Halide and amine influence in the equilibrium formation of palladium tris(o-tolyl)phosphine mono(amine) complexes from palladium aryl halide dimers

Article abstract of DOI:10.1021/om9509608

The relative binding constants (K_b) for the coordination of amines to the palladium fragment Pd[P(o-tol)₃](p-C₆H₄Me)Cl were determined by ¹H NMR spectroscopy and decrease in the order hexylamine > ben

Full text of DOI:10.1021/om9509608

Organometallics 1996, 15, 2755-2763
2755
Ha lid e a n d Am in e In flu en ce in th e Equ ilibr iu m
F or m a tion of P a lla d iu m Tr is(o-tolyl)p h osp h in e
Mon o(a m in e) Com p lexes fr om P a lla d iu m Ar yl Ha lid e
Dim er s
Ross A. Widenhoefer and Stephen L. Buchwald*
Department of Chemistry, Massachusetts Institute of Technology,
Cambridge, Massachusetts 02139
Received December 14, 1995X
The relative binding constants (Kb) for the coordination of amines to the palladium fragment
1
Pd[P(o-tol)3](p-C6H4Me)Cl were determined by H NMR spectroscopy and decrease in the
order hexylamine > benzylamine ≈ cyclohexylamine ≈ piperidine > dibutylamine ≈
diethylamine ≈ N-benzylmethylamine > morpholine > diisobutylamine > dibenzylamine ≈
tert-octylamine >> diisopropylamine > N-methylaniline. The palladium halide dimers {Pd-
[P(o-tol)3](p-C6H4Me)(µ-X)}2 (X ) Cl (1), Br (2), I (3)) react reversibly with dibenzylamine to
generate the corresponding 1:1 palladium amine adducts Pd[P(o-tol)3](p-C6H4Me)(HNBn2)X
(X ) Cl (12), K ) 6 ( 1 × 103 M-1; X ) Br (18), K ) 3.5 ( 0.5 × 103 M-1; X ) I (22), K ) 90
( 20 M-1), respectively. The related reaction of dibenzylamine with the iodide dimer {Pd-
[P(o-tol)3](p-C6H4OMe)(µ-I)}2 (21) to form Pd[P(o-tol)3](p-C6H4OMe)(HNBn2)I (24) provided
the thermodynamic parameters ∆G298 K ) -3.1 ( 0.1 kcal mol-1, ∆H298 K ) -11.9 ( 0.1 kcal
mol-1, and ∆S298 K ) -30 ( 4 eu. Dimers 1-3 also react reversibly with diisopropylamine
at 25 °C to form the amine adducts Pd[P(o-tol)3](p-C6H4Me)[HN(i-Pr2)]X (X ) Cl (17), K )
14 ( 3 M-1; X ) Br (19), K ) 2.8 ( 0.5 M-1; X ) I (26), K ) 6 ( 2 × 10-3 M-1), respectively.
Sch em e 1
In tr od u ction
Palladium tris(o-tolyl)phosphine complexes catalyze
the cross-coupling of aryl bromides with aminostan-
nanes1 and also catalyze the conversion of aryl bromides
to anilines via reaction with free amine and sodium tert-
butoxide.2 The corresponding reaction of aryl iodides
and amines required higher temperatures and produced
somewhat lower yields of cross-coupled product than did
aryl bromides.3 Likewise, unbranched secondary amines
such as N-benzylmethylamine were considerably more
efficient as coupling partners than were either primary
amines or bulky secondary amines. The tin-free reac-
tion is believed to occur via initial oxidative addition of
the aryl halide to a palladium mono(phosphine) species
to form the palladium halide dimer A,1c,4 which reacts
with amine to form the corresponding palladium amine
monomer B (Scheme 1).5 Deprotonation of B and
reductive elimination from the three-coordinate pal-
ladium amido complex C1c forms the corresponding
aniline derivative and regenerates the catalytically
active mono(phosphine) species Pd[P(o-tol)3] (Scheme
1).4
reaction to the nature of the halide and the amine. For
example, we have recently shown that the palladium
halide dimers {Pd[P(o-tol)3](p-C6H4Me)(µ-X)}2 (o-tol )
o-tolyl; X ) Cl, (1) Br (2), I (3)) react with N-benzyl-
methylamine to generate the corresponding 1:1 amine
adducts Pd[P(o-tol)3](p-C6H4Me)[HN(Me)Bn]X (X ) Cl,
(4) Br (5), I (6)) (Scheme 2).6 The iodide mono(amine)
adduct 6 was approximately 35% dissociated in CDCl3
at room temperature, forming iodide dimer 3 and free
amine, while the corresponding chloride derivative 4
and bromide derivative 5 revealed no evidence for amine
dissociation.6 Although formation of 6 was clearly less
favorable than formation of 4 or 5, the high binding
affinity of N-benzylmethylamine precluded the quanti-
tative thermodynamic comparison of the reaction of
N-benzylmethylamine with dimers 1-3. Here we report
We have been investigating the coordination chem-
istry of the 1:1 palladium amine adducts B in an effort
to gain insight into the palladium-catalyzed amination
reaction, particularly the observed sensitivity of the
X Abstract published in Advance ACS Abstracts, April 15, 1996.
(1) (a) Kosugi, M.; Kameyama, M.; Migita, T. Chem. Lett. 1983, 927.
(b) Guram, A. S.; Buchwald, S. L. J . Am. Chem. Soc. 1994, 116, 7901.
(c) Paul, F.; Patt, J .; Hartwig, J . F. J . Am. Chem. Soc. 1994, 116, 5969.
(2) (a) Guram, A. S.; Rennels, R. A.; Buchwald, S. L. Angew. Chem.,
Int. Ed. Engl. 1995, 34, 1348. (b) Louie, J .; Hartwig, J . F. Tetrahedron
Lett. 1995, 36, 3609.
(3) Wolfe, J . N.; Buchwald, S. L. J . Org. Chem. 1996, 61, 1133.
(4) Hartwig, J . F.; Paul, F. J . Am. Chem. Soc. 1995, 117, 5373.
(5) Paul, F.; Patt, J .; Hartwig, J . F. Organometallics 1995, 14, 3030.
(6) Widenhoefer, R. A.; Zhong, A. H.; Buchwald, S. L. Organome-
tallics 1996, 15, 2745.
S0276-7333(95)00960-5 CCC: $12.00 © 1996 American Chemical Society

2756 Organometallics, Vol. 15, No. 12, 1996
Widenhoefer and Buchwald
Sch em e 4
Sch em e 2
Sch em e 3
and Pd[P(o-tol)3](p-C6H4OMe)(HNBn2)I (24), respec-
tively. Anisyl dimer 21 also reacted with tert-octyl-
amine to form Pd[P(o-tol)3](p-C6H4OMe)(H2NCMe2CH2-
CMe3)I (25). Addition of diisopropylamine to a CD2Cl2
solution of iodide dimer 3 formed the corresponding
adduct Pd[P(o-tol)3](p-C6H4Me)[HN(i-Pr)2]I (26), as de-
the binding constants (Kb) for the coordination of amines
of varying pKa and steric bulk to the palladium chloride
fragment Pd[P(o-tol)3](p-C6H4Me)Cl. We also report the
quantitative comparison of the equilibrium constants
(K) for the reaction of halide dimers 1-3 with weakly
binding amines.
1
termined by H and 31P NMR spectroscopy (see below),
although the unfavorable equilibrium precluded isola-
tion of 26. For example, addition of hexane to a solution
of dimer 3 in neat diisopropylamine followed by cooling
to -20 °C led to the precipitation of dimer 3 only; the
Resu lts
1
corresponding amine adduct 26 was not detected by H
Syn th esis of P alladiu m Mon o(am in e) Com plexes.
Palladium amine adducts were synthesized by reaction
of free amine and the desired palladium aryl halide
dimers, as has been described previously (Scheme 3,
Table 1).6 For example, reaction of piperidine and
chloride dimer 1 in CH2Cl2 gave a clear solution which
was concentrated and diluted with hexane to give the
1:1 palladium-amine adduct Pd[P(o-tol)3](p-C6H4Me)-
(piperidine)Cl (7) in 81% yield (Table 1). Likewise, 1
reacted with dibutylamine, diethylamine, morpholine,
diisobutylamine, dibenzylamine, hexylamine, benzyl-
amine, cyclohexylamine, tert-octylamine, and diisopro-
pylamine to form the corresponding amine adducts Pd-
[P(o-tol)3](p-C6H4Me)[HN(n-Bu)2]Cl (8), Pd[P(o-tol)3](p-
C6H4Me)(HNEt2)Cl (9), Pd[P(o-tol)3](p-C6H4Me)(mor-
pholine)Cl (10), Pd[P(o-tol)3](p-C6H4Me)[HN(i-Bu)2]Cl
(11), Pd[P(o-tol)3](p-C6H4Me)(HNBn2)Cl (12), Pd[P(o-
tol)3](p-C6H4Me)(H2N-n-hex)Cl (13),7 Pd[P(o-tol)3](p-
C6H4Me)(H2NBn)Cl (14),6,7 Pd[P(o-tol)3](p-C6H4Me)-
(H2NCy)Cl (15), Pd[P(o-tol)3](p-C6H4Me)(H2NCMe2CH2-
CMe3)Cl (16), and Pd[P(o-tol)3](p-C6H4Me)[HN(i-Pr)2]Cl
(17), respectively.
Amine adducts derived from cleavage of the halide
bridge in palladium bromide and iodide dimers were
isolated by a similar procedure (Scheme 3, Table 1). For
example, bromide dimer 2 reacted with dibenzylamine
and diisopropylamine to form Pd[P(o-tol)3](p-C6H4Me)-
(HNBn2)Br (18) and Pd[P(o-tol)3](p-C6H4Me)[HN(i-Pr)2]-
Br (19), respectively. Reaction of dibenzylamine with
iodide dimers 3, {Pd[P(o-tol)3](p-C6H4CMe3)(µ-I)}2 (20),6
and Pd[P(o-tol)3](p-C6H4OMe)(µ-I)}2 (21)6 gave the cor-
responding amine adducts Pd[P(o-tol)3](p-C6H4Me)-
(HNBn2)I (22), Pd[P(o-tol)3](p-C6H4CMe3)(HNBn2)I (23),
NMR spectroscopy.
Deter m in a tion of Am in e Bin d in g Con sta n ts.
Rapid displacement of the coordinated N-benzylmethyl-
amine of 4 by free amine6 allowed the determination of
the relative binding constants (Kb) for the coordination
of amines to the three-coordinate palladium fragment
Pd[P(o-tol)3](p-C6H4Me)Cl by 1H NMR spectroscopy.8
For example, addition of a 3.0:1.0 mixture of N-benzyl-
methylamine and piperidine (8 × 10-2 mmol total) to a
solution of chloride dimer 1 (6 × 10-3 mmol) in CDCl3
formed a ∼1:1 mixture of p-tolyl resonances correspond-
1
ing to 4 (δ 2.03) and 7 (δ 2.08) in the H NMR spectrum
at 55 °C.8 With Kb for N-benzylmethylamine arbitrarily
set to unity, the binding constant for piperidine relative
to N-benzylmethylamine was calculated to be Kb ) [7]
[HN(Me)Bn]/[4][piperidine] ) 3.1 ( 0.4 (∆Gb ) -0.8 (
0.1 kcal mol-1) (Scheme 4, Table 2). Likewise, the
binding constants for dibutylamine (Kb ) 1.1 ( 0.2),
diethylamine (Kb ) 1.1 ( 0.2), morpholine (Kb ) 0.56 (
0.09), diisobutylamine (Kb) 0.28 ( 0.04), hexylamine
(Kb ) 7 ( 1), benzylamine (Kb ) 3.6 ( 0.6), cyclohexyl-
amine (Kb ) 3.3 ( 0.5), and tert-octylamine (Kb ) 0.12
( 0.02) were determined relative to N-benzylmethyl-
amine by a procedure analogous to that used to deter-
mine Kb for piperidine (Table 2).
The p-tolyl resonances for N-benzylmethylamine ad-
duct 4 and dibenzylamine adduct 12 were not well-
separated in the 1H NMR spectrum. Conversely, the
(8) At 55 °C in CDCl3, the p-tolyl methyl resonances of the amine
adducts were sharp and well-resolved. At lower temperatures, the
p-tolyl methyl resonances of the amine adducts were partially obscured
by the broad P(o-tol)3 methyl resonance. Likewise, the p-tolyl methyl
resonances of the palladium amine adducts derived from palladium
bromide dimer 2 or palladium iodide dimer 3 were considerably less
sharp than were the corresponding chloride derivatives.
(9) (a) Albert, A.; Serjeant, E. P. The Determination of Ionization
Constants, 3rd ed.; Chapman and Hall: London, 1984. (b) Smith, R.
M.; Martell, A. E. Critical Stability Constants; Plenum Press: New
York, 1989; Vol. 6, 2nd supplement. (c) Perrin, D. D. Stability Constants
of Metal-Ion Complexes: Part B- Organic Ligands; IUPAC Pergamon
Press: London, 1979.
(7) The syntheses of the benzylamine derivative 11 and hexylamine
derivative 13 were complicated by formation of the bis(amine) com-
plexes Pd(p-C6H4Me)[H2NBn]2Cl and Pd(p-C6H4Me)[H2N-n-hex]2Cl,
respectively. The rate and extent of formation of bis(amine) complexes
is halide dependent. We are currently investigating the formation of
palladium bis(amine) complexes in detail as a potential chain-limiting
step in the catalytic cross-coupling of aryl bromides and primary
amines.
(10) Seligson, A. L.; Trogler, W. C. J . Am. Chem. Soc. 1991, 113,
2520.

Pd Tris(o-tolyl)phosphine Mono(amine) Complexes
Organometallics, Vol. 15, No. 12, 1996 2757
Ta ble 1. P a lla d iu m Am in e Ad d u cts Isola ted fr om Rea ction of Am in e a n d P a lla d iu m Ha lid e Dim er s
a
Reference 6.
p-tolyl resonances for morpholine adduct 10 (δ 2.07) and
Equ ilibr iu m F or m a tion of P a lla d iu m Am in e
Addu cts fr om Reaction of P alladiu m Halide Dim er s
w ith Am in es. In contrast to the corresponding N-
benzylmethylamine adduct 4,6 palladium chloride com-
plexes of weakly binding amines (Table 2, Kb < 0.3)
dissociated in CDCl3 to form an equilibrium mixture of
chloride dimer, amine adduct, and free amine. For
dibenzylamine adduct 12 (δ 2.03) were well-resolved
with an equilibrium constant of K ) [12][morpholine]/
[10][HNBn2] ) 0.22 ( 0.04 (Scheme 5), which corre-
sponds to a binding constant of Kb ) 0.13 ( 0.04 relative
to HN(Me)Bn (Table 2). The 1H NMR spectrum of
N-benzylmethylamine adduct 4 (15 mM) in CDCl3 which
contained 0.15 M N-methylaniline showed no evidence
for N-benzylmethylamine displacement. If we make the
conservative assumption that g10% of the N-benzyl-
methylamine dissociation could be detected by 1H NMR
spectroscopy, a binding constant of e0.001 can be
estimated for N-methylaniline. In addition, although
the low affinity of diisopropylamine precluded direct
comparison with N-benzylmethylamine, a binding con-
stant of Kb ) ∼1.6 × 10-3 was estimated from the
equilibrium constants (K) for the reactions of 1 with
diisopropylamine and dibenzylamine to form 17 and 12,
respectively (see below).
1
example, the H NMR spectrum of dibenzylamine ad-
duct 12 (16 mM) in CDCl3 at 53 °C displayed a 21:79
ratio of p-tolyl resonances for 1 (δ 1.98) and 12 (δ 2.03)
(1:12 ) 12:88), which corresponds to an equilibrium
constant for formation of 12 of K ) [12]2/[1][NHBn2]2 )
6 ( 1 × 103 M-1 (∆G326 ) -5.6 ( 0.2 kcal mol-1
)
K
(Scheme 6, Table 3). The equilibrium constant for the
formation of bromide adduct 18 from 2 and dibenzyl-
amine was slightly less favorable (K ) (3.5 ( 0.5) × 103
M-1; ∆G326
) -5.1 ( 0.1 kcal mol-1) than the
K
corresponding reaction of dibenzylamine and chloride
dimer 1. In addition, the equilibrium constant for the

2758 Organometallics, Vol. 15, No. 12, 1996
Widenhoefer and Buchwald
Sch em e 6
Ta ble 2. Com p a r ison of Equ ilibr iu m Bin d in g
Con sta n ts (Kb) a n d Con e An gles for th e
Coor d in a tion of Am in es w ith th e P a lla d iu m
F r a gm en t P d [P (o-tol)3](p-C6H4Me)Cl, Deter m in ed
a t 55 °C in CDCl3
solvent dependence of K was determined for the equi-
librium formation of 23 from iodide dimer 20 and
dibenzylamine and ranged from 75 ( 15 M-1 in CDCl3
to 20 ( 4 M-1 in dioxane-d8 (Scheme 6, Table 3). In
addition, K was determined for the equilibrium forma-
tion of 24 from 21 and dibenzylamine over the temper-
ature range 9-58 °C (Scheme 6, Table 3). A van’t Hoff
plot of the data allowed calculation of the thermody-
namic parameters ∆G298 ) -3.1 ( 0.1 kcal mol-1
,
K
∆H298 K ) -11.9 ( 0.1 kcal mol-1, and ∆S298 K ) -30 (
4 eu (Figure 1). Similarly, K for the formation of 25
from reaction of tert-octylamine and 21 was determined
over the temperature range 14-57 °C to provide the
thermodynamic parameters ∆G298 K ) -2.8 ( 0.1 kcal
mol-1, ∆H298 K ) -12.0 ( 0.1 kcal mol-1, and ∆S298 K
)
-31 ( 4 eu (Scheme 7, Table 4, Figure 1). The
temperature dependence of the reactions of 21 with
dibenzylamine and tert-octylamine reveal that ∆S is
relatively insensitive to the amine.
a
Determined relative to HN(Me)Bn; each Kb value represents
b
The equilibrium constants were also determined for
the reactions of palladium halide dimers 1-3 with
diisopropylamine to form 17, 19, and 26, respectively.
These equilibria displayed similar halide dependence
but were considerably less favorable than the corre-
sponding reactions of 1-3 with dibenzylamine. For
example, K for the reaction of 1 with diisopropylamine
to form 17 was 14 ( 3 M-1 at 25 °C (∆G298 K ) -1.6 (
0.2 kcal mol-1) (Scheme 8). When the temperature
dependence of K determined for reaction of 21 with
dibenzylamine is employed, an equilibrium constant of
∼0.8 M-1 (∆G328 K ≈ 0.1 kcal mol-1) can be estimated
for the reaction of 1 with diisopropylamine to form 17
at 55 °C, which is ∼6 kcal mol-1 (∼3 kcal mol-1 per
Pd-N bond formed) less favorable than reaction of 1
with dibenzylamine to form 12. The equilibrium con-
stant for the reaction of bromide dimer 2 with diisopro-
pylamine to form 19 (K ) 2.8 ( 0.5 M-1; ∆G298 K ) -0.6
( 0.1 kcal mol-1) was slightly less favorable than for
the corresponding reaction of diisopropylamine with 1.
The equilibrium constant for the reaction of iodide dimer
3 with diisopropylamine to form 26 (K ) (6 ( 2) × 10-3
M-1; ∆G298 K ) 3.0 ( 0.3 kcal mol-1) was considerably
less favorable than reaction of diisopropylamine with
either 1 or 2 (Scheme 8). In addition, the free energy
difference between the reaction of 1 and 3 with diiso-
the average of two separate measurements. pKa values taken
from ref 9. c Cone angles taken from ref 10. Cone angle estimated
d
to be equivalent to that of HN(n-Pr)2. e Cone angle estimated to
be equivalent to that of piperidine. f pKa estimated to be equivalent
g
to that of HNBu2. Determined relative to morpholine (see text).
h
i
pKa estimated to be equivalent to that of H2N-t-Bu. Cone angle
estimated to that of be equivalent to that of adamantylamine. j Kb
estimated from reactions of HNBn2 and HN(i-Pr)2 with 1 (see text).
Sch em e 5
formation of iodide adduct 22 from 3 and dibenzylamine
(K ) 90 ( 20 M-1; ∆G326 ) -2.9 ( 0.1 kcal mol-1
)
K
was considerably less favorable than for the reaction of
dibenzylamine with 1 or 2 (Scheme 6, Table 3).
Determination of the solvent and temperature depen-
dence of K for reactions of dibenzylamine with dimers
1-3 was precluded by the poor resolution of the respec-
tive p-tolyl resonances at temperatures lower than ∼55
°C and in solvents other than CDCl3. However, the
propylamine (∆∆G298
) 4.6 ( 0.5 kcal mol-1) was
K
significantly larger than the free energy difference
between the reactions of 1 and 3 with dibenzylamine
(∆∆G328 ) 2.7 kcal mol-1).
K

Pd Tris(o-tolyl)phosphine Mono(amine) Complexes
Organometallics, Vol. 15, No. 12, 1996 2759
Ta ble 3. Equ ilibr iu m Con sta n ts for Rea ction of HNBn 2 w ith P a lla d iu m Ha lid e Dim er s
entry no.
equilibrium
T (°C)
solvent
CDCl3
CDCl3
CDCl3
CDCl3
C6D6
THF-d8
dioxane-d8
CDCl3
CDCl3
CDCl3
CDCl3
CDCl3
K (M-1
)
∆G (kcal/mol)
1
2
3
4
5
6
7
8
9
[1] + 2[HNBn2] h 2[12]
[2] + 2[HNBn2] h 2[18]
[3] + 2[HNBn2] h 2[22]
[20] + 2[HNBn2] h 2[23]
[20] + 2[HNBn2] h 2[23]
[20] + 2[HNBn2] h 2[23]
[20] + 2[HNBn2] h 2[23]
[21] + 2[HNBn2] h 2[24]
[21] + 2[HNBn2] h 2[24]
[21] + 2[HNBn2] h 2[24]
[21] + 2[HNBn2] h 2[24]
[21] + 2]HNBn2] h 2[24]
53
53
53
24
24
24
24
9
22
32
45
58
(6 ( 1) × 103
(3.5 ( 0.5) × 103
90 ( 20
-5.6 ( 0.2
-5.1 ( 0.1
-2.9 ( 0.1
-2.5 ( 0.1
-2.1 ( 0.1
-2.0 ( 0.1
-1.8 ( 0.1
-3.3 ( 0.1
-3.1 ( 0.1
-2.8 ( 0.1
-2.4 ( 0.1
-1.9 ( 0.1
75 ( 15
30 ( 5
26 ( 4
20 ( 4
(3.7 ( 0.7) × 102
(2.1 ( 0.4) × 102
(1.0 ( 0.1) × 102
43 ( 7
10
11
12
18 ( 3
Sch em e 8
and steric bulk to the kinetic or thermodynamic binding
affinity of amines to transition metals.15 While pKa has
typically been employed as a measure of amine basicity,
several measures of amine steric bulk have been devel-
oped.16 For example, Trogler has determined amine
cone angles (θ, in deg) from space-filling CPK models10
by the method of Tolman.17 Molecular mechanics were
employed to determine a series of amine solid cone
angles (θ′) in an effort to more accurately assess the
steric bulk of unsymmetric amines.18 Likewise, molec-
ular mechanics were employed to determine the ligand
energy repulsive parameter (ER, in kcal mol-1) for a
series of amines, which estimates the van der Waals
repulsion between the lowest energy conformation of the
amine and the carbonyl ligands in the hypothetical
amine complexes Cr(CO)5(amine).19 The conformation
of the amine determined from molecular mechanics
minimization may differ significantly from the confor-
F igu r e 1. van’t Hoff plots for the reaction of 21 with
dibenzylamine to form 24 (denoted by ×) and reaction of
tert-octylamine with 21 to form 25 (denoted by O).
Sch em e 7
Ta ble 4. Tem p er a tu r e Dep en d en ce of K for th e
Rea ction of ter t-Octyla m in e w ith 21 to for m 25 in
CDCl3
(12) (a) Baralt, E.; Smith, S. J .; Hurwitz, J .; Horvath, I. T.; Fish, R.
H. J . Am. Chem. Soc. 1992, 114, 5187. (b) Gray, S. D.; Smith, D. P.;
Bruck, M. A. J . Am. Chem. Soc. 1992, 114, 5462. (c) Fish, R. H.; Baralt,
E.; Smith, S. J . Organometallics 1991, 10, 54. (d) Satterfield, C. N.
Heterogeneous Catalysis in Practice, 2nd ed.; McGraw-Hill: New York,
1991.
(13) Ball, G. E.; Cullen, W. R.; Fryzuk, M. D.; Henderson, W. J .;
J ames, B. R.; MacFarlane, K. S. Inorg. Chem. 1994, 33, 1464.
(14) Volt, J .; Pasek, J . In Catalytic Hydrogenation: Studies in
Surface Science and Catalysis; Cerverry, L., Ed.; Elsevier: Amsterdam,
1986; Vol. 27, Chapter 4.
T (°C)
K (M-1
)
∆G (kcal/mol)
14
22
32
42
45
57
(1.6 ( 0.3) × 102
(1.1 ( 0.1) × 102
47 ( 7
-2.9 ( 0.1
-2.8 ( 0.1
-2.3 ( 0.1
-2.1 ( 0.1
-1.9 ( 0.1
-1.6 ( 0.1
29 ( 5
20 ( 3
12 ( 2
(15) (a) Canovese, L.; Visentin, F.; Uguagliati, P.; Di Bianca, F.;
Antonaroli, S.; Crociani, B. J . Chem. Soc., Dalton Trans. 1994, 3113.
(b) Dennenberg, R. J .; Darensbourg, D. J . Inorg. Chem. 1972, 11, 72.
(c) Canovese, L.; Visentin, F.; Uguagliati, P.; Crociani, B.; Di Bianca,
F. Inorg. Chim. Acta 1995, 235, 45. (d) Uruska, I.; Zielkiewicz, J .;
Szpakowska, M. J . Chem. Soc., Dalton Trans. 1990, 733. (e) Therien,
M. J .; Trogler, W. C. J . Am. Chem. Soc. 1987, 109, 5127. (f) Burkey,
T. J . Polyhedron 1989, 8, 2681. (g) Uruska, I.; Zielkiewicz, J .;
Szpakowska, M. J . Chem. Soc., Dalton Trans. 1990, 733. (h) Brown,
H. C. Boranes in Organic Chemistry: Cornell University Press: Ithaca,
NY, 1972. (i) Deeming, A. J .; Rothwell, I. P.; Hursthouse, M. B.; New,
L. J . Chem. Soc., Dalton Trans. 1978, 1490. (j) Odell, A. L.; Raethel,
H. A. J . Chem. Soc., Chem. Commun. 1968, 1323.
Discu ssion
Bin d in g Con sta n ts for Coor d in a tion of Am in es
to th e P a lla d iu m F r a gm en t P d [P (o-tol)3](p-C6H4-
Me)Cl. The coordination or dissociation of an amine
serves as a key step in a variety of catalytic processes
such as the amination of aryl halides,2 hydroamination
of olefins,11 hydrodenitrogenation,12 asymmetric imine
hydrogenation,13 and nitrile reduction.14 As a result,
there has been an effort to correlate both amine basicity
(16) Brown, T. L.; Lee, K. J . Coord. Chem. Rev. 1993, 128, 89.
(17) Tolman, C. A. Chem. Rev. 1977, 77, 313.
(11) Gagne, M. R.; Stern, C. L.; Marks, T. J . J . Am. Chem. Soc. 1992,
114, 275.
(18) White, D.; Coville, N. J . Adv. Organomet. Chem. 1994, 36, 95.
(19) Choi, M. G.; Brown, T. L. Inorg. Chem. 1993, 32, 1548.

2760 Organometallics, Vol. 15, No. 12, 1996
Widenhoefer and Buchwald
the three-coordinate palladium fragment Pd[P(o-tol)3]-
(p-C6H4Me)Cl and an amine bound to the five-coordinate
chromium fragment Cr(CO)5.
Equ ilibr iu m F or m a tion of P a lla d iu m -Am in e
Ad d u cts fr om Ar yl Ha lid e Dim er s. Reaction of
dibenzylamine with palladium chloride dimer 1 to form
the corresponding monomer 12 cleaves two dative Pd-
(µ-Cl) bonds and forms two dative Pd-N bonds. The
enthalpy for the reaction can therefore be expressed as
∆H ) 2E[Pd(µ-Cl)] - 2E[Pd-N], where E designates
the energy of the respective dative bond. Assuming that
∆S ) -30 eu (as determined for the reaction of diben-
zylamine with 21), the enthalpy for the reaction of
dibenzylamine with 1 is estimated to be -14.5 kcal
mol-1. Because the dative Pd-N(piperidine) bond is 2.1
kcal mol-1 stronger than the Pd-N(dibenzylamine)
dative bond (Table 2, assuming ∆S is amine indepen-
dent), the enthalpy for the formation of 7 from 1 and
piperidine is estimated to be -18.7 kcal mol-1, which
is slightly larger than the bridge-cleavage enthalpies of
related reactions.22 For example, the enthalpy for the
cleavage of a dative dichloride bridge by piperidine
ranges from -15.7 kcal mol-1, for the reaction of
piperidine with the rhodium dicarbonyl complex
[Rh(CO)2Cl]2 to form Rh(CO)2Cl(piperidine),22a,23 to -9.4
kcal mol-1, for the reaction of piperidine with the
palladium π-allyl complex [Pd(η3-CH2C(Me)CH2)Cl]2 to
form Pd(η3-CH2C(Me)CH2)Cl(piperidine).24
The equilibrium constants (K) for the reaction of
dibenzylamine and diisopropylamine with halide dimers
1-3 to form the corresponding amine adducts were
halogen dependent and decreased in the order Cl > Br
>> I. The resistance of a palladium iodide dimer to
undergo bridge cleavage relative to the corresponding
bromide and chloride derivatives has been previously
observed. For example, reaction of PdI2 and excess P(o-
tol)3 led to the exclusive formation of the mono(phos-
phine) iodide dimer {Pd[P(o-tol)3](µ-I)I}2.6 Conversely,
the corresponding reactions of P(o-tol)3 with PdBr2 or
PdCl2 formed only the respective bis(phosphine) mono-
mers Pd[P(o-tol)3]2X2 (X ) Cl, Br).6,25 Similarly, Chatt
and co-workers observed that the palladium chloride
bridged dimer {Pd[P(n-Pr)3](µ-Cl)Cl}2 reacted with p-
toluidine to form the stable amine monomer Pd[P(n-Pr)3]-
(p-toluidine)Cl2. Conversely, addition of p-toluidine to
the palladium iodide bridged dimer {Pd[P(n-Pr)3](µ-I)I}2
(27) generated an equilibrium mixture of 27 and the
corresponding amine monomer Pd[P(n-Pr)3](p-toluidi-
ne)I2 (28). Isolation of monomeric 28 required both the
use of excess p-toluidine in the reaction with 27 and
recrystallization of 28 from a p-toluidine solution of light
petroleum ether.26
F igu r e 2. Plot of log Kb versus amine cone angle for
hexylamine, cyclohexylamine, piperidine, diethylamine,
dibutylamine, diisobutylamine, tert-octylamine, and diiso-
propylamine. Numerical labels correspond to entries in
Table 2.10
mation of the free amine or the folded-back conformation
assumed in the cone angle determination.19
The equilibrium binding constants (Kb) for coordina-
tion of amines to the palladium fragment Pd[P(o-tol)3]-
(p-C6H4Me)Cl were dependent on both the basicity and
steric bulk of the amine (Table 2). For example, Kb for
dibutylamine was >1100 times larger (∆∆Gb g 4.5 kcal
mol-1) than Kb for N-methylaniline, which possessed a
nearly identical cone angle ((1°) but has a pKa 6 units
lower than that of dibutylamine. Likewise, Kb for
hexylamine was ∼4000 times larger (∆∆Gb ∼5.4 kcal
mol-1) than Kb for diisopropylamine, which has a
comparable pKa but possesses a cone angle 30° larger
than that of hexylamine. In addition, a plot of log Kb
versus cone angle for amines with comparable pKa (10.9
( 0.25) revealed a nonlinear dependence of log Kb on
cone angle; log Kb decreased only slightly with increas-
ing cone angle below ∼120° but dropped rapidly with
increasing cone angle above ∼120° (Figure 2). The
dependence of log Kb on the cone angle is consistent with
the presence of a steric threshold which has been
previously observed in the correlation of both amine20
and phosphine21 binding constants with the respective
cone angles.
Although diisobutylamine and diisopropylamine pos-
sess nearly identical pKa values and cone angles, the
Kb value for diisobutylamine was ∼170 times larger
(∆∆Gb ≈ 3.5 kcal mol-1) than Kb for diisopropylamine.
Likewise, dibenzylamine possesses a cone angle 2°
larger than for diisopropylamine but has a Kb ∼100
times greater (∆∆Gb ≈ 3 kcal mol-1). These discrep-
ancies suggest that, despite the similarity in cone
angles, bound diisopropylamine possesses greater steric
bulk than either diisobutylamine or dibenzylamine. In
accord with these observed binding constants, the ligand
repulsive factor (ER) for diisopropylamine (105 kcal
mol-1) is considerably larger than ER for diisobutyl-
amine (85 kcal mol-1) (ER for dibenzylamine has not
been determined).19 However, little overall correlation
was observed between log Kb and ER, which may point
to different steric requirements for an amine bonded to
The enthalpy equation for cleavage of the halide
bridge of a palladium halide dimer by amine to form
(22) (a) Pribula, A. J .; Drago, R. S. J . Am. Chem. Soc. 1976, 98, 2784.
(b) Wang, K.; Rosini, G. P.; Nolan, S. P.; Goldman, A. S. J . Am. Chem.
Soc. 1995, 117, 5082. (c) Van Gaal, H. L. M.; Moers, F. G.; Steggerda,
J . J . J . Organomet. Chem. 1974, 65, C43. (d) Van Gaal, H. L. M.; Van
Den Bekerom, F. L. A. J . Organomet. Chem. 1977, 134, 237. (e)
Partenheimer, W.; Durham, B. Inorg. Nucl. Chem. Lett. 1974, 10, 1143.
(f) Gregory, N. W.; MacLaren, R. O. J . Phys. Chem. 1955, 59, 110.
(23) Drago, R. S.; Miller, J . G.; Hoselton, M. A.; Farris, R. D.;
Desmond, M. J . J . Am. Chem. Soc. 1983, 105, 444.
(24) Li, M. P.; Drago, R. S.; Pribula, A. J . J . Am. Chem. Soc. 1977,
99, 6900.
(25) Bennett, M. A.; Longstaff, P. A. J . Am. Chem. Soc. 1969, 91,
6266.
(20) Romeo, R.; Arena, G.; Scolaro, L. M.; Plutino, M. R.; Bruno, G.;
Nicolo, F. Inorg. Chem. 1994, 33, 4029.
(21) (a) Liu, H. Y.; Eriks, K.; Prock, A.; Giering, W. P. Organome-
tallics 1990, 9, 1758. (b) Rahman, M. M.; Liu, H. Y.; Prock, A.; Giering,
W. P. Organometallics 1987, 6, 650. (c) Romeo, R.; Arena, G.; Scolaro,
L. M. Inorg. Chem. 1992, 31, 4879.
(26) Chatt, J .; Venanzi, L. M. J . Chem. Soc. 1957, 2445.

Pd Tris(o-tolyl)phosphine Mono(amine) Complexes
Organometallics, Vol. 15, No. 12, 1996 2761
the corresponding 1:1 palladium amine adduct (∆H )
2E[Pd-(µ-halogen)] - 2E[Pd-N]) indicates that K will
decrease with increasing Pd-(µ-halogen) bond strength
and with decreasing Pd-N bond strength (assuming ∆S
is constant). The halide ligand can potentially affect
both the Pd-(µ-halogen) and Pd-N bond strengths
although the relative contributions of the two effects to
the halide dependence of K cannot be determined. For
example, the dative Pd-(µ-halogen) bond strength is
expected to parallel the polarizability and hence donicity
of the respective halide, which increases in the order
Cl < Br < I. Similar donicity trends have been observed
for the binding affinity of halocarbons to low-valent
transition metals, which increases in the order RF <<
RCl < RBr < RI.27
such as tert-butylamine, tert-octylamine, and diisopro-
pylamine were also slow and produced lower yields of
aniline.32
The equilibrium constant for bridge cleavage of the
palladium iodide dimer 3 by amine to form the corre-
sponding 1:1 palladium amine adduct is considerably
less favorable than the corresponding reaction of pal-
ladium bromide dimer 2. This halide effect on K may
be responsible for the lower efficiency observed for aryl
iodides relative to aryl bromides in the catalytic ami-
nation reaction. The correlation between K and cata-
lytic proficiency is supported by the observation that
both iodoamines and bromoamines undergo intramo-
lecular cross-coupling with high efficiency, as chelation
presumably renders amine coordination to palladium
favorable for both halides.3 For example, while acyclic
iodide amine adducts such as 6 dissociate readily at
room temperature, the corresponding chelated amine
complexes such as Pd[P(o-tol)3[2-C6H4(CH2)2N(H)Bn)]I
show no evidence for amine dissociation even at 55 °C.6
Due to unfavorable steric interaction between the cis
amine and halide ligands of the palladium amine
adducts, the Pd-N bond strength should depend in-
versely on the van der Waals radii of the halide, which
increases in the order Cl (1.75 Å) < Br (1.85 Å) < I (1.96
Å).28,29 Unfavorable steric interaction between the cis
amine and halide ligands of the palladium amine
adducts is supported by the greater halide sensitivity
of K in the case of diisopropylamine relative to the less
bulky dibenzylamine. For example, the free energy
difference for the reactions of 1 and 3 with diisopropyl-
Exp er im en ta l Section
Gen er a l Meth od s. All manipulations were performed
under an atmosphere of nitrogen or argon in an inert-
atmosphere glovebox or by standard Schlenk techniques.
Preparative-scale reactions were performed in flame- or oven-
dried Schlenk tubes equipped with a stirbar, side-arm joint,
and septum. NMR spectra were obtained in CDCl3 at 55 °C
amine (∆∆G298 ) 4.6 kcal mol-1) was considerably
K
larger than the free energy difference for the reactions
1
of 1 and 3 with dibenzylamine (∆∆G328 ) 2.7 kcal
K
on a Varian XL-300 spectrometer unless otherwise noted. H
mol-1). Likewise, steric interaction between the cis
amine and halide ligands is supported by the observed
halide effect on energy barriers for P-C and Pd-P bond
rotation in palladium amine adducts, which increased
in the order 4 < 5 < 6.6 Likewise, steric interaction of
the cis-bromide ligand and the o-fluorine atoms of the
C6F5 ring in the S,N-bonded palladium bromide complex
(Pd(C6F5)Br(SPPyPh2)) [Py ) pyridin-2-yl] became so
severe (2.02 Å) upon rotation of the C6F5 ring through
the coordination plane that halide dissociation was
required for complete aryl rotation.30
NMR spectra were referenced relative to the residual proton
resonance of the solvent. Room-temperature 31P NMR (96
MHz) spectra were referenced relative to external 85% H3PO4;
low-temperature 31P NMR spectra were referenced relative to
external PPh3 in CDCl3 (δ -4.69). IR spectra were recorded
on a Perkin-Elmer Model 6000 FTIR spectrometer. Elemental
analyses were performed by E+R Analytical Laboratories
(Corona, NY). Hexane, pentane, benzene, and benzene-d6 were
distilled from purple solutions of sodium and benzophenone.
THF-d8 and dioxane-d8 were distilled from Na/K alloy, meth-
ylene chloride and methylene chloride-d2 were distilled from
CaH2, and CDCl3 was distilled from P2O5; all solvents were
distilled under argon or nitrogen. All amines were purchased
from Aldrich and were distilled from CaH2 under N2 prior to
use. Compounds 1-6, 20, and 21 were prepared according to
published procedures.6
Con clu sion s
Although the efficiency and rate of the Pd2(DBA)3/
P(o-tol)3-catalyzed amination reaction are clearly af-
fected by the binding constant (Kb) of the amine, no
simple correlation between amine Kb value and catalytic
proficiency was observed. For example, the reaction
rate of 5-bromo-m-xylene with amines was inversely
related to the amine binding constant, where kobs
decreased in the order HN(Me)Ph >> morpholine >
N-benzylmethylamine ≈ dibutylamine > piperidine >>
hexylamine.31 Conversely, the reactions of aryl bro-
mides with bulky amines which possess a low Kb value
The error limits for equilibrium constants (K and Kb) and
the corresponding free energy values were estimated via
extrapolation of (10% error in the ratio of the respective
palladium complexes determined by 1H NMR spectroscopy.
Duplicate determinations of equilibrium constants were con-
sistently within the estimated error limits. The error limits
for enthalpy values correspond to the standard deviation of
the least-squares fit of the corresponding van’t Hoff plot. The
error limits for entropy values were estimated by extrapolation
of the errors in the corresponding free energy and enthalpy
values.
P d[P (o-tol)3](p-C6H4Me)[HN(CH2)5]Cl (7). Piperidine (200
mg, 2.3 mmol) was added to a yellow solution of 1 (210 mg,
0.18 mmol) in 5 mL of CH2Cl2 and stirred at room temperature
for 10 min. The resulting colorless solution was concentrated
to 1 mL under vacuum and diluted with 20 mL of hexane.
Cooling the solution via concentration to 10 mL under vacuum
formed a white precipitate, which was filtered, washed with
hexane, and dried under vacuum to give 7 (194 mg, 81%) as a
pale yellow microcrystalline solid which contained traces of
(27) Kulawiec, R. J .; Crabtree, R. H. Coord. Chem. Rev. 1990, 99,
89.
(28) Porterfield, W. W. Inorganic Chemistry: A Unified Approach;
Addison-Wesley: Reading, MA, 1984; p 168.
(29) The Lewis acidity of the palladium center is expected to parallel
the Pauling electronegativity of the bound halide ligand, which
increases in the order I (2.66) < Br (2.96) < Cl (3.16) (Allred, A. J .
Inorg. Nucl. Chem. 1961, 17, 215). However, the enhanced Lewis
acidity of palladium should increase the strength of the dative Pd-N
bond and the dative Pd-(µ-halogen) bond to a similar degree and
therefore have no net effect on K.
1
hexane (<5% as determined by H NMR analysis). 1H NMR:
δ 7.78 (br, 3 H), 7.33 (t, 3 H, J ) 7 Hz), 7.15 (m, 6 H), 6.76 (br,
(30) Cusares, J . A.; Coco, S.; Espinet, P.; Liu, Y. S. Organometallics
1995, 14, 3058.
(31) Widenhoefer, R. A.; Buchwald, S. L. Unpublished results.
(32) Wolfe, J . N.; Buchwald, S. L. Unpublished results.

2762 Organometallics, Vol. 15, No. 12, 1996
Widenhoefer and Buchwald
2 H), 6.49 (d, J ) 7.5 Hz, 3 H), 3.28 (br s, 1 H, HN), 3.22 (d,
J ) 13 Hz, 2 H, R-CH2), 2.55 (d, J ) 13 Hz, 2 H, R-CH2), 2.26
[s, 9 H, P(o-tol)3], 2.08 (s, 3 H, C6H4Me), 1.60 (m, 3 H), 1.36
(m, 3 H). 31P{1H} NMR: δ 25.6. IR (Nujol): 3200 cm-1. Anal.
Calcd (found) for C33H39ClNPPd: C, 63.67 (63.46); H, 6.31
(6.53); N, 2.25 (2.39).
The procedure employed in the synthesis of 7 was applied
to the syntheses of complexes 8-16 and 18 by utilizing the
appropriate palladium aryl halide dimer and amine. All
complexes were isolated as cream or pale yellow powders;
yields are given in Table 1.
P d [P (o-tol)3](p-C6H4Me)(HNBu 2)Cl (8). 1H NMR: δ 7.75
(br s, 3 H), 7.28-7.11 (m, 9 H), 6.6 (br s, 2 H), 6.43 (d, J ) 7.7
Hz, 2 H), 3.55 (br s, 1 H, NH), 2.25 [br s, 9 H, P(o-tol)3], 2.06
(s, 3 H, C6H4Me), 1.80 (br s, 6 H), 1.30 (m, 6 H), 0.93 (t, J )
7.5 Hz, 6 H, -CH2CH3). 31P{1H} NMR: δ 27.4. IR (Nujol):
3215, 3200 cm-1. Anal. Calcd (found) for C36H47ClNPPd: C,
64.87 (65.01); H, 7.11 (7.30); N, 2.10 (2.15).
Hz, HNCH2Ph), 3.76 (br, 2 H, HNCH2Ph), 3.24 (br, 1 H, HN),
2.19 [s, 9 H, P(o-tol)3], 2.04 (s, 3 H, C6H4Me). 31P{1H} NMR:
δ 28.2. Anal. Calcd (found) for C42H43BrNPPd: C, 64.75
(64.74); H, 5.56 (5.55); N, 1.80 (1.80).
P d [P (o-tol)3](p-C6H4Me)[HN(i-P r )2]Cl (17). A solution of
1 (120 mg, 0.22 mmol) and diisopropylamine (770 mg, 7.6
mmol) in CH2Cl2 (3 mL) was stirred for 5 min at room
temperature. The yellow solution was concentrated to 1 mL
under vacuum, diluted with 5 mL of hexane, and cooled to -20
°C overnight. The resulting precipitate was collected, washed
with hexane, and dried under vacuum to give 17 (88 mg, 74%)
as yellow blocks. 1H NMR (CD2Cl2 [0.1 M HN(i-Pr)2], 25 °C;
17:1 ≈ 2:1): δ 2.05 (C6H4Me); additional resonances corre-
sponding to 17 were obscured by the resonances for 1 and free
HN(i-Pr)2. 31P{1H} NMR (CD2Cl2 [0.5 M HN(i-Pr)2], 25 °C):
δ 27.3. Anal. Calcd (found) for C34H43ClNPPd: C, 63.95
(64.05); H, 6.79 (6.63).
P d [P (o-tol)3](p-C6H4Me)[HN(i-P r )2]Br (19). Reaction of
diisopropylamine (1.4 g, 14 mmol) and 2 (100 mg, 0.17 mmol)
employing a procedure analogous to that used to prepare 17
led to the isolation of 19 (76 mg, 64%) as yellow crystals. 1H
NMR (CD2Cl2 [0.1 M HN(i-Pr)2], 25 °C; 19:2 ≈ 2:1): δ 2.05
(C6H4Me); additional resonances corresponding to 19 were
obscured by the resonances for 2 and free HN(i-Pr)2. 31P{1H}
NMR (CD2Cl2 [0.5 M HN(i-Pr)2], 25 °C): δ 26.6. Anal. Calcd
(found) for C34H43BrNPPd: C, 59.79 (59.77); H, 6.35 (6.21).
P d [P (o-tol)3](p-C6H4Me)(HNEt2)Cl (9). 1H NMR: δ 7.75
(br s, 3 H), 7.29 (t, J ) 7.4 Hz, 3 H), 7.12 (t, J ) 6.1 Hz, 6 H),
6.66 (br s, 2 H), 6.43 (d, J ) 7.7 Hz, 2 H), 3.40 (br s, 1 H, NH),
2.61 (br s, 2 H, NCH2CH3), 2.42 (m, 2 H, NCH2CH3), 2.26 [br
s, 9 H, P(o-tol)3], 2.06 (s, 3 H, C6H4Me), 1.44 (t, J ) 7.05, 6 H,
NCH2CH3). 31P{1H} NMR: δ 27.3. Anal. Calcd (found) for
C
32H39ClNPPd: C, 62.96 (62.77); H, 6.44 (6.28); N, 2.29 (2.22).
P d [P (o-t ol)3](p-C6H4Me)[HN(CH2)2O]Cl (10). 1H NMR
(20 °C): δ 7.30, 7.11, 6.45, 3.69 (br s, 1 H, HN), 3.64 (d, J )
11 Hz, 2 H, R-CH2), 3.40 (dt, J ) 2, 11 Hz, 2 H, R-CH2), 2.93
(br, 2 H, â-CH2), 2.65 (br, 2 H, â-CH2), 2.21 [s, 9 H, P(o-tol)3],
2.07 (s, 3 H, C6H4Me). 31P{1H} NMR: δ 25.9. IR (Nujol): 3174
P d [P (o-tol)3](p-C6H4Me)(HNBn 2)I (22). Dibenzylamine
(205 mg, 1.04 mmol) was added to an orange solution of 3 (56
mg, 0.09 mmol) in CH2Cl2 (5 mL), and the resulting yellow
solution was stirred for 5 min. Solvent was evaporated under
vacuum to give an orange oil, which was diluted with 5 mL of
hexane and cooled to -20 °C overnight. The resulting
precipitate was collected, washed with hexane (10 mL), and
dried under vacuum to give 22 (61 mg, 83%) as a yellow
microcrystalline solid. 1H NMR (CDCl3 (0.09 M HNBn2); 22:3
≈ 4:1): δ 6.31 (d, J ) 7.3 Hz), 6.10, 4.40 (br t, J ) 7 Hz) and
2.06 (s, C6H4Me); additional resonances corresponding to 22
were obscured by the resonances for free HNBn2 and 3. 31P-
{1H} NMR (CDCl3 (0.09 M HNBn2), 25 °C; 22:3 g 10:1): δ
28.0. Anal. Calcd (found) for C42H43INPPd: C, 61.07 (61.16);
H, 5.25 (5.44).
cm-1
. Anal. Calcd (found) for C32H37ClNOPPd: C, 61.55
(61.38); H, 5.97 (5.95); N, 2.24 (2.33).
P d [P (o-t ol)3](p-C6H 4Me)[HN(i-Bu )2]Cl (11). 1H NMR
(11:1 ≈ 12:1): δ 7.70 (br, 3 H), 7.29 (t, J ) 6.7 Hz, 3 H), 7.12
(br t, J ) 6.3 Hz, 6 H), 6.65 (br s, 2 H), 6.44 (d, J ) 7.6 Hz, 2
H), 3.65 (br, 1 H, NH), 2.24 [br s, 9 H, P(o-tol)3], 2.06 (s, 3 H,
C6H4Me), 0.94 [d, J ) 6.7 Hz, 6 H, NCH2CH(CH3)2], 0.86 [d, J
) 6.7 Hz, 6 H, NCH2CH(CH3)2]. 31P{1H} NMR (25 °C): δ 27.9.
Anal. Calcd (found) for C36H47ClNPPd: C, 64.87 (65.03); H,
7.11 (6.96); N, 2.10 (1.99).
P d [P (o-tol)3](p-C6H4Me)(HNBn 2)Cl (12). 1H NMR (20 °C;
12:1 > 20:1): δ 7.60 (br, 3 H), 7.32 (m, 13 H), 7.09 (m, 6 H),
6.31 (d, J ) 7.2 Hz, 2 H), 6.03 (br, 2 H), 4.35 (dd, 2 H, J ) 8.4,
13.8 Hz, HNCH2Ph), 3.70 (br, 2 H, HNCH2Ph), 3.20 (br, 1 H,
HN), 2.23 [s, 9 H, P(o-tol)3], 2.03 (s, 3 H, C6H4Me). 31P{1H}
The procedure employed in the synthesis of 22 was applied
to the syntheses of complexes 23-25, utilizing the appropriate
halide and amine; yields are given in Table 1.
NMR: δ 27.2. IR (Nujol): 3221 cm-1
. Anal. Calcd (found)
P d[P (o-tol)3](p-C6H4CMe3)(HNBn 2)I (23): orange needles.
1H NMR [CDCl3 (0.09 M HNBn2); 23:20 ≈ 15:1]: δ 6.55, 6.20,
4.42 (br, HNCH2Ph), 3.37, 1.14 (C6H4CMe3); additional reso-
nances corresponding to 23 were obscured by the resonances
for free HNBn2 and 20. 31P{1H} NMR (CDCl3 (0.09 M HNBn2),
25 °C): δ 28.3. Anal. Calcd (found) for C45H49INPPd: C, 62.26
(62.19); H, 5.69 (5.65).
for C42H43ClNPPd: C, 68.67 (68.55); H, 5.90 (6.11); N, 1.91
(2.01).
P d [P (o-tol)3](p-C6H4Me)[H2N(CH2)5Me]Cl (13). 1H NMR
(20 °C): 7.76 (m, 3 H), 7.30 (t, J ) 7 Hz, 3 H), 7.13 (m, 6 H),
6.63 (br s, 2 H), 6.47 (d, J ) 7.5 Hz, 2 H), 2.60 (br s, 2 H,
H2N), 2.56 (br s, 2 H, H2NCH2-), 2.21 [s, 9 H, P(o-tol)3], 2.07
(s, 3 H, C6H4Me), 1.40 (t, J ) 6 Hz, 2 H), 1.20 (t, J ) 7 Hz, 2
H), 1.11 (br s, 4 H), 0.80 [t, J ) 7.5 Hz, 3 H, H2N(CH2)5Me].
P d [P (o-tol)3](p-C6H4OMe)(HNBn 2)I (24): yellow powder.
1H NMR [CDCl3 (0.09 M HNBn2); 24:21 ≈ 15:1]: δ 6.24, 6.10,
4.43 (br, HNCH2Ph), 3.60 (s, C6H4OMe); additional resonances
corresponding to 24 were obscured by the resonances for free
HNBn2 and 21. 31P{1H} NMR (CDCl3 (0.09 M HNBn2), 25
°C): δ 28.1. Anal. Calcd (found) for C42H43INOPPd: C, 59.91
(59.66); H, 5.15 (5.35).
31P{1H} NMR: δ 25.8. Anal. Calcd (found) for C34H43
-
ClNPPd: C, 63.95 (64.11); H, 6.79 (6.75); N, 2.19 (2.21).
P d [P (o-tol)3](p-C6H4Me)(H2NCy)Cl (15). 1H NMR:
δ
7.76 (br, 3 H), 7.29 (t, 3 H, J ) 7.13 Hz), 7.10 (m, 6 H), 6.62
(m, 2 H), 6.46 (d, 2 H, J ) 7.46 Hz), 2.66, 2.47, 2.18 [br s, 9 H,
P(o-tol)3], 2.07 (s, 3 H, C6H4Me), 1.93 (d, J ) 9.06 Hz), 1.60
(m), 1.46 (m), 1.05 (m). 31P{1H} NMR: δ 27.9. Anal. Calcd
(found) for C34H41ClNPPd: C, 64.16 (63.97); H, 6.49 (6.50); N,
2.20 (2.18).
P d [P (o-t ol)3](p-C6H4Me)(H2N-t-oct yl)Cl (16). 1H NMR
(16:1 ≈ 8:1): δ 7.70 (br, 3 H), 7.28 (t, 6 H, J ) 6.04 Hz), 7.24
(m, 3 H), 6.65 (m, 2 H), 6.42 (d, 2 H, J ) 6.49 Hz), 3.14 (br, 2
H, H2N-t-octyl), 2.18 [br s, 9 H, P(o-tol)3], 2.07 (s, 3 H, C6H4Me),
1.09 (m). 31P{1H} NMR: δ 30.0. Anal. Calcd (found) for
P d [P (o-tol)3](p-C6H4OMe)(H2NCMe2CH2CMe3)I (25): or-
ange crystals. 1H NMR (CDCl3 (0.09 M tert-octylamine), 25:
21 ≈ 15:1): δ 3.63 (s, C6H4OMe); additional resonances
corresponding to 25 were obscured by the resonances for free
tert-octylamine and 21. 31P{1H} NMR (CDCl3 (0.09 M tert-
octylamine), 25 °C): δ 28.3. Anal. Calcd (found) for C36H47
-
INOPPd: C, 55.86 (55.60); H, 6.12 (6.35).
P d [P (o-t ol)3](p-C6H4Me)[HN(i-P r )2]I (26). Complex 26
was generated by addition of diisopropylamine to a solution
of 3 in CD2Cl2 and was analyzed without isolation. 1H NMR
(300 MHz, CD2Cl2 [0.5 M HN(i-Pr)2], 25 °C; 26:3 ≈ 1:2): δ 2.05
(C6H4Me); additional resonances corresponding to 26 were
C
36H47ClNPPd: C, 64.87 (64.82); H, 7.11 (7.28); N, 2.10 (2.00).
P d [P (o-tol)3](p-C6H4Me)(HNBn 2)Br (18). 1H NMR (20
°C; 18:2 ≈ 20:1): δ 7.38-7.27 (13 H), 7.12-7.01 (6 H), 6.31
(d, J ) 7.8 Hz, 2 H), 6.05 (br, 2 H), 4.38 (dd, 2 H, J ) 7.7, 12.2

Pd Tris(o-tolyl)phosphine Mono(amine) Complexes
Organometallics, Vol. 15, No. 12, 1996 2763
obscured by the resonances for 3 and free HN(i-Pr)2. 31P{1H}
NMR (96 MHz, CD2Cl2 [3 M HN(i-Pr)2], 25 °C; 26:3 ≈ 3:1): δ
26.2.
K ) [22]2/[3][HNBn2]2. K for the formation of 18 from reaction
of bromide dimer 2 and HNBn2 was determined by an
analogous procedure.
Deter m in a tion of th e Bin d in g Con sta n t (Kb) for P ip -
er id in e Rela tive to N-Ben zylm eth yla m in e. Piperidine (2.0
µL, 0.02 mmol) and N-benzylmethylamine (8.0 µL, 0.06 mmol)
were added via syringe to an NMR tube containing a solution
of 1 (5.9 mg, 5.5 × 10-3 mmol) in CDCl3 (0.63 mL). The tube
was shaken, and its contents were analyzed by 1H NMR
spectroscopy at 55 °C. The concentrations of 4 and 7 were
determined from the area (A ) hω1/2) of the p-tolyl methyl
resonances for 4 (2.03) and 7 (2.08) and from the mass balance.
The concentrations of piperidine and N-benzylmethylamine
were determined from the mass balance. The binding constant
for piperidine relative to that of N-benzylmethylamine was
determined from two separate experiments according to the
formula Kb ) [7][HN(Me)Bn]/[4][piperidine]. The binding
constants for dibutylamine, morpholine, diisobutylamine, ben-
zylamine, cyclohexylamine, and tert-octylamine were deter-
mined by an analogous procedure. Likewise, the binding
constant of dibenzylamine relative to that of morpholine was
determined by an analogous procedure.
Equ ilibr iu m F or m a tion of 23 fr om 20 a n d HNBn 2.
Dibenzylamine (2.0 µL, 0.016 mmol) was added via syringe to
an NMR tube containing a solution of 20 (9.8 mg, 6.4 × 10-3
mmol) and PhSiMe3 (0.74 mg, 5.0 × 10-3 mmol) in CDCl3 (0.68
mL). The tube was shaken, and its contents were analyzed
1
by H NMR spectroscopy at 24 °C. The concentrations of 20
and 23 were determined by integrating the t-Bu resonance for
20 (δ 1.08) and 23 (δ 1.14) relative to the trimethylsilyl
resonance for PhSiMe3 (δ 0.25). The concentration of diben-
zylamine was determined from the mass balance. The equi-
librium constant for the conversion of 20 + dibenzylamine to
23 was determined according to the formula K ) [23]2/[20]-
[HNBn2]2. K values for formation of 23 from 20 + HNBn2 in
C6D6, dioxane-d8, and THF-d8 were determined by an analo-
gous procedure. Likewise, the equilibrium formation of 24
from 21 and HNBn2 was analyzed in an analogous manner
by integrating the methoxy resonances for 21 (δ 3.53) and 24
(δ 3.60). The thermodynamic parameters for the conversion
of 21 to 24 were determined from a van’t Hoff plot of the data
over the temperature range 9-58 °C (Figure 1).
Equ ilibr iu m F or m a tion of 12 fr om 1 a n d HNBn 2.
A
solution of 12 (8.2 mg, 0.01 mmol) and PhSiMe3 (1.8 mg, 0.01
Equ ilibr iu m betw een 17 a n d 1 + HN(i-P r )2. Diisopro-
pylamine (5.0 µL, 3.9 mmol) was added via syringe to an NMR
tube containing a solution of 1 (7.5 mg, 7.0 × 10-3 mmol) in
1
mmol) in CDCl3 was analyzed by H NMR spectroscopy at 53
°C. The concentrations of 1, 12, and dibenzylamine were
determined by integrating the p-tolyl resonances for 12 (δ 2.03)
and 1 (δ 1.96) and the benzyl resonance of HNBn2 (δ 3.82)
relative to the trimethylsilyl resonance for PhSiMe3 (δ 0.25).
The equilibrium constant for the conversion of 1 + dibenzyl-
amine to 12 was determined according to the formula K )
[12]2/[1][HNBn2]2.
1
CD2Cl2 (0.96 mL). The tube was shaken and analyzed by H
NMR spectroscopy at 25 °C. The concentrations of 1 and 17
were determined by cutting and weighing the p-tolyl methyl
resonances for 1 (2.02) and 17 (2.05) and from the mass
balance. The concentration of diisopropylamine was deter-
mined from the mass balance. The equilibrium constant for
the conversion of 1 + diisopropylamine to 17 was determined
as the average of three separate experiments according to the
formula K ) [17]2/[1][HN(i-Pr)2]2. K for the formation of 19
from 2 and diisopropylamine and K for the formation of 26
from 3 and diisopropylamine were determined analogously.
Equ ilibr iu m F or m a tion of 22 fr om 3 a n d HNBn 2.
A
solution of 3 (9.8 mg, 7.8 × 10-3 mmol) and PhSiMe3 (0.74 mg,
5.0 × 10-3 mmol) in CDCl3 (0.68 mL) was analyzed by 1H NMR
spectroscopy. The concentration of 3 was determined by
integrating the P(o-tol)3 + p-tolyl resonances of 3 relative to
the trimethylsilyl resonance for PhSiMe3. Dibenzylamine (1.5
µL, 0.012 mmol) was then added to the tube via syringe, the
1
tube was shaken, and its contents were analyzed by H NMR
Ack n ow led gm en t. We thank the National Science
Foundation, the National Institutes of Health, Dow
Chemical, and Pfizer for their support of this work.
R.A.W. was an NCI Postdoctoral Trainee supported by
NIH Cancer Training Grant CI T32CA09112.
spectroscopy at 53 °C. The concentrations of 3 and 22 were
determined from the area of the p-tolyl methyl resonances for
3 (δ 1.98) and 22 (δ 2.06) and from the mass balance. The
concentration of HNBn2 was determined from the mass
balance. The equilibrium constant for the conversion of 3 +
dibenzylamine to 22 was determined according to the formula
OM9509608