High-activity catalysts for Suzuki coupling and amination reactions with deactivated aryl chloride substrates: Importance of the palladium source

Article abstract of DOI:10.1021/om020841+

A range of ortho-metalated catalysts with alkylphosphine ligands of the general formula [Pd(X)(kappa;²N,C-C₆H₄CH₂NMe₂ )(PR₃)] have been synthesized, and the crystal structures of five examples (R = Cy, X = TFA, OTf, Cl, I; PR₃ = PCy₂(o-biphenyl), X = TFA) have been determined. The crystal structures of two dimeric precursor complexes, [{Pd(μ-TFA)(κ²N,C-C₆H₄CH₂NM e₂)}₂] and [{Pd(TFA)(κ²N,C-C₆H₄CH=NⁱPr)} ₂], have also been determined. The application of the phosphine adducts to both Suzuki coupling and Buchwald-Hartwig amination reactions with aryl chloride substrates was examined, and the performance of these catalysts versus conventional palladium sources was evaluated. In general the palladacyclic complexes show considerably enhanced activity. Typically, the best activity is seen with tricyclohexylphosphine adducts in Suzuki coupling and tri-tert-butylphosphine analogues in amination reactions. In nearly all the amination reactions performed, small amounts of a second product species were observed, namely 4,6-bis(aryl)-3,4-dihydro-2H-[1,4]oxazines. The crystal structure of one example, 4,6-bis(4-methoxyphenyl)-3,4-dihydro-2H-[1,4]oxazine, was determined. Studies on the activation of palladacyclic precatalysts in the coupling of morpholine led to the isolation of a morpholine adduct, [Pd(TFA)(κ²N,C-C₆H₅CH₂NMe ₂){NH(CH₂CH₂)O}], which was structurally characterized by X-ray analysis.

Full text of DOI:10.1021/om020841+

Organometallics 2003, 22, 987-999
987
High -Activity Ca ta lysts for Su zu k i Cou p lin g a n d
Am in a tion Rea ction s w ith Dea ctiva ted Ar yl Ch lor id e
Su bstr a tes: Im p or ta n ce of th e P a lla d iu m Sou r ce
Robin B. Bedford* and Catherine S. J . Cazin
School of Chemistry, University of Exeter, Exeter EX4 4QD, U.K.
Simon J . Coles, Thomas Gelbrich, Peter N. Horton, Michael B. Hursthouse, and
Mark E. Light
Department of Chemistry, EPSRC National Crystallography Service,
University of Southampton, Highfield Road, Southampton SO17 1BJ , U.K.
Received October 9, 2002
A range of ortho-metalated catalysts with alkylphosphine ligands of the general formula
[Pd(X)(κ2N,C-C6H4CH2NMe2)(PR3)] have been synthesized, and the crystal structures of five
examples (R ) Cy, X ) TFA, OTf, Cl, I; PR3 ) PCy2(o-biphenyl), X ) TFA) have been
determined. The crystal structures of two dimeric precursor complexes, [{Pd(µ-TFA)(κ2N,C-
C6H4CH2NMe2)}2] and [{Pd(TFA)(κ2N,C-C6H4CHdNiPr)}2], have also been determined. The
application of the phosphine adducts to both Suzuki coupling and Buchwald-Hartwig
amination reactions with aryl chloride substrates was examined, and the performance of
these catalysts versus conventional palladium sources was evaluated. In general the
palladacyclic complexes show considerably enhanced activity. Typically, the best activity is
seen with tricyclohexylphosphine adducts in Suzuki coupling and tri-tert-butylphosphine
analogues in amination reactions. In nearly all the amination reactions performed, small
amounts of a second product species were observed, namely 4,6-bis(aryl)-3,4-dihydro-2H-
[1,4]oxazines. The crystal structure of one example, 4,6-bis(4-methoxyphenyl)-3,4-dihydro-
2H-[1,4]oxazine, was determined. Studies on the activation of palladacyclic precatalysts in
the coupling of morpholine led to the isolation of a morpholine adduct, [Pd(TFA)(κ2N,C-
C6H5CH2NMe2){NH(CH2CH2)O}], which was structurally characterized by X-ray analysis.
Sch em e 1. Su zu k i Bia r yl Cou p lin g Rea ction
Sch em e 2. Am in a tion of Ar yl Ha lid es
In tr od u ction
Palladium-catalyzed coupling reactions are extremely
powerful tools for the synthesis of new carbon-carbon
and carbon-heteroatom bonds. The Suzuki coupling
reaction (Scheme 1) is a versatile technique for the
formation of unsymmetrical biaryls,1 while the amina-
tion of aryl halides (Scheme 2) has emerged as an
excellent technique for the synthesis of N-substituted
anilines.2A major recent focus in coupling chemistry has
been on the development of catalysts that are able to
activate aryl chloride substrates, since these tend to be
cheaper and more readily available than their bromide
and iodide counterparts, factors that make them par-
ticularly relevant to the industrial sector. Unfortunately
the comparatively high C-Cl bond strength makes their
activation by oxidative addition comparatively difficult.3
While there are now many reports on the use of aryl
chlorides in both Suzuki coupling and amination reac-
tions,4,5 in most cases the catalysts employed need to
be used in relatively high loadings. Therefore, the
advantages associated with the use of aryl chlorides may
be negated by the high cost of the catalyst systems and
the need to remove palladium residues from the prod-
ucts down to the ppm level for use in fine chemicals and
pharmaceutical applications. Consequently, there is
considerable interest in the development of catalysts
that can activate aryl chlorides at low loadings. Palla-
dacyclic complexes have played a particular role in this
regard. For instance, the complexes 1-3 (Chart 1) all
show some activity in the Suzuki coupling of aryl
chlorides.6-8 We have recently demonstrated that the
complexes 4a and 5 show very high activity in the
Suzuki coupling reaction of both deactivated and acti-
* To whom correspondence should be addressed. E-mail: r.bedford@
ex.ac.uk. Fax: +44 (0) 1392 263434.
(1) Recent reviews: (a) Miyaura, N.; Suzuki, A. Chem. Rev. 1995,
95, 2457. (b) Stanforth, S. P. Tetrahedron 1998, 54, 263. (c) Suzuki, A.
J . Organomet. Chem. 1999, 576, 147.
(2) For reviews see: (a) Wolfe, J . P.; Wagaw, S.; Marcoux, J .-F.;
Buchwald, S. L. Acc. Chem. Res. 1998, 31, 805. (b) Hartwig, J . F.
Angew. Chem., Int. Ed. 1998, 37, 2046. (c) Yang, B. H.; Buchwald, S.
L. J . Organomet. Chem. 1999, 576, 125.
(3) For a discussion see: Grushin, V. V.; Alper, H. Chem. Rev. 1994,
94, 1047.
10.1021/om020841+ CCC: $25.00 © 2003 American Chemical Society
Publication on Web 02/01/2003

988 Organometallics, Vol. 22, No. 5, 2003
Bedford et al.
Ch a r t 1
Sch em e 3a
a
Conditions: (i) C6H5CH2NMe2, THF, 40-70 °C, 1-5 h. (ii)
NaBr, acetone, room temperature, 1.5 h.
activity has been observed with phosphine-free systems
based on ortho-metalated imines and related ligands.4b,c,11
The new dimeric precursor 7a was synthesized in 94%
yield by heating palladium trifluoroacetate with N,N-
dimethylbenzylamine in THF (Scheme 3). The dimeric
complexes [{Pd(µ-X)(κ2N,C-C6H4CH2NMe2)}2] (7b, X )
Cl; 7c, X ) I) and the imine-containing dimer [{Pd(µ-
TFA)(κ2N,C-C6H4CHdNiPr)}2] (8) were prepared as
described previously,12-14 while the complex 7d (X ) Br)
was synthesized in 90% yield by the reaction of the
complex 7b with sodium bromide in acetone at room
vated aryl chlorides,4e while Na´jera has shown that the
phosphine-free oxime-containing palladacycles 6 show
good activity in the presence of water and tetrabutyl-
ammonium bromide.4b,c We report herein full details of
the synthesis of the complexes of the types 4 and 5 and
related catalysts and their application to Suzuki coup-
ling and amination reactions. In particular we address
the role of the ortho-metalation on catalyst performance.
Aspects of this work have been communicated previous-
ly.4e
1
temperature. The H NMR spectra of complexes 7a ,d
were very similar to those reported for the complexes
7b,c.12,13 The crystal structures of both complexes 7a
and 8 were determined, and the molecules are shown
in Figures 1 and 2, while selected data are given in
Tables 1 and 2, respectively. When the two structures
are compared, it can be seen that, with the exception of
the Pd-N bonds, all other bonds to the palladium
centers are identical or very close in length. The Pd-N
bond of 8 is shorter than that of the analogous bond in
7a . This may be due to a contribution to the bond by
retrodonation into the ortho-metalated imines CdN π*-
LUMO. The Pd‚‚‚Pd distance in 8 is somewhat shorter
than that in 7a , presumably as a result of lower steric
repulsion between the substituents on the N-donor and
the ortho-metalated aromatic ring on the opposite half
of the molecule. The Pd-C bond lengths of 7a are
essentially the same as those found in the complexes
[{Pd(µ-X){κ2N,C-C6H4CH2NH(CH2Ph)}}2] (X ) Br, ac-
etate).15,16 The Pd-N bonds lengths of 7a are the same
as those found for the latter bromide complex but
shorter than those in the acetate analogue.15,16 The
Pd-C and Pd-N bonds of complex 8 are very similar
to those of the closely related complex [{Pd(µ-TFA)-
(κ2N,C-C6H4C(Me)dNiPr)}2].11a
Resu lts a n d Discu ssion
Syn th esis a n d Ch a r a cter iza tion of Dim er ic P r e-
cu r sor Com p lexes. We chose to synthesize phosphine
adducts of palladacyclic complexes based on ortho-
metalated N,N-dimethylbenzylamine, as we and, later,
others have shown that the phosphine-free parent
dimers show good activity in coupling reactions with
aryl bromide and iodide substrates.9,10 Similarly good
(4) For recent examples of Suzuki couplings of aryl chlorides see:
(a) Li, G. Y. J . Org. Chem. 2002, 67, 3643. (b) Alonso, D. A.; Na´jera,
C.; Pacheco, M. C. J . Org. Chem. 2002, 67, 5588. (c) Botella, L.; Na´jera,
C. Angew. Chem. Int. Ed. 2002, 41, 179. (d) Liu, S.-Y.; Choi, M. J .; Fu,
G. C. Chem. Commun. 2001, 2408. (e) Bedford, R. B.; Cazin, C. S. J .
Chem. Commun. 2001, 1540. (f) Li, G. Y.; Zheng, G.; Noonan, A. F. J .
Org. Chem. 2001, 66, 8677. (g) Li, G. Y. Angew. Chem., Int. Ed. 2001,
40, 1513. (h) Netherton, M. R.; Fu, G. C. Org. Lett. 2001, 3, 4295. (i)
Zapf, A.; Ehrentraut, A.; Beller, M. Angew. Chem., Int. Ed. 2000, 39,
4153. (j) Andreu, M. G.; Zapf, A.; Beller, M. Chem. Commun. 2000,
2475. (k) Littke, A. F.; Dai, C.; Fu, G. C. J . Am. Chem. Soc. 2000, 122,
4020. (l) Old, D. W.; Wolfe, J . P.; Buchwald, S. L. J . Am. Chem. Soc.
1998, 120, 9722. (m) Wolfe, J . P.; Singer, R. A.; Yang, B. H.; Buchwald,
S. L. J . Am. Chem. Soc. 1999, 121, 9550. (n) Wolfe, J . P.; Buchwald, S.
L. Angew. Chem., Int. Ed. 1999, 38, 2413. (o) Bei, X.; Turner, H. W.;
Weinberg, W. H.; Guram, A. S. J . Org. Chem. 1999, 64, 6797. (p) Zhang,
C.; Huang, J .; Trudell, M. L.; Nolan, S. P J . Org. Chem. 1999, 64, 3804.
(q) Littke, A. F.; Fu, G. C. Angew. Chem., Int. Ed. 1998, 37, 3387.
(5) For examples of aryl chloride amination reactions see: (a) Viciu,
M. S.; Kissling, R. M.; Stevens, E. D.; Nolan, S. P. Org. Lett. 2002, 4,
2229. (b) Wolfe, J . P.; Tomori, H.; Sadighi, Y. P.; Buchwald, S. L. J .
Org. Chem. 2000, 65, 1158. (c) Stauffer, S. R.; Lee, S.; Stambuli, J . P.;
Hauck, S. I.; Hartwig, J . F. Org. Lett. 2000, 2, 1423. (d) Hartwig, J .
F.; Kawatsura, M.; Hauk, S. I.; Shaughnessy, K. H.; Alcazar-Roman,
L. M. J . Org. Chem. 1999, 64, 5575. (e) Hamann, B. C.; Hartwig, J . F.
J . Am. Chem. Soc. 1998, 120, 7369. (f) Reddy, N. P.; Tanaka, M.
Tetrahedron Lett. 1997, 38, 4807. See also ref 4f,g,l,n.
(9) Albisson, D. A.; Bedford, R. B.; Scully, P. N. Tetrahedron Lett.
1998, 39, 9793.
(10) Beletskaya, I. P.; Kashin, A. N.; Karlstedt, N. B.; Mitin, A. V.;
Cheprakov, A. V.; Kazankov, G. M. J . Organomet. Chem. 2001, 622,
89.
(11) (a) Ohff, M.; Ohff, A.; Milstein, D. Chem. Commun. 1999, 357.
(b) Weissman, H.; Milstein, D. Chem. Commun. 1999, 1901. (c)
Rocaboy, C.; Gladysz, J . A., Org. Lett. 2002, 4, 1993.
(12) Pfeffer, M.; Goel, A. B. Inorg. Synth. 1989, 26, 211.
(13) Maasarani, F; Pfeffer, M.; Le Borgne, G. Organometallics 1987,
6, 2029.
(6) Beller, M.; Fischer, H.; Herrmann, W. A.; O¨ fele, K.; Brossmer,
(14) Bedford, R. B.; Cazin, C. S. J .; Hursthouse, M. B.; Light, M. E.;
Pike, K. J .; Wimperis, S. J . Organomet. Chem. 2001, 633, 173.
(15) Vincete, J .; Saura-Llamas, I.; Turp´ın, J .; Ram´ız de Apellano,
M. C.; J ones, P. G. Organometallics 1999, 18, 2683.
(16) Fuchita, Y.; Yoshinaga, K.; Hanaki, T.; Kawano, H.; Kinoshita-
Nagaoka, J . J . Organomet. Chem. 1999, 580, 273.
C. Angew. Chem., Int. Ed. 1995, 34, 1848.
(7) Bedford, R. B.; Draper, S. M.; Scully, P. N.; Welch, S. L. New J .
Chem. 2000, 24, 745.
(8) Zim, D.; Gruber, A. S.; Ebeling, G.; Dupont, J .; Monteiro, A. L.
Org. Lett. 2000, 2, 2881.

Catalysts for Suzuki Coupling
Organometallics, Vol. 22, No. 5, 2003 989
Ta ble 2. Selected Bon d Len gth s or In ter a tom ic
Dista n ces (Å) a n d An gles (d eg) for
[{P d (µ-TF A)(K2N,C-C6H4CHdNiP r )}2] (8)
Pd1-C10
Pd1-N1′
Pd1-O1′
Pd1-O2
Pd1‚‚‚Pd2
1.964(5)
2.020(3)
2.066(3)
2.202(3)
2.8996(4)
Pd2-C20
Pd2-N1
Pd2-O1
Pd2-O2′
1.960(4)
2.002(3)
2.061(3)
2.182(3)
C10-Pd1-N1′
C10-Pd1-O2
C10-Pd1-O1′
N1′-Pd1-O1′
N1′-Pd1-O2
O1′-Pd1-O2
81.24(17)
179.35(17)
93.16(16)
174.38(13)
98.23(13)
87.38(12)
C20-Pd2-N1
C20-Pd2-O2′
C20-Pd2-O1
N1-Pd2-O1
N1-Pd2-O2′
O1-Pd2-O2′
81.55(17)
177.59(16)
93.58(15)
173.27(13)
96.47(13)
88.28(12)
Sch em e 4a
F igu r e 1. Molecular structure of [{Pd(µ-TFA)(κ2N,C-C6H4-
CH2NMe2)}2] (7a ).
a
Conditions: (i) 1.1 PCy3/Pd, CH2Cl2, room temperature,
45 min; (ii) AgOTf, CH2Cl2, room temperature, 1 h.
phine adducts 4a -d in 36-91% yields (Scheme 4). The
triflate-containing complex 4e was prepared by the
metathesis reaction of the chloride complex 4b, formed
in situ, with silver triflate in dichloromethane.
The 31P{1H} NMR spectra of the complexes 4 each
show a single peak in the range 40.4-46.4 ppm. The
1H NMR spectra of all of the tricyclohexylphosphine
adducts 4 show a distinctive multiplet, shifted slightly
downfield with respect to the rest of the aromatic
protons, corresponding to the proton ortho to the meta-
lated carbon, H-6. The signals corresponding to the
methylene bridge of the ortho-metalated dimethylben-
zylamine ligand are seen at between 3.91 and 3.98 ppm
and are typically broad singlets, except for that of 4e,
which is a doublet with a J PH value of 1.5 Hz. The
resonances for the protons of the methyl groups of the
ortho-metalated dimethylbenzylamine ligands are seen
between 2.57 and 2.76 ppm and are doublets with J PH
couplings in the range 2.2-2.5 Hz. A 1H{31P} spectrum
of 4a confirmed that the coupling observed was indeed
due to a P-H interaction.
F igu r e 2. Molecular structure of [{Pd(µ-TFA)(κ2N,C-
C6H4CHdNiPr)}2] (8).
Ta ble 1. Selected Bon d Len gth s or In ter a tom ic
Dista n ces (Å) a n d An gles (d eg) for
[{P d (µ-TF A)(K2N,C-C6H4CH2NMe2)}2] (7a )
Pd1-C3′
1.954(3)
2.059(3)
2.174(2)
2.072(2)
3.0588(4)
Pd2-C3
Pd2-N1
Pd2-O1
Pd2-O2′
1.960(4)
2.062(3)
2.163(3)
2.064(2)
Pd1-N1′
Pd1-O1′
Pd1-O2
Pd1‚‚‚Pd2
The 13C{1H} NMR spectra of the complexes 4 show
two peaks for quaternary aromatic carbons with phos-
phorus couplings: 4a , 147.6 (d, J PC ) 4.5 Hz), 148.7 ppm
(d, J PC ) 1.5 Hz); 4b, 148.7 (d, J PC ) 2.3 Hz), 153.3 ppm
(d, J PC ) 1.5 Hz); 4c, 148.6 (d, J PC ) 2.3 Hz), 155.5 (s)
C3′-Pd1-N1′
C3′-Pd1-O2
C3′-Pd1-O1′
N1′-Pd1-O1′
N1′-Pd1-O2
O1′-Pd1-O2
82.73(14)
92.29(13)
174.67(12)
94.00(11)
171.92(11)
90.47(10)
C3-Pd2-N1
C3-Pd2-O2′
C3-Pd2-O1
N1-Pd2-O1
N1-Pd2-O2′
O1-Pd2-O2′
82.16 (14)
92.77(14)
174.48(13)
94.12(12)
171.93(12)
90.46(11)
ppm; 4d , 148.7 (d, J PC ) 1.5 Hz), 159.1 ppm (d, J PC
)
1.5 Hz); 4e, 143.6 (d, J PC ) 2.3 Hz), 148.4 ppm (d, J PC
) 2.3 Hz). From these data it is apparent that one peak
remains constant at 148.55 ( 0.15 ppm with varying
“X” ligand, which is assignable as the C-CH2 aromatic
carbon with J PC couplings in the range 1.5-2.3 Hz. The
position of the second quaternary aromatic peak, which
corresponds to the metalated carbon, is dependent on
Syn th esis a n d Ch a r a cter iza tion of P h osp h in e-
Con ta in in g Com p lexes. The reaction of the dimeric
complexes 7a -d with 1.1 equiv per palladium of tricy-
clohexylphosphine in dichloromethane gave the phos-

990 Organometallics, Vol. 22, No. 5, 2003
Bedford et al.
Sch em e 5a
F igu r e 3. Molecular structure of [Pd(TFA)(κ2N,C-C6H4-
CH2NMe2)(PCy3)] (4a ).
a
Conditions: (i) 1.2 PCy3/Pd, CH2Cl2, room temperature,
30 min; (ii) 1.1 PCy2(o-biphenyl)/Pd, CH2Cl2, room tempera-
ture, 2 h; (iii) 1 PtBu3/Pd, CD2Cl2, room temperature, 1 h.
Ta ble 3. Selected Bon d Len gth s (Å) a n d An gles
(d eg) for th e Com p lexes
[P d (X)(K2N,C-C6H4CH2NMe2)(P Cy3)] (4a , X ) TF A;
4b‚CHCl3, X ) Cl; 4d , X ) I; 4e, X ) OTf).
larger the X, the further the P and N atoms move away
from the palladium center.
4a
4b‚CHCl3
4d
4e
A method similar to that used for the synthesis of
complexes 4a -d was employed to generate the tricy-
clohexylphosphine adduct 5 from the ortho-metalated
imine complex 8 (Scheme 5). The 31P NMR spectrum of
5 shows a singlet at 39.4 ppm. In addition to the peaks
associated with the tricyclohexylphosphine ligand, the
1H NMR spectrum of complex 5 shows a distinct
multiplet at 8.07 ppm, downfield of the aromatic signals,
corresponding to the benzylidene proton. The methine
Pd-P
Pd-C
Pd-N
Pd-X
2.2618(4)
2.2777(7)
2.3203(10) 2.2847(8)
1.9943(17) 2.010(3)
2.1672(15) 2.159(2)
2.1295(12) 2.4259(7)
2.104(4)
2.317(3)
2.6814(4)
1.992(3)
2.145(3)
2.208(2)
C-Pd-N
C-Pd-P
81.37(7)
97.46(5)
N-Pd-X 92.36(5)
81.44(9)
99.74(7)
91.30(6)
89.59(2)
166.67(7)
168.61(6)
89.59(14)
91.78(11)
84.34(9)
98.38(3)
164.52(9)
162.03(8)
81.79(13)
96.72(10)
91.29(10)
92.10(7)
169.76(12)
162.89(8)
P-Pd-X
C-Pd-X
N-Pd-P
88.91(4)
173.60(6)
162.47(5)
i
protons of the Pr groups are observed as a multiplet at
3.98 ppm, while the methyl groups are a doublet at 1.36
ppm with a J HH value of 6.6 Hz.
The general method for the synthesis of phosphine
adducts from the dimer 7a was applied to the prepara-
tion of the 2-(dicyclohexylphosphino)biphenyl and tri-
tert-butylphosphine adducts 9a ,b (Scheme 5).
3
the nature of the group trans to it. The TFA and OTf
complexes show even greater upfield shift than the
chloride, in line with the lower trans influence of oxygen
donors. There does not seem to be a discernible pattern
for the variation in J PC (none to 4.5 Hz). By contrast,
the parent dimer 7a shows two singlets at 141.8 and
2
The 31P{1H} NMR spectrum of complex 9a shows a
broad singlet at 52.8 ppm. The breadth of the signal may
indicate that the ligand is reasonably labile. The 13C-
{1H} NMR spectrum shows four low-field quaternary
aromatic carbons as doublets between 142.3 and 148.5
ppm, two of which correspond to the ortho-metalated
ring; the other two are C2 and C1′ of the o-biphenyl
group. The C1 of the o-biphenyl group is seen as a
doublet at 125.5 ppm with a 34.3 Hz 1J PC coupling. The
1H NMR spectrum shows the characteristic downfield
shift for the proton ortho to the palladium atom ob-
served for the complexes 4. The protons of the methyl-
ene on the ortho-metalated dimethylbenzylamine group
are seen as a singlet at 3.97 ppm, while the methyl
protons are seen as a doublet at 2.42 ppm with a J PH
value of 2.6 Hz. These data are again very similar to
those observed for the complexes 4.
The structure of 9a was confirmed by X-ray analysis.
The asymmetric unit contains two distinct molecules
and a molecule of diethyl ether. The structure of one of
the molecules (molecule A) is shown in Figure 4, while
selected data for both molecules are given in Table 4.
All of the lengths of the bonds to the palladium fall in
the range of those observed for the complexes 4a ,b,d ,e,
147.3 ppm for these two quaternary carbons.
The structures of the tricyclohexylphosphine adducts
4a ,b,d ,e were determined by X-ray analysis. The mo-
lecular structure of one example (4a ) is shown in Figure
3, while selected bond lengths and angles for all of the
adducts are given in Table 3. In all cases the palladium
adopts an approximately square planar configuration
in which the PCy3 ligand is trans to the nitrogen donor.
The complex 4b has a molecule of chloroform hydrogen-
bonded to the chloride ligand, with a H‚‚‚Cl distance of
2.403(6) Å. The Pd-C bond lengths of 4a (X ) TFA)
and 4e (X ) OTf) are essentially identical and the same
as that reported in [Pd(κ2N,C-C6H4CH2NMe2)(κ2P,O-
Ph2PCH2CO2)].17 The Pd-C bond length of 4b (X ) Cl)
is only marginally longer, but that of 4d (X ) I) is
significantly longer. When the Pd-P bond lengths are
compared, it can be seen that they fall in the order 4d
> 4e > 4b > 4a , while the Pd-N bond lengths decrease
in the order 4d > 4a > 4e ≈ 4b. Taken together, these
data seem to reflect the steric bulk of the X ligand; the
(17) Braunstein, P.; Matt, D.; Nobel, D.; Bouaoud, S.-E.; Granjean,
D. J . Organomet. Chem. 1986, 301, 401.

Catalysts for Suzuki Coupling
Organometallics, Vol. 22, No. 5, 2003 991
complexes 4, a distinctive multiplet is seen downfield
of the other aromatic protons corresponding to the
proton ortho to the metalated carbon. The distinctive
peaks for the two quaternary aromatic carbons seen in
the 13C NMR spectra of the complexes 4 are also seen
in the spectrum of 9b. Thus, a doublet at 148.2 ppm
with a 1.9 Hz coupling to phosphorus is assignable as
the C-CH2 group of the aromatic ring, while a doublet
at 144.4 ppm with a 3.0 Hz coupling to phosphorus
corresponds to the metalated carbon. Considered to-
gether, the NMR data show that not only does 9a form
but also that its structure and stereochemistry are very
similar to those of the complexes 4.
Ca ta lysis. (a ) Su zu k i Cou p lin g. We have previ-
ously shown that complex 4a functions as an excellent
precatalyst in the coupling of a range of aryl chloride
substrates.4e Here we wished to compare its activity
with other catalyst systems and palladium precursors.
To standardize the results, we decided to concentrate
primarily on the coupling of 4-chloroanisole with phen-
ylboronic acid, as this is an electronically “challenging”
reaction due to the reluctance of the electron-rich aryl
chloride to undergo oxidative addition reactions. Three
phosphine ligands were chosen for the study: PCy3,
PCy2(o-biphenyl), and PtBu3. The last two ligands were
chosen as they have proven to be useful in a range of
aryl chloride coupling reactions, including the Suzuki
reaction.4h,i,k-n,q,5b,d The palladium sources investigated
in the first instance were palladium acetate, dipalla-
dium tris(dibenzylideneacetone), the dimeric complex
7a , and its PCy3 and PCy2(o-biphenyl) adducts 4a and
9a . Since the preformed complex 9b is highly labile and
could not be isolated pure, it was not used in the
catalytic studies. Instead, the catalyst was formed in
situ from 7a and PtBu3. The range of catalyst precursors
and ligands chosen allows for the results to be “stan-
dardized” with literature examples of their use. The
results of these catalytic studies are summarized in
Table 5.
The initial reactions were performed using potassium
phosphate as the base in toluene at 100 °C (Table 5,
entries 1-8), as these conditions were found previously
to give good results when palladium acetate and PCy2-
(o-biphenyl) are used in aryl chloride Suzuki reactions.4l
However, we found that changing to cesium carbonate
in 1,4-dioxane gave much better results with the
reported catalyst system, especially at lower catalyst
loadings. Therefore, the rest of the Suzuki coupling
reactions were performed using this solvent/base com-
bination.
F igu r e 4. Structure of one of the molecules (molecule A)
of the two molecules of [Pd(TFA)(κ2N,C-C6H4CH2NMe2)-
{PCy2(o-biphenyl)}]‚0.5Et2O (9a ‚0.5Et2O).
Ta ble 4. Selected Bon d Len gth s (Å) a n d An gles
(d eg) for [P d (TF A)(K2N,C-C6H4CH2NMe2)-
{P Cy2(o-bip h en yl)}]‚0.5Et2O (9a ‚0.5Et2O)
molecule
A (shown)
B (not shown)
Pd-P
Pd-C
Pd-N
Pd-O
2.2781(11)
2.002(5)
2.134(3)
2.144(3)
2.2792(12)
2.007(5)
2.143(4)
2.136(4)
C-Pd-N
C-Pd-P
N-Pd-O
P-Pd-O
C-Pd-O
N-Pd-P
82.4(2)
81.9(2)
98.35(14)
91.84(17)
87.35(9)
173.98(17)
177.99(18)
98.54(15)
92.76(18)
86.71(10)
174.36(18)
178.9(2)
with the Pd-O bond lengths being directly comparable
with that of 4a . The C-Pd-N and C-Pd-P bond angles
again fall in the range for those observed for 4a ,b,d ,e,
while the N-Pd-O and C-Pd-O angles are compa-
rable with those of 4a . The P-Pd-O angles in the two
molecules of 9a are slightly more acute than that for
4a , while the N-Pd-P angles in 9a are considerably
larger than that in 4a .
The isolation and purification of the tri-tert-butylphos-
phine adduct, 9b, proved to be problematic. The phos-
phine ligand is extremely labile, presumably due to its
high steric bulk. Even removal of solvents from the
product under reduced pressure led to the regeneration
of 7a and the loss of PtBu3 by sublimation. However,
NMR spectroscopy of a sample prepared in CD2Cl2
demonstrated conclusively that it does indeed form at
room temperature. The 31P NMR spectrum shows a
Dipalladium tris(dibenzylideneacetone) proved early
on to be a poor palladium precursor (Table 5, entries 1
and 2). This is probably not surprising, in light of
Amatore and J utland’s observation that DBA can in fact
be a tenacious ligand during coupling reactions and is
not as “innocent” a spectator ligand as previously
supposed.18 Therefore, this catalyst system was not
examined further. When PtBu3 is used as a ligand, there
is no advantage in the use of the dimeric palladium
precursor 7a compared with palladium acetate (Table
5, entries 25-28, 31, and 32). With PCy2(o-biphenyl) at
lower catalyst concentrations (e0.1 mol %) there is some
1
singlet at 75.8 ppm. The H NMR spectrum shows, in
addition to peaks associated with the phosphine ligand,
a doublet at 2.42 ppm with a J PH value of 2.6 Hz
corresponding to the methyl groups of the ortho-meta-
lated N,N-dimethylbenzylamine ligand. This is directly
comparable with the data for all of the complexes 4. The
methylene peaks for the ortho-metalated ligand are seen
as a singlet at 3.97 ppm, again in the range for the
analogous peaks of the complexes 4. As with the
(18) For leading references see: Amatore, C.; J utland, A.; Thuilliez,
J . Organomet. Chem. 2002, 643-644, 416.

992 Organometallics, Vol. 22, No. 5, 2003
Bedford et al.
Ta ble 5. In vestiga tion in to th e Su zu k i Cou p lin g of 4-Ch lor oa n isole w ith P h en ylbor on ic Acid w ith Va r yin g
P a lla d iu m Sou r ces a n d P h osp h in esa
entry
Pd source (amt, mol % Pd)
added phosphine (amt, mol %)
base
solvent
conversn (%)b
1
2
3
4
5
6
7
8
Pd2(dba)3 (1.0)
Pd2(dba)3 (1.0)
Pd(OAc)2 (1.0)
Pd(OAc)2 (1.0)
7a (1.0)
7a (1.0)
9a (1.0)
4a (1.0)
Pd(OAc)2 (1.0)
Pd(OAc)2 (0.1)
Pd(OAc)2 (0.01)
Pd(OAc)2 (0.01)
7a (0.1)
7a (0.01)
7a (0.01)
9a (0.01)
9a (0.01)
9a (0.01)
9a (0.01)
4a (0.1)
4a (0.01)
4a (0.01)
4a (0.01)
4a (0.01)
7a (0.01)
7a (0.01)
7a (0.01)
7a (0.01)
Pd(OAc)2 (0.01)
Pd(OAc)2 (0.01)
Pd(OAc)2 (0.01)
Pd(OAc)2 (0.01)
PCy2(o-biphenyl) (1.0)
PCy2(o-biphenyl) (2.0)
PCy2(o-biphenyl) (1.0)
PCy2(o-biphenyl) (2.0)
PCy2(o-biphenyl) (1.0)
PCy2(o-biphenyl) (2.0)
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
K3PO4
toluene
toluene
toluene
toluene
toluene
toluene
toluene
toluene
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
dioxane
3.5
8.5
28.5
50
17.5
39.5
49
15
100
77
40
15
88
19.5
46.5
45
64.5
72
71
88
71
78
4
2.5
19
33
6
4.5
2.5
7
19
46
9
PCy2(o-biphenyl) (2.0)
PCy2(o-biphenyl) (0.2)
PCy2(o-biphenyl) (0.02)
PCy2(o-biphenyl) (0.01)
PCy2(o-biphenyl) (0.2)
PCy2(o-biphenyl) (0.01)
PCy2(o-biphenyl) (0.02)
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
Cs2CO3
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
PCy2(o-biphenyl) (0.01)
PCy2(o-biphenyl) (0.02)
PCy2(o-biphenyl) (0.03)
PCy3 (0.01)
PCy3 (0.02)
PCy3 (0.03)
PtBu3 (0.01)
PtBu3 (0.02)
PtBu3 (0.03)
PtBu3 (0.04)
PCy3 (0.01)
PCy3 (0.02)
PtBu3 (0.01)
PtBu3 (0.02)
a
b
Conditions: MeOC6H4-4-Cl (1.0 mmol), PhB(OH)2 (1.5 mmol), base (2.0 mmol), solvent (3 mL), 100 °C, 2 h. Conversion to
4-methoxybiphenyl, average of two reactions, determined by GC (hexadecane standard).
advantage to using the dimeric precursor 7a (Table 5;
compare entries 10-12 with 13-15); however, this
advantage becomes greatly pronounced when tricyclo-
hexylphosphine is employed. Indeed, at 0.01 mol %
catalyst loading, tricyclohexylphosphine goes from show-
ing virtually no activity when palladium acetate is used
to giving the best catalyst systems. The overall order of
activity when the ortho-palladated precursor is used is
PCy3 > PCy2(o-biphenyl) > PtBu3. The advantage of
using tricyclohexylphosphine is compounded by the fact
that it is much cheaper than either PCy2(o-biphenyl)
or PtBu3 and it is much easier to handle than the latter
phosphine. In general we find it is not necessary to use
the preformed catalyst 4a , since the catalyst formed in
situ from the dimer 7a plus 2 equiv of phosphine shows
very similar activity. However, catalyst 4a has the
F igu r e 5. Effect of P:Pd ratio on the performance of the
advantage of being air and moisture stable in solution
ortho-metalated complexes [Pd(TFA)(κ2N,C-C6H4CH2NMe2)-
(at least >1 month) as well as in the solid state; thus,
handling is facilitated compared with the air-sensitive
phosphine.
(PR3)] in the Suzuki coupling of 4-chloroanisole with
phenylboronic acid. PR3: (green) PCy3; (blue) PtBu3; (red)
PCy2(o-biphenyl). Conditions: MeOC6H4-4-Cl (1.0 mmol),
PhB(OH)2 (1.5 mmol), Cs2CO3 (2.0 mmol), 1,4-dioxane (3
mL), 100 °C, 2 h.
While varying the tricyclohexylphosphine to pal-
ladium ratio from 1:1 to 2:1 does not unduly affect the
activity under these conditions, increasing the ratio
further is highly deleterious, with essentially a complete
loss of activity observed. This can be more easily seen
in the plot of P:Pd vs conversion for PCy3, PtBu3, and
PCy2(o-biphenyl) shown in Figure 5. The data for PCy3
and PCy2(o-biphenyl) were obtained by adding the
phosphines to the preformed complexes 4a and 9a , while
the data for PtBu3 were obtained by adding the phos-
phine to the dimeric precursor 7a . With PtBu3 a similar
trend is observed to that seen with PCy3, whereby the
addition of more than 2 equiv of phosphine per pal-
ladium is highly deleterious, whereas with PCy2(o-
biphenyl) there appears to be no diminution in activity.
This tends to imply that the optimum active catalysts,
when both tricyclohexylphosphine and tri-tert-butylphos-
phine are used, are low-coordinate and that overcoor-
dination effectively “switches off” the catalysis. Indeed,
a recurrent feature of chloride coupling catalysts is the
observation that in many cases the active catalysts are

Catalysts for Suzuki Coupling
Organometallics, Vol. 22, No. 5, 2003 993
Ta ble 6. In vestiga tion in to th e Su zu k i Cou p lin g of
4-Ch lor oa n isole w ith P h en ylbor on ic Acid w ith
Va r iou s P a lla d a cyclic P r eca ta lystsa
entry catalyst conversn (%)b
entry catalyst conversn (%)b
1
2
3
4a
5
4e
71
81
72
4
5
6
4b
4c
4d
80.5
81.5
64.5
a
Conditions: MeOC6H4-4-Cl (1.0 mmol), PhB(OH)2 (1.5 mmol),
Cs2CO3 (2.0 mmol), catalyst (0.01 mol % Pd), 1,4-dioxane (3 mL),
b
100 °C, 2 h. Conversion to 4-methoxybiphenyl, average of two
reactions, determined by GC (hexadecane standard).
probably monophosphine complexes.19 With ligands
such as PCy2(o-biphenyl) it is possible that, regardless
of the amount of added phosphine, the formation of
monophosphine adducts is favored. Such a preference
for monocoordination may be due in part to a secondary
π-coordination of the aromatic ring to the palladium
center.20 In this scenario excess phosphine would simply
act to force any dissociation equilibria back to the
formation of complex and thus increase catalyst longev-
ity. In contrast, the use of an excess of a phosphine that
can have no secondary interactions may tip the balance
in favor of higher coordinate complexes, thus diminish-
ing activity.
Having established that tricyclohexylphosphine shows
by far the best activity, provided it is used in conjunction
with an appropriate palladium precursor, we decided
to investigate the effect of simple structural changes of
the ortho-palladated precursor on activity. The results
from this study are summarized in Table 6. As can be
seen, the ortho-palladated imine complex 5 (entry 2)
shows slightly higher activity than the dimethylbenzyl-
amine-containing complex 4a (entry 1) under the condi-
tions employed here; however, the need to presynthesize
the ortho-palladated ligand in this case detracts from
its overall appeal. The effect of varying the nature of
the anionic coligand, “X”, in the ortho-palladated com-
plexes 4 was also investigated (entries 1 and 3-6).
While the range of activity varied, the values obtained
are not sufficiently disparate to allow detailed inter-
pretation.
The formation of poly-ortho-substituted biaryls by
Suzuki coupling is obviously particularly challenging.20a
To test whether the complex 4a was able to catalyze
such reactions, we briefly investigated its use in the
coupling of the sterically hindered substrates 2-chloro-
m-xylene and 2-tolylboronic acid. When the same condi-
tions as those in Table 5 were used with 0.1 mol % Pd,
quantitative conversion was observed. Lowering the
catalyst loading to 0.01 mol % Pd gave a 20% conversion,
which is equivalent to a TON of 2000.
We have previously found that the coupling of 4-chlo-
roanisole with phenylboronic acid catalyzed by 4a (0.01
mol % Pd) in 1,4-dioxane with cesium carbonate acting
as base can be performed under air with, if anything, a
slight increase in activity.4e In this case the reaction was
set up under air and an oil bubbler was placed on the
F igu r e 6. Effect of added water on the coupling of
4-chloroanisole with phenylboronic acid catalyzed by com-
plex 4a . Conditions: MeOC6H4-4-Cl (1.0 mmol), PhB(OH)2
(1.5 mmol), Cs2CO3 (3.0 mmol), 1,4-dioxane (10 mL), 100
°C, 17 h.
top of the reflux condenser to provide a closed system.
Numerous repetitions of this reaction show it to be
highly capricious, with two contrasting outcomes ob-
served: either complete activity or complete inactivity!
To examine this phenomenon further, we performed a
series of experiments. In the first instance reactions
were performed under air, but with a drying tube
containing calcium chloride on the top of the condenser.
Here we found complete inactivity. Next we examined
reactions performed under nitrogen with controlled
amounts of water added. The results of this are shown
in Figure 6. As can be seen, the added water is
deleterious to catalyst performance at all the levels
investigated. Since both oxygen and water appear to be
separately deleterious, we are unable to account for the
fact that sometimes very high activity is observed under
aerobic conditions.
(b) Am in a tion . We again find that the palladium
source, in addition to the correct choice of phosphine
ligand, can play a pivotal role in Buchwald-Hartwig
amination reactions. To illustrate this point, the data
in Table 7 show the coupling of 4-chloroanisole with
morpholine with a range of palladium precursors and
the phosphines used above in the Suzuki coupling as
well as the ligand PtBu2(o-biphenyl).4m,n,5b As can be
seen, the relative ordering of ligand performance when
palladium acetate is used as a precursor is PCy2(o-
biphenyl) > PtBu2(o-biphenyl) > PCy3 > PtBu. In all
cases, except with PCy3, changing from palladium
acetate to an ortho-metalated palladium source gives
an increase in TON of between about 2- and 6.5-fold.
The fact that no advantage is seen on changing pal-
ladium source when tricyclohexylphosphine is employed
(compare entries 5 and 6 with 11 and 12) is a little
surprising, since the maximum enhancement in activity
in the Suzuki couplings is observed with this ligand.
Conversely, the greatest activity enhancement seen in
the amination reaction was for PtBu3 (compare entries
7 and 8 with 9 and 10), which showed essentially no
enhancement in the Suzuki coupling reaction.
(19) See: Stambuli, J . P.; Bu¨hl, M.; Hartwig, J . F. J . Am. Chem.
Soc. 2002, 124, 9346 and references therein.
(20) For examples of Pd-aryl π-interactions see: (a) Yin, J .; Rainka,
M. P.; Zhang, X.-X.; Buchwald, S. J . J . Am. Chem. Soc. 2002, 124, 1162.
(b) Kocˇovsky´, P.; Yyskocˇil, S.; C´ısarˇovrˇa´, I.; Sejbal, J .; Tisˇlerova´, I.;
Smrcˇina, M.; Lloyd-J ones, G. C.; Stephen, S. C.; Butts, C. P.; Murray,
M.; Langer, V. J . Am. Chem. Soc. 1999, 121, 7714. (c) Hayashi, T.;
Iwamura, H.; Naito, M.; Matsumoto, Y.; Uozumi, Y.; Miki, M.; Yanagi,
K. J . Am. Chem. Soc. 1994, 116, 775.
In all cases, regardless of phosphine or palladium
source, except in the coupling catalyzed by 7a (0.1 mol

994 Organometallics, Vol. 22, No. 5, 2003
Bedford et al.
Ta ble 7. Am in a tion of 4-Ch lor oa n isole w ith
Mor p h olin ea
Pd source
(amt, mol
% Pd)
added
phosphine (amt,
mol %)
conversn
(%)b
entry
TON
1
2
3
4
5
6
7
8
Pd(OAc)2 (0.1) PtBu2(o-biphenyl) (0.1)
Pd(OAc)2 (0.1) PtBu2(o-biphenyl) (0.2)
Pd(OAc)2 (0.1) PCy2(o-biphenyl) (0.1)
Pd(OAc)2 (0.1) PCy2(o-biphenyl) (0.2)
Pd(OAc)2 (0.1) PCy3 (0.1)
7
26
23.5
62.5
5.5
7
70
260
235
625
55
70
10
20
F igu r e 7. Molecular structure of 4,6-bis(4-methoxyphen-
yl)-3,4-dihydro-2H-[1,4]oxazine (10a ).
Pd(OAc)2 (0.1) PCy3 (0.2)
aromatic protons of the N-(4-methoxyphenyl)morpho-
line. The very small amount of 10a isolated meant that
we were unable to obtain full 13C NMR characterization;
in particular, we were not able to see any of the
quaternary carbons. However, we did observe the ap-
pearance of seven of the expected eight aromatic CH’s
at 113.8, 114.8, 116.5, 123.6, 125.3, 128.2, and 129.0
ppm. The eighth is presumably obscured by one of the
signals for the N-(4-methoxyphenyl)morpholine’s aro-
matic CHs. The alkenic CH is seen at 106.6 ppm. Two
distinct peaks are observed at 55.7 and 55.3 ppm
corresponding to the 4-methoxyphenyls’ methyl groups.
For comparison, the methyl group of N-(4-methoxyphen-
yl)morpholine is seen at 55.6 ppm. Two CH2 peaks were
observed at 64.4 and 53.4 ppm corresponding to the
methylenes next to the nitrogen and oxygen atoms,
respectively. By comparison, the equivalent peaks in
N-(4-methoxyphenyl)morpholine are seen at 67.0 and
50.8 ppm. The proposed structure of 10a was confirmed
by X-ray crystallography, and the molecule is shown in
Figure 7. Unfortunately crystallographic disorder is
present, where the asymmetric unit is half a molecule,
with the second half being generated by the symmetry
of the space group about the C9-C10 axis causing an
overlap of atom types on the N1/C8 and C11/O2 sites.
As a result it is not possible to rely on geometric data
in the oxazine region of the molecule.
An amination reaction between 4-chlorotoluene and
morpholine catalyzed by 4a at 0.5 mol % catalyst
loading gave an isolated yield of 52% of N-(4-tolyl)-
morpholine, and again the GC-MS spectrum of the
product mixture showed a peak with a mass of 265.1,
consistent with the presence of a species that could be
tentatively assigned as 4,6-bis(4-tolyl)-3,4-dihydro-2H-
[1,4]oxazine (10b).23 Attempts to isolate and further
characterize 10b have so far proved unsuccessful.
Obviously the formation of compounds of the type 10
requires three processes to occur: (i) an amination
reaction, (ii) the formation of a double bond, and (iii)
the arylation of the carbon. That the amination occurs
first is demonstrated by the fact that when pure N-(4-
methoxyphenyl)morpholine is treated with 4-chloroani-
sole in the presence of NaOtBu and the catalyst 4a , the
formation of small amounts of 10a is observed by GC-
MS.24 It is probable that the second aryl group is
introduced by a Heck reaction on a preformed C-C
double bond. The double bond presumably results from
the oxidation of the N-(4-methoxyphenyl)morpholine,
but we do not yet know the precise mechanism by which
Pd(OAc)2 (0.1) PtBu3 (0.1)
1
2
Pd(OAc)2 (0.1) PtBu3 (0.2)
9
7a (0.1)
7a (0.05)
7a (0.05)
4a (0.5)
9a (0.5)
7a (0.05)
7a (0.05)
PtBu3 (0.1)
PtBu3 (0.1)
PCy3 (0.1)
92
17
3.5
25
63
41
22.5
920
340
70
10
11
12
13
14
15
50
126
820
450
PCy2(o-biphenyl) (0.1)
PtBu2(o-biphenyl) (0.1)
a
Conditions: MeOC6H4-4-Cl (1.0 mmol), morpholine (1.26
mmol), NaOtBu (1.39 mmol), toluene (3 mL), 110 °C, 17 h.
b
Conversion to N-(4-methoxyphenyl)morpholine, average of two
reactions, determined by GC (hexadecane standard).
Sch em e 6. F or m a tion of
4,6-Bis(a r yl)-3,4-d ih yd r o-2H-[1,4]oxa zin es fr om
Ar yl Ha lid es a n d Mor p h olin e
% Pd) and 1 equiv of PtBu3 (entry 9), we see a
substantial (ca. 20-45%) loss of 4-chloroanisole besides
that which is incorporated into the coupled product. We
suspect that much of this “missing chloride” is lost by a
hydrodehalogenation reaction of the substrate. In ad-
dition, at least two minor peaks are observed in the GC
traces, one of which we have as yet been unable to
characterize. The mass spectrum of the other compound
is consistent with the formation of 4,6-bis(4-methoxy-
phenyl)-3,4-dihydro-2H-[1,4]oxazine (10a ) (Scheme 6).21
We were unable to isolate the new compound cleanly,
but careful column chromatography of the product
mixture for a reaction between 4-chloroanisole and
morpholine gave a small (<4 mg), impure sample of the
new compound 10a in which the major impurity was
N-(4-methoxyphenyl)morpholine.22
1
The H NMR spectrum of the mixture indicated the
presence of an alkenic proton as a broad singlet at 6.44
ppm. The methylene protons are seen as two broad
signals at 4.28 and 3.66 ppm, and the methyl groups
appear as two broad singlets at 3.82 and 3.80 ppm. A
broad multiplet was observed at 7.41 ppm, and a
sharper multiplet was seen at 7.18 ppm, each integrated
to two protons and assignable as aromatic CH’s. An
additional peak was observed under the CDCl3 peak,
and further aromatic signals may be buried under a
large multiplet at 6.90 ppm, which corresponds to the
(21) GC-MS (EI) data for 10a : m/z 297.2 (53%) [M+], 283.2 (3%)
[M+ - Me], 148.7 (20%), 134.1 (100%), 107.1 (18), 92.1 (14%), 77.1
(35%), 64.1 (10%).
(22) Conditions: MeOC6H4-4-Cl (1.0 mmol), morpholine (1.26 mmol),
NaOtBu (1.39 mmol), 4a (1.0 mol % Pd), toluene (3 mL), 110 °C, 17 h.
Column chromatography (silica, hexane/ethyl acetate/methanol 90:10:
2) gave impure 10a (<4 mg) and N-(4-methoxyphenyl)morpholine
(0.099 g, 51%).
(23) Conditions: MeC6H4-4-Cl (1.0 mmol), morpholine (1.26 mmol),
NaOtBu (1.39 mmol), 4a (0.5 mol % Pd), toluene (3 mL), 110 °C, 17 h.
Yield of N-(4-tolyl)morpholine: 52%. GC-MS (EI) data for compound
tentatively assigned as 10b: m/z 265.1 (74%) [M+], 174.0 (8%), 132.9
(19%), 117.9 (100%), 90.9 (90%), 65.0 (57%).
(24) Conditions: N-(4-methoxyphenyl)morpholine (1.15 mmol), 4-chlo-
roanisole (1.31 mmol), NaOtBu (5.75 mmol), 4a (2.0 mol % Pd), toluene
(3 mL), 110 °C, 17 h.

Catalysts for Suzuki Coupling
Organometallics, Vol. 22, No. 5, 2003 995
Sch em e 7. P ossible Mech a n ism for th e F or m a tion
of Active Ca ta lysts fr om th e P r eca ta lysts 4 a n d 9
on 4a ), indicating that such a process almost certainly
occurs under catalytic conditions.26 Compound 12a could
also be prepared, for comparison purposes, by the
reaction of dimethylamine with 2-phenylbenzyl bromide.
A related ring-opening process of a palladacyclic catalyst
has been reported in the Stille coupling of aryl bromides
catalyzed by complex 1,27 while the formation of pal-
ladium(0) catalysts from Pd(II) N-donor palladacycles
has been implicated in coupling reactions.14,28
In effect the ortho-metalated ligand plays a “sacrifi-
cial” role, and the palladacycle therefore acts as a
precatalyst that is able to deliver essentially all of the
catalyst into the catalytic manifold early on in the
process, where, by definition, the highest rate of cataly-
sis occurs. As it is a simple molecular activation
pathway, the process would deliver large amounts of
relatively uniform active catalyst species in a “clean’
manner. The de-ortho-metalated N-donor ligands, such
as 12a , are relatively labile ligands on palladium(II) and
very poor ligands for Pd(0) and are therefore not
expected to compete effectively with the incoming
substrates, unlike, for instance, DBA.18 In addition, the
liberated N-donor ligands will act as a base and will be
“mopped-up” by the arylboronic acid during the course
of the reaction, preventing them from reversibly coor-
dinating to the palladium centers.
this occurs. Further studies on this novel reaction are
ongoing in our group.
Mech a n istic Con sid er a tion s. (a ) Su zu k i Cou p -
lin g. From the above catalytic studies it seems apparent
that the source of palladium can play at least as
important a role as the nature of the phosphine ligand
in aryl chloride coupling reactions. Additionally it
appears that palladacyclic complexes make excellent
catalyst precursors. What, therefore, is the role of the
ortho-metalation and why does it lead to such high
activity in catalysis? To address these questions, we
performed the following mechanistic studies.
Induction times of up to 1 h have been noted previ-
ously when using 4a as a catalyst in Suzuki couplings
of aryl chlorides.29 This is followed by rapid catalysis
for about 30 min and then fast catalyst decomposition,
with no further activity noted after 2 h. If the model is
correct, then it may be anticipated that the induction
process would be influenced by changing the leaving
group “X”. Unfortunately, attempts to prove this by
following the reaction catalyzed by 4d over time have
been repeatedly unsuccessful, with no activity observed.
We are not able to explain why careful sampling of this
system should lead to such a major disruption in
catalytic activity compared with the catalyst 4a .
(b) Am in a tion Rea ction s. Since the complexes 4
show no tendency to react with aryl chlorides, we
instead investigated their reactions with morpholine.
31P and 1H NMR spectra of a solution of 4a and 1.3 equiv
of morpholine in d8-toluene at room temperature showed
no apparent reaction. Heating a toluene solution of 4a
with 1.3 equiv of morpholine again showed no reaction.
However, when 9b is formed in situ in toluene and then
reacted with 1.25 equiv of morpholine at room temper-
ature, then PtBu3 is liberated and a new complex results
which contains no phosphine ligands. The new complex,
13, can be more conveniently prepared by the reaction
of dimer 7a with 1.25 equiv of morpholine per palladium
center (Scheme 8).
It seems highly unlikely that a Pd(II)/Pd(IV) catalytic
cycle operates when complexes of the type [Pd(X)(κ2N,C-
C6H4CH2NMe2)(PR3)] are used as precatalysts, particu-
larly when 4-chloroanisole, which is comparatively
resistant to oxidative addition reactions, is used as the
substrate. Indeed, we see no reaction of complex 4a with
either aryl chlorides or considerably more reactive aryl
bromides in the absence of boronic acids under the
conditions used in the catalytic reactions. It is far more
likely that the active catalysts formed from such precur-
sors are low-coordinate palladium(0) monophosphine
species. Monocoordinate phosphine complexes have been
implicated previously as catalysts in coupling. Indeed,
Hartwig recently isolated and structurally characterized
monophosphine adducts of the type [Pd(X)(Ar)(PtBu3)],
formed by oxidative addition of aryl halides to pal-
ladium(0) precursors in the presence of phosphine.19 The
isolated complexes react smoothly with a range of
coupling partners to generate coupled organic products.
Zerovalent monophosphine species may conceivably
be formed from the palladacyclic catalysts 4 and 9 by
the process outlined in Scheme 7, which involves (a)
nucleophilic attack of the aryl boronic acid at the metal
center,25 (b) rearrangement, and (c) reductive elimina-
tion of the aryl and the metalated N,N-dimethylbenzyl-
amine functions from the putative intermediates 11. We
have previously demonstrated that such an activation
process is likely to be operative in the Suzuki coupling
of aryl bromides catalyzed by imine dimers of the type
5 and their triphenylphosphine adducts.14 However, the
previous study was performed under stoichiometric
conditions. GC and GC-MS analysis of a reaction
mixture for the coupling of 4-chloroanisole with phen-
ylboronic acid catalyzed by 1.0 mol % of 4a at 60 °C
showed the presence of substantial amounts of N,N-
dimethyl-(2-phenyl)benzylamine (12a; ∼63% yield, based
1
The H NMR spectrum of 13 shows a large downfield
shift of ca. 2.9 ppm for the proton corresponding to the
N-H compared with morpholine and four environments
(26) Conditions: MeOC6H4-4-Cl (2.0 mmol), PhB(OH)2 (3.0 mmol),
Cs2CO3 (6.0 mmol), 4a (1.0 mol % Pd), 1,4-dioxane (20 mL), 60 °C, 17
h. Conversion to 4-methoxybiphenyl determined as 97% by GC
(hexadecane standard).
(27) Louie, J .; Hartwig, J . F. Angew. Chem., Int. Ed. Engl. 1996,
35, 2359.
(28) Nowotny, M.; Hanefeld, U.; van Koningsveld, H.; Maschmeyer,
T. Chem. Commun. 2000, 1877.
(29) Bedford, R. B.; Cazin, C. S. J .; Hazelwood, S. L. Angew. Chem.,
Int. Ed. 2002, 41, 4120.
(25) This process is sometimes referred to as “transmetalation”.

996 Organometallics, Vol. 22, No. 5, 2003
Bedford et al.
Sch em e 8a
organic products from its thermal decomposition in the
presence of sodium tert-butoxide have, as yet, proved
inconclusive.
Given that there is no reaction between 4a and
morpholine and that catalyst activation may occur via
the initial coordination of the morpholine to the pal-
ladium center, then it is possible that the poor activity
observed with the PCy3-containing complex may be, in
part, due to effective competition between the phosphine
and the substrate, preventing catalyst activation.
a
Conditions: (i) toluene, room temperature; (ii) toluene, ∆;
Con clu sion s
(iii) morpholine, toluene, room temperature.
In conclusion, we have found that the performance
of catalysts in the Suzuki coupling and amination
reactions of deactivated aryl chlorides is dependent not
only on the nature of the phosphines employed but also
strongly on the choice of palladium source. Ortho-
palladated amine and imine complexes typically show
enhanced activity compared with “classical” precursors
such as palladium acetate and [Pd2(dba)3]. It seems
likely that the ortho-metalated ligands play a sacrificial
role in the activation of the catalysts and that the true
active catalysts are low-coordinate palladium(0) mono-
phosphine adducts. During the course of studies on the
amination of aryl chlorides we have identified new
byproducts, namely the 4,6-bis(aryl)-3,4-dihydro-2H-
[1,4]oxazines 10a ,b, which require three sequential
palladium-mediated transformations to occur during
their formation. GC-MS analysis of a range of experi-
ments show such species to be present in almost all
reactions, regardless of palladium source or phosphine.
We are currently investigating the possibility of opti-
mizing conditions for this new transformation and
exploiting it in the synthesis of a range of compounds.
F igu r e 8. Structure of one (molecule A) of the two
molecules of [Pd(TFA)(κ2N,C-C6H5CH2NMe2){NH(CH2-
CH2)O}] (13).
Ta ble 8. Selected Bon d Len gth s (Å) a n d An gles
(d eg) for th e Tw o Molecu les of
[P d (TF A)(K2N,C-C6H4CH2NMe2)-
{NH(CH2CH2)2O}] (13)
Exp er im en ta l Section
molecule
Gen er a l Meth od s. Unless otherwise stated, all reactions
were carried out under nitrogen using standard Schlenk
techniques. Catalytic reactions were performed on a Radleys
Carousel Reactor. This consists of 12 ca. 45 mL tubes which
are fitted with screw-on Teflon caps that are equipped with
valves for the introduction of inert gas and septa for the
introduction of reagents. The 12 reaction tubes sit in 2 stacked
aluminum blocks; the lower one fits on a heater-stirrer and
can be maintained at a constant temperature with a thermo-
stat, while the upper block has water circulating which cools
the top of the tubes, allowing reactions to be performed at
reflux temperature.
A (shown)
B (not shown)
Pd-C
1.998(12)
2.073(10)
2.101(9)
2.173(8)
1.992(12)
2.095(9)
2.106(10)
2.144(8)
Pd-Na (benzylamine)
Pd-Nb (morpholine)
Pd-O
C-Pd-Na
C-Pd-Nb
C-Pd-O
Na-Pd-Nb
Na-Pd-O
Nb-Pd-O
81.6(4)
93.4(4)
172.7(4)
174.8(4)
91.2(3)
93.8(3)
81.7(4)
92.7(4)
173.3(4)
174.5(4)
91.7(3)
93.8(3)
THF, toluene, petroleum ether (40-60), 1,4-dioxane, and
diethyl ether were dried over Na/benzophenone and freshly
distilled prior to use. Dichloromethane and CDCl3 were
distilled from CaH2, and ethanol was dried over molecular
sieves (4 Å) and degassed before use. All other chemicals were
used as received. The complexes 7b,c and 8 and the imine
N-benzylideneisopropylamine were prepared according to
literature methods.12-14,30 GC analyses were performed on a
Varian 3800 GC fitted with a WCOT fused silica 25 m column,
for the morpholine methylene protons compared with
only two observed in the free amine. The structure of
complex 13 was determined by single-crystal X-ray
analysis. There are two molecules in the asymmetric
unit, one of which (molecule A) is shown in Figure 8,
while selected data for both molecules are given in Table
8. The Pd-C bond lengths for the two molecules are
essentially the same as those for 4a , and the Pd-O bond
lengths are also very close. The major difference in
comparable bond lengths is seen for the Pd-N(benzy-
lamine) bonds of the two molecules compared with those
of 4a , with the latter being substantially longer.
and data were recorded on
measurements were performed on
a
Star workstation. GC-MS
ThermoQuest Trace
a
spectrometer fitted with a 15 m Restek RTX-5MS column,
operating in EI mode. HRMS measurements were run on a
Micromass GCT spectrometer operating in CI mode.
Obviously complex 13 represents a likely candidate
(30) Me´lot, J .-M.; Texier-Boullet, F.; Foucaud, A. Synthesis 1998,
for catalyst activation. However, attempts to identify
558.

Catalysts for Suzuki Coupling
Organometallics, Vol. 22, No. 5, 2003 997
Syn th esis of Com p lexes. P r ep a r a tion of [{P d (µ-TF A)-
(K2N,C-C6H4CH2NMe2)}2] (7a ). N,N-Dimethylbenzylamine
(2.00 mL, 13.38 mmol) and Pd(TFA)2 (4.046 g, 12.18 mmol)
were stirred in THF (130 mL) at 40-70 °C for 1-5 h until a
dark green coloration appeared. The solution was filtered
though Celite and concentrated on a rotary evaporator, leading
to the precipitation of the product. Recrystallization from
tetrahydrofuran/ethanol gave the product as a yellow solid.
synthesis of complex 4a to give complex 4b as a pale yellow
solid. Yield: 0.291 g (36%). H NMR (CDCl3): δ 1.19 (m, 9H,
1
PCy3), 1.69 (m, 15H, PCy3), 2.00 (m, 6H, PCy3), 2.42 (m, 3 H,
PCy3), 2.62 (d, J (PH) ) 2.5 Hz, 6 H, CH3), 3.94 (br s, 2 H,
CH2), 6.92 (m, 3 H, Me ortho-metalated ring), 7.21 (m, 1 H,
H-6 ortho-metalated ring). 31P{1H} NMR (CDCl3): δ 43.4 (s,
PCy3). 13C{1H} NMR (CD2Cl2): δ 26.8 (d, J (PC) ) 1.5 Hz, CH2,
PCy3), 28.0 (d, J (PC) ) 10.5 Hz, CH2, PCy3), 30.55 (s, CH2,
1
1
3
Yield: 4.046 g (94%). H NMR (CDCl3): δ 2.04 (s, 3 H, CH3),
PCy3), 34.6 (d, J (PC) ) 21.85 Hz, CH, PCy3), 49.8 (d, J (PC)
) 2.3 Hz, CH3), 73.4 (d, 3J (PC) ) 3.1 Hz, CH2N), 122.9 (s, CH,
ortho-metalated ring), 123.75 (s, CH, ortho-metalated ring),
125.5 (d, J ) 3.8 Hz, CH, ortho-metalated ring), 136.7 (d, J (PC)
) 6.0 Hz, CH, ortho-metalated ring), 148.7 (d, J (PC) ) 2.3
Hz, CCH2, ortho-metalated ring), 153.3 (d, J (PC) ) 1.5 Hz,
Pd-C). Anal. Calcd for C27H45ClNPPd: C, 58.3; H, 8.15; N,
2.5. Found: C, 58.6; H, 8.3; N, 2.3.
2.87 (s, 3 H, CH3), 3.17 (d, 2J (HH) ) 14.05 Hz, 1 H, NCH),
2
3.62 (d, J (HH) ) 14.05 Hz, 1 H, NCH), 6.91 (m, 3 H, ortho-
metalated ring), 7.05 (m, 1 H, ortho-metalated ring).13C{1H}
NMR (CD2Cl2): δ 51.5 (s, CH3), 53.2 (s, CH3), 72.6 (s, CH2),
116.0 (q, 1J (CF) ) 287.8 Hz, CF3), 122.2 (s, CH, ortho-
metalated ring), 125.6 (s, CH, ortho-metalated ring), 125.75
(s, CH, ortho-metalated ring), 131.3 (s, CH, ortho-metalated
ring), 141.8 (s, C, ortho-metalated ring), 147.3 (s, C, ortho-
metalated ring), 165.6 (q, 2J (CF) ) 37.7 Hz, CCF3). Anal. Calcd
for C22H24F6N2O4Pd2: C, 37.4; H, 3.4; N, 4.0. Found: C, 37.2;
H, 3.3; N, 3.8.
P r ep a r a t ion of [P d (Br )(K2N,C-C6H 4CH 2NMe2)(P Cy3)]
(4c). This was prepared from complex 7d (0.326 g, 0.510 mmol)
using a scaled-down version of the method employed for the
synthesis of complex 4a to give complex 4c as a yellow solid.
Yield: 0.314 g (51%). 1H NMR (CDCl3): δ 1.18 (m, 9H, PCy3),
1.70 (m, 15H, PCy3), 2.01 (m, 6H, PCy3), 2.42 (m, 3 H, PCy3),
2.67 (d, J (PH) ) 2.45 Hz, 6 H, CH3), 3.95 (br s, 2 H, CH2),
6.93 (m, 3 H, ortho-metalated ring), 7.21 (m, 1 H, H-6 ortho-
metalated ring). 31P{1H} NMR (CDCl3): δ 44.6 (s, PCy3). 13C-
{1H} NMR (CD2Cl2): δ 26.8 (d, J ) 1.5 Hz, CH2, PCy3), 28.0
(d, J (PC) ) 10.5 Hz, CH2, PCy3), 30.7 (br s, CH2, PCy3), 35.6
(d, 1J (PC) ) 21.85 Hz, CH, PCy3), 50.3 (d, 3J (PC) ) 2.3 Hz,
CH3), 73.5 (d, 3J (PC) ) 2.3 Hz, CH2N), 123.0 (s, CH, ortho-
metalated ring), 123.9 (s, CH, ortho-metalated ring), 125.65
(d, J ) 4.5 Hz, CH, ortho-metalated ring), 136.45 (d, J (PC) )
6.0 Hz, CH, ortho-metalated ring), 148.6 (d, J (PC) ) 2.3 Hz,
CCH2, ortho-metalated ring), 155.5 (s, Pd-C). Anal. Calcd for
P r ep a r a t ion of [{P d (µ-TF A)(K2N,C-C6H 4CH dNiP r )}2]
(8). A solution of N-benzylideneisopropylamine (1.316 g, 8.93
mmol) and Pd(TFA)2 (2.125 g, 6.39 mmol) in THF (80 mL) was
stirred at 50 °C for 20 min, cooled to room temperature, and
then filtered through Celite. The solvent was removed on a
rotary evaporator, and the residue was recrystallized from
dichloromethane/ethanol to give the title product as a green
solid. Yield: 1.762 g (75%). 1H NMR (CDCl3): δ 0.71 (d, 3J (HH)
3
) 6.6 Hz, 3 H, CH3), 1.25 (d, J (HH) ) 6.6 Hz, 3H, CH3), 3.31
(m, 1 H, CHCH3), 6.93 (m, 1 H, ortho-metalated ring), 7.09
(m, 3 H, ortho-metalated ring), 7.40 (s, 1 H, NdCH). Anal.
Calcd for C24H24F6N2O4Pd2: C, 39.4; H, 3.3; N, 3.8. Found: C,
39.25; H, 3.2; N, 3.7.
P r epar ation of [{P d(µ-Br )(K2N,C-C6H4CH2NMe2)}2] (7d).
Complex 7b (0.460 g, 1.67 mmol) and sodium bromide (0.343
g, 3.33 mmol) were stirred in acetone (40 mL) for 1.5 h. The
solvent was removed in vacuo, and dichloromethane was added
(40 mL). Filtration through Celite and removal of the solvent
in vacuo gave the title complex as a yellow solid. Yield: 0.480
g (90%). 1H NMR (CDCl3): δ 2.87 (s, 3H, CH3), 2.91 (s, 3H,
CH3), 3.98 (br s, 2H, CH2), 6.79-7.06 (m, 3H, ortho-metalated
ring), 7.36 (m, 1H, ortho-metalated ring). Anal. Calcd for
C
27H45BrNPPd: C, 54.0; H, 7.55; N, 2.3. Found: C, 54.0; H,
7.9; N, 2.1.
P r epar ation of [P d(I)(K2N,C-C6H4CH2NMe2)(P Cy3)] (4d).
This was prepared from complex 7c (0.490 g, 0.670 mmol)
using a scaled-down version of the method employed for the
synthesis of complex 4a to give complex 4d as a yellow solid.
Yield: 0.657 g (76%). 1H NMR (CDCl3): δ 1.18 (m, 9H, PCy3),
1.71 (m, 15H, PCy3), 2.04 (m, 6H, PCy3), 2.43 (m, 3 H, PCy3),
2.76 (d, J (PH) ) 2.45 Hz, 6 H, CH3), 3.98 (br s, 2 H, CH2),
6.92 (m, 3 H, ortho-metalated ring), 7.20 (m, 1 H, H-6 ortho-
metalated ring). 31P{1H} NMR (CDCl3): δ 46.4 (s, PCy3). 13C-
{1H} NMR (CD2Cl2): δ 26.7 (br s, CH2, PCy3), 27.95 (d, J (PC)
C
18H24Br2N2Pd: C, 33.7; H, 3.8; N, 4.4. Found: C, 33.6; H,
3.6; N, 4.0.
P r ep a r a tion of [P d (TF A)(K2N,C-C6H4CH2NMe2)(P Cy3)]
(4a ). A solution of complex 7a (4.046 g, 5.72 mmol) and
tricyclohexylphosphine (3.626 g, 12.93 mmol) in dichlo-
romethane (50 mL) was stirred at room temperature for 45
min and concentrated to about 10 mL on a rotary evaporator,
and EtOH (15 mL) was added. Further concentration gave a
precipitate that was collected by filtration, washed with cold
ethanol (3 × 10 mL), and then recrystallized from dichlo-
romethane/ethanol to give the title complex as a colorless solid.
Yield: 6.600 g (91%). 1H NMR (CDCl3): δ 1.22 (m, 9H, PCy3),
1.72 (m, 21H, PCy3), 2.21 (m, 3 H, PCy3), 2.57 (d, J (PH) ) 2.2
Hz, 6 H, CH3), 3.92 (br s, 2 H, CH2), 6.92 (m, 3 H, ortho-
metalated ring), 7.11 (m, 1 H, H-6 ortho-metalated ring). 31P-
{1H} NMR (CDCl3): δ 41.3 (s, PCy3). 13C{1H} NMR (CD2Cl2):
δ 26.8 (br s, CH2, PCy3), 28.0 (d, J (PC) ) 11.3 Hz, CH2, PCy3),
30.2 (s, CH2, PCy3), 32.2 (d, 1J (PC) ) 21.85 Hz, CH, PCy3),
49.1 (d, 3J (PC) ) 2.3 Hz, CH3), 72.4 (d, 3J (PC) ) 3.0 Hz, CH2N),
117.2 (q, 1J (CF) ) 293.1 Hz, CF3), 123.3 (br s, CH, ortho-
metalated ring), 124.1 (s, CH, ortho-metalated ring), 125.8 (d,
J ) 3.8 Hz, CH, ortho-metalated ring), 137.0 (d, J (PC) ) 6.0
Hz, CH ortho-metalated ring), 147.6 (d, J (PC) ) 4.5 Hz, Pd-
C), 148.7 (d, J (PC) ) 1.5 Hz, CCH2 ortho-metalated ring), 161.4
(q, 2J (CF) ) 34.7 Hz), CCF3). Anal. Calcd for C29H45F3NO2-
PPd: C, 54.9; H, 7.15; N, 2.2. Found: C, 54.6; H, 7.2; N, 2.0.
P r ep a r a t ion of [P d (Cl)(K2N,C-C6H 4CH 2NMe2)(P Cy3)]
(4b). This was prepared from complex 7b (0.399 g, 0.720 mmol)
using a scaled-down version of the method employed for the
1
) 11.3 Hz, CH2, PCy3), 31.0 (s, CH2, PCy3), 36.9 (d, J (PC) )
3
21.85 Hz, CH, PCy3), 51.4 (d, J (PC) ) 1.5 Hz, CH3), 73.4 (d,
3J (PC) ) 2.3 Hz, CH2N), 123.1 (s, CH, ortho-metalated ring),
124.0 (s, CH, ortho-metalated ring), 125.7 (d, J ) 4.5 Hz), CH
ortho-metalated ring), 136.0 (d, J (PC) ) 6.8 Hz, CH, ortho-
metalated ring), 148.7 (d, J (PC) ) 1.5 Hz, CCH2, ortho-
metalated ring), 159.1 (d, J (PC) ) 1.5 Hz, Pd-C). Anal. Calcd
for C27H45INPPd: C, 50.05; H, 7.0; N, 2.2. Found: C, 50.4; H,
7.25; N, 1.8.
P r ep a r a tion of [P d (OTf)(K2N,C-C6H4CH2NMe2)(P Cy3)]
(4e). A mixture of complex 7b (0.245 g, 0.44 mmol) and
tricyclohexylphosphine (0.274 g, 0.98 mmol) in dichloromethane
(40 mL) was stirred at room temperature for 2.5 h, after which
time it was transferred to a flask containing silver trifluo-
romethanesulfonate (0.228 g, 0.89 mmol). The resultant
mixture was stirred at room temperature for 1 h with the
exclusion of light and then filtered through Celite. Petroleum
ether (10 mL) was added, and the solution was concentrated
in vacuo to about 5 mL. The resultant precipitate was collected
by filtration, washed with petroleum ether (3 × 10 mL), and
recrystallized from dichloromethane/petroleum ether to give
1
the title complex as a colorless solid. Yield: 0.214 g (36%). H
NMR (CD2Cl2): δ 1.24 (m, 9H, PCy3), 1.73 (m, 15H, PCy3),
2.03 (m, 6H, PCy3), 2.24 (m, 3 H, PCy3), 2.69 (d, J (PH) ) 2.6
Hz, 6 H, CH3), 3.94 (d, J (PH) ) 1.5 Hz, 2 H, CH2), 6.96 (m, 3
H, ortho-metalated ring), 7.13 (m, 1 H, H-6 ortho-metalated

998 Organometallics, Vol. 22, No. 5, 2003
Bedford et al.
2
ring). 31P{1H} NMR (CD2Cl2): δ 42.85 (s, PCy3). 19F{1H} NMR
(CD2Cl2): δ -78.4 (s, OTf). 13C{1H} NMR (CD2Cl2): δ 26.7 (d,
J (PC) ) 1.5 Hz, CH2, PCy3), 27.9 (d, J (PC) ) 10.6 Hz, CH2,
PCy3), 30.6 (s, CH2, PCy3), 34.1 (d, 1J (PC) ) 21.1 Hz, CH PCy3),
49.3 (d, 3J (PC) ) 2.3 Hz, CH3), 72.1 (d, J (PC) ) 3.0 Hz, CH2N),
120.4 (q, 1J (CF) ) 319.5 Hz, CF3), 123.6 (s, CH, ortho-
metalated ring), 124.8 (s, CH, ortho-metalated ring), 126.3 (d,
J ) 4.5 Hz, CH, ortho-metalated ring), 136.7 (d, J (PC) ) 6.8
Hz, CH, ortho-metalated ring), 143.6 (d, J (PC) ) 2.3 Hz, Pd-
C),148.4 (d, J (PC) ) 2.3 Hz, CCH2, ortho-metalated ring). Anal.
Calcd for C28H45F3NO3PPdS: C, 50.2; H, 6.8; N, 2.1. Found:
C, 49.9; H, 6.8; N, 2.0.
aromatic), 161.3 (q, J (CF) ) 34.7 Hz, CCF3). Anal. Calcd for
35H43F3NO2PPd: C, 59.7; H, 6.2; N, 2.0. Found: C, 59.4; H,
6.4; N, 1.8.
C
P r ep a r a tion of [P d (TF A)(K2N,C-C6H4CH2NMe2)(P tBu 3)]
(9b). In a glovebox, an NMR tube was charged with the dimer
7a (0.142 g, 0.20 mmol), PtBu3 (0.081 g, 0.40 mmol), and CD2-
Cl2 (0.8 mL). The 1H, 31P, and 13C NMR spectra were then
recorded. The high lability of the phosphine prevented the
isolation of pure 9b. 1H NMR (CD2Cl2): δ 1.53 (d, 3J (PH) )
t
12.5 Hz, 27H, Bu), 2.42 (d, J (PH) ) 2.6 Hz, 6H, NCH3), 3.97
(s, 2H, CH2N), 6.85 (m, 1H, CH ortho-metalated ring), 6.91-
7.01 (m, 2H, ortho-metalated ring), 7.42 (dd, 4J (HH) ) 3.8 Hz,
3J (HH) ) 7.9 Hz, 1H, ortho-metalated ring). 31P{1H} NMR
(CD2Cl2): δ 75.8 (s, PtBu3). 13C{1H} NMR (CD2Cl2): δ 32.9 (d,
P r ep a r a tion of [P d (TF A)(K2N,C-C6H4CHdNiP r )(P Cy3)]
(5). A mixture of complex 8 (0.096 g, 0.13 mmol) and tricyclo-
hexylphosphine (0.091 g, 0.32 mmol) in CH2Cl2 (20 mL) was
stirred at room temperature for 30 min. The resultant solution
was filtered through Celite, ethanol (40 mL) was added, the
solution was concentrated to ca. 5 mL on a rotary evaporator,
hexane (20 mL) was added, and the solution was oncentrated
to induce precipitation. The resultant precipitate was collected
by filtration and then recrystallized from CH2Cl2/EtOH to give
2J (PC) ) 3.4 Hz, CH3, Bu), 40.3 (d, J (PC) ) 7.9 Hz, CMe3),
t
1
3
3
48.9 (d, J (PC) ) 2.2 Hz, N(CH3)2), 73.5 (d, J (PC) ) 2.3 Hz,
CH2N), 117.2 (q, 1J (CF) ) 272.7 Hz, CF3), 124.1 (s, CH, ortho-
metalated ring), 124.3 (s, CH, ortho-metalated ring), 125.0 (d,
J (PC) ) 3.8 Hz, CH, ortho-metalated ring), 138.25 (d, J (PC)
) 6.4 Hz, CH, ortho-metalated ring), 144.4 (d, J (PC) ) 3.0
Hz, Pd-C), 148.2 (d, J (PC) ) 1.9 Hz, CCH2, ortho-metalated
2
ring), 161.2 (q, J (CF) ) 34.7 Hz, C, C-CF3).
1
the title complex as a colorless solid. Yield: 0.076 g (45%). H
Ca ta lysis. P r ep a r a tion of Stock Ca ta lyst a n d P h os-
p h in e Solu tion s. In a glovebox, volumetric flasks (10 mL,
class A) were charged with the palladium source (0.50 mol %
or 1.00 mol %, based on the amount of aryl chloride to be used
in reactions) or with the phosphine (1.00 mol % or 2.00 mol
%) and the appropriate solvent (i.e. the solvent used for the
catalytic run) was added to obtain 10 mL solutions. When
necessary, further dilutions were performed in order to obtain
the appropriate catalyst amount in 1.00 mL of solvent.
Gen er a l P r oced u r e for th e Su zu k i Rea ction . In a
glovebox, a Radleys tube was charged with the appropriate
aryl chloride (1.0 mmol), boronic acid (1.5 mmol), base (2.0
mmol), and solvent (1 or 2 mL, to obtain a total volume of 3
mL after addition of catalyst/phosphine solution(s)). Then the
freshly prepared stock solutions (see above) of phosphine (1.00
mL) and/or complex (1.00 mL) were added. The reaction
mixture was heated at 100 °C for 2 h, cooled in an ice bath for
5 min, and then HCl(aq) (5 mL, 2.3 M) was added to quench
the reaction. The aqueous layer was extracted with toluene
(30 mL) and CH2Cl2 (30 mL). The combined organic extracts
were dried over MgSO4, and the solution was filtered through
Celite. The solvent was removed on a rotary evaporator. The
residue was dissolved in toluene (4 mL), hexadecane (internal
standard, 0.034 M in toluene, 1.00 mL) was added, and the
mixture was analyzed by GC. Isolation and purification of
coupled products from representative reactions by column
chromatography (silica, hexane/ethyl acetate) gave the prod-
ucts with satisfactory microanalyses and with spectroscopic
data consistent with genuine examples in yields no more than
5-11% lower than the conversions determined by GC.
Gen er a l P r oced u r e for th e Bu ch w a ld -Ha r tw ig Am i-
n a tion Rea ction . NaOtBu (0.134 g, 1.39 mmol), 4-chloroani-
sole (0.12 mL, 1.0 mmol), and morpholine (0.11 mL, 1.26 mmol)
were added to an oven-dried Radleys tube, which was then
flushed with nitrogen. Then freshly prepared stock solutions
(see above) of phosphine (1.00 mL) and/or complex (1.00 mL)
and the necessary amount of toluene to bring the total solvent
volume to 3 mL were added. The reaction mixture was then
heated to reflux temperature for 17 h. The solution was cooled
to 50 °C, quenched with HCl(aq) (5 mL, 2.3 M), and basified
with NaOH(aq) (5 mL, 5 M). The aqueous layer was extracted
with toluene (2 × 30 mL), the combined organic extracts were
dried over MgSO4, the solution was filtered through Celite,
and the solvent was removed on a rotary evaporator. The
residue was dissolved in toluene (9 mL), hexadecane (internal
standard, 0.068 M in toluene, 1.00 mL) was added, and the
mixture was analyzed by GC. Isolation of the products of
selected couplings by column chromatography (silica, hexane/
ethyl acetate/methanol 90:10:2) gave microanalytically pure
NMR (CDCl3): δ 1.28 (m, 6 H, PCy3), 1.36 (d, 3J (HH) ) 6.6
Hz, 6 H, CH3), 1.70 (m, 12 H, PCy3), 1.83 (m, 6 H, PCy3), 2.01
(m, 6 H, PCy3), 2.30 (m, 3 H, PCy3), 3.98 (m, 1 H, CHMe2),
7.05 (m, 2 H, ortho-metalated ring), 7.18 (m, 1 H, ortho-
metalated ring), 7.27 (m, 1 H, ortho-metalated ring), 8.07 (m,
1 H, NdCH). 31P{1H} NMR (CDCl3): δ 39.35 (s, PCy3). Anal.
Calcd for C30H45F3NO2PPd: C, 55.8; H, 7.0; N, 2.2. Found: C,
55.2; H, 7.1; N, 1.95. This analysis was repeated on several
samples after multiple recrystallizations, but better data could
not be obtained.
P r ep a r a tion of [P d (TF A)(K2N,C-C6H4CH2NMe2){P Cy2-
(o-bip h en yl)}] (9a ). Complex 7a (0.182 g, 0.26 mmol) and
2-(dicyclohexylphosphino)biphenyl (0.198 g, 0.56 mmol) were
stirred in dichloromethane (30 mL) at room temperature for
2 h. The solution was concentrated to 5 mL, and diethyl ether
(15 mL) was added. Further concentration led to the precipita-
tion of the product. The supernatant solution was removed
with a syringe, and the solid was washed with diethyl ether
(2 × 10 mL) and then recrystallized from CH2Cl2/Et2O to give
the title compound as a colorless solid. Yield: 0.165 g (49%).
1H NMR (CDCl3): δ 1.00 (m, 2H, PCy2), 1.12 (m, 4H, PCy2),
1.34 (m, 4H, PCy2), 1.61 (m, 4H, PCy2), 1.69 (m, 2H, PCy2),
1.93 (m, 6H, PCy2), 2.64 (d, J (PH) ) 2.3 Hz, 6 H, CH3), 3.87
(br s, 2 H, CH2), 6.28 (m, 1H, H-6 of ortho-metalated ring),
3
3
4
6.51 (ddd, J (H5H4) ) 7.3 Hz, J (H5H6) ) 7.6 Hz, J (H5H3)
) 1.5 Hz, 1H, H-5 of ortho-metalated ring), 6.83 (ddd,
3
4
3J (H4H3) ) 7.3 Hz, J (H4H5) ) 7.3 Hz, J (H4H6) ) 0.9 Hz,
1H, H-4 of ortho-metalated ring), 6.89 (dd, 3J (H3H4) ) 7.3 Hz,
4J (H3H5) ) 1.5 Hz, 1H, H-3 of ortho-metalated ring), 7.17
(ddd, J ) 1.3 Hz, J ) 3.4 Hz, J ) 7.6 Hz, 1H, biphenyl), 7.24
(tt, J ) 1.2 Hz, J ) 7.6 Hz, 1H, biphenyl), 7.38 (m, 4H,
biphenyl), 7.52 (m, 2H, biphenyl), 7.79 (ddd, J ) 0.8 Hz, J )
7.9 Hz, J ) 11.3 Hz, 1H, biphenyl) 31P{1H} NMR (CDCl3): δ
52.75 (br s, PCy2(o-biphenyl). 13C{1H} NMR (CD2Cl2), δ 26.4
(d, J (PC) ) 1.5 Hz, CH2, PCy2), 27.25 (d, J (PC) ) 12.4 Hz,
CH2, PCy2), 27.6 (d, J (PC) ) 11.7 Hz, CH2, PCy2), 29.7 (s, CH2,
PCy2), 31.4 (s, CH2, PCy2), 34.1 (d, 1J (PC) ) 22.2 Hz, PCH),
49.6 (d, J (PC) ) 2.3 Hz, CH3, NMe2), 71.8 (d, J (PC) ) 3.0 Hz,
CH2N), 117.1 (q, 1J (CF) ) 293.1 Hz, CF3), 122.7 (s, CH,
aromatic), 123.95 (s, CH, aromatic), 124.7 (d, J (PC) ) 4.5 Hz,
CH, aromatic), 125.5 (d, J (PC) ) 34.3 Hz, C, aromatic), 126.3
(d, J (PC) ) 10.9 Hz, CH, aromatic), 128.0 (s, CH, aromatic),
128.2 (s, CH, aromatic), 129.8 (d, J (PC) ) 2.7 Hz, CH,
aromatic), 130.0 (s, CH, aromatic), 132.6 (d, J (PC) ) 6.8 Hz,
CH, aromatic), 136.9 (d, J (PC) ) 14.3 Hz, CH, aromatic), 137.9
(d, J (PC) ) 8.7 Hz, CH, aromatic), 142.3 (d, J (PC) ) 2.6 Hz,
C, aromatic), 145.0 (d, J (PC) ) 1.9 Hz, C, aromatic), 147.5 (d,
J (PC) ) 4.1 Hz, C, aromatic), 148.5 (d, J (PC) ) 2.3 Hz, C,

Catalysts for Suzuki Coupling
Organometallics, Vol. 22, No. 5, 2003 999
Ta ble 9. Cr ysta l Da ta for th e Com p lexes 7a , 8, 4a , 4b‚CHCl3, 4d , 4e, 9a ‚0.5Et2O, 10a , a n d 13
7a
8
4a
4b‚CHCl3
4d
4e
9a ‚0.5Et2O
10a
13
empirical
formula
fw
C
22H24F6N2- C24H24F6N2- C29H45F3N- C28H46Cl4N- C27H45IN-
C28H45F3N- C74H96F6N2- C18H19NO3
C15H20F3N2-
O3Pd
439.73
O4Pd
707.23
O4Pd2
731.25
O2PPd
634.03
PPd
PPd
O3PPdS O5P2Pd2
670.08 1482.27
675.83
647.91
297.34
cryst syst,
orthorhombic, orthorhombic, monoclinic,
monoclinic,
P21/n
monoclinic,
P21/n
monoclinic,
P21
monoclinic,
P21
orthorhombic, monoclinic,
space
P212121
P212121
P21/c
Cmc21
P21/c
group
a, Å
b, Å
c, Å
10.05000(10) 11.0425(2)
14.7341(2)
11.2352(2)
19.3377(3)
90
111.8780(10) 91.353010)
90
11.6996(2)
15.5253(2)
16.7499(3)
90
10.5363(2)
15.0361(3)
17.3668(3)
90
9.2892(2)
9.6862(2)
16.7860(6
90
11.05970(10) 30.742(2)
9.3727(9)
19.1440(15)
20.2492(19)
90
101.41
90
14.6781(2)
17.2111(3)
90
15.4028(3)
16.0274(4)
90
19.3488(3)
16.9257(3)
90
7.4158(4)
6.4650(3)
90
R, deg
â, deg
γ, deg
V, Å3
Z; calcd
density,
Mg m-3
abs coeff,
mm-1
90
90
99.7760(10) 92.1110(10) 99.2200(10) 90
90
2711.38(9)
4; 1.587
90
90
90
90
90
90
2538.89(6)
4; 1.850
2726.03(10)
4; 1.782
2970.62(8)
4; 1.418
3041.60(8)
4; 1.476
1509.33(7)
2; 1.474
3575.17(9)
2; 1.377
1473.86(14)
4; 1.340
3561.5(6)
8; 1.640
1.491
1.392
0.723
1.033
1.897
0.785
0.613
0.091
1.086
cryst size,
0.20 × 0.15 × 0.20 × 0.15 × 0.25 × 0.25 × 0.14 × 0.10 × 0.10 × 0.08 × 0.28 × 0.05 × 0.36 × 0.18 × 0.15 × 0.10 × 0.32 × 0.01 ×
0.10 0.10 0.20 0.08 0.05 0.02 0.12 0.03 0.005
26.00 25.50 27.48 25.03 27.50 27.50 25.03 25.03 23.03
mm3
θ
max, deg
no. of rflns 11 995/4909 15 811/5039 17 439/6656 25 311/5321 35 615/5499 11 588/6436 20 230/10 061 3095/1257
9643/2323
collected/
unique
Rint
0.0395
0.0267,
0.0708
0.0557
0.0296,
0.0671
0.0349
0.0284,
0.0714
0.0728
0.0310,
0.0677
0.0924
0.0385,
0.0951
0.0348
0.0341,
0.0734
0.0486
0.0366,
0.0829
0.0804
0.0475,
0.1189
0.0537
0.0535,
0.1097
final R
indices
(I > 2σ(I))
R1, wR2
R indices
0.0276,
0.0713
0.0343,
0.0688
0.0322,
0.0745
0.0457,
0.0728
0.0513,
0.1026
0.0420,
0.0772
0.0437,
0.0858
0.0538,
0.1251
0.0682,
0.1147
(all data)
R1, wR2
F
max, Fmin
,
0.584,
-0.685
0.673,
-0.526
0.509,
-1.097
0.516,
-0.623
1.100,
-1.264
0.801,
-0.762
0.807,
-0.573
0.285,
-0.204
0.546,
-0.643
e Å-3
samples of the coupled products with spectroscopic data
consistent with genuine samples in yields no more than ca.
7-12% lower than the conversions determined by GC.
(CH2)2O), 72.9 (s, CH2NMe2), 117.1 (q, 1J (CF) ) 292.0 Hz, CF3),
122.6 (s, CH, ortho-metalated ring), 125.1 (s, CH, ortho-
metalated ring), 125.6 (s, CH, ortho-metalated ring), 130.3 (s,
CH, ortho-metalated ring), 142.5 (s, C, ortho-metalated ring),
148.6 (s, C, ortho-metalated ring), 162.8 (q, 2J (CF) ) 35.3 Hz,
CCF3). Anal. Calcd for C15H20F3N2O3Pd: C, 41.0; H, 4.6; N,
6.4%. Found: C, 40.6; H, 4.7; N, 6.1.
X-r a y Str u ctu r e Deter m in a tion s. All data were collected
on a Bruker-Nonius KappaCCD area detector diffractometer
at 120 K with Mo KR radiation (λ ) 0.710 73 Å) produced by
a Bruker-Nonius FR591 rotating anode generator. The struc-
tures were solved by direct methods and refined by full-matrix
least squares using the SHELX-97 suite of programs.31 Non-
hydrogen atoms were refined anisotropically, while hydrogen
atoms were refined isotropically in geometric positions using
the riding model. Data were corrected for absorption by
comparison of equivalent reflections using the program SOR-
TAV.32 All figures were prepared using the program PLA-
TON.33 Crystallographic data for the complexes 7a , 8, 4a , 4b‚
CHCl3, 4d , 4e, 9a ‚0.5Et2O, 10a , and 13 are given in Table 9.
Full details of the crystallographic analyses are given in the
Supporting Information as CIF files.
Mech a n istic Stu d ies. P r ep a r a tion of N,N-Dim eth yl(2-
p h en yl)ben zyla m in e (12a ). A solution of 2-phenylbenzyl
bromide (1.9 mL, 10.40 mmol) and aqueous dimethylamine
(25-30% w/v, 10 mL) in ethanol (13 mL) was heated at reflux
temperature for 4 h, cooled, and acidified to pH 1 with HCl-
(aq). The solution was concentrated on a rotary evaporator to
ca. 15 mL and then basified to pH 14 with NaOH(aq). The
aqueous phase was extracted with diethyl ether (3 × 20 mL),
and the combined organic phase was washed with saturated
NaCl(aq) (3 × 30 mL) and water (3 × 30 mL) and then dried
(MgSO4). The solution was filtered through Celite, and the
solvent was removed on a rotary evaporator. Pentane (10 mL)
was added, the solution was filtered through Celite, and the
solvent was removed on a rotary evaporator to give the title
compound as a yellow oil. Yield: 1.080 g (49%). 1H NMR
(CDCl3): δ 2.17 (s, 6H, CH3), 3.39 (s, 2H, CH2), 7.24-7.45
(complex overlapping multiplets, 8H, aromatic), 7.58 (br, d of
m, J HH ≈ 7 Hz, 1H, aromatic). HRMS (CI): [M]+ calcd for
C
15H17N 211.1361, found 211.1364.
P r ep a r a tion of [P d (TF A)(K2N,C-C6H5CH2NMe2){NH-
Ack n ow led gm en t. We thank the University of
Exeter for funding (bursary for C.S.J .C.) and J ohnson
Matthey for the loan of palladium salts.
(CH2CH2)O}] (13). Complex 7a (0.386 g, 0.55 mmol) and
morpholine (0.12 mL, 1.38 mmol) were stirred in toluene (20
mL) for 15 min at room temperature. The solvent was removed
in vacuo, and the residue was recrystallized from toluene to
give the title product as a colorless solid. Yield: 0.377 g (79%).
1H NMR (CD2Cl2): δ 2.66 (s, 6H, CH3), 3.15 (br d, J (HH) )
13.6 Hz, 2H, (CH2)2NH), 3.38 (m, 2H, (CH2)2NH), 3.63 (ddd,
J (HH) ) 12.25 Hz, J (HH) ) 2.4 Hz, 2H, (CH2)2O), 3.85 (dd,
J (HH) ) 12.25 Hz, J (HH) ) 3.5 Hz, 2H, (CH2)2O), 3.89 (s, 2H,
CH2NMe2), 4.67 (br t, 1H, NH), 6.91 (m, 1H, ortho-metalated
ring), 7.00-7.10 (m, 3H, ortho-metalated ring). 13C{1H} NMR
(CD2Cl2): δ 50.4 (s, CH2, (CH2)2NH), 51.2 (s, NMe2), 68.3 (s,
Su p p or tin g In for m a tion Ava ila ble: Full details of the
crystallographic analyses as CIF files. This material is avail-
able free of charge via the Internet at http://pubs.acs.org.
OM020841+
(31) Sheldrick, G. M. SHELX-97; University of Go¨ttingen, Go¨ttingen,
Germany, 1997.
(32) Blessing, R. H. J . Appl. Crystallogr. 1997, 30, 421-429.
(33) Spek, A. L. Acta Crystallogr. 1990, A46, C34.