Platinum complexes of tertiary amine functionalised phosphines

Article abstract of DOI:10.1016/S0277-5387(02)01329-3

The reaction of several tertiary amine functionalised phosphines with [Pt(COD)Cl₂] has been studied. Reactions with ratios of L:Pt = 2 proceeded cleanly to give either cis or trans complexes depending on the cone angle of the phosphines. The trans complexes could also be prepared by reaction with Zeise's salt. Reactions of [Pt(COD)Cl₂] with L:Pt = 1 are less straightforward, and a mixture of cis and trans monodentate complexes, and in one example, a chelate complex have been isolated. The X-ray crystal structures of [PtCl₂(η²-ⁱPr₂PN(Me) CH₂CH₂NMe₂)] and a monodentate complex [trans-PtCl₂(η¹-ⁱPr₂ PNH-c-NC₅H₁₀)₂] are reported.

Full text of DOI:10.1016/S0277-5387(02)01329-3

Polyhedron 22 (2003) 19ꢁ26
/
www.elsevier.com/locate/poly
Platinum complexes of tertiary amine functionalised phosphines
Matthew L. Clarke a,*, Alexandra M.Z. Slawin b, J. Derek Woollins b
a School of Chemistry, Cantocks Close, University of Bristol, Bristol BS8 1TS, UK
b School of Chemistry, University of St. Andrews, St. Andrews, Fife KY16 9ST, UK
Received 22 May 2002; accepted 6 August 2002
Abstract
The reaction of several tertiary amine functionalised phosphines with [Pt(COD)Cl2] has been studied. Reactions with ratios of
L:Ptꢀ2 proceeded cleanly to give either cis or trans complexes depending on the cone angle of the phosphines. The trans complexes
could also be prepared by reaction with Zeise’s salt. Reactions of [Pt(COD)Cl2] with L:Ptꢀ1 are less straightforward, and a mixture
/
/
of cis and trans monodentate complexes, and in one example, a chelate complex have been isolated. The X-ray crystal structures of
[PtCl2(h2-i Pr2PN(Me)CH2CH2NMe2)] and a monodentate complex [trans-PtCl2(h1-i Pr2PNH-c-NC5H10)2] are reported.
# 2002 Elsevier Science Ltd. All rights reserved.
Keywords: Co-ordination chemistry; Platinum; Catalysis; Hemi-labile; Phosphines; Chelate ligands
1. Introduction
who demonstrated that palladium complexes of ligand 2
were particularly suited to this reaction [5]. Further
study revealed that the more readily synthesised ligand 3
could be used at least as effectively, even at room
temperature (it is possible that ligand 3 acts a phos-
Hemilabile ligands continue to provide novel applica-
tions in catalysis and organometallic chemistry. Phos-
phorus-amine ligands have proved to be very useful in
several reactions. For example, there are a large range of
chiral P,N ligands that have been shown to provide
excellent enantioselectivity in several asymmetric pro-
phorusꢁ
/
areneꢁalkene chelate) [6].
/
It is thought that the catalytic reaction proceeds
through palladium complexes that contain only one
phosphine ligand, and are, therefore, very reactive in
both oxidative addition and transmetallation steps. The
ligands 2 and 3 have also shown utility in palladium
catalysed amination and etherifications of aryl chlorides
[7]. We envisaged that reaction of a di-amine with
cesses [1] the use of phosphineꢁ
fruitful in palladium catalysed Cꢀ
/
imine ligands has proved
C bond forming
/
reactions [2] and pyridylphosphine palladium complexes
catalyse the carbonylation of alkynes thousands of times
more effectively than triphenylphosphine palladium
catalysts [3]. This reaction is an industrially viable
process.
In the last few years, several catalytic systems contain-
ing potentially hemilabile phosphine ligands have been
developed that allow the Suzuki cross coupling reaction
to be carried out using more readily available and less
costly aryl chlorides. Two of the ligand types are shown
in Fig. 1. The P,O chelate ligand 1 can catalyse the
iPr2PCl or Cy2PCl could deliver a range of Pꢁ
N ligands
/
that might be useful in this reaction. Work described in
our preliminary communication established that palla-
dium complexes prepared (in situ) from 4 and 5 are a
cheap and readily available catalyst for the Suzuki
reaction of aryl chlorides. Ligand 6 and an analogue
of 5 that lacked the second N atom, gave much lower
yields (Fig. 2, Scheme 1) [8].
Our initial attempts at isolating the catalytic inter-
mediates in these reactions have not been successful.
However, it seems reasonable that a difference in co-
ordination chemistry is responsible for the differing
catalytic activity seen from the various phosphino-amine
ligands. In order to gain greater understanding of the
coupling of aryl chlorides (1 mol% Pd2dba3×
105 8C) [4] while a more active catalytic system was
simultaneously developed by Buchwald and co-workers,
/CHCl3,
* Corresponding author. Fax: ꢂ
E-mail address: matt.clarke@bristol.ac.uk (M.L. Clarke).
/
44-117-929-0509
0277-5387/02/$ - see front matter # 2002 Elsevier Science Ltd. All rights reserved.
PII: S 0 2 7 7 - 5 3 8 7 ( 0 2 ) 0 1 3 2 9 - 3

20
M.L. Clarke et al. / Polyhedron 22 (2003) 19ꢁ26
/
2.2. X-ray crystallography
Crystal structure 12 was obtained using a Bruker
SMART diffractometer with graphite-monochromated
˚
0.71073 A). Intensity data were
Mo Ka radiation (lꢀ
/
collected using 0.38 or 0.158 width v steps accumulating
area detector frames spanning a hemisphere of recipro-
cal space for all structures.
Fig. 1.
Crystal structure 13 was determined by a Ryder CCD
diffractometer using Mo Ka radiation.
All data were corrected for Lorentz, polarisation and
long term intensity fluctuations. Absorption effects were
corrected on the basis of multiple equivalent reflections
or by semi-empirical methods. Structures were solved by
direct methods and refined by F2 (SHELXTL) [14] for all
data with I ꢀ3s(I) (Table 1).
/
Fig. 2.
2.3. N-Di-isopropylphosphino-N?-aminopiperidine (7)
co-ordination chemistry of these types of PꢁN ligand,
/
we have studied the reaction of a range of tertiary amine
functionalised phosphines with well known platinum
This ligand has not been reported previously and is
prepared by a similar method to that used for the other
phosphino-amines. Dropwise addition of neat di-iso-
propylchlorophosphine to a solution of N-amino-piper-
adine and triethylamine in diethylether gives a white
precipitate of Et3NHCl and a solution of the desired
ligand within 1 h. Filtration (using a cannula or filter
stick) and removal of solvent gave the desired ligand
precursors. Platinum complexes derived from PꢁN
/
ligands have been extensively studied [9]. However,
these generally feature a phosphine that contains a
more strongly co-ordinating sp2 hybridised nitrogen
donor. Late transition metals derived from PꢁN ligands
/
in which the N atom is an sp3 hybridised tertiary amine
are far less common [10,11], thus, providing the impetus
for the study described here.
Table 1
Crystal data for compounds 4 and 5
12
13
2. Experimental
Empirical formula
M
C
11H27Cl2N2PPt
C22H50Cl2N4P2Pt
698.59
triclinic
484.31
orthorhombic
Pna21
20.5504(10)
8.9428(5)
9.1679(4)
90
Crystal system
Space group
2.1. General
¯
P
/1/
˚
a (A)
7.976(2)
10.169(3)
10.234(3)
70.952(11)
81.716(14)
72.550(9)
747.5(4)
1
All manipulations were carried out under an atmo-
sphere of nitrogen, unless stated otherwise. All solvents
were either freshly distilled from an appropriate drying
agent (THF, Et2O, DCM) or obtained as anhydrous
grade (Aldrich Chemical Co.). 1H and 31P NMR spectra
were recorded using a ‘Varian 2000’ 300 MHz spectro-
meter. IR spectra were recorded as KBr discs (prepared
˚
b (A)
˚
c (A)
a (8)
b (8)
g (8)
U (A )
Z
m (mmꢃ1
90
90
3
˚
1684.86(14)
4
8.724
)
4.995
4606
Reflections measured
Independent reflections
Final R1
6764
2361
in air) on a Perkinꢁ
/
Elmer PE2000 FTIRꢁRAMAN
/
3865
0.0249
0.0714
spectrometer. Ligands 4, 6 and 8 have been reported
elsewhere [8] [12]. [Pt(COD)Cl2] was prepared by
literature methods [13].
0.0.261
0.0523
WR2 [I ꢀ2s(I)]
Scheme 1.

M.L. Clarke et al. / Polyhedron 22 (2003) 19ꢁ
/26
21
(colourless oil) in essentially pure form, and quantitative
yield.
br.), 3.25 (4H, s, br.), 3.7 (2H, s, br.), 5.3 (0.7H, s
CH2Cl2), 6.58 (2H, m), 7.12 (4H, t, Jꢀ7.0 Hz), 7.25
(2H, m), 7.44 (1H, t, Jꢀ7.4 Hz), 7.70 (4H, t, Jꢀ8.6
Hz), 8.13 (1H, d, Jꢀ3.9 Hz). M. S. (ESꢂ) (MꢃCl)ꢂ
req’s 952. 2752; Found: 952.2753.
/
31P NMR (121.4 MHz; CDCl3): d 62.9. 1H NMR
/
/
(300 MHz; CDCl3): d 1.0 (12H, m), 1.1ꢁ
(4H, m), 2.6 (4H, s, br.), 3.0 (1H, s, br.). M. S. (ESꢂ
MHꢂreq’s 217.1833; Found: 217.1833.
/1.3 (4H, m), 1.6
/
/
/
/
/)
/
Addition of 1 or 2 equiv. of AgBF4 to this compound
provides a new species which shows a 31P NMR
spectrum that strongly suggests the piperazine nitrogen
is co-ordinated to the platinum as shown in structure 14.
2.4. Synthesis of trans complexes from Zeise’s salt
31P NMR (121.4 MHz; CDCl3): d ꢃ
Hz; satellites weak and slightly broad), ꢃ
/
9.1 (1JPꢀPt
61.1 (1JPꢀPt
ꢀ3252
/
A solution of ligand (2 equiv.) in CH2Cl2 was added
dropwise to an acetone solution of Zeise’s salt. After
stirring for 5 min, the solvent was removed and the
complex extracted with CH2Cl2, filtered through Celite
and had solvent removed in vacuo. Triturating the pale
yellow precipitate with Et2O and drying in vacuo gave
the desired complex in essentially quantitative yield.
/
ꢀ
/
3140 Hz; satellites weak and slightly broad).
2.8. Cis-dichloro-(N-Di-isopropylphosphino-N,N?Nƒ-
trimethylethylenediamine, P,N)platinum(II) (12)
2.5. Trans-dichloro-bis-(N-Di-isopropylphosphino-N-
methylpiperazine)platinum(II) (9)
Addition of a dichloromethane solution containing 1
equiv. of N-Di-isopropylphosphino-N,N?Nƒ-trimethy-
lethylenediamine to a dichloromethane suspension of
[Pt(COD)Cl2] gave initially a mixture of products. The
major species had the following NMR spectrum typical
of a cis chelate complex.
Yellow powder. Anal. Calc. for C22H50N4P2Pt1Cl2: C,
37.82; H, 7.21; N, 8.02. Found: C, 38.05; H, 7.16; N,
7.86%. IR(nmax cmꢃ1) 2968, 2932, 2841, 2786, 1604,
1453, 1370, 1288, 1256, 1217, 1288, 1147, 1098, 1002,
31P NMR (121.4 MHz; CDCl3): d 61.8 (1JPꢀPt
ꢀ4001
/
961, 924, 884, 784, 696, 673, 635, 583, 498, 406, 338. 31
P
Hz). M.S. 449 (Mꢃ
/
Cl)ꢂ Recrystallisation by slow
NMR (121.4 MHz; CDCl3): d 77.3 (1JPꢀPt
ꢀ2678 Hz).
/
1H NMR (300 MHz; CDCl3): d 1.2 (12H, m), 2.25, (3H,
s), 2.42 (4H, s, br.), 2.75 (2H, m), 3.42 (4H, s, br.). M. S.
diffusion of diethyl ether into a concentrated CH2Cl2
solution of the mixture gave a few crystals of 12. There
was only enough pure material for a X-ray crystal-
lography and elemental analysis. Anal. Calc. for
C11H27N2P1Pt1Cl2: C, 27.28; H, 5.62; N, 5.78. Found:
C, 27.63; H, 5.63; N, 5.60%.
(ESꢂ
/
) MHꢂreq’s 698.2614; Found: 698.2619.
/
2.6. Trans-dichloro-bis-(N-Di-isopropylphosphino-
NN?Nƒ-trimethylethylenediamine)platinum(II) (10)
This complex, which is very soluble was isolated by
cooling a hexane solution in the freezer overnight to give
yellow crystals. Anal. Calc. for C22H54N4P2Pt1Cl2: C,
37.61; H, 7.76; N, 7.97. Found: C, 38.20; H, 8.29; N,
2.9. Trans-dichloro-bis-(N-Di-isopropylphosphino-N?-
aminopiperidine)platinum(II) (13)
7.80%. 31P NMR (121.4 MHz; CDCl3): d 82.2 (1JPꢀPt
ꢀ
/
Addition of a solution of N-Di-isopropylphosphino-
N?-aminopiperidine in CH2Cl2 to a suspension of
[Pt(COD)Cl2] in CH2Cl2 yields a yellow solution, which
1
2641 Hz). H NMR (300 MHz; CDCl3): d 1.25 (12H,
m), 2.14 (6H, s), 2.45 (2H, t, app., Jꢀ
m), 2.80 (2H, t, Jꢀ4.3 Hz), 2.22 (3H, br.). M. S. (ESꢂ
MHꢂreq’s 702. 2927; Found: 702.2938.
/7 Hz), 2.65 (2H,
/
/)
when analysed by 31P NMR shows several peaks with
/
1
one major product (dꢀ
/
59, JPꢀPt
ꢀ2485 Hz). Slow
/
diffusion (in air) of Et2O to a concentrated CH2Cl2
solution of this mixture reproducibly gave a few large
yellow blocks of the trans complex. This complex was
also prepared in high yield by the reaction of ligand 7
with Zeise’s salt in 2:1 ratio.
2.7. Cis-[(8)2PtCl2] (11)
Addition of CH2Cl2 to a schlenk tube containing 2
equiv. of ligand 8 and 1 equiv. of [Pt(COD)Cl2] and
stirring for 1 h gives quantitative conversion to the cis
complex. Removal of solvent to near dryness, and
addition of diethylether produces a white precipitate
that was collected on a frit (in air), washed with diethyl
ether, and dried in vacuo.
Anal. Calc. for C22H54N4P2Pt1Cl2: C, 37.82; H, 7.21;
N, 8.02. Found: C, 38.15; H, 7.03; N, 7.92%. 31P NMR
(121.4 MHz; CDCl3): d 59.1 (1JPꢀPt
NMR (300 MHz; CDCl3): d: 1.31 (12H, m), 1.55 (6H, s,
ꢀ
2485 Hz). 1H
/
br), 2.68 (4H, m, br.), 3.48 (1H, s, v.br.), 4.75 (2H, m,
Anal. Calc. for C44H48N6P2Pt1Cl2×
/
0.5 CH2Cl2: C,
51.83; H, 4.79; N, 8.15. Found: C, 52.46; H, 5.07; N,
br.). M. S. (FABꢂ
/
): 697 (Mꢂ
/
), 663, (Mꢃ
/
Clꢂ). This
/
7.86%. 31P NMR (121.4 MHz; CDCl3): d 1.2 (1JPꢀPt
3641 Hz). H NMR (300 MHz; CDCl3): d: 2.45 (4H, s,
ꢀ
/
compound was additionally characterised by X-ray
crystallography.
1

22
M.L. Clarke et al. / Polyhedron 22 (2003) 19ꢁ26
/
3. Results and discussion
reactions conditions. It is noteworthy that cis and trans
L2PtCl2 complexes have not been efficiently prepared
i
t
The amino-phosphine ligands 4, 6 and 7 are obtained
in essentially pure form by the slow addition of neat di-
isopropyl-chlorophosphine to an Et2O solution of the
corresponding diamine and triethylamine. The reactions
are typically complete in about 1 h. All of the ligands are
air and moisture sensitive liquids or oils, and were
characterised by multinuclear NMR, IR and high
resolution mass spectroscopies and were described in
communication form. Ligand 7, which still contains an
NH functionality, does not react further with iPr2PCl to
give a PNP type ligand. Ligand 8 has been reported
elsewhere [12].
The reaction of Zeises salt with diisopropylpho-
sphino-N-methylpiperazine, 4 gave the P-monodentate
trans-(h1-L)2PtCl2 compound (9) (100% conversion).
Analytically pure material was obtained after extraction
with DCM, filtration, and precipitation with diethyl
ether.
from [Pt(COD)Cl2] using either Pr3P, Cy3P and Bu3P
as ligands. We estimate that the cone angles of ligands 4,
6 and 7 to be similar to the parent tri-isopropylpho-
sphine (uꢀ
likely to be considerably less sterically demanding (uꢀ
1358ꢁ1458). PꢁN type ligands have been observed to
give cis or trans-[(h1-L)2PtCl2], cis-[(h1-L)(h2-L) PtCl2],
or cis-[(h2-L)2Pt]×
2Cl complexes depending on choice of
ligand [9,10]. It was, therefore, of interest to determine
how ligands 4ꢁ8 react with this well known precursor.
/
1608ꢁ
/
1708). Ligand 8, on the other hand is
/
/
/
/
/
The reaction of ligands 4 and 6 with [Pt(COD)Cl2] in
a 2:1 ratio gives the trans complexes 9 and 10 (100%
conversion by 31P NMR), which had also been prepared
from Zeises salt (Schemes 2 and 3). The reaction of 8
with [Pt(COD)Cl2] in a 2:1 ratio gives a quantitative
1
yield of the cis complex 11. The magnitude of JPꢀPt
(3641 Hz) is typical of a cis phosphine complex. In each
case, there is no evidence of nitrogen co-ordination
within the complexes. Ligand 7 gave a mixture of
products, in which the proportion of trans-[L2Pt Cl2]
increases over time. None of the peaks observed in the
Reaction
of
diisopropylphosphino-N,N?,Nƒ-tri-
methylethylenediamine, 10 with Zeises salt also gave a
P-monodentate complex in 100% conversion. Ligand 6
has three very flexible alkyl groups, and consequently its
platinum complex is significantly more soluble than 9.
Crystals of trans-(h1-L)2PtCl2 (10) could be obtained by
cooling a hexane solution in the freezer over night.
Complex 13 can also be prepared by the reaction of di-
isopropylphosphino-N-amino-piperidine, 7 and Zeise’s
salt (2:1 ratio).
1
31P NMR spectrum show JPꢀPt coupling constants
diagnostic for P trans chloride.
We have also studied the reactions of these ligands
and [Pt(COD)Cl2] in an equimolar ratio.
When diisopropylphosphino-N-methylpiperazine, 4
was reacted with [Pt(COD)Cl2] in a 1:1 ratio, a complex
mixture of products is formed. One of these is assumed
to be trans-L2PtCl2, 9 as it shares a similar 31P NMR
spectrum. We have not been able to separate any new
complexes from this mixture.
These complexes show the expected singlet with
satellites in their 31P NMR spectrum. The magnitude
1
of JPꢀPt (2678, 2642 and 2485 Hz, respectively) is
typical for alkyl phosphines located trans to each other.
The FAB mass spectra of the two complexes supports
the formulation suggested by the microanalytical data
The situation is somewhat different with diisopropyl-
phosphino-N,
N?,Nƒ-trimethylethylenediamine,
6.
Although the reaction does not proceed cleanly, there
is clearly a major product present. It was possible to
isolate a small sample of this complex in pure form by
by showing peaks due to MHꢂ
/
, Mꢃ
/
Cl and Mꢃ2Cl as
/
the highest mass ions present.
The reaction of simple mono- and di- phosphines with
[Pt(COD)Cl2] is a tried and tested method to form cis
platinum complexes [LPtCl2] or [L2PtCl2]. The products
are often formed in quantitative yields and purity after
removal of cyclo-octadiene (COD). An examination of
the literature reveals that ligands that are significantly
more bulky than triphenylphosphine (Tolman cone
recrystallisation from dichloromethaneꢁEt2O. The mag-
/
1
nitude of JPꢀPt (4001 Hz) is strongly indicative of a
phosphorus positioned trans to chloride and a complex
of formula [(6)PtCl2] is supported by chemical analyses
(Scheme 4).
To fully establish the structure of 12, an X-ray crystal
structure determination was carried out (Fig. 3, Table
2). This unambiguously confirms that the major product
angle [15] uꢀ1458) will generally selectively give trans
/
L2PtCl2 complexes. It is likely that the formation of the
trans complexes is predominantly due to steric factors as
no cis complexes are observed when electron rich Cy3P
from the [Pt(COD)Cl2]ꢂ1 equiv. Compound 6 is a 1:1
/
(uꢀ
/
1708) [16] or electron poor (3,5-(CF3)2C6H3)3P
(uꢀ
/
1708) [17] are reacted with [Pt(COD)Cl2]. However,
these reactions are also sensitive to electronic effects as
more complicated reactions are generally observed with
bulky electron rich ligands: a mixture of cis or trans-bi-
nuclear complexes, trans mononuclear complexes and
unidentified products can be formed depending on the
Scheme 2.

M.L. Clarke et al. / Polyhedron 22 (2003) 19ꢁ
/26
23
Scheme 3.
chelate complex. The platinum complex is of distorted
square planar geometry with the PꢀPtꢀN angle within
/
/
the chelate ring enlarged to 98.0(2)8. The corresponding
angle between the chloride ligands is reduced 85.70(10)8.
As phosphorus ligands have higher trans influence than
amines, the Ptꢀ
/
Cl bond trans to P is elongated by 0.077
A compared with the chloride bound trans to N. The
PtꢀN bond lengths in 12 are similar to, but slightly
longer, than those found in platinum Pꢁ
N(sp2 nitrogen)
chelate structures [9]. If the structure is compared with
the related structure, [Pt(Cl)(h2-Ph2PN(Me)ꢀ
CH2CH2ꢀ
NMe2, P,N) (Ph2PN(Me)ꢀCH2CH2ꢀNMe2CoCl3] re-
ported by Burrows and co-workers [11] many of the
parameters are similar. PꢀN and PtꢀN bond lengths
in the more electron rich phosphine complex, 12, are
˚
/
/
/
/
/
/
/
/
˚
0.02(5) and 0.05(5) A shorter, although this small
variation could be ascribed to other factors not related
Fig. 3. Molecular structure of complex 12.
to the ligand. The Pꢀ
/
Ptꢀ
/
N angle is significantly smaller
Table 2
(90.85(13)8) in the Burrows structure, but this is most
likely due to the more sterically crowded metal centre
that contains an extra phosphine ligand.
˚
Selected bond lengths (A) and angles (8) for 12
Bond lengths
It should be noted that the co-ordination behaviour of
6 in the presence of [Pt(COD)Cl2] is very different from
the phenyl substituted ligand. Whereas, the bulky
ligand, 6 generates a trans-monodentate complex in
the presence of 0.5 equiv. [Pt(COD)Cl2], the Burrows
ligand gave a complex of type cis-[(h2-L-P,N)(h1-L-
P)PtCl]Cl which was in equilibrium with the cis-
monodentate species. Meanwhile, equimolar reactions
of diphenylphosphino-N,N?,Nƒ-trimethylethylenedia-
mine and [Pt(COD)Cl2] occur cleanly to give the chelate
complexes in high yield. It, therefore, appears that
increasing steric bulk or basicity can reduce a ligands
tendency to chelate. We favour the former explanation,
and suggest that this may be due to an indirect effect:
Pt(1)ꢀCl(1)
Pt(1)ꢀP(1)
P(1)ꢀN(1)
2.295(2) Pt(1)ꢀCl(2)
2.224(2) Pt(1)ꢀN(9)
1.650(5)
2.371(2)
2.118(8)
Bond angles
Cl(1)ꢀPt(1)ꢀCl(2)
P(1)ꢀPt(1)ꢀCl(2)
N(9)ꢀPt(1)ꢀCl(1)
C(7)ꢀN(1)ꢀP(1)
C(7)ꢀN(1)ꢀC(8)
85.7(1) P(1)ꢀPt(1)ꢀCl(1)
172.3(2) N(9)ꢀPt(1)ꢀP(1)
174.1(6) N(9)ꢀPt(1)ꢀCl(2)
132.0(8) N(1)ꢀP(1)ꢀPt(1)
113.1(7)
87.02(8)
98.0(2)
89.1(2)
113.2(3)
The nitrogen in a cis-L2PtCl2 complex displaces a
chloride ligand that is labilised by the strong trans
influence of the P atom. The chloride ligands in trans-
L2PtCl2 complexes formed from bulky ligands are
bound more strongly to the platinum centre and less
likely to dissociate and allow N co-ordination.
The different behaviour that ligands 4 and 6 show
when reacted with Pt(COD)Cl2 may well have some
significance. Palladium complexes derived from di-
isopropylphosphino-N,N?,Nƒ-trimethylethylenediamine
do not show any catalytic activity in the Suzuki coupling
Scheme 4.
reaction of the tricky substrate chloroꢁtoluene. It is
/

24
M.L. Clarke et al. / Polyhedron 22 (2003) 19ꢁ26
/
rus ligands are part of the same hydrogen bonded ring.
Hydrogen bonding between halide ligands and NH in
phosphino-amine metal complexes is often observed,
although most reports are of intramolecular hydrogen
bonds [18,9]. It is possible the increased steric bulk of the
ligand and the trans geometry of complex 13 favours the
intermolecular hydrogen bonds observed here. The Ptꢀ
P distances within this complex are longer than those
found in the PꢁN chelate structure as befits the greater
trans influence of phosphorus over chloride. The PtꢀCl
/
/
/
bond lengths lie in between the values found for the Cl
trans P and Cl trans N in the chelate complex, and are in
line with arguments that chloride is slightly more trans
labilising than amine ligands. The presence of two
Scheme 5.
tempting to suggest that the formation of a strong P,N-
chelate could deactivate the palladium complex towards
oxidative addition or transmetallation steps. Di-isopro-
pylphosphino-N-methylpiperazine, which does not yield
heteroatom substituents on N(1) does not have any
˚
N bond length (1.668(3) A),
major effect on the Pꢀ
/
which is planar at nitrogen in common with most
phosphino-amines [9].
an isolatable PꢁN chelate complex with platinum shows
/
Reaction of ligand 8 with [Pt(COD)Cl2] in an
equimolar ratio gives an equimolar mixture of
[Pt(COD)Cl2] and cis[h1-(8)2PtCl2, (11), strong evidence
that ligand 8 strongly prefers monodentate cis co-
ordination. However, when 1 or 2 equiv. of silver
perchlorate were added to a NMR tube containing 11,
a new complex is formed quantitatively. Although we
have not isolated this new complex, it is worthy of
comment, as the 31P NMR spectrum observed is highly
good catalytic activity in the Suzuki reaction. There is
not sufficient evidence to make definite conclusions, but
it does seem likely that a subtle difference in co-
ordination strength is responsible for the differing
catalytic behaviour.
Addition of 1 equiv. of di-isopropylphosphino-N-
amino-piperidine to [Pt(COD)Cl2] does not proceed
cleanly either. Fortunately, slow diffusion of Et2O into
DCM solutions of the resulting mixture repeatedly gives
large yellow blocks of trans-L2PtCl2 as determined by
31P NMR spectroscopy and chemical analyses (Scheme
5).
Table 3
˚
Selected bond lengths (A) and angles (8) for 13
The crystal structure of 13 (Fig. 4, Table 3) confirmed
our spectroscopic and analytical data that suggested a
complex of type trans-L2PtCl2. The structure has
slightly distorted square planar geometry at platinum,
and a notable feature is the presence of two intermole-
Bond lengths
Pt(1)ꢀCl(1)#
Pt(1)ꢀP(1)
P(1)ꢀN(1)
2.306(1)
Pt(1)ꢀCl(1)
2.306(1)
2.3272(8)
1.428(4)
2.37(5)
2.3272(8) Pt(1)ꢀP(1)#
1.668(3)
3.098(3)
N(1)ꢀN(7)
NH(1)ꢀCl(1)
N(1)ꢀCl(1)#
cular hydrogen bonds between the NH moiety and a
˚
Bond angles
Cl(1)#ꢀPt(1)ꢀCl(1)
Cl(1)ꢀPt(1)ꢀP(1)
Cl(1)ꢀPt(1)ꢀP(1)#
N(7)ꢀN(1)ꢀP(1)
N(1)ꢀP(1)ꢀPt(1)
180.00(5) Cl(1)#ꢀPt(1)ꢀP(1)
88.82(4) Cl(1)#ꢀPt(1)ꢀP(1)
91.17(4) P(1)ꢀPt(1)ꢀP(1)#
91.18 (4)
88.83(4)
180.0
chloride ligand (N(1)ꢀ
/
HÁ Á ÁClꢀ
/
2.37(5) A; N(1)ꢀ
/
˚
Cl(1)ꢀ
are symmetrical, and the angle Clꢀ
88.82(4)8) is slightly larger if the chloride and phospho-
/
3.098(3) A. The Clꢀ
/
Ptꢀ
/
Pꢀ
/
Nꢀ
/
H ring systems
/
Ptꢀ
/
P (91.18(3)8 vs.
122.5
N(1)ꢀN(7)ꢀC(8)
110.0(3)
109.5(1)
Fig. 4. Molecular structure of complex 13.

M.L. Clarke et al. / Polyhedron 22 (2003) 19ꢁ
/26
25
it@ccdc.cam.ac.uk or www: http://www.ccdc.cam.ac.uk)
on request quoting deposition codes: CCDC 185583 and
CCDC 185584.
Acknowledgements
The authors wish to thank Johnson Matthey for loan
of platinum salts, the EPRSC mass spectrometry service
(Swansea), EPSRC, and the JREI for equipment grants.
The authors also wish to thank Professor D.J. Cole-
Hamilton and Dr. D.F. Foster who collaborated in our
earlier catalytic studies with these ligands.
Scheme 6.
distinctive of a particular co-ordination type. Two
slightly broadened singlets with shortened Pt satellites
(satellites were ca. 5ꢁ/10% of the main peak area,
possible due to relaxation effects) are observed in the
References
1
31P NMR spectrum (dPꢀ
/
ꢃ
/
9.1, JPꢀPt
3140 Hz). There was no change in this
spectrum when cooled to ꢃ60 8C. A shift of 60 ppm
ꢀ
/
3252 Hz; dpꢀ
/
1
ꢃ
/
61.1, JPꢀPt
ꢀ
/
[1] (a) A. Lightfoot, P. Schnider, A. Pfaltz, Angew. Chem., Int. Ed.
Engl. 37 (1998) 2897;
/
(b) J.M.J. Williams, Synlett (1996) 705 and references therein.;
(c) G.J. Dawson, C.G. Frost, J.M.J. Williams, Tetrahedron Lett.
34 (1993) 3149;
upfield, along with a reduction in the magnitude of
1JPꢀPt suggests that a four membered phosphorus
containing ring had formed [19,13]. This is most easily
(d) J. Sprintz, G. Helmchen, Tetrahedron Lett. 34 (1993) 1769;
(e) P. Von Matt, A. Pfaltz, Angew. Chem., Int. Ed. Engl. 32
(1993) 566;
explained by structure 14. The lack of Pꢀ
/
P coupling is
somewhat surprising but can be ascribed to the some-
what broad resonances. Addition of silver salts to other
complexes described in this paper generally led to a
complex mixture of products that could not be identified
(Scheme 6).
(f) M. Sawamura, Y. Ito, Chem. Rev. 92 (1992) 857;
(g) J.M. Brown, D.I. Hulmes, T.P. Layzell, J. Chem. Soc., Chem.
Commun. (1993) 1673.;
(h) I. Beletskaya, A. Pelter, Tetrahedron 35 (1997) 4957.
[2] H. Yoshida, E. Shirakawa, T. Kurahashi, Y. Nakao, T. Hiyama,
Organometallics 19 (2000) 5671 (and references therein).
[3] E. Drent, P. Arnoldy, P.H.M. Budzelaar, J. Organomet. Chem.
455 (1993) 247 (and references therein).
4. Summary and conclusions
[4] (a) X. Bei, H. Turner, W.H. Weinberg, A.S. Gurran, J. Org.
Chem. 64 (1999) 6797;
(b) X. Bei, H. Turner, W.H. Weinberg, A.S. Gurran, Tetrahedron
Lett. 40 (1999) 1237.
This study has highlighted the different co-ordination
chemistry observed when bulky electron rich ligands are
used in place of more common diphenylphosphino
substituted ligands. We have also determined the
propensity of the ligands 4, 6, 7 and 8 to form chelate
complexes with platinum. It is clear that these ligands
are far less willing to act as bidentate ligands with
platinum than sp2 nitrogen functionalised phosphines.
Sp2 nitrogen functionalised phosphines have been
shown to readily form four, five and six membered
chelate rings when added to [Pt(COD)Cl2] or similar
precursors [10]. Further applications of weakly chelating
electron rich ligands in homogeneous catalysis will be
reported in due course.
[5] D. Old, J.P. Wolfe, S.L. Buchwald, J. Am. Chem. Soc. 120 (1998)
9722.
[6] J.P. Wolfe, R.A. Singer, B.H. Yang, S.L. Buchwald, J. Am.
Chem. Soc. 121 (1999) 9550.
[7] (b) A. Aranyos, D.W. Old, A. Kiyomori, J.P. Wolfe, J.P. Sadighi,
S.L. Buchwald, J. Am. Chem. Soc. 121 (1999) 4369;
(c) D. Old, J.P. Wolfe, S.L. Buchwald, J. Am. Chem. Soc. 120
(1998) 9722.
[8] M.L. Clarke, D.J. Cole-Hamilton, J.D. Woollins, J. Chem. Soc.,
Dalton Trans. (2001) 2721.
[9] Many P,N ligands readily form chelate complexes when reacted
with soluble sources of PtCl2. For examples: (a) S.A. Aucott,
A.M.Z. Slawin, J.D. Woollins, J. Chem. Soc., Dalton Trans.
(2001) 2559. (b) S.M. Aucott, M.L Clarke, A.M.Z. Slawin, J.D.
Woollins, J. Chem. Soc., Dalton Trans. (2001) 972. (c) A.J.
Blacker, M.L. Clarke, M.A. Humphries, M.S. Loft, M.F. Mahon,
J.M.J Williams, Chem.: A. Eur. J. 6 (2000) 353. (d) For a review
of Ph2Ppy chemistry: H. Cheng, Z. Zhang, Coord. Chem. Rev.
147 (1996) 1.
5. Supplementary data
[10] A.D. Burrows, M.F. Mahon, M.T. Palmer, J. Chem. Soc., Dalton
Trans. (2000) 3615.
Full crystallographic data has been deposited at the
Cambridge crystallographic database and is available
from The Director, CCDC, 12 Union Road, Cambridge,
[11] G.K. Anderson, R. Kumar, Inorg. Chem. 23 (1984) 4064.
[12] M.L Clarke, A.M.Z. Slawin, M.V. Wheatley, J.D. Woollins, J.
Chem. Soc., Dalton Trans. (2002) 3421.
CB2 1EZ, UK (fax: ꢂ
/
44-1223-336033; e-mail: depos-
[13] H.C. Clark, L.E. Manzer, J. Organomet. Chem. 59 (1973) 411.

26
M.L. Clarke et al. / Polyhedron 22 (2003) 19ꢁ26
/
[14] SHELXTL, structure solution and determination package, Bruker
AXS, Madison, WI, 1999.
[18] Q. Zhang, A.M.Z. Slawin, J.D. Woollins, J. Chem. Soc., Dalton
Trans. (2001) 621 and 3598.
[15] C.A. Tolman, Chem. Soc. Rev. 1 (1972) 337.
[16] I.M. Al-Najjar, Inorg. Chem. Acta. 128 (1987) 93.
[17] M.L. Clarke, D. Ellis, K.L. Mason, A.G. Orpen, P.G. Pringle, in
preparation.
[19] (a) P.E. Garrou, Chem. Rev. 81 (1981) 229;
(b) S. Hietkamp, D.J. Stufkens, K. Vrieze, J. Organomet. Chem.
169 (1979) 107.