allyl)Cl]2 occurred in higher yields than reactions catalyzed by
the combination of Xantphos and [Pd(CH3CN)4](BF4)2.
General Procedure for the Preparation of [(Bisphosphine)Pd-
(η3-allyl)]Cl. Into a 50 mL round-bottom flask equipped with a
magnetic stirbar was placed [Pd(η3-allyl)Cl]2 (200 mg, 0.547 mmol).
Into a second 50 mL round-bottom flask equipped with a magnetic
stirbar was placed 2 equiv of bisphosphine (1.09 mmol). Benzene (15
mL) was added to each flask, and the resulting solutions stirred for 10
min at 50 °C to ensure complete dissolution. The solutions were
subsequently combined. After stirring for 30 min at 50 °C, the resulting
solution/suspension was allowed to cool to room temperature. Com-
plexes that precipitated were isolated by filtration in air and were
washed 2 times with 50 mL of Et2O and dried in vacuo. Complexes
that were soluble in benzene were isolated by removing all volatiles
and recystallizing the residue from dichloromethane and Et2O or
pentane. The resulting crystals were isolated from the supernatant,
washed with 2 × 10 mL of Et2O or pentane, and dried in vacuo.
General Procedure for the Preparation of [(Bisphosphine)Pd-
(η3-allyl)]X (X ) BF4 or OTF). Into a 20 mL scintillation vial equipped
with a magnetic stirbar was placed [Pd(η3-allyl)Cl]2 (29.4 mg, 0.0804
mmol) and 2 equiv of bisphosphine (0.161 mmol). Dichloromethane
(4 mL) was added, and the solution was stirred for 10 min. The
appropriate silver salt (2 equiv, 0.161 mmol) was added to the stirring
solution, and the resulting suspension was stirred for 30 min. Filtration
through Celite yielded a clear solution, which was concentrated to 1
mL under reduced pressure, layered with Et2O or pentane, and cooled
at -35 °C for 12 h. The resulting crystals were washed 2 times with
3 mL of Et2O or pentane and dried in vacuo.
General Procedure for the Preparation of [(Bisphosphine)Pd-
(benzyl)Cl]. Into a 20 mL scintillation vial equipped with a magnetic
stirbar was placed (COD)Pd(CH2Ph)Cl (50.0 mg, 0.147 mmol) and 1
equiv of bisphosphine (0.147 mmol). Dichloromethane (4 mL) was
added, and the solution was stirred for 10 min. Filtration through Celite
yielded a clear solution, which was concentrated to 1 mL under reduced
pressure, layered with pentane, and cooled at -35 °C. The resulting
crystals were isolated from the supernatant, washed with 2 × 5 mL of
pentane, and dried in vacuo.
General Procedure for the Preparation of [(Bisphosphine)Pd-
(η3-benzyl)]X (X ) BF4 or OTF). Into a 20 mL scintillation vial
equipped with a magnetic stirbar was placed (COD)Pd(CH2Ph)Cl (50.0
mg, 0.147 mmol) and 1 equiv of bisphosphine (0.147 mmol). Dichlo-
romethane (5 mL) was added, and the solution was stirred for 10 min.
The appropriate silver salt (1 equiv, 0.147 mmol) was added to the
stirring solution, and the resulting suspension was stirred for 10 min.
Filtration through Celite yielded a clear solution, which was concen-
trated to 1 mL under reduced pressure, layered with Et2O or pentane,
and cooled to -35 °C. The resulting crystals were isolated from the
supernatant, washed with 2 × 5 mL of Et2O or pentane, and dried in
vacuo.
Conclusions
Highly active late transition metal catalysts for hydroamina-
tions are needed to accomplish olefin hydroamination in a
synthetically valuable fashion. Although much progress must
be made to develop transition metal-catalyzed hydroamination,
we have demonstrated that improved catalysts for additions of
arylamines to vinylarenes and dienes can be developed and that
these improved catalysts can tolerate the types of functionality
commonly found in natural products and pharmaceutically active
materials. This work has shown that palladium complexes of
Xantphos are much more reactive catalysts for the hydroami-
nation of dienes and vinylarenes than catalysts reported previ-
ously for this reaction. These reactions occur selectively to add
the amine N-H bond of arylamines to vinylarenes in the
presence of nitrile, nitro, ester, amide, carboxylic acid, phenolic
hydroxyl, hydroxyalkyl, and enolizable keto groups.
In addition to demonstrating increased activity and high
functional group tolerance from catalysts with Xantphos as
ligand, we have revealed some of the factors that control the
rates of the catalytic reactions and the rate of the C-N bond-
forming step. By monitoring the reactions of isolated bisphos-
phine-ligated allyl- and benzylpalladium complexes with amines,
we demonstrated a correlation between bite angle and rate of
nucleophilic attack on η3-benzyl complexes and unsymmetrical
η3-allyl complexes. We also showed that the population of active
catalyst in the hydroaminations of dienes depends on the bite
angle of the ligand and that the concentration of the allylpal-
ladium complex may be the major factor that influences the
rate of the catalytic hydroamination of symmetrical dienes.
Finally, while the origin of the differences in the rate of attack
as a function of counterion remain to be delineated, we have
shown that the rate of attack is affected significantly by the
counterion, even when it is not directly coordinated to the metal.
Experimental Procedures
General Experimental Procedure and Reagent Availability. All
manipulations were carried out under an inert atmosphere using a
nitrogen-filled glovebox or standard Schlenk techniques. All glassware
was oven or flame dried immediately prior to use. THF and diethyl
ether were obtained as HPLC grade without inhibitors; benzene, toluene,
dichloromethane, and pentane were obtained as ACS reagent grade.
THF, diethyl ether, benzene, toluene, dichloromethane, and pentane
were degassed by purging with nitrogen for 45 min and dried with a
solvent purification system containing a 1 m column containing
activated alumina. All reagents were obtained from commercial sources
General Procedure for the Preparation of [(Bisphosphine)Pd-
(η3-1,1-dimethylallyl)]OTf. Into a 20 mL scintillation vial equipped
with a magnetic stirbar was placed [Pd(η3-1,1-dimethylallyl)Cl]2 (40.0
mg, 0.0948 mmol) and 2 equiv of bisphosphine (0.190 mmol).
Dichloromethane (5 mL) was added, and the solution was stirred for
10 min. Silver trifluoromethanesulfonate (2 equiv, 48.7 mg, 0.190
mmol) was added to the stirring solution, and the resulting suspension
was stirred for 30 min. Filtration through Celite yielded a clear solution,
which was concentrated to 1 mL under reduced pressure, layered with
Et2O or pentane, and cooled at -35 °C for 12 h. The resulting
precipitate was washed 2 times with 3 mL of Et2O or pentane and
dried in vacuo.
1
and used without further purification. H NMR spectra were obtained
at 400 or 500 MHz and recorded relative to residual protio-solvent.
13C NMR spectra were obtained at 101 or 126 MHz, and chemical
shifts were recorded relative to the solvent resonance. 31P NMR spectra
were obtained at 122, 162, or 202 MHz, and chemical shifts are reported
relative to 85% H3PO4. [Pd(η3-allyl)Cl]261 and [Pd(η3-1,1-dimethylallyl)-
(60) Malaise, G.; Barloy, L.; Osborn, J. A.; Kyritsakas, N. C. R. Chim. 2002, 5,
(62) Auburn, P. R.; Mackenzie, P. B.; Bosnich, B. J. Am. Chem. Soc. 1985,
107, 2033-46.
(63) Stockland, R.; Anderson, G.; Rath, N.; Braddock-Wilking, J.; Ellegood, J.
Can. J. Chem. 1996, 74, 1990-1997.
289-296.
(61) Palenik, R. C.; Palenik, G. J. Synth. Inorg. Met. Org. Chem. 1992, 22,
1395-1399.
9
1838 J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006

Active Pd Catalyst for Intermolecular Hydroamination
A R T I C L E S
General Procedure for the Hydroamination of 1,3-Cyclohexa-
diene with Functionalized Anilines (Table 3). In a drybox, Xant-
phosPd(η3-allyl)Cl (1) (15.2 mg, 2.00 × 10-2 mmol), 1,3-cyclohexa-
diene (191 µL, 2.00 mmol), and the aniline (1.00 mmol) were placed
into a small vial, dissolved in 2.00 mL of tetrahydrofuran, and sealed
with a cap containing a PTFE septum. The reaction mixture was stirred
at 25 °C for 24 h. Upon completion, the reaction mixture was directly
adsorbed onto silica gel and purified by flash column chromatography.
General Procedure for the Hydroamination of Styrene with
Functionalized Anilines (Table 4). In a drybox, XantphosPd(η3-allyl)-
OTf (2) (17.5 mg, 2.00 × 10-2 mmol), styrene (229 µL, 2.00 mmol),
and the aniline (1.00 mmol) were placed into a small vial, dissolved in
1.00 mL of 1,4-dioxane, and sealed with a cap containing a PTFE
septum. The reaction mixture was stirred at 100 °C for 12 h. Upon
completion, the reaction mixture was directly adsorbed onto silica gel
and purified by flash column chromatography.
Determination of the Rate Constants (kobs) for the Nucleophilic
Attack of Aniline and Substituted Anilines on Benzylpalladium
Complexes. Benzylpalladium complexes (4-10) (0.0108 mmol), PPh3
(0.0108 mmol), and 1,3,5-trimethylbenzene (0.00719 mmol, internal
standard) were dissolved in a mixture of CD2Cl2 and THF-d8 (0.2 mL:
0.4 mL), and the solution was transferred to a screw-capped NMR tube.
The arylamines (0.054 mmol) in 0.2 mL THF-d8 were added by syringe
into the NMR tube immediately before the tube was placed in a probe
at the appropriate temperature. Decay of the CH2Ph resonance was
monitored by 1H NMR spectroscopy over at least three half-lives with
an automated data collection program. Kinetic data were fit to the
expression y ) m1 + m2 exp-kt, in which k is the first-order rate constant
tetrahydrofuran were loaded into a screw-capped NMR tube, followed
by 0.50 mL of CH2Cl2. Aniline (1.98 mmol) was added by syringe
into the NMR tube immediately before it was placed in a probe at the
appropriate temperature. Decay of the allylpalladium complexes was
monitored by 31P NMR spectroscopy over at least three half-lives with
an automated data collection program. Kinetic data were fit to the
expression y ) m1 + m2 exp-kt, in which k is the first-order rate constant
kobs
.
Determination of the Rate Constants (kobs) for the Nucleophilic
Attack of Aniline on 1,1-Dimethylallylpalladium Complexes. 1,1-
Dimethylallylpalladium complexes (19-22) (0.0197 mmol), diphenyl-
acetylene (0.296 mmol), and a capillary containing PPh3 in tetrahy-
drofuran were loaded into a screw-capped NMR tube, followed by 0.50
or 0.17 mL of CH2Cl2. Aniline (0.295 or 3.95 mmol) was added by
syringe into the NMR tube immediately before it was placed in a probe
at the appropriate temperature. Decay of the 1,1-dimethylallylpalladium
complexes was monitored by 31P NMR spectroscopy over at least three
half-lives with an automated data collection program. Kinetic data were
fit to the expression y ) m1 + m2 exp-kt, in which k is the first-order
rate constant kobs
.
Acknowledgment. M.U. thanks Mitsubishi Chemical Cor-
poration for support. We thank the NIH-NIGMS for support,
Union Carbide and Boehringer-Ingelheim for unrestricted gifts,
and Johnson-Matthey for palladium.
Supporting Information Available: Spectroscopic and ana-
lytical data of new compounds, information on procedures, and
details of crystal structure analysis of complexes 1, 6, 7, 9, 12,
18, and 24 (PDF and CIF). This material is available free of
charge via the Internet at http://pubs.acs.org.
kobs
.
Determination of the Rate Constants (kobs) for the Nucleophilic
Attack of Aniline and Substituted Anilines on Allylpalladium
Complexes. Allylpalladium complexes (1, 2, 11-15) (0.0197 mmol),
diphenylacetylene (0.296 mmol), and a capillary containing PPh3 in
JA056003Z
9
J. AM. CHEM. SOC. VOL. 128, NO. 6, 2006 1839