Development of a Synthetic Route to Unsymmetrical Triphosphine Ligands and an Investigation of Their Coordination Chemistry with Nickel and Palladium

Article abstract of DOI:10.1021/ic960136r

Chiral tridentate phosphines, R₂P(CH₂)₃PPh(CH₂) ₂PPh₂ where R = C₆H₅, p-ClC₆H₄, and p-FC₆H₄, can be prepared from simple starting materials, (R₃P, I(CH₂)₃I, and Ph₂P(CH₂)₂PPh₂), in a few stages involving phosphonium salts and phosphine oxides as intermediates. Crystalline diamagnetic complexes of nickel(II) and palladium(II) have been isolated. In solution these show first-order 12 line ³¹P NMR spectra consistent with three nonequivalent phosphorus nuclei coupled to one another in a square planar geometry. A single X-ray crystallographic study of NiI₂{P(CH₂)₃PPh(CH₂) ₂PPh₂} showed that this was square pyramidal in the solid state with a weakly held apical iodo ligand.

Full text of DOI:10.1021/ic960136r

4098
Inorg. Chem. 1996, 35, 4098-4102
Ar ticles
Development of a Synthetic Route to Unsymmetrical Triphosphine Ligands and an
Investigation of Their Coordination Chemistry with Nickel and Palladium
G. Dyer† and J. Roscoe*,‡
School of Chemical Sciences, University of East Anglia, Norwich NR4 7TJ, U.K., and Department of
Chemistry, University of Central Lancashire, Preston PR1 2HE, U.K.
ReceiVed February 8, 1996X
Chiral tridentate phosphines, R2P(CH2)3PPh(CH2)2PPh2 where R ) C6H5, p-ClC6H4, and p-FC6H4, can be prepared
from simple starting materials, (R3P, I(CH2)3I, and Ph2P(CH2)2PPh2), in a few stages involving phosphonium
salts and phosphine oxides as intermediates. Crystalline diamagnetic complexes of nickel(II) and palladium(II)
have been isolated. In solution these show first-order 12 line 31P NMR spectra consistent with three nonequivalent
phosphorus nuclei coupled to one another in a square planar geometry. A single X-ray crystallographic study of
NiI2{P(CH2)3PPh(CH2)2PPh2} showed that this was square pyramidal in the solid state with a weakly held apical
iodo ligand.
Introduction
Scheme 1. Preparation of Eptp
The unsymmetrical triphosphine ligand R2P(CH2)3PPh(CH2)2-
PPh2 (R ) C6H5), “eptp”,1 is not well-known. Its synthesis
was reported by Meek and co-workers in 1975.2 It was prepared
on a small scale by heating (3-(diphenylphosphino)propyl)-
phenylphosphine with vinyldiphenylphosphine in the presence
of 2,2-azoisobutyronitrile (AIBN). Later it was shown that the
ligand could also be synthesized rapidly on platinum by an Et3N-
catalyzed reaction of vinyldiphenylphosphine with the coordi-
nated secondary phosphine:3
These preparations use expensive and difficult-to-prepare
starting materials. In contrast, the aim here was to produce the
ligand, and some variations with substituted phenyl groups, from
readily available starting materials, in a few simple stages
involving phosphonium salts and phosphine oxides as interme-
diates.
This ligand is a simple tridentate analogue of the well-known
bidentate ligands Ph2P(CH2)nPPh2, (n ) 2, 3, etc.). It is able
to chelate to three of the four coordination sites of a square
planar complex. Unlike the much studied symmetrical “ttp”
ligand, PhP[(CH2)3PPh2]2,2-4 the ligand eptp contains a chiral
phosphorus center and three nonequivalent phosphorus nuclei.
These features are of course imparted to its square planar
complexes which could therefore have applications as homo-
geneous catalysts for asymmetric syntheses since molecular
models show that the third group on the central phosphorus atom
must be directed to one side of the square plane. This causes
an incoming fifth ligand to attach to the metal on the other side
of the square plane and experience a chiral array of four different
ligand groups in a given clockwise order. The complexes are
also ideally suited to 31P NMR studies. The only complexes
reported by Meek and co-workers3,5 were [Pt(eptp)Cl]AsF6 and
[RhNO(eptp)]. In this study we report analogous nickel(II) and
palladium(II) complexes.
† University of Central Lancashire.
‡ University of East Anglia.
X Abstract published in AdVance ACS Abstracts, June 15, 1996.
(1) The ligand abbreviations are derived from the lengths of the alkyl
chains connecting the phosphorus atoms. The nonsymmetric triphos-
phine Ph2P(CH2)3PPh(CH2)2PPh2, (2-(diphenylphosphino)ethyl)(3-
(diphenylphosphino)propyl)phenylphosphine then becomes abbreviated
to eptp.
Results and Discussion
The new synthetic route to eptp is shown in Scheme 1. Each
stage was monitored by 31P{1H} NMR spectroscopy, which
confirmed that each reaction proceeded effectively quantita-
tively. The eptp ligand was collected as an oil.
(2) Dubois, D. L.; Myers, W. H.; Meek, D. W. J. Chem. Soc., Dalton
Trans. 1975, 1011.
(3) Waid, R. D.; Meek, D. W. Inorg. Chem. 1984, 23, 778-782.
(4) King, R. B.; Cloyd, J. C. Inorg. Chem. 1975, 14, 114.
(5) Mazanec, T. J.; Tau, K. D.; Meek, D. W. Inorg. Chem. 1980, 19,
85-91.
S0020-1669(96)00136-X CCC: $12.00 © 1996 American Chemical Society

Unsymmetrical Triphosphine Ligands
Inorganic Chemistry, Vol. 35, No. 14, 1996 4099
Table 1. Summary of 31P{1H} NMR Spectra (CDCl3 at 25 °C)
chemical shift (ppm)
coupling const (Hz)
coord shift (ppm)
P(3)
complex
P(2)
P(3)
P(1)
P(2-3)
P(3-1)
P(2-1)
P(2)
P(1)
P(2)+(CH2)3P(3)+(CH2)2P(1)
P(2)(O)(CH2)3P(3)(O)(CH2)P(1)
P(2)(CH2)3P(3)(CH2)2P(1), eptp
eptp-Cl2
eptp-F2
o-anisyl2-eptp
[NiI(eptp)]I
[NiI(eptp)]ClO4
[NiBr(eptp)]ClO4
[PdBr(eptp)]ClO4
[PdBr(eptp)]PF6
23.9
32.5
-16.7
-18.9
-18.7
-34.8
0.2
29.0
41.5
-20.9
-22.0
-21.2
-20.6
53.3
54.9
56.9
54.8
53.9
52.2
53.9
57.2
56.4
52.8
59.5
-11.1
-11.4
-12.6
-13.4
-11.6
-11.9
60.6
60.9
60.0
58.5
58.2
62.4
63.2
60.9
59.6
62.5
66.1
10.2
47.3
45.5
29.8
29.4
28.0
26.3
20.9
28.8
47.7
0
72.3
80.9
80.0
38.6
38.5
69.8
71.4
80.2
36.8
71.5
73.7
281.8
276.9
276.6
414.0
413.4
285.2
283.1
280.4
415.2
283.8
273.1
16.9
18.4
19.8
17.8
17.6
19.5
20.0
23.1
20.0
18.8
41.0
74.2
75.8
77.8
75.7
74.8
74.2
75.9
79.2
78.4
74.0
80.1
73.2
73.5
72.6
71.1
70.8
75.8
76.6
74.3
73.0
74.1
78.0
1.7
3.1
1.1
0.9
0.6
1.1
4.2
1.1
0
[NiI(eptp-Cl2)]I
18.1
21.3
45.2
0
20.2
24.0
[NiI(eptp-Cl2)]ClO4
[NiBr(eptp-Cl2)ClO4
[PdBr(eptp-Cl2)]PF6
[NiI(eptp-F2)]I
0.1
6.2
[NiI(eptp-o-anisyl2)]I
The 31P NMR spectrum comprised two doublets, (-20.9 and
-12.6 ppm with a coupling constant of 29.8 Hz), and a singlet,
(-16.7 ppm), in the phosphine region, the chemical shifts and
coupling constants being identical to those reported by Meek
and Tau.2,6 Surprisingly no trace of phosphine oxide was
detectable in the 31P NMR spectrum even after many months
of exposure to air.
Using this synthetic route, eptp is easily prepared from very
simple starting materials, in a few stages which do not require
the use of organometallic reagents, dried solvents, or even a
nitrogen atmosphere. All intermediates are air stable and readily
characterized by 31P NMR spectroscopy and elemental analysis
as pure materials. Since dppe can be quickly made from PPh3
by a slight modification of the usual literature method,7 all three
phosphorus atoms originate simply from PPh3, while the
-(CH2)2- and -(CH2)3- backbones originate from dichloro-
ethane and diiodopropane respectively.
Figure 1. 31P{1H} NMR spectrum of NiI2(eptp) (25 °C, CDCl3) (60.6
ppm (d of d), 53.3 ppm (d of d), 0.2 ppm (d of d), J ) 72.3, 20.9,
281.8 Hz).
ethanol solution. Melting points and elemental analysis data
are given in Table 3. The orange-brown colors and diamag-
netism strongly indicate a square planar structure in solution.
Consistent with this, the 31P{1H} NMR spectra (Table 1 and
Figure 1) showed 12 sharp lines, i.e. a doublet of doublets for
each phosphorus atom.
Tridentate ligands R2P(CH2)3PPh(CH2)2PPh2, where R )
p-ClC6H4 and p-FC6H4 were made with the same ease by
substituting PR3 into the first step of the reaction sequence in
place of triphenylphosphine. All observations and intermediates
were similar and 31P NMR spectra almost identical, with the
exception of the final tridentate phosphine products. Here the
peak for the phosphorus atom at the end of the trimethylene
backbone was shifted upfield by about 2 ppm due to the
substitution of a halogen onto the phenyl ring. It is expected
that other unsymmetrical tridentate phosphine ligands with
different substituents will be made with similar ease. The
ligands were isolated as oils and used for complexation after
recrystallization from methanol by cooling to -20 °C.
The triphosphine eptp has interesting steric features resulting
from its lack of symmetry, including three nonequivalent
phosphorus nuclei, one of which is chiral. This chirality will
inevitably be induced onto the metal center of a complex, so
that the metal and any substrate molecule which attaches to it
will exist in a rigidly defined, highly asymmetric environment.
Suitable substituents introduced into the phenyl rings could
enhance chiral recognition.
Nickel(II) phosphine complexes have been widely used as
the basis for studying steric effects in terms of “cone angle”,8
and for investigating the “ring contribution” to coordination
shifts. However problems can arise when polydentate ligands
are examined which have the potential to form different size
chelate rings as in eptp. The very large coordination shifts
observed when phosphorus is incorporated in a five-membered
chelate ring are well-known and usually attributed to some form
of “ring strain”.9 The series of nickel(II) complexes prepared
here are ideal for investigating such phenomena since of the
three phosphorus atoms, one is in a five-membered chelate ring,
one is in a six-membered ring, and one is in both.
From Table 1 it can be seen that the coordination shifts are
about the same for palladium(II) as for nickel(II), being 18-
25 ppm for P(2) and 70-83 ppm for P(3) and P(1) (the lettering
system used is Ph2P(2)(CH2)3P(3)Ph(CH2)2P(1)Ph2 as shown
on Figure 2). The last compound in the table is exceptional
because of the o-anisyl groups on P(2), which obviously affects
the coordination shift of P(2) very markedly.
To compliment the solution studies, which firmly indicated
that the unsymmetrical chiral tridentate phosphine produced
chelate mononuclear square planar species, it was decided to
grow crystals for X-ray analysis. X-ray crystallography (see
Orange-brown nickel(II) complexes containing the [NiX-
(eptp)]+ cation (X ) Br or I) were easily prepared in 50-90%
yield by reaction of the appropriate nickel(II) salts in ethanol
solution with the stoichiometry quantity of eptp. The complexes
are stable to air and water, and may be crystallized from
dichloromethane/ethanol. They behave as 1:1 electrolytes in
(8) Butler, I. S.; Harrod, J. F. Inorganic Chemistry-Principles and
Applications; Benjamin/Cummings: Reading, MA, 1989; p 678.
(9) Bell, C. F. Principles and Applications of Metal Chelation; Oxford
University Press: Oxford, England, 1977.
(6) Tau, K. D. Ph.D. Thesis; The Ohio State University, 1978.
(7) Hewertson, W.; Watson, H. R. J. Chem. Soc. 1962, 1490.

4100 Inorganic Chemistry, Vol. 35, No. 14, 1996
Dyer and Roscoe
plane, and these may be termed “axial” phenyl groups in contrast
to the other two which are “equatorial”. Such a conformation
leads to considerable steric blocking on one side of the nickel
by three bulky, nonpolar, hydrophobic phenyl groups, while the
other side is remarkably uncluttered. Coordination of the apical
iodine naturally occurs on the sterically unhindered side.
(d) Unlike the solution studies, the X-ray result proved that
the [NiI(eptp)]+ species is capable of weakly binding a fifth
ligand in an apical position, the conventional first stage of many
catalytic cycles. It is important that the binding is not too strong,
otherwise the process either stops completely or proceeds but
with a slow turnover rate. The 298 pm bond distance to the
apical iodine indicates significant interaction with the nickel,
but the binding is obviously much weaker than the other iodine
at 254 pm from the nickel.
(e) The central phosphorus of the tridentate ligand is
significantly closer to the nickel (217.0 pm) than the other two
phosphorus atoms (222.0, 223.5 pm), presumably because it is
required to be simultaneously part of a five-membered and a
six-membered ring. The “double chelate effect” could encour-
age stronger bonding, and the bicyclo structure could restrict
vibrational freedom and/or force this phosphorus atom closer
to nickel.
(f) There are several possibilities for the conformation of the
C3 backbone in the six-membered chelate ring. The arrangement
observed here is loosely analogous to the “chair” conformation
of cyclohexane.
Figure 2. X-ray crystal structure of [NiI(eptp)]I.10 Table 2 shows
important bond lengths and bond angles.
Table 2. Summary of X-ray Data for NI2(eptp)
(a) Intramolecular Distances (Å)
Ni-I(1)
Ni-I(2)
Ni-P(1)
2.544(2)
2.974(2)
2.217(4)
Ni-P(2)
Ni-P(3)
2.233(4)
2.164(5)
(b) Intramolecular Bond Angles (deg)
The ligands R2P(CH2)3PPh(CH2)2PPh2, where R ) p-ClC6H4
and p-FC6H4, were used to prepare the analogous complexes
with NiI2, which gave analogous 31P{1H} NMR spectra. All
the spectra showed the typical 12 lines, and coupling was similar
with all three ligands. The fluoro-substituted ligands are
probably the most interesting since the nonequivalence of the
two p-fluoro substituents on the terminal phosphorus can be
determined using 19F NMR spectroscopy. The 75 MHz 19F
NMR spectrum of the NiI2 complex of eptp-F2 showed two
sharply-resolved signals separated by 2.15 ppm corresponding
to a rigid structure with two different fluorine environments
which may loosely be described as “axial” and “equatorial”,
(attached to C(29) and C(22A) in Figure 2).
I(1)-Ni-I(2)
I(1)-Ni-P(1)
I(1)-Ni-P(2)
I(1)-Ni-P(3)
I(2)-Ni-P(1)
105.08(8)
92.1(1)
91.3(1)
163.0(1)
91.1(1)
I(2)-Ni-P(2)
I(2)-Ni-P(3)
P(1)-Ni-P(2)
P(1)-Ni-P(3)
P(2)-Ni-P(3)
96.5(1)
91.6(1)
170.6(2)
84.0(2)
90.2(2)
Figure 2 and Table 2) of [NI2(eptp)]10 surprisingly revealed five
coordination (an only slightly distorted square pyramid), in the
solid state. However, the Ni-I bond lengths show that the
second iodide ligand in the apical position is only weakly held
which explains the square planar structure found in solution.
As is usual in square pyramidal stereochemistry, the nickel lies
just above the basal plane.
A number of interesting features revealed in the solid-state
structure deserve further comment.
Palladium(II) complexes based on the same structure as the
nickel complexes were prepared with eptp by addition of the
ligand to a solution of palladium(II) bromide perchlorate or
palladium(II) bromide hexafluorophosphate. The 31P NMR
spectra of these complexes were slightly different from the
pattern set by the nickel complexes. The spectra showed only
(a) Five-coordinate complexes of nickel(II) and other d8 metal
ions are quite well-known, but the majority are trigonal
bipyramidal, rather than square pyramidal (e.g. Fe(CO)5).
Where ligands are not all equivalent, distortion of the trigonal
bipyramidal structure necessarily occurs, sometimes toward an
approximately square pyramidal arrangement, although crystal
packing forces may sometimes favor this so that the structure
in solution may be markedly different.
-
eight sharp peaks, (PF6 being ignored). The phosphorus at
the end of the two carbon chain was only coupled to the other
terminal phosphorus through the metal and did not show
coupling to its phosphorus neighbor along the two-carbon chain.
This is presumably due to the fact that the coupling along the
carbon chain is equal to the coupling through the palladium
but opposite in sign.
Many other complexes can be envisaged with interesting
structures based on this unusual asymmetrical tridentate ligand
system. The complexes made to date with nickel and palladium
are all crystalline solids, stable in air and stereochemically rigid
at room temperature on the NMR time scale making investiga-
tion and characterization relatively straight forward.
(b) The chirality of the molecule is clearly accentuated by
the coordination of the apical iodo ligand which “sees” either
a clockwise or an anticlockwise arrangement of C2 backbone,
C3 backbone, and the other ligand. Chiral recognition between
the molecules is presumably very strong, and the X-ray study
shows a regular array of alternating R- and S-enantiomers in
the solid crystal, which is therefore not optically active.
(c) The structure clearly shows that three phenyl groups, one
on each phosphorus, are directed to the side of the molecule
opposite to the apical iodide. The planes of these three phenyl
rings are approximately perpendicular to that of the NiP3I square
Experimental Section
All Fourier-mode, proton-decoupled phosphorus-31 NMR spectra
were collected on a Bruker 250 MHz spectrometer at 101 Hz in CDCl3
at 25 °C. Elemental analyses were performed by The University of
Manchester Microanalytical Laboratory.
(10) The X-ray crystallographic study was carried out by R. G. Pritchard
and C. A. McAuliffe of the University of Manchester Institute of
Science and Technology. The X-ray data will be published at a later
date.

Unsymmetrical Triphosphine Ligands
Inorganic Chemistry, Vol. 35, No. 14, 1996 4101
Preparation of the Tridentate Phosphine Ligand, eptp. Prepara-
tion of [Ph3P+(CH2)3I]I- (I). A solution of triphenylphosphine (5.25
g, 0.02 mol) in toluene (30 mL) was added dropwise over 15 min, by
way of a dropping funnel, to a refluxing, stirred solution of diiodopro-
pane (5.92 g, 0.02 mol) in toluene (10 mL). Reflux was maintained
for 1.5 h, and then the resulting white precipitate was filtered on a
sinter from the hot toluene (yield 7.60 g, 68%). 31P{1H} NMR: 25.2
ppm (s).
Preparation of [Ph3P+(CH2)3P+Ph2(CH2)2PPh2](I-)2 (II). Bis(1,2-
diphenylphosphino)ethane, “dppe”, Ph2P(CH2)2PPh2 (11.95 g, 0.03 mol),
was dissolved in hot dimethylformamide, DMF (60 mL, 120 °C).
Product I (5.58 g, 0.01 mol) was added as a powder to the hot DMF
solution and the resulting clear colorless solution was heated at 120
°C for 30 min and then cooled to room temperature with continuous
stirring, during which time the excess bidentate phosphine crystallized
out and could be filtered from the cooled DMF solution (5.58 g, i.e.
73% recovery of the 0.02 mol excess of bidentate phosphine used).
Dry diethyl ether (180 mL) was added to the filtrate to precipitate the
product. The solvent mixture was decanted off and the product washed
with warm toluene (100 mL, 60 °C) to remove any remaining dppe
starting material. The pure product was collected by filtration from
warm toluene as a white powder and washed with diethyl ether (2 ×
10 mL) (yield 9.16 g, 96%). 31P{1H} NMR: 29.0 ppm (d of d), 23.9
ppm (d), -11.1 ppm (d) J ) 10.2, 47.3 Hz.
Preparation of Ph2P(O)(CH2)3P(O)Ph(CH2)2PPh2 (III). Product
II (9 g, 0.009 mol) was added to sodium hydroxide solution (5 g of
NaOH in 50 mL of H2O), in a 250 mL conical flask equipped with a
magnetic stirrer bar and a reflux condenser. The sodium hydroxide
solution was brought to reflux. On heating the solid phosphonium salt
became a brown oil and was stirred vigorously for 2 h at reflux. After
this time the sodium hydroxide solution was cooled to below room
temperature and decanted off to leave the product as a viscous light
brown oil which was washed with cold water (2 × 15 mL), and dried
in an oven (120 °C) for 2 h. The water-free product became solid on
cooling to room temperature (yield 4.02 g, 77%). 31P{1H} NMR: 32.5
ppm (s), 41.5 ppm (d), -11.4 ppm (d), J ) 45.5 Hz.
Preparation of Ph2P(CH2)3PPh(CH2)2PPh2, eptp (IV). The dry
product III (4.21 g, 0.0073 mol) was dissolved in acetonitrile (65 mL)
in a conical flask equipped with a magnetic stirrer bar and a long air
condenser. Triethylamine (5.87 g, 0.0584 mol) was added to the
solution with stirring. Trichlorosilane (7.86 g, 0.0584 mol) was added
dropwise with care to the stirring solution producing white fumes. The
white mixture in the flask was heated at 76 °C for 30 min. After this
time the now yellow mixture was heated to gentle reflux (82 °C), for
24 h and then cooled to room temperature. The mixture was poured
slowly, in portions, into sodium hydroxide solution (12.50 g of NaOH
in 50 mL of H2O) and stirred for 30 min, resulting in two clear phases
with the product dissolved in the organic layer. The organic layer was
evaporated to dryness, yielding the product as a clear brown oil. No
phosphine oxide or other impurities were detected by 31P and 1H NMR
spectroscopy (yield 3.37 g, 84%). 31P{1H} NMR: -12.6 ppm (d),
-16.7 ppm (s), -20.9 ppm (d), J ) 29.8 Hz. The oil was recrystallized
by dissolving in hot methanol, and on cooling to -20 °C the product
was precipitated as a white solid. This was separated and dried in a
vacuum; on warming to room temperature, the white solid became a
clear brown oil.
Table 3. Elemental Analyses of Nickel and Palladium Tridentate
Phosphine Complexes
found (calculated)
complex
[NiI(eptp)]I
NiI(eptp)]ClO4
NiBr(eptp)]ClO4
[PdBr(eptp)]PF6
[PdBr(eptp)]ClO4
[NiI(eptp-Cl2)]I
[NiI(eptp-F2)]I
[NiBr(eptp-Cl2)ClO4
[NiI(eptp-Cl2)]ClO4
mp (°C)
% C
% H
% P
276
242
242
191
218
212
267
168
215
48.8 (48.8) 4.0 (4.1) 10.3 (10.7)
50.0 (50.4) 4.3 (4.2) 10.7 (11.1)
53.5 (53.4) 4.4 (4.4) 11.6 (11.8)
47.6 (47.8) 4.2 (4.0) 12.8 (14.1)
47.4 (50.4) 4.0 (4.2) 10.1 (11.1)
46.5 (45.2) 4.1 (3.6) 10.1 (10.0)
44.9 (46.8) 3.7 (3.7) 10.2 (10.4)
50.1 (49.1) 4.6 (3.9) 10.8
46.2 (46.6) 4.2 (3.7)
9.7 (10.3)
ethanol (30 mL). On addition of each drop of ligand solution, a dark
brown powder precipitated. The dichloromethane was removed and
the product collected on a sinter. The product was recrystallized from
dichloromethane and ethanol giving a black crystalline solid (mp 276
°C; yield 2.72 g, 93%). 31P{1H} NMR: 60.6 ppm (d of d), 53.3 ppm
(d of d), 0.2 ppm (d of d), J ) 72.5, 20.9, 281.8 Hz.
Preparation of [NiI(eptp)]ClO4.11 NiI2 (500 mg, 1.6 mmol)
dissolved in ethanol (16 mL) was mixed with Ni(ClO4)2 (590 mg, 1.6
mmol) dissolved in ethanol (10 mL), resulting in a pale green solution.
Recrystallized eptp (IV) (194 mg, 3.5 mmol) dissolved in dichloro-
methane (5 mL) was added dropwise to the nickel(II) iodide perchlorate
solution, resulting in the precipitation of a deep brown colored powder
in a brown/red solution. The dichloromethane was removed and the
product collected on a sinter. The product was recrystallized from
dichloromethane and ethanol (mp 242 °C; yield 1.25 g, 52%). 31P{1H}
NMR: 1.7 ppm (d of d), 54.9 ppm (d of d), 60.9 ppm (d of d), J )
80.9, 28.8, 276.9 Hz.
Preparation of [NiBr(eptp)]ClO4.11 NiBr2 (452 mg, 1.7 mmol)
was dissolved in warm ethanol (10 mL) to give a deep green solution.
Ni(ClO4)2 (610 mg, 1.7 mmol) dissolved in warm ethanol (10 mL)
was added to the NiBr2 solution resulting in a pale green solution when
cool. Ligand eptp (1.82 g, 3.3 mmol), dissolved in dichloromethane
(5 mL), was added to the NiBrClO4 solution. A brown/orange
precipitate was formed immediately. The dichloromethane was re-
moved and the product collected on a sinter. The product was
recrystallized from dichloromethane and ethanol (mp 242 °C; yield 1.65
g, 60%). 31P{1H} NMR: 3.1 ppm (d of d), 56.9 ppm (d of d), 60.0
ppm (d of d), J ) 80.0, 47.7, 276.6 Hz.
Preparation of [PdBr(eptp)]PF6. PdCl2 (301 mg, 1.7 mmol) was
made into a suspension in warm ethanol (40 mL, 30 °C). LiBr (590
mg, 6.8 mmol), dissolved in ethanol (10 mL), was added dropwise to
the palladium chloride suspension which became more orange/red in
color on addition of each drop. The solution was shaken for 10 min
until no further color change took place. Ammonium hexafluorophos-
phate (277 mg, 1.7 mmol), dissolved in ethanol (5 mL), was added to
the red palladium solution with no visible change. The solution was
now decanted off any excess palladium chloride. Ligand eptp (940
mg, 1.7 mmol), dissolved in dichloromethane (5 mL), was added
dropwise to the red solution precipitating a pale brown precipitate in a
clear yellow solution. The dichloromethane was removed and the pale
brown oily precipitate recrystallized from dichloromethane and ethanol.
The product was precipitated as a yellow powder (mp 191 °C; yield
0.81 g, 54%). 31P{1H} NMR: 0.9 ppm (d of d), 53.9 ppm (d), 58.2
ppm (d), -145 ppm (sep), J ) 38.5, 413.4 Hz.
Preparation of (2-(Diphenylphosphino)ethyl)(3-(dichlorophos-
phino)propyl)phenylphosphine. Tri(p-chlorophenyl)phosphine was
substituted into the preparation used for I in place of triphenylphosphine.
The product after the silane reduction (IV) was stored under nitrogen,
as a clear brown oil (yield 2.11 g, 47%). 31P{1H} NMR: -13.4 ppm
(d), -18.9 ppm (s), -22.0 ppm (d), J ) 29.4 Hz.
Preparation of (2-(Diphenylphosphino)ethyl)(3-(difluorophos-
phino)propyl)phenylphosphine. Tri(p-fluorophenyl)phosphine was
substituted into the preparation used for I in place of triphenylphosphine.
The product was isolated as a clear brown oil (yield 3.50 g, 82%).
31P{1H} NMR: -11.6 ppm (d), -18.7 ppm (s), -21.2 ppm (d), J )
28.0 Hz.
Preparation of [PdBr(eptp)]ClO4.11 Lithium perchlorate (180 mg,
1.7 mmol), in ethanol (5 mL), was used in the above reaction in place
of ammonium hexafluorophosphate. The product was precipitated as
a yellow powder in ethanol and was recrystallized from dichloromethane
and ethanol (mp 218 °C; yield 0.92 g, 65%). 31P{1H} NMR: 1.1 ppm
(d of d), 54.8 ppm (d), 58.5 ppm (d), J ) 38.6, 414.0 Hz. The
complexes with the fluoro- and chloro-substituted ligands were prepared
in the same way, and data are tabulated in Tables 1 and 3.
X-ray Structure of [NiI(eptp)]I.10 Crystals of [Ni(eptp)]I were
grown, and a black plate crystal of C35H35NiI2P3 having approximate
Preparation of Tridentate Phosphine Complexes. Preparation
of [NiI(eptp)]I. The tridentate ligand eptp (IV) (1.84 g, 0.0034 mol)
was recrystallized from ethanol and dissolved in dichloromethane (5
mL) and added dropwise to a solution of NiI2 (1.04 g, 0.0034 mol) in
(11) Perchlorate salts of metal complexes with organic ligands are
potentially explosive and should be handled with great caution. [J.
Chem. Educ. 1973, 50, A335.]

4102 Inorganic Chemistry, Vol. 35, No. 14, 1996
Dyer and Roscoe
dimensions of 0.400 by 0.400 by 0.050 mm was mounted on a glass
fiber. All measurements were recorded on a Rigaku AFC6S diffrac-
tometer with graphite-monochromated Mo KR radiation and a 12 kW
rotating anode generator.
The data were collected at a temperature of 23 ( 1 °C using the ω
scan technique to a maximum value 2θ value of 50.0°. A total of 6745
reflections were collected. The intensities of six representative reflec-
tions which were measured after every 150 reflections declined by
2.02%. A linear correction factor was applied to the data to account
for this phenomenon. The linear absorption coefficient for Mo KR is
24.9 cm-1. An empirical absorption correction, based on azimuthal
scans of several reflections, was applied which resulted in transmission
factors ranging from 0.42 to 1.00. The data were corrected for Lorentz
and polarization effects.
Supporting Information Available: Crystallographic data for (2-
(diphenylphosphino)ethyl)(3-(diphenylphosphino)propyl)-
phenylphosphine, including text giving data collection and experimental
details, tables of crystallographic and refinement data, positional
parameters, intramolecular bond angles for non-hydrogen atoms, and
conformation angles (21 pages). Ordering information is given on any
current masthead page.
IC960136R