2-Pyridyl-phosphine and -diphosphine complexes of nickel(0), their reactivity (including aqueous solution chemistry), and some related, incidental methylphosphonium iodides

Article abstract of DOI:10.1016/j.ica.2015.04.003

The chemistry of Ni⁰-dicarbonyl(pyridylphosphine) complexes of the type Ni(CO)₂L₂, where L is either P-bonded PPh_3-npy_n (n = 1-3, py = 2-pyridyl; abbreviated PN_x, x = 1-3, species 1a-c), or L₂ is (P-P)-chelated py₂P(CH₂)₂Ppy₂ or, is further developed from earlier studies by our group [the P-P ligands are abbreviated, respectively, as d(py)pe and d(py)pcp]. The complexes are synthesized from C₆H₆ solutions of Ni(CO)₂(PPh₃)₂, and the Ni(CO)₂(PPh₃)(PN_x) intermediates (1a-c) are detected; Ni(CO)₂[d(py)pcp] (2b) is shown by X-ray analysis to have a distorted tetrahedral structure; and the Ni^II species [Ni₂(CO)₄(μ-PN₂)₂]Cl₄ is isolated from a light-induced reaction in CDCl₃ solution. Complex 2b dissolves in water at ambient conditions via a net double protonation of pyridyl N-atoms, the {Ni(CO)₂[2H-d(py)pcp]}²⁺ being isolated as the bis(triflate) salt; the dication decomposes in minutes with formation of [Ni(H₂O)₆]²⁺, CO, the phosphine dioxide, and deprotonated d(py)pcp. Some twenty-two Ni⁰ complexes, exemplified by Ni(P-P)₂, Ni(PN_x)₂(P-P), Ni(PN_x)₄, and related PPh₃- and Ph₂P(CH₂)₂PPh₂ (dppe)-containing species, are synthesized from Ni(1,5-COD)₂ and their reactivity studied; for example, oxidative addition of MeI generates trans-Ni(Me)(I)(PN₃)₂ and trans-Ni(Me)(I)(P-P)₂ but, with non-pyridyl containing reactants such as Ni(PPh₃)₄ and Ni(dppe)₂, only (monomethyl)phosphonium iodides are formed. Such iodides, and the bis(methyl) analogues [(CH₃)₂(diphosphine)]I₂, are then studied for clarification of some observed Ni chemistry. The NMR trends (a)-(d) are noted within the series of Ni⁰ complexes, and are rationalized: (a) the ²J_PP values in 1a-c, and the separation between the two doublets, parallel the number of N-atoms present; (b) the ³¹P{¹H} signals in the Ni(PN_x)₄ and Ni(PN_x)₂(P-P) complexes shift downfield in the order PN₁ < PN₂ < PN₃ within linear dependences; (c) the ²J_PP values for the Ni(PR₃)₂(P-P) complexes (R = Ph and PN_x) decrease in the order R = Ph > PN₁ > PN₂ > PN₃; (d) and the separation between the two ³¹P{¹H} triplets of the Ni(PN_x)₂(P-P) complexes generally depends on the relative numbers of phenyl and pyridyl groups.

Full text of DOI:10.1016/j.ica.2015.04.003

Inorganica Chimica Acta 431 (2015) 276–288
Contents lists available at ScienceDirect
Inorganica Chimica Acta
journal homepage: www.elsevier.com/locate/ica
2-Pyridyl-phosphine and -diphosphine complexes of nickel(0), their
reactivity (including aqueous solution chemistry), and some related,
incidental methylphosphonium iodides
1
⇑
Matthew D. Le Page, Brian O. Patrick, Steven J. Rettig , Brian R. James
Department of Chemistry, University of British Columbia, Vancouver, BC V6T 1Z1, Canada
a r t i c l e i n f o
a b s t r a c t
The chemistry of Ni0-dicarbonyl(pyridylphosphine) complexes of the type Ni(CO)2L2, where
either P-bonded PPh3ꢀnpyn (n = 1–3, py = 2-pyridyl; abbreviated PNx, x = 1–3, species 1a–c), or
L2 is (P–P)-chelated py2P(CH2)2Ppy2 or , is further developed from earlier studies
L is
Article history:
Received 27 January 2015
Received in revised form 6 April 2015
Accepted 7 April 2015
by our group [the P–P ligands are abbreviated, respectively, as d(py)pe and d(py)pcp]. The complexes are
synthesized from C6H6 solutions of Ni(CO)2(PPh3)2, and the Ni(CO)2(PPh3)(PNx) intermediates (1a⁄–c⁄) are
detected; Ni(CO)2[d(py)pcp] (2b) is shown by X-ray analysis to have a distorted tetrahedral structure;
Available online 15 April 2015
Dedicated to the memory of Professor
Fausto Calderazzo, a close friend of BRJ.
and the NiII species [Ni2(CO)4(
l-PN2)2]Cl4 is isolated from a light-induced reaction in CDCl3 solution.
Complex 2b dissolves in water at ambient conditions via a net double protonation of pyridyl N-atoms,
the {Ni(CO)2[2H-d(py)pcp]}2+ being isolated as the bis(triflate) salt; the dication decomposes in minutes
with formation of [Ni(H2O)6]2+, CO, the phosphine dioxide, and deprotonated d(py)pcp. Some twenty-two
Ni0 complexes, exemplified by Ni(P–P)2, Ni(PNx)2(P–P), Ni(PNx)4, and related PPh3- and Ph2P(CH2)2PPh2
(dppe)-containing species, are synthesized from Ni(1,5-COD)2 and their reactivity studied; for example,
oxidative addition of MeI generates trans-Ni(Me)(I)(PN3)2 and trans-Ni(Me)(I)(P–P)2 but, with non-
pyridyl containing reactants such as Ni(PPh3)4 and Ni(dppe)2, only (monomethyl)phosphonium iodides
are formed. Such iodides, and the bis(methyl) analogues [(CH3)2(diphosphine)]I2, are then studied for
clarification of some observed Ni chemistry. The NMR trends (a)–(d) are noted within the series of Ni0
Keywords:
Nickel(0) carbonyl complexes
Nickel(0)-2-pyridylphosphine complexes
Water-solubility of Ni(0) complexes
Methyl phosphonium iodides
2
complexes, and are rationalized: (a) the JPP values in 1a⁄–c⁄, and the separation between the two
doublets, parallel the number of N-atoms present; (b) the 31P{1H} signals in the Ni(PNx)4 and
Ni(PNx)2(P–P) complexes shift downfield in the order PN1 < PN2 < PN3 within linear dependences;
2
(c) the JPP values for the Ni(PR3)2(P–P) complexes (R = Ph and PNx) decrease in the order
R = Ph > PN1 > PN2 > PN3; (d) and the separation between the two 31P{1H} triplets of the Ni(PNx)2(P–P)
complexes generally depends on the relative numbers of phenyl and pyridyl groups.
Ó 2015 Elsevier B.V. All rights reserved.
1. Introduction
such pyridyl-substituted phosphines, which increases with the
number of pyridyl groups and where protonation of the pyridyl-
We reported recently on the synthesis, characterization,
and aqueous solution chemistry of NiII complexes containing
2-pyridyl-phosphine and -diphosphine ligands [1]. The specific
phosphines were PPh3ꢀnpyn, py2P(CH2)2Ppy2 [abbreviated
nitrogen plausibly plays a role [2]. Qualitative observations on
the dissolution in water of such NiII complexes with ancillary
halogen ligands were first made by our group 20 years ago [3],
and the recent paper described details of the aqueous solution
chemistry of such complexes where the halogen ligands are
substituted by aquo ligands to generate dicationic NiII derivatives
[1]. The early report [3] also discussed aspects of the Ni0 com-
plexes: Ni(PPh3-npyn)4, Ni[d(py)pe]2, and the carbonyl species
Ni(CO)2(PPh2py)2, Ni(CO)2(PPhpy2), and Ni(CO)2[d(py)pe].
d(py)pe], and
[abbreviated d(py)pcp] that
contains a cyclopentane ring bridge; py = 2-pyridyl, and n = 1–3.
These studies stemmed from interest in homogeneous catalytic
systems in aqueous solution because of the water-solubility of
This current paper examines in detail the syntheses, character-
ization, reactivity (including protonation in aqueous solution) of
Ni0 species containing 2-pyridyl-phosphine and -diphosphine
ligands, including the five complexes listed above; oxidation
⇑
Corresponding author.
E-mail address: brj@chem.ubc.ca (B.R. James).
Deceased October 27, 1998.
1
http://dx.doi.org/10.1016/j.ica.2015.04.003
0020-1693/Ó 2015 Elsevier B.V. All rights reserved.

M.D. Le Page et al. / Inorganica Chimica Acta 431 (2015) 276–288
277
addition reactivity of some of these Ni0 complexes toward CH3I led
us to investigate some methylphosphonium salts.
Ni-(PN3)2), AB doublets for Ni(CO)2(PPh3)(PN3) seen at d 35.7 and
43.1 (2JPP ꢂ115). IR (CH2Cl2): 1943, 2004 (s, CO).
2.2.4. [Ni2(CO)4(
Heating Ni(CO)2(PPh3)2 (4.2 mg, 6.57
13.2 mol) in CDCl3 (1 mL) at 60 °C for 90 min gave a yellow solu-
l-PN2)2]Cl4 (1d)
2. Experimental
l
mol) and PN2 (3.51 mg,
l
2.1. General
tion and a purple precipitate. The purple product was isolated by
column chromatography (300-mesh silica gel, Fisher) using ace-
tone to elute PPh3, PN2, and Ni(CO)2(PPh3)2, and then a CH2Cl2/
MeOH (20:1) mixture to elute the purple band. Yield: 1.12 mg
(38%). Anal. Calc. for C36H26Cl4N4O4P2Ni2ꢁ½CH2Cl2: C, 46.53; H,
2.89; N, 5.95. Found: C, 46.9; H, 2.9; N, 5.5%. 1H NMR (CD3OD): d
7.13 (t, J ꢂ1.5), 7.15 (m), 7.20 (m), 7.25 (m), 7.34 (m), 7.45–7.52
(m), 7.56 (t, J ꢂ3.8), 7.60 (t, J ꢂ3.0), 7.80 (m), 8.20 (tt, J ꢂ10 and
ꢂ1.5), 8.79 (dt, H6, J ꢂ7.5 and ꢂ1.5), 9.22 (dt, H⁄6, J ꢂ7.5 and
ꢂ1.5). 31P{1H} NMR (CD3OD): d 24.1 (s). UV–Vis (MeOH): 572
[122]. IR (CH2Cl2): 1971 (s, CO). +LSIMS (m/z): 757 [M+]. KM: 429
(MeOH).
The sources of the general chemicals, solvents, gases, Ni
reagents [Ni(CO)2(PPh3)2, Ni(COD)2, NiCl2ꢁ6H2O], and instrumenta-
tion used in our recently published studies, as well as details of the
pyridylphosphines, have been described [1]; other reagents used in
this current work are the Mattheson products Cl2, C2H4 (that was
passed through an anhydrous P2O5 column prior to use), and
CH3I (an Aldrich product). The 2-pyridylphosphines PPh3ꢀnpyn
are designated as PN1, PN2, and PN3, where the subscript refers
to the number of pyridyl groups; the commercially used
Ph2P(CH2)2PPh2 is abbreviated dppe. Unless indicated otherwise,
syntheses and investigations were carried out at room temperature
(r.t., ꢂ20 °C) in an N2-flushed Schlenk tube, or in an N2-flushed,
J-Young NMR tube for in situ syntheses. NMR spectra were
acquired at room temperature on Varian XL300 and Bruker
AC200 spectrometers. 31P{1H} data are reported with respect to
85% aq. H3PO4; s = singlet, d = doublet, t = triplet, qn = quintet,
m = multiplet, br = broad, ps = pseudo, with J values given in Hz.
Chart 1 shows the proton labelling used for the pyridyl-phosphines
and -diphosphines, and monomethylated diphosphines. UV–Vis,
IR, and conductivity data, are given, respectively, as kmax in nm
2.2.5. Ni(CO)2[d(py)pe] (2a)
Ni(CO)2(PPh3)2 (12.1 mg, 18.9 lmol) was dissolved in C6H6
(5 mL) and the mixture was stirred for 10 min at 85 °C, and then fil-
tered to remove trace material; to the pale-yellow filtrate, d(py)pe
(7.81 mg, 19.4 lmol) was added, and the mixture was stirred for
2 h at 85 °C. The resulting bright-yellow solution was filtered,
reduced in volume to ꢂ1 mL; addition of hexanes (ꢂ1 mL) precip-
itated a pale-yellow product that was collected, washed with cold
Et2O (2 ꢃ 1 mL), and dried in vacuo. Yield: 3.52 mg (36%). Anal.
Calc. for C24H20N4NiO2P2: C, 55.77; H, 3.90; N, 10.83. Found: C,
55.7; H, 3.9; N, 10.5%. 1H NMR (C6D6): d 2.8 (m, 4Ha), 7.1–8.6 (m,
16H, pyridyl). 31P{1H} NMR (C6D6): d 55.3 (s). UV–Vis (CH2Cl2):
240 [3270] and 268 [3430], 322 [4010], 334 [3875], 352 [3360].
IR (thin CH2Cl2 film on KBr): 1999, 1937 (s, CO); 1569, 1450,
1420 (s, py skeletal bands). m.pt.: ꢂ185 °C (dec.). KM: 0 (MeOH),
3.6 (H2O). A crystal was grown by diffusion of hexanes into an
EtOH solution of the complex, but it was not of X-ray quality.
[e
, Mꢀ1 cmꢀ1], cmꢀ1 (s = strong, m = medium), and KM (ohmꢀ1
cm2 molꢀ1). For IR data of in situ species, samples formed in
CH2Cl2 were placed as a film between KBr discs, the data being
noted as ‘IR (CH2Cl2)’. Gas chromatographic analyses for CO and
H2 were performed using a 10-foot molecular sieve column in an
HP 5890A instrument equipped with a thermal conductivity
detector and an HP 3392A integrator, the carrier gas being He.
2.2. Ni(0) carbonyl 2-pyridylphosphine complexes
2.2.6. Ni(CO)2[d(py)pcp] (2b)
This procedure was as for 2a, but using Ni(CO)2(PPh3)2 (185 mg,
290 lmol) in C6H6 (40 mL), and d(py)pcp (118 mg, 267 lmol).
NMR spectra for 1a–c and 2a–b were measured at 200 MHz;
those for 1d and protonated 2b were done at 300 MHz.
After being filtered, the bright-yellow solution was reduced in vol-
ume to ꢂ5 mL, and then 50 mL hexanes were then added to give
the pale-yellow product that was washed with Et2O (4 ꢃ 5 mL),
prior to being dried in vacuo. Yield: 58.9 mg (40%). Anal. Calc. for
2.2.1. Ni(CO)2(PN1)2 (1a) in situ
C6D6 (1 mL) was added to Ni(CO)2(PPh3)2 (4.2 mg, 6.57
and two equivalents of PN1 (3.47 mg, 13.2 mol). A yellow solution
lmol)
l
was obtained on heating the mixture for 3 h at 70 °C. In situ 31P{1H}
NMR: d ꢀ3.4 (s, free PPh3), ꢀ3.5 (s, free PN1), 32.6 (s, Ni-(PPh3)2),
31.3 (s, Ni-(PN1)2); AB signal (dd) for Ni(CO)2(PPh3)(PN1) seen at
d 35.8 and 37.4 (2JPP ꢂ37). IR (CH2Cl2): 1941, 2001 (s, CO) [litera-
ture data (KBr): 1946, 1998 [3]]. NMR data have not been reported
previously.
C27H24N4NiO2P2: C, 58.21; H, 4.34; N, 10.06. Found: C, 58.2; H,
4.4; N, 9.8%. 1H NMR (C6D6): d 1.59 (m, 2Ha), 2.15 (dm, 4Hb),
3.74 (m, 2Hc), 6.41–8.55 (m, 16H, pyridyl). 31P{1H} NMR (C6D6): d
35.0 (s). UV–Vis (CH2Cl2): 236 [2365], 268 [2520], 322 [2910],
344 [2740], 364 [2600]. IR (CH2Cl2): 1997, 1927 (s, CO); 1569,
1450, 1420 (s, py skeletal bands). EIMS (m/z): 500 [M+ꢀ2CO].
m.pt.: ꢂ205 °C (dec.). KM: 0.0 (MeOH), 4.4 (H2O). An X-ray quality
crystal was grown by diffusion of hexanes into a CH2Cl2 solution of
the complex.
2.2.2. Ni(CO)2(PN2)2 (1b) in situ
The above procedure was used with Ni(CO)2(PPh3)2 (4.7 mg,
7.35 lmol) and PN2 (3.90 mg, 14.8 l
mol). In situ (C6D6): 31P{1H}
2.2.7. Protonated 2b, {Ni(CO)2[2H-d(py)pcp]}(SO3CF3)2,
[2b⁄2H](SO3CF3)2
NMR: d ꢀ3.4 (s, free PPh3), ꢀ1.9 (s, free PN2), 32.6 (s, Ni-(PPh3)2),
33.1 (s, Ni-(PN2)2); AB doublets at d 36.2 and 39.7 (2JPP ꢂ82) are
attributed to Ni(CO)2(PPh3)(PN2). IR (CH2Cl2): 1942, 2003 (s, CO).
In situ 31P{1H} NMR (CDCl3) literature data [3]: d Ni-(PN2)2 = 31.2,
and AB signals at d 32.9 and 37.5. IR and 2JPP data were not reported
previously.
Addition of 2 equivalents (22
lL) of CF3SO3H (0.113 M solution
in C6H6) to a solution of Ni(CO)2[d(py)pcp] (1.4 mg, 2.51
l
mol) in
C6H6 (2 mL) resulted in precipitation within 5 min of a bright-
yellow solid that was collected and dried in vacuo. Yield:
ꢂ1.6 mg (ꢂ75%). Anal. Calc. for C29H26F6N4O8P2S2Ni: C, 40.63; H,
3.06; N, 6.54. Found: C, 40.9; H, 3.3; N, 6.4%. 1H NMR (acetone-
d6): d 1.39 (m, 2Ha), 2.20 (dm, 4Hb), 3.48 (m, 2Hc), 8.08–9.36 (m,
16H, pyridyl), 10.5 (s, 2H, N–H). 31P{1H} NMR (acetone-d6): d
25.5 (s). IR (KBr): 3095 (s, N–H); 2045, 1996 (s, CO); 1522, 1447,
1420 (s, py skeletal bands); 1170, 765 [d(py)pcp]; 1623, 1400,
2.2.3. Ni(CO)2(PN3)2 (1c) in situ
The procedure was as for 1a, but using Ni(CO)2(PPh3)2 (4.6 mg,
7.20
d ꢀ3.4 (s, free PPh3), ꢀ0.1 (s, free PN3), 32.7 (s, (Ni-PPh3)2), 34.1 (s,
lmol), PN3 (3.83 mg, 14.4 l
mol). In situ (C6D6): 31P{1H} NMR:

278
M.D. Le Page et al. / Inorganica Chimica Acta 431 (2015) 276–288
9
I-
8
7
8
7
H
c
H
c
H
H
b
H
H
b
b
b
(a)
(b)
(c)
P
3
*
*
CH3
PR2
N
H
H
a
a
H
H
a2
a1
R2P
py2P
Ppy2
6
4
3
N
5
5
4
6
Chart 1. Numbering scheme for pyridyl-phosphines and -diphosphines: (a) e.g., PN2; (b) d(py)pe and d(py)pcp; (c) monomethylated phosphonium iodides.
1277, 1256, 1031, 641 (s, CF3SOꢀ3 ). Syntheses using 4 or 6 equiva-
lents of d(py)pcp gave similar yields of the same product.
which addition of hexanes (20 mL) gave a dark-red precipitate that
was collected, washed with hexanes (4 ꢃ 5 mL), and dried in vacuo.
Yield: 84.0 mg (42%). Anal. Calc. for C50H48N8NiP4: C, 63.65; H,
5.13; N, 11.88. Found: C, 63.5; H, 5.0; N, 11.6%. 1H NMR: d 1.72
(m, 4Ha), 2.13 (dm, 8Hb), 3.89 (m, 4Hc), 6.32–8.51 (m, 32H, pyri-
dyl). 31P{1H} NMR: d 38.7 (s). UV–Vis (C6H6): 409 [2325]. EIMS
(m/z): 942–944 [M+]. m.pt.: 112–115 °C (dec.).
2.3. Protonated [d(py)pcp], [2H-d(py)pcp](SO3CF3)2
To a solution of d(py)pcp (6.8 mg, 16.9
was added Et2O (3 mL) followed by 2 equivalents (3
l
mol) in EtOH (0.5 mL)
L) of neat
l
CF3SO3H; a finely divided white precipitate was formed. (Of note,
addition of CF3SO3H prior to that of Et2O generated a pale-yellow
oil.) The solid was collected and dried in vacuo. Yield: 8 mg
(70%). Anal. Calc. for C27H26F6N4O6P2S2: C, 43.67; H, 3.53; N, 7.55.
Found: C, 43.5; H, 3.4; N, 7.2%. 1H NMR (300 MHz, acetone-d6): d
1.44 (m, 2Ha), 2.27 (dm, 4Hb), 3.49 (m, 2Hc), 7.98–9.27 (m, 16H,
pyridyl), 10.0 (s, 2H, N–H). 31P{1H} NMR (300 MHz, acetone-d6): d
ꢀ13.6 (s). IR (KBr): 2690 (s, N–H); 1524, 1447, 1420 (s, py skeletal
bands); 1172, 762 [d(py)pcp]; 1623, 1399, 1277, 1259, 1031, 638
(s, CF3SOꢀ3 ). EIMS (m/z): 402 [d(py)pcp+], 149 [CF3SOꢀ3 ]. Syntheses
using 4 or 6 equivalents of d(py)pcp gave similar yields of the same
product.
2.4.5. Ni[d(py)pe]2 (6b)
A sample made previously by our group [3] was available. The
31P{1H} NMR singlet in C6D6 (d 55.90) was the same as the reported
value [3]; the UV–Vis spectrum (389 [1970]) was not recorded
previously.
2.4.6. Ni(dppe)2 (6c)
Ni(COD)2 (4.37 mg, 15.9 lmol) and dppe (15.4 mg, 38.6 lmol)
were added to C6D6 (1 mL); 2 min heating at 60 °C resulted in a
yellow solution. 31P{1H} NMR: d 44.0 (s) (literature value, 42.4 in
toluene-d8 [4]).
2.4.7. Ni(P–P)(P0–P0) [P–P = dppe, P0–P0 = d[py]pe (7a); P–P = dppe,
P0–P0 = d[py]pcp (7b); P–P = d[py]pe, P0–P0 = d[py]pcp (7c)]
Use of the ‘3 procedure’ for these systems generated orange
solutions.
2.4. Ni(0) phosphine complexes
Complexes 3–5, 6b–c and 7 were prepared in situ; 6a and 8
were isolated. The quantities of reagent used in the in situ synthe-
ses of 3–5 are listed in Table S1; those for 7 and 8 are given in
Tables S2 and S3, respectively.
2.4.8. Ni(PR3)4 [PR3 = PPh3 (8a); PN1 (8b); PN2 (8c); PN3 (8d)]
These known complexes [3,5] were prepared by the published
procedure from Ni(COD)2 and 4 equivalents of PR3 in chilled
hexanes solution; 8a and c are orange, 8b is red, and 8d is yellow.
NMR spectra for 3–8 were measured at 200 MHz. 31P{1H} NMR
measurements were carried out in C6D6 solutions, except for the
isolated 6a (in CDCl3); data are given in Table 1 for 3–5 and 7,
and Table 2 for 6 and 8.
2.5. Reactions of CH3I with Ni(0) pyridylphosphine complexes
NMR data for the 9a–e systems were measured at 300 MHz.
2.5.1. trans-Ni(Me)(I)(PN3)2 (9a)
2.4.1. Ni(PR3)2(dppe): PR3 = PPh3, (3a); PN1, (3b); PN2, (3c); PN3, (3d)
One equivalent of dppe, and two equivalents of PR3 were added
to Ni(COD)2 in C6D6 (1 mL); heating to 60 °C resulted in orange (3a
and b), dark-orange (3c), or red (3d) solutions.
Ni(PN3)4 (25.2 mg, 22.5
to give a dark-yellow solution. One equivalent of CH3I (1.4
22.5 mol) was added and the solution warmed to r.t. with stir-
l
mol) was added to C6H6 (6 mL) at 0 °C
2.4.2. Ni(PR3)2[d(py)pe)]: PR3 = PPh3, (4a); PN1, (4b); PN2, (4c); PN3,
(4d)
The above procedure (for 3) was used, and in each case gave an
orange solution.
lL,
l
ring; further stirring for ꢂ20 min at 80 °C gave a red solution and
a tan precipitate that was collected by filtration, and dried in
vacuo. Yield: 6.9 mg (42%). Anal. Calc. for C31H27IN4NiP2: C,
50.93; H, 3.72; N, 11.49. Found: C, 50.8; H, 3.7; N, 11.2%. 1H NMR
(CDCl3): d 0.6 (t, 3H, 3JPH 6.2, Ni-CH3), 7.45–9.21 (m, 24H, pyridyl).
31P{1H} NMR (CDCl3): d 31.5 (s). Hexanes (10 mL) was added to the
filtrate to yield a reddish-brown solid that was shown by 31P{1H}
NMR spectroscopy to be a mixture of ꢂ2 equivalents of PN3 and
a trace amount of Ni(PN3)4.
2.4.3. Ni(PR3)2[d(py)pcp]: PR3 = PPh3, (5a); PN1, (5b); PN2, (5c); PN3,
(5d)
Use of the ‘3 procedure’ for these systems generated red
solutions.
2.4.4. Ni[d(py)pcp]2 (6a)
This compound was prepared following a modification of the
literature method for Ni[d(py)pe)]2 [3]. Ni(COD)2 (29.3 mg,
2.5.2. trans-Ni(Me)(I)[d(py)pe]2 (9b)
107
lmol) and d(py)pcp (93.4 mg, 211
lmol) were mixed in C6H6
Use of the ‘9a procedure’, but with Ni[d(py)pe]2 (68.7 mg,
(10 mL), this giving a rapid color change from bright-yellow to
80.3
lmol) and CH3I (5.0 lL, 80.3 lmol), gave an orange solid.
orange. Stirring at 60 °C for 3 h generated a dark red solution, to
Yield: 29.9 mg (37%). Anal. Calc. for C45H43IN8NiP4: C, 53.76; H,

M.D. Le Page et al. / Inorganica Chimica Acta 431 (2015) 276–288
279
Table 1
Selected 31P{1H} NMR data for complexes 3–5 and 7 in C6D6.
Complex
Ligand:
Column A
Column B
Column C
Column D
Column E
Column F
L dP in species:
L0 dP in species:
a
b
b
a
c
d
L
L0
NiL4
NiL2L0
NiL2L0
NiL02
2JPP
Dt-t
3a
4a
5a
PPh3
dppe
d(py)pe
d(py)pcp
24.4
24.4
24.4
31.6
35.2
25.6
32.2
39.6
34.3
44.0
55.9
38.7
19.9
26.6
27.7
76
540
1059
3b
4b
5b
PN1
PN2
PN3
dppe
d(py)pe
d(py)pcp
16.0
16.0
16.0
33.7
38.2
29.3
35.1
47.5
35.6
44.0
55.9
38.7
23.8
24.5
25.9
170
1133
767
3c
4c
5c
dppe
d(py)pe
d(py)pcp
24.9
24.9
24.9
36.2
42.7
32.8
39.7
48.4
43.6
44.0
55.9
38.7
22.8
23.3
25.7
427
696
1306
3d
4d
5d
dppe
d(py)pe
d(py)pcp
28.9
28.9
28.9
38.3
47.5
34.0
44.9
49.4
49.9
44.0
55.9
38.7
22.2
21.9
25.8
806
221
1929
e
f
g
h
7a
7b
7c
dppe
dppe
d(py)pe
d(py)pe
d(py)pcp
d(py)pcp
44.0
44.0
55.9
45.2
45.1
56.1
53.0
37.0
38.9
55.9
38.7
38.7
27.5
23.2
10.3
955
985
2092
e
f
g
h
e
f
g
h
a
b
c
Singlet.
Triplet.
In Hz.
d
e
f
Separation in Hz between the centre of the two triplets.
dP refers to singlet of NiL2 species.
dP refers to L triplet of Ni(L)(L0) species.
dP refers to L0 triplet of Ni(L)(L0) species.
dP refers to singlet of Ni(L0)2 species.
g
h
Table 2
31P{1H} NMR singlets for homoleptic complexes 6 and 8.
3
assignments). 31P{1H} NMR (CDCl3): d 27.4 (d, JPP 44.0), ꢀ11.1
a
(d, JPP 43.7). 31P{1H} NMR of the filtrate (CDCl3): d 44.0 (s,
3
Complex
dP
Complex
dP
3
Ni(dppe)2), 32.9 (s, dppe(O)2), 28.5 (d, JPP 48.6, dppe(O)), ꢀ12.2
b
c
e
f
3
Ni[d(py)pcp)]2 6a
Ni[d(py)pe]2 6b
Ni(dppe)2 6c
38.7
55.9
44.0
Ni(PPh3)4 8a
Ni(PN1)4 8b
Ni(PN2)4 8c
Ni(PN3)4 8d
24.4
16.0
24.9
28.9
(d, JPP 48.6, dppe(O)), ꢀ13.0 (s, free dppe). The NMR data for the
filtrate agreed well with known values for the di- and mono-oxides
[6]. After 4 months, the Schlenk tube contained a pale-pink precip-
itate and purple crystals; the latter were identified as NiI2(dppe)
[1,7], and the former was a mixture of white dppe(O)2 and traces
of NiI2(dppe).
d
g
h
a
b
c
In C6D6.
In CDCl3.
dP (C6D6) = 55.9, Ref. [3].
dP (toluene-d8) = 42.4, Ref. [4].
dP (toluene-d8) = 25.5, Ref. [38b].
The value of dP (toluene-d8) = 31.4 in Ref. [3] is incorrect.
dP (solid state) = 41.9, Ref. [3].
dP (C6D6) = 28.9, Ref. [3].
d
e
f
2.5.5. Attempted synthesis of trans-Ni(Me)(I)(PPh3)2 (9e)
g
h
The ‘9a procedure’, but with Ni(PPh3)4 (10.0 mg, 9.0 lmol) and
CH3I (0.6 lL, 9.6 lmol), gave an off-white solid identified by NMR
spectroscopy as [(CH3)(PPh3)]I (see 10e below). 1H NMR (CDCl3):
2
d 3.46 (d, 3H, JPH 12.9, P-CH3), 7.60–8.20 (m, 15H, phenyl).
31P{1H} NMR (CDCl3): d 22.2 (s, br).
4.31; N, 11.14. Found: C, 53.9; H, 4.5; N, 10.8%. 1H NMR (CDCl3):
3
d 0.9 (qn, 3H, JPH 7.1, Ni-CH3), 2.7 (t, 8Ha), 7.51–8.78 (m, 32H,
pyridyl). 31P{1H} NMR (CDCl3): d 29.0 (s). No d(py)pe was detected
2.6. Monomethylated phosphonium iodides
in the filtrate.
The in situ preparation of these salts in C6D6 was monitored
using 31P{1H} NMR, the data being labelled ‘soln’. The precipitate
which eventually formed was collected and re-analyzed by 1H
and 31P{1H} NMR in CDCl3, these data being labelled ‘ppt’; all
NMR data for 10a–e were recorded at 300 MHz. 10a and 10e were
the sole products in their in situ spectra, whereas traces of free
ligand or ligand oxides were also detected in spectra of 10b–d.
Proton labelling is shown in Chart 1.
2.5.3. trans-Ni(Me)(I)[ d(py)pcp]2 (9c)
Use of the ‘9a procedure’, but with Ni[d(py)pcp]2 (59.3 mg,
69.3 lmol) and CH3I (4.3 lL, 69.3 lmol), gave a yellow solid.
Yield: 34.6 mg (46%). Anal. Calc. for C51H51IN8NiP4: C, 56.43; H,
4.73; N, 10.32. Found: C, 56.6; H, 4.8; N, 10.1%. 1H NMR (CDCl3):
3
d 0.7 (qn, 3H, JPH 7.3, Ni-CH3), 7.22–8.60 (m, 32H, pyridyl), 1.79
(m, 4Ha), 1.91 (m, 4Hb), 2.34 (m, 4Hb), 3.90 (m, 4Hc). 31P{1H}
NMR (CDCl3): d 36.5 (s). No d(py)pcp was detected in the filtrate.
2.6.1. [CH3(PN3)]I (10a)
2.5.4. Attempted synthesis of trans-Ni(Me)(I)(dppe) (9d)
Addition of excess CH3I (15
lL, 169 lmol) to PN3 (23.6 mg,
The ‘9a procedure’, but with Ni(dppe)2 (110.7 mg, 129
l
mol)
89.0
l
mol) in C6D6 (2 mL) gave a tan precipitate. 1H NMR
2
and CH3I (8.4
to be [(CH3)(dppe)]I. Yield: 10.7 mg (19.7
l
L, 135
l
mol), gave an off-white solid that proved
(ppt): d 3.29 (d, 3H, JPH 15.0, P-CH3), 6.85 (t, 3H5), 7.30
(t, 3H4), 7.80 (d, 3H3), 8.82 (d, 3H6). 31P{1H} NMR (ppt): d 10.0
(s, br). 31P{1H} NMR (soln): d 10.3 (s, br). For [(CH3)(PN1)]I in
C6D6, dP = 17.5 [8].
l
mol). 1H NMR
2
(CDCl3): d 2.41 (t, 2Ha1), 3.02 (m, 2Ha2), 3.25 (d, 3H, JPH 15,
P-CH3), 7.15–8.10 (m, 20H, phenyl) (see Chart 1 for proton

280
M.D. Le Page et al. / Inorganica Chimica Acta 431 (2015) 276–288
2.6.2. [CH3(dppe)]I (10b)
m
(P-CH3)), 907 (s,
m
(P-CH3)), 2944 (m,
m
" (C–H)), 2867 (s,
m(C–H)),
The ‘10a method’, but with dppe (22.7 mg, 57.0
l
mol) and CH3I
1568, 1448, 1417 (s, py skeletal bands). EIMS (m/z): 559 [MꢀI+].
m.pt.: 232–234 °C. KM: 210.
(15
lL, 169
l
mol), gave a white precipitate. 1H NMR (ppt): d 2.41
2
(ps t, 2Ha1), 3.02 (m, 2Ha2), 3.27 (d, 3H, JPH 13.6, P-CH3),
7.25–8.15 (m, 20H, phenyl). 31P{1H} NMR (ppt): d 27.4 (d, JPP
2.7.3. {[(CH3)2[d(py)pcp]}I2 (11c)
3
3
3
44.0), ꢀ11.1 (d, JPP 43.7). 31P{1H} NMR (soln): d 26.1 (d, JPP
A solution of d(py)pcp (24.1 mg, 54.5
lmol) in CDCl3 (4 mL),
3
41.2), ꢀ12.4 (d, JPP 41.3).
stirred with 10 equivalents of CH3I (34 L, 0.55 mmol) for 6 h at
l
55 °C, generated a pale-yellow suspension that was collected and
dried in vacuo. Yield: 24.5 mg (62%). Anal. Calc. for C27H30N4P2I2:
C, 44.65; H, 4.16; N, 7.71. Found: C, 45.0; H, 4.3; N, 7.6%. 1H NMR
2.6.3. {CH3[d(py)pe]}I (10c)
The ‘10a method’, but with d(py)pe (13.4 mg, 33.3
lmol) and
2
CH3I (3.0 L, 48.2
l
l
mol), gave a tan precipitate. 1H NMR (ppt): d
(CD3OD): d 1.52 (ps qn, 2Ha), 2.48 (m, 4Hb), 3.40 (d, 6H, JPH 13.2,
2
2.57 (m, 2Ha1), 3.11 (m, 2Ha2), 3.26 (d, 3H, JPH 13.7, P-CH3), 7.30
P-CH3), 5.12 (m, 2Hc), 7.56 (m, 4H5), 8.01 (m, 4H3), 8.71 (m, 4H6),
(m, 4H5), 7.60 (m, 4H3), 8.15 (m, 4H4), 8.75 (m, 4H6). 31P{1H}
8.78 (m, 4H4). 31P{1H} NMR (CD3OD): d 23.4 (s). IR: 1311 (m,
m(P-
3
3
NMR (ppt): d 21.7 (d, JPP 38.5), ꢀ8.0 (d, JPP 38.6). 31P{1H} NMR
CH3)), 900 (s, m(P-CH3)), 2965 (m, m(C–H)), 2872 (s, m(C–H)),
3
3
(soln): d 21.7 (d, JPP 38.5), ꢀ7.9 (d, JPP 38.5).
1572, 1450, 1424 (s, py skeletal bands). EIMS (m/z): 599 [MꢀI+],
473 [Mꢀ2I+]. m.pt.: 221–224 °C. KM: 185.
2.6.4. {CH3)[d(py)pcp)]}I (10d)
The ‘10a method’, but with d(py)pcp (11.6 mg, 26.2
lmol) and
2.8. Attempted syntheses of Ni(II)-acyl complexes
CH3I (2.0 lL, 32.1 l
mol), gave a tan precipitate. 1H NMR (ppt): d
1.67 (m, Ha1), 2.00 (m, Ha2), 2.02 (m, 2Hb), 2.46 (m, 2Hb), 3.18 (d,
The Bruker AC200 NMR spectrometer was used for NMR spectra
data given in this Section.
2
3H, JPH 13.5, P-CH3), 4.22 (m, 2Hc), 7.31 (m, 4H5), 7.96 (m, 4H3),
3
8.44 (m, 4H4), 8.70 (m, 4H6). 31P{1H} NMR (ppt): d 23.6 (d, JPP
3
3
15.4), ꢀ0.9 (d, JPP 15.6). 31P{1H} NMR (soln): d 23.4 (d, JPP 15.3),
2.8.1. Attempted synthesis of Ni(COCH3)(I)(PN3)2 via carbonylation
Bubbling CO at 1 atm for 2 h through a solution of trans-
Ni(CH3)(I)(PN3)2 (9a) (30.8 mg, 42.2 lmol) in hexanes (6 mL) gen-
3
ꢀ1.1 (d, JPP 15.2).
2.6.5. [CH3(PPh3)]I (10e)
erated a tan precipitate that was collected by anaerobic filtration
and dried in vacuo. Yield: 34.4 mg. 1H NMR (CDCl3): d 2.45 (s,
CH3I), 7.4–8.7 (m, pyridyl). 31P{1H} NMR (CDCl3): d ꢀ0.7 (s, PN3),
10.0 (s, br, [(CH3)(PN3)]I), 14.6 (s, OPN3). Replacing the hexanes
by CD3OD and by C6D6 gave similar results. A similar procedure
in hexanes but with Zn dust added prior to CO treatment again
gave a tan precipitate; addition of CDCl3, and filtering to remove
the Zn, gave a filtrate that by 31P{1H} NMR analysis was a mixture
of PN3, OPN3, and the reactant 9a.
A white precipitate was formed via use of the ‘10a procedure’
with PPh3 (13.5 mg, 51.5 lmol) and CH3I (4.0 lL, 64.2 l
mol). 1H
2
NMR (ppt): d 3.46 (d, 3H, JPH 13.2, P-CH3), 7.95–8.15 (m, 15H,
phenyl). 31P{1H} NMR (ppt): d 21.8 (s, br), (literature value, 22.2,
CDCl3 [9]). 31P{1H} NMR (soln): d 21.6 (s, br).
2.7. Bismethylated phosphonium iodides
Attempts to prepare the bismethylated salts 11a–c in situ in
C6D6 were unsuccessful (see Section 3.5). Use of CDCl3 instead gave
a mixture of the mono and bismethyl salts (for 11a and b); 11a, b,
and c were prepared as the sole-products using CD3OD, MeOH, and
CDCl3, respectively. NMR data were measured at 300 MHz; IR spec-
tra are obtained from KBr discs; and KM data refer to MeOH
solutions.
2.8.2. Attempted syntheses of Ni(COCH3)(I)(PPh3)2 and
Ni(COCH3)(CO)(I)[d(py)pcp] via methylation
CH3I (3
lL, 48.2 lmol) was added to Ni(CO)2(PPh3)2 (7.7 mg,
12.0 mol) in CDCl3. Heating at 55 °C for 2 h generated
l
a
dark-green solution and an off-white precipitate; the latter was
analyzed by 31P{1H} NMR (CDCl3) to be a mixture of PPh3 (d ꢀ5.0
s), [(CH3)PPh3]I (d 22.4 s, br), OPPh3 (d 29.4 s), and
Ni(CO)2(PPh3)2 (d 32.8 s).
2.7.1. [(CH3)2(dppe)]I2 (11a)
A procedure like that for 10b, but using dppe and 5 equivalents
of CH3I in CDCl3, gave a white precipitate that was found by 31P{1H}
NMR (CD3OD) to be a ꢂ3:1 mixture of the mono and bismethyl
salts.
A similar procedure using Ni(CO)2[d(py)pcp] (1.7 mg, 3.1
l
mol)
and 9 equivalents of CH3I (1.75
CD3OD gave no reaction.
lL, 0.0275 mmol) in CDCl3 or
A similar procedure, but using 20 equivalents of CH3I in CD3OD,
and heating the NMR tube at 55 °C for 2 h, gave a precipitate of
pure (by NMR) 11a. 1H NMR (CD3OD): d 3.03 (dm, 4Ha), 3.41 (d,
2.9. Reactivity of Ni(0) 2-pyridylphosphine complexes with small
molecules
2
6H, JPH 5.2, P-CH3), 7.25–8.15 (m, 20H, phenyl). 31P{1H} NMR
Systems discussed in this Section were studied using the Varian
XL300 NMR spectrometer.
(CD3OD): d 26.6 (s) (literature value, 26.9 in CD3OD [10]).
2.7.2. {(CH3)2[d(py)pe)]}I2 (11b)
2.9.1. Reactivity with ethylene
A procedure like that for 11a, but using d(py)pe and 15 equiva-
lents of CH3I in CDCl3, gave a cream precipitate that 31P{1H} NMR
(CD3OD) revealed it to be a ꢂ 1:1 mixture of the mono and bis-
Ni(PN3)4 (8d) (48.8 mg, 43.6 lmol) in C6D6 (2 mL) at 0 °C was
unreactive toward 1 atm C2H4, 31P{1H} analysis of the dark-orange
solution showing just the singlet of 8d.
methyl salts. A solution of d(py)pe (82.2 mg, 204
lmol) in MeOH
(10 mL), stirred with 20 equivalents of CH3I (254
l
L, 4.08 mmol)
2.9.2. Reactivity with chlorine, oxygen, and water
for 2 h at 55 °C, became pale-yellow. Addition of hexanes (25 mL)
gave a cream precipitate of 11b that was collected, washed with
hot THF (4 ꢃ 5 mL), and dried in vacuo. Yield: 58.0 mg (41%).
Anal. Calc. for C24H26N4P2I2: C, 42.01; H, 3.82; N, 8.16. Found: C,
42.2; H, 3.8; N, 7.9%. 1H NMR (CD3OD): d3.27 (ps t, 4Ha), 4.45 (d,
Addition of 1 equivalent of Cl2 (150
tight syringe to a red C6D6 solution (1 mL) of Ni[d(py)pcp)]2 (6a)
(6.26 mg, 6.63
mol) changed the color to dark-brown; 31P{1H}
analysis revealed a mixture containing d(py)pcp, the monoxide
and dioxide of d(py)pcp, and 6a [6a]. The same procedure
lL, 6.14 lmol) via a gas-
l
2
6H, JPH 5.7, P-CH3), 7.80 (m, 4H5), 8.15 (m, 4H3), 8.26 (m, 4H4),
used for the Cl2-reaction, but using 6a (3.75 mg, 3.98
lmol) and
8.93 (m, 4H6). 31P{1H} NMR (CD3OD): d21.5 (s). IR: 1297 (m,
1 equivalent of O2 (90 L, 3.69 mol), gave a pale-yellow solution;
l
l

M.D. Le Page et al. / Inorganica Chimica Acta 431 (2015) 276–288
281
31P{1H} analysis showed a mixture of d(py)pcp and its dioxide.
Reaction of 6a (2.50 mg, 2.65 mol) with 1 equivalent of degassed
H2O (0.1 L, 3.0 mol) also yielded a pale-yellow solution, contain-
ligand would be expected to give a dP value similar to that of
NiCl2(P,N-PN3) (dP = ꢀ17.0 [1]), and a monodentate P-coordinated
PN2 would give a dP value closer to that of 1b (dP = 33.1). HT and
l
l
l
ing d(py)pcp and its dioxide.
head-to-head (HH) isomers for related Pt2X2(l-PN2)2 complexes
2
(X = halogen) can be distinguished by 195Pt NMR and JPtP values
[12]. The P-centre in a bridging PN2 is chiral, and thus 1d should
show two dP singlets: one for the (R,R) and (S,S) enantiomers, and
one for the (R,S) enantiomers. The lone singlet detected at r.t. is
3. Results and discussion
3.1. Ni0 carbonyl pyridylphosphine complexes
thought to be an averaged signal, as reported for Pd2Cl2(l-PN2)2
complexes where two singlets, seen at ꢀ20 °C, coalesce at r.t. to
one [14]. The 1H NMR spectrum of 1d is complex (Fig. 1). The d
7.1 to 8.2 resonances are assigned to the phenyl and pyridyl
H3–H5 protons (N-coordinated and uncoordinated), with the
d 8.79 and 9.22 assigned, respectively, to the H6 protons of the
uncoordinated and coordinated pyridyls, the triplet component of
these being consistent with the singlet seen for the two P-atoms;
these data and assignments are comparable to those reported for
Complexes 1a–c (Chart 2) were not isolated as pure solids, but
were detected in situ by treatment of Ni(CO)2(PPh3)2 with two
equivalents of PNx in C6D6 solution. The 31P{1H} data show singlets
in the d 31–34 range for 1a–c, a singlet for Ni(CO)2(PPh3)2, and AB
doublets of doublets in the d 35–43 range for mixed PPh3/PNx
species; these are consistent with the equilibria shown in Eq. (1),
which we noted previously for the x = 2 system [3]. As well as
C6D6 solutions, DMSO-d6 and acetone-d6 were also tested for the
in situ syntheses in attempts to increase the yields of 1b, but this
species was not detected (Table S4). Further data from C6D6
solutions (Table S5) indicate that an equilibrium mixture in the
PN2 system is attained after heating at 55 °C for ꢂ12 h, with
amounts of up to 9% 1b, ꢂ45% mixed species, and ꢂ45%
Ni(CO)2(PPh3)2. An increase in [PN2] naturally shifts the equilibria
to the right (Table S6): maximum yields of ꢂ60% of 1b and ꢂ55%
Ni(CO)2(PPh3)(PN2) were attained at 8 and 4 equivalents of PN2,
respectively.
Os3(CO)10(l-P,N-PN2) [15], and the downfield shifts for such H6
protons are well established [1,16]. The conductivity data in
MeOH are consistent with the 4:1 electrolyte formulation [17], as
is formation of 0.95 mg of AgCl by addition of excess AgNO3 to
1.55 mg of the complex in MeOH: i.e. 15.2% Cl, in good agreement
with the theoretical value of 15.76%. The elemental analysis is
satisfactory if ½CH2Cl2 solvate is added to the formula, noting that
the complex was extracted by column chromatography using
largely this solvent. The associated chloride anions must derive
from the CDCl3 solvent used in synthesis of the complex from
Ni(CO)2(PPh3)2; the synthesis is comparable to the laboratory
light-induced syntheses of the tetrahedral NiIICl2(P,N-PN3) and
tripodal [NiII(N,N0,N00-PN3)2]Cl2 from the same precursor with three
equivalents of PN3 [1].
PNx
ꢀ
PNx
ꢀ
NiðCOÞ2ðPPh3Þ2
NiðCOÞ2ðPPh3ÞðPNxÞ
NiðCO2ÞðPNxÞ2 ð1Þ
—PPh3
—PPh3
1aꢄ—cꢄ
1a—c
2
Of note, the JPP values for the mixed PPh3/PNx species in C6D6
(ꢂ37, 82, and 115 Hz, for x = 1–3, respectively) are of relative order
1.0:2.2:3.1, paralleling the 1:2:3 ratio of N-atoms present; further,
Of the two Ni(CO)2(P–P) complexes 2a and 2b (also made from
Ni(CO)2(PPh3)2), the former has been isolated previously as a dihy-
the separation between the centres of the two doublets (Dd–d
)
drate [3]. The 1H- and 31P{1H} NMR and
m(CO) data for 2a agree
increases reasonably linearly with the number of N-atoms present
(Fig. S1). Such trends are qualitatively consistent with greater dis-
similarity of electronic environments at the P nuclei with increas-
with the earlier data, and new data, including UV–Vis bands, the
expected zero molar conductivity, and more detailed IR, are
presented. Species 2b is equally well characterized here, and an
X-ray crystal structure has also been done. The ORTEP (Fig. 2)
shows the R,R-enantiomer with the chiral C(1) and C(2) atom,
the not shown S,S-enantiomer also being present within the unit
cell. Selected bond lengths and angles are listed in Table 3. The
geometry is considerably distorted from tetrahedral, as indicated
by the P–Ni–P and CO–Ni–CO bond angles of 90.49 and 116.94°,
respectively. The former is only 2° greater than the P–Ni–P angle
of the square-planar NiCl2[d(py)pcp] [1], the compression in 2b
being attributed to the limited bite of d(py)pcp, since the same
P–Ni–P angle in Ni(CO)2(PPh3)2 is 117° in which the CO–Ni–CO
angle is 113° [18]; the P–Ni–P angle in Ni(CO)2(PN1)2 (1a) is 113°
[11]. The bond lengths of 2b are overall close to those of
Ni(CO)2(PPh3)2 [18] and 1a [11], while the Ni–P bonds are
ꢂ0.05 Å longer than in NiCl2[d(py)pcp)] [1], perhaps implying that
Ni–P bonds are weaker in the Ni0- than in the NiII-d(py)pcp)
ing number of N-atoms. The two
m(CO) values of 1a–c both
increase marginally (from 1941 to 1943, and 2001 to 2004 cmꢀ1
)
as x increases from 1 to 3 and, together with the 1932 and
1998 cmꢀ1 values measured similarly for Ni(CO)2(PPh3)2, imply
stronger
PNx ligands relative to those of PPh3; less electron density on the
metal available for
-back-donation into the p⁄ orbitals of the CO
p-acceptor and/or weaker r-donor properties for the
p
ligands strengthens the C„O bond.
A previous attempt to isolate 1a via reduction of NiBr2 by Zn/CO
in THF gave a solid contaminated with paramagnetic impurities
(probably NiII species) [3]; nevertheless, a crystal structure of 1a
has been reported, although details of the synthesis were unavail-
able to us [11].
Of interest, the equilibria of Eq. (1) had first been established
using the in situ reaction in CDCl3 at ambient temperature, but
the system was kinetically slow, taking about 5 days at ambient
conditions to reach equilibrium [3]. A sample of Ni(CO)2(PPh3)2
with two equivalents of PN2 in CDCl3 was thus heated at 60 °C,
and this generated a purple precipitate of 1d (in 38% yield) along
with the same, in situ yellow solution products (Section 2.2.4).
The precipitate is only seen in this solvent, and is thought to be
complexes; in conjunction with the IR data, where msym(CO) for
2b is ꢂ5 cmꢀ1< that of Ni(CO)2(PPh3)2, the data are consistent with
greater
Both
p-acidity of d(py)pcp vs. that of PPh3.
m(CO) values for the dicarbonyl(pyridylphosphine) com-
plexes (in a thin CH2Cl2 film on a KBr disc) decrease regularly by
few cmꢀ1 (1943 ? 1927, and 2004 ? 1997) in the order:
PN3 > PN2 > PN1 > d(py)pe > d(py)pcp, illustrative of increasing
-acceptor ability at the P-atom, implying pyridyl ligand -acidity
a
[Ni2(CO)4(l-PN2)2]Cl4, a dimetallic nickel(II) species (Chart 3).
The +LSIMS pattern, with a major m/z 757 peak, is identical to that
simulated for ‘Ni2O4P2N4C36H27’ [Ni2(CO)4(PN2)2 + 1H+] (Fig. S2).
p
p
The dP singlet at 24.1 and one m
(CO) band at 1971 cmꢀ1 are consis-
increases in the reverse order. The dP singlets of the free ligands in
CDCl3 are 0.0, ꢀ0.7, ꢀ2.6, ꢀ4.0, ꢀ6.1, respectively, for d(py)pcp,
PN3, PN2, PN1, and d(py)pe [3], which could reflect increasing
tent with equivalent P-atoms, trans-CO ligands, and the so-called
head-to-tail (HT) arrangement of the PN2 ligands [12]. The down-
r
-basicity at the P-atom. Theoretical studies on these trends
field coordination shift (dP
D = +26.7 relative to free PN2) is charac-
would be informative.
teristic of a 5-membered chelate ring [13]. A P,N-chelated PN2

282
M.D. Le Page et al. / Inorganica Chimica Acta 431 (2015) 276–288
Ni(CO)2(P−P)
P−P = d(py)pe 2a
P−P = d(py)pcp 2b
Ni(CO)2(PNx)2
x = 1 1a
x = 2 1b
P
PN
x
Ni
Ni
OC
OC
OC
OC
P
PN
x
x = 3 1c
Chart 2. Complexes 1a–c, and 2a–b.
-
The ‘immediate’ (within 5 min of sample preparation) 31P{1H}
NMR spectrum of 2b in D2O (containing trace H2O) under N2 con-
sists of a singlet at dP 25.2 and trace singlets attributable to
d(py)pcp bisoxide (dP 31.9) [1] and remaining 2b (dP 37.4); after
a further 5 min, only the d(py)pcp(O)2 singlet and a trace singlet
at dP ꢀ12.9 were seen. Dissolution of d(py)pcp in D2O/aq.HBF4
gives a dP singlet at ꢀ13.0, attributed to [nH-d(py)pcp]n+, a species
with n protonated pyridyls (see below), and thus 2b dissolves in
D2O to form an intermediate (dP 25.2) that rapidly decomposes
to d(py)pcp(O)2 and [nH-d(py)pcp]n+. The data below establish that
4Cl
*
pyPhP
N
CO
CO
Ni
Ni
OC
OC
N
PPhpy
*
Chart 3. Proposed structure for the purple complex 1d; N = o-py; ⁄ indicates chiral
centre.
n = 2; i.e. the intermediate is {Ni(CO)2[2H-d(py)pcp]}2+ (2b⁄2H2+
)
and that the dP ꢀ12.9 species is [2H-d(py)pcp]2+
.
The above 31P{1H} NMR experiment, repeated in H2O/acetone-
d6 (1:9), gave the same results [dP of d(py)pcp(O)2 = 32.7, dP of
2b = 37.7, dP of nH-d(py)pcpn+ = ꢀ13.5, and a singlet at dP 25.7].
The 1H NMR spectrum revealed, aside from the resonances of 2b,
a broadened peak at dH 10.25 that integrated (vs. two Hc protons)
to show ꢂ2 N–H protons (dH ꢂ10 is typical for an N–H proton
[22]), implying that n = 2, and that the water-soluble intermediate
is 2b⁄2H2+. Examination of the 1H NMR spectrum (acquired after
10 min) confirms these findings: the H3–H6 signals [dH 6.70, (m,
H5), 7.28 (m, H4), 7.70 (m, H3), 8.47 (m, H6)] are essentially the
same as those of d(py)pcp(O)2 in D2O [1], and not of coordinated
d(py)pcp [1]; and the broadened N–H peak of 2b⁄2H2+ has almost
completely disappeared. The loss of N–H resonance intensity indi-
cates that the d(py)pcp(O)2 is not diprotonated, and hence the
quaternized N-atoms in 2b⁄2H2+ must deprotonate when this
intermediate decomposes; the remaining trace N–H peak confirms
that the dP signal at about ꢀ13 must correspond to N-protonated
The UV–Vis maxima at ꢂ240 and 268 nm for 2a and 2b are sim-
ilar to those seen for free d(py)pe) and d(py)pcp [1], whereas the
bands in the 320 to 365 nm region are probably Ni(e) ? CO(p⁄
)
charge-transfer bands [19]. The IR bands in the 1570–1420 cmꢀ1
region are pyridyl skeletal bands [1,3].
Pale-yellow crystals of 2a were obtained but they did not dif-
fract properly. A search of the Cambridge Structural Database
reveals that the X-ray structures of Ni(CO)2(dppe) and
Ni(CO)2(dpcp) (the phenyl-phosphine analogues of 2a and 2b,
respectively) have not been reported. Ab initio MO calculations
suggest that Ni(CO)2(PR3)2 complexes will be pseudo-Td, with the
ligand
r-donor/p-acceptor electronic properties determining the
degree of distortion [20].
Attempts to make 2a or 2b via Zn-reduction and CO-carbonyla-
tion of known NiCl2(P–P) precursors, as used to synthesize
Ni(CO)2(PPh3)2 [21], yielded only the respective precursors as
determined by 31P{1H} NMR spectroscopy [1,3].
[2H-d(py)pcp]2+
.
Chart 4 shows the proposed structure for the dicationic NiII
complex; the sharp dP singlet signifies equivalent P nuclei, and
implies that the two protons are ‘evenly distributed’ among the
four pyridyl N-atoms, likely via rapid proton exchange.
Phosphorus protonation is excluded because no P–H resonances,
typically in the dH = 0 to ꢀ10 range [22,23], are seen and, of course,
the N-atoms are more basic than the P-atoms. Of note,
3.2. Aqueous solution chemistry of Ni(CO)2[d(py)pcp] and d(py)pcp
Both Ni(CO)2[d(py)pe] (2a) and Ni(CO)2[d(py)pcp)] (2b), pale
yellow solids, dissolve in water to give initially pale yellow solu-
tions. The aqueous solution chemistry of 2b was studied in detail;
pure Ni(CO)2(PNx)2 complexes were not isolated, and could not be
similarly studied.
H3-H5, H3*-H5*, phenyl
H6
H6*
9.4 9.2 9.0 8.8 8.6 8.4 8.2 8.0 7.8 7.6 7.4 7.2 7.0
4
9.2
9.0
8.8
8.6
8.4
8.2
8.0
7.8
7.6
7.4
7.2
7.0 ppm
Fig. 1. 1H NMR spectrum (CD3OD) of the pyridyl region of [Ni2(CO)4(
l
-PN2)2]Cl4 (1d); H = protons of non-coordinated pyridyl group, H⁄ = protons of coordinated pyridyl
group. The relative integrations for the H⁄6 and H6 signals versus the H3–H5, H3⁄–H⁄5 signals are correct.

M.D. Le Page et al. / Inorganica Chimica Acta 431 (2015) 276–288
283
a C6H6 solution of 2b, and an EtOH solution of d(py)pcp. The
elemental analyses of the Ni salt and the protonated pyridyl
phosphine, together with spectroscopic data, fully establish their
identities.
In the IR spectra, a
m
(N–H) band is observed at ꢂ2690 cmꢀ1 for
[2H-d(py)pcp](OTf)2, and at 3095 cmꢀ1 for the NiII salt. The values
are typical of those reported for amine salts (2250–2700 cmꢀ1
[22], and for metal complexes containing protonated N-atoms
(2950–3150 cmꢀ1
[25]; no (P–H) band, typically 2250–
)
)
m
2450 cmꢀ1 [22], was seen. The
m(CO) band shifts from
1997 ? 2045 and 1927 ? 1996 cmꢀ1 upon protonation of 2b are
a manifestation of the electrophilic N-atoms balancing their
positive charge by both decreasing the
r-basicity and increasing
the -acidity of the P-atoms (see Section 3.1). Comparison of IR
p
data for 2b and 2b⁄2H2+, with those for pyridine and pyridinium,
further supports the N-protonation: a pyridyl skeletal band of 2b
at 1569 shifts to 1522 cmꢀ1 for 2b⁄2H2+, and a corresponding shift
(1587 ? 1538 cmꢀ1) is seen between pyridine and a pyridinium
salt [26]. IR bands of the triflate anion (ꢂ1630, 1400, 1280, 1256,
1031, and 640 cmꢀ1 [22]) are seen for the triflate salts of 2b⁄2H2+
and [2H-d(py)pcp)]2+
.
The dP 25.7 and ꢀ13.5 singlets in the 31P{1H} NMR spectrum of
2b in H2O/acetone-d6 correspond to the respective values of 25.5
and ꢀ13.6 for the triflate salts of 2b and d(py)pcp in acetone-d6.
Similarly, the respective 1H NMR spectra of the salts show the pres-
ence of ꢂ2 N–H protons at dH 10.5 and 10.0, in comparison to the
10.25 value for 2b in H2O/acetone-d6. No P–H protons were
detected in the 1H NMR spectra of the triflate salts. Further support
for the diprotonated formulations is that on addition of six equiv-
alents of CF3SO3H to acetone-d6 solutions of 2b and of d(py)pcp,
four equivalents of free CF3SO3H (dH 7.7) are detected in each
system.
The overall findings establish that 2b dissolves in aqueous
media via quaternization of the pyridyl N-atoms to yield initially
the diprotonated species 2b⁄2H2+ that decomposes in minutes to
d(py)pcp(O)2 and traces of N-protonated [2H-d(py)pcp]2+. Other
decomposition products are CO (detected by GC analysis of
the headspace of the NMR tube) and possibly [Ni(H2O)6]2+; the
UV–Vis spectrum is broad but shows a kmax at 396 nm, the same
as for NiCl2ꢁ6H2O in 1:9 H2O/acetone. The hexa(aquo) cation is
paramagnetic, and thus other paramagnetic NiII species containing
the diphosphine or its oxides that are not detected by NMR could
be present. Eq. (2), where P–P is d(py)pcp, shows the reactant,
intermediate, and identified products. Ni(CO)2[d(py)pe)] (2a)
almost certainly behaves similarly. However, details of the
stoichiometry remain unsolved.
Fig. 2. ORTEP (50% probability) of the molecular structure of Ni(CO)2[d(py)pcp]
(2b).
Ni(CO)2(PNP), where PNP is P,P-bonded bis(diphenylphosphi-
nomethyl)pyridine, similarly undergoes protonation at the pyri-
dine-N to give [Ni(CO)2(HPNP)]+ [24]. In contrast, the NiIIX2(P–P)2
complexes [X = halide, P–P = d(py)pe, d(py)pcp)] dissolve in water
to form [Ni(H2O)2(P–P)2]2+ 2Xꢀ [1].
3.3. Triflate salts of {Ni(CO)2[2H-d(py)pcp]}2+ and [2H-d(py)pcp]2+
The protonated species were subsequently isolated as
the bis(triflate) salts, {Ni(CO)2[2H-d(py)pcp]}(SO3CF3)2 and
[2H-d(py)pcp](SO3CF3)2, by addition of triflic acid to, respectively,
Table 3
Selected bond lengths and angles for Ni(CO)2[d(py)pcp] (2b).
Bond
Length (Å)
Bond
Length (Å)
Bond lengths (Å)
Ni–P(1)
Ni–P(2)
Ni–C(26)
Ni–C(27)
C(26)–O(1)
C(27)–O(2)
P(1)–C(1)
P(1)–C(6)
P(1)–C(11)
P(2)–C(2)
P(2)–C(16)
2.2097(6)
2.2085(7)
1.758(3)
1.758(3)
1.151(3)
1.147(3)
1.834(2)
1.837(2)
1.828(2)
1.836(2)
1.835(2)
P(2)–C(21)
N(1)–C(6)
N(2)–C(11)
N(3)–C(16)
N(4)–C(21)
C(1)–C(2)
C(2)–C(3)
C(3)–C(4)
C(4)–C(5)
C(5)–C(1)
1.835(2)
1.342(3)
1.341(3)
1.336(3)
1.344(3)
1.541(3)
1.529(3)
1.528(3)
1.554(3)
1.528(3)
O
NiðCOÞ2ðP—PÞ H!2 2bꢄ2H2þ ! ½NiðH2OÞ ꢅ2þ þ ðOÞP—PðOÞ
6
5 min ðsee Chart 4Þ 5 min
þ ½2HðP—PÞꢅ2þ þ 2CO
ð2Þ
The water in Eq. (2) is thought to be the source of O-atoms for
the diphosphine oxidation, as strictly anaerobic conditions (N2 and
de-aerated H2O) were employed. Such chemistry is well docu-
mented for oxidation of a phosphine to its monooxide, definitively
by use of 17O-enriched H2O and 17O NMR [27], within redox reac-
tions of the type shown in Eq. (3) and where the metal is 2-electron
Bond
Angle (°)
Bond
Angle (°)
Bond angles (°)
P(1)–Ni–P(2)
90.49(2)
P(1)–C(11)–N(2)
P(1)–C(1)–C(2)
P(2)–C(16)–N(3)
P(2)–C(21)–N(4)
P(2)–C(2)–C(1)
C(1)–C(2)–C(3)
C(2)–C(3)–C(4)
C(3)–C(4)–C(5)
C(4)–C(5)–C(1)
H(C1)–C(1)–C(2)–H(C2)
118.91(18)
110.00(14)
113.16(19)
117.10(17)
110.38(15)
102.56(18)
104.8(2)
106.84(18)
103.69(19)
1.97(8)
+
2
C(26)–Ni–C(27)
C(26)–Ni–P(1)
C(27)–Ni–P(2)
C(26)–Ni–P(2)
C(27)–Ni–P(1)
Ni–C(26)–O(1)
Ni–C(27)–O(2)
P(1)–C(6)–N(1)
P(1)–Ni–P(2)–C(2)a
116.94(13)
114.66(8)
108.77(10)
113.40(9)
109.45(9)
179.4(2)
177.2(3)
113.25(17)
10.03(9)
CO
OC
Ni
P
P
P
P
P
P
=
N
H
N
H
N
N
a
Chart 4. Diprotonated form of 2b (labelled 2b⁄2H2+).
Torsion angle.

284
M.D. Le Page et al. / Inorganica Chimica Acta 431 (2015) 276–288
reduced. Examples of such redox chemistry have been found in NiII
chemistry [28], as well as in PdII and PtII systems [29]. Of interest,
we reported recently that NiX2(P–P) species, where X is a halide
and P–P is d(py)pcp or d(py)pe, undergo slow decomposition (over
24 h at r.t.) in water to give [Ni(H2O)6]Cl2 and the diphosphine
dioxide with no change in pH, and the reaction was attributed to
trace O2 [1]. The presence of the CO ligand raises the possibility
of OHꢀ attack at the CO and the 2e-redox process outlined in Eq.
(4), which is well established for higher valent metal ions such as
RhIII [30], but this redox process generates CO2 which was not
detected. The system outlined in Eq. (2) for 2b is different in show-
ing no reduction of the metal ion; indeed, as written there is an
overall 6e-process (Ni0 ? NiII, and two PIII ? PV) that requires a
concomitant 6e-reduction! The most plausible reduction product
is H2, but only traces of H2 (< 1%, based on the reactant
Ni(CO)2[d(py)pcp)]) were seen by GC, and so the redox chemistry
related to Eq. (2) remains unresolved; the hydrogen could possibly
be utilized in hydrogenation of the pyridyl groups. More detailed
studies on the precise stoichiometry of Eq. (2) are needed. The pos-
sibility of trace O2 being the oxidant remains a possibility, although
the rapidity of the oxidation process, which would likely have to
involve formation of a Ni-dioxygen intermediate [31], argues
somewhat against this; however, protonation of a coordinated
oxygen (e.g., present as superoxide) could lead to accelerated
free-radical oxidation processes, as demonstrated at least for late
transition metals including Pd and Pt systems [32]. Worth noting
is that the protonated complex [Ni(CO)2(HPNP)]+ mentioned above
is reported to react stoichiometrically with aqueous HCl to form
[Ni(PNP)Cl]+, H2 and 2CO, although details of the reaction were
not given in the short communication [24].
Table 4
Tetrahedral Ni0 complexes with P-coordinated ligands.a
Category
Complex type
No. of complexes
A
B
C
D
E
F
G
H
I
Ni(PR3)4
4
Ni(PR3)3(PR03)
Ni(PR3)2(PR03)2
Ni(PR3)2(PR03)(PR300 )
Ni(PPh3)(PN1)(PN2)(PN3)
Ni(PR3)2(P–P)
Ni(PR3)(PR03)(P–P)
Ni(P–P)(P0–P0)
Ni(P–P)2
12
12
24
1
12
36
3
3
a
Combination of PR3, PR30 , PR300 (i.e. PPh3, PN1, PN2, PN3) and/or P–P or P0–P0 [dppe,
d(py)pe, d(py)pcp] ligands.
and 7 were made in situ, whereas 6 and 8 were isolated; 6b [3],
6c [4,37] and 8a–d [3,5,38] have been reported previously. The
NMR spectra of species containing PR3 (3–5 and 8) were obtained
rapidly because of decomposition to black ‘solutions’ within
10–30 min, even at 0 °C. Amounts of reagents used for syntheses
of 3–5, 7 and 8 (not given in Sections 2.4.1–2.4.3, 2.4.7, and
2.4.8) are available in Tables S1–S3.
The only new homoleptic complex is 6a that was characterized
by NMR, UV–Vis and EI mass spectra, microanalysis, and m.pt. The
d(py)pcp used is a racemic mixture of R,R- and S,S-enantiomers [1],
and thus 6a should give rise to two singlets, one for the Ni(RR)2 and
Ni(SS)2 enantiomers (where the P-nuclei are equivalent), and one
for the (RR)-Ni-(SS) species (where the P-nuclei are enantiotopi-
cally related by an inversion centre at Ni); however, just one
singlet (dP = 38.70) is seen (Table 2). The {M[d(py)pcp]2}X2
complexes (M = Pd, Pt; X = Cl, PF6) show two dP singlets separated
by <0.1 ppm [39], suggesting that for 6a the two singlets are essen-
tially identical, rather than the one singlet being due to formation
of just one set of enantiomers. The successful in situ formation of
the other known homoleptic complexes 6b [3], 6c [4,37], 8a
[5,38], and 8b–d [3]) was confirmed by comparison of 31P{1H}
NMR data to those reported (Table 2). Of note, Ni(PPh3)4 (8a) is
known to be in solution equilibria with Ni(PPh3)3 and PPh3 at r.t.,
whereas at ꢀ80 °C it is the sole species present [38]; in contrast,
and as pointed out previously, such phosphine dissociation from
8b to 8d is not seen at ambient conditions, and is thought to be
PIII þ OHꢀ ! PVO þ Hþ þ 2e
ð3Þ
MN—CO þ OHꢀ ! ½MN—COOHꢅꢀ ! MNꢀ2 þ CO2 þ Hþ
ð4Þ
Noting that Ni(CO)2(PPh3)2 has been found to have marginal
activity for the Water–Gas-Shift reaction (ꢂ1.0 turnover for an
aqueous 2 mM HCl system after 8 h at 150 °C under 1 atm CO
[33]), we tested 2b for such activity. Because of rapid decomposi-
tion of this complex via protonation in aqueous solution, Eq. (2),
activity was tested in 0.2 M KOH under 1 atm CO; however, after
24 h, no H2 was seen, implying zero activity. Of note, a system
based on the dicarbonyl derivative Ni(CO)2(PNP), was shown to
have weak catalytic activity for the WGS reaction [24].
Related to our 2a and 2b complexes are the water-soluble
Ni(CO)4ꢀn(PTA)n species, PTA being the well-known 1,3,5-triaza-
7-phosphaadamantane, another N-containing phosphine ligand
[34]. Studies on catalysis in aqueous media using metal-PTA sys-
tems are extensive [35], but not (to our knowledge) utilizing Ni
complexes. Ni(CO)4 ‘dissolves’ in water above 50 °C, but this is
results from decomposition to metal [36].
due to the stronger p-acid character of the pyridylphosphines [3].
Characterization of 3–5 and 7 is based solely on 31P{1H} NMR
data (Table 1); a typical spectrum, shown for Ni(PN3)2[d(py)pe]
(4d), is shown in Fig. 3. The expected triplets (columns B and C)
arise from coupling of the two P nuclei of one ligand type (L) to
the two P nuclei of the other ligand type (L0); the assignments
are based on the dP values for the appropriate homoleptic com-
plexes [Ni(L)4 or Ni(L)2 (Column A) and Ni(L0)2 (Column D)], with
2
the JPP values listed in Column E, and the separation in Hz
between each triplet pair (Dt–t) listed in Column F. Other dP signals
of unreacted phosphines, phosphine oxides, and the homoleptic
Ni0 complexes were also usually detected. The regions of the
31P{1H} NMR spectrum containing the triplets for 3–5 and 7 are
depicted in Figs. S3–S6. Of note, the 31P coordination chemical shift
for Ni(PPh3)4 (29.0 ppm) is similar to that of the Ni(PNx)4 com-
plexes 8b and 8d (29.5 and 26.9 ppm, respectively), whereas that
for Pt(PPh3)4 (15.5 ppm, [40]) differs substantially from those of
Pt(PN1)4 (28.1 ppm, [8]), Pt(PN2)4 (28.7 ppm, [41]) and Pt(PN3)4
(29.8 ppm, [8]).
The dP singlet for the Ni(PNx)4 complexes, and the PNx and P–P
triplets for the Ni(PNx)2(P–P) complexes (Table 1, Columns A, B, and
C, respectively) shift downfield in the order PN1 < PN2 < PN3, the
deshielding increasing with more electron-withdrawing N-atoms.
Indeed, a linear correlation exists between both the PNx and P–P
3.4. Ni0 tertiary phosphine complexes
3.4.1. Synthesis and characterization
A total of 107 unique tetrahedral Ni0 complexes, with just P-co-
ordinated ligands, can be synthesized by combination of the four
PR3 ligands (PPh3, PN1, PN2, PN3) and the three P–P ligands [dppe,
d(py)pe, d(py)pcp)] (Table 4). The 22 complexes of categories (A),
(F), (H), and (I) (Chart 5) were prepared in situ or isolated, the
prime (0) and double-prime (00) notations in Table 4 denoting differ-
ent PR3 ligands within the same complex.
Complexes 3–7 were prepared under N2 at 60 °C in C6D6 or
C6H6, using appropriate amounts of phosphine and Ni(1,5-COD)2,
while syntheses of type 8 were conducted similarly at 0 °C in
hexanes to prevent decomposition of the products. Species 3–5
triplets values and
x for complexes 3–5 (Fig. 4). Increased

M.D. Le Page et al. / Inorganica Chimica Acta 431 (2015) 276–288
285
PR3
PR3
Ni
P
Ni
Ni
R3P
R3P
P
'P
PR3
PR3
P
P
'P
−
−
21-23
24-25
26
Ni(PR3)2(dppe)
PR3 = PPh3 (3a)
PR3 = PN1 (3b)
PR3 = PN2 (3c)
PR3 = PN3 (3d)
Ni(PR3)2[d(py)pe]
PR3 = PPh3 (4a)
PR3 = PN1 (4b)
PR3 = PN2 (4c)
PR3 = PN3 (4d)
Ni(PR3)2[d(py)pcp)]
PR3 = PPh3 (5a)
PR3 = PN1 (5b)
PR3 = PN2 (5c)
PR3 = PN3 (5d)
Ni(P−P)2
Ni(P−P)(P'−P')
Ni(PR3)4
PR3 = PPh3 (8a)
PR3 = PN1 (8b)
PR3 = PN2 (8c)
PR3 = PN3 (8d)
P−P = d(py)pcp (6a)
(P−P)(P'−P') = dppe & d(py)pe
(7a)
(7b)
P−P = d(py)pe (6b)
(P−P)(P'−P') = dppe & d(py)pcp
P−P = dppe
(6c)
(P−P)(P'−P') = d(py)pe & d(py)pcp (7c)
Chart 5. The twenty-two synthesized Ni0 phosphine complexes.
Fig. 3. 31P{1H} NMR spectrum of Ni(PN3)2[d(py)pe] (4d) in C6D6.
deshielding of the P–P phosphorus nuclei could result from a
d(py)pe and two phenyls and one pyridyl per P in PN1) is
1133 Hz. This trend is seen in nine (3a–d, 4b–d, and 5a–b) of the
twelve Ni(PR3)2(P–P) complexes, but not for 4a and 5c–d. The
Ni(P–P)(P0–P0) complexes (7) also deviate from the trend, as Dt–t
is ꢂ1000 Hz for 7a and b, which have four phenyl and four pyridyl
groups, whereas for 7c, which contains only pyridyl groups, Dt–t is
ꢂ2100 Hz.
decrease in PNx P ? Ni
increase (and deshielding) in the P ? Ni
r
-donation inducing a compensating
-donation from the
r
two P-nuclei of the P–P ligand. The dP triplet values for the Ni(P–
P)(P0–P0) complexes (7) (Table 1, Columns B and C) are within
ꢂ1–3 ppm of the dP singlets of the corresponding homoleptic spe-
cies (Table 1, Columns A and D, respectively).
2
The separation between the two triplets
(
Dt–t
)
of the
The JPP value for the Ni(PR3)2(P–P) complexes 3–5 (Table 1,
Ni(PR3)2(P–P) complexes (Table 1, Column F) shows similarities
to the two doublets separation (Dd–d) in the Ni(CO)2(PPh3)(PNx)
complexes (see above). When the P-nuclei of both PR3 and P–P
are ligated by a similar number of either phenyl or pyridyl sub-
stituents, Dt–t is generally smaller then when the respective liga-
tions are significantly different in numbers of such substituents.
For example, 3a contains ten P-ligated phenyl groups and Dt–t is
76 Hz, in comparison to 806 Hz for 3d where the two P-nuclei
are each ligated by three pyridyls and the other two P nuclei are
each ligated by two phenyl groups. Similarly, Dt–t for 4d (ten pyr-
idyl groups) is 221 Hz, while that for 4b (two pyridyls per P in
Column E) decreases, with the exception of 3a, in the order
PPh3 > PN1 > PN2 > PN3, possibly due to the reduction in electron
density available at the P-atom of PR3 for through-bond overlap
with the P-atoms of P–P as the number of electron-withdrawing
N-atoms increases. Similarly for the Ni(P–P)(P0–P0) complexes 7a
2
and b that possess only four N-atoms, JPP = 27.4 and 23.2 Hz,
2
respectively, whereas for 7c, with eight N-atoms, JPP = 10.3 Hz.
2
The lower JPP for 3a compared to those of 3b–d could possibly
be due to
ligands; such stacking of two aryl rings in cis-PdCl2(PN2)2 has been
shown to cause a weak charge-transfer interaction between
p-stacking of phenyl rings on adjacent phosphine
p

286
M.D. Le Page et al. / Inorganica Chimica Acta 431 (2015) 276–288
Fig. 4. Plot of N-atom count for PNx versus increase in dP triplet of PNx (left) and P–P (right); s = 3b–d, h = 4b–d, 4 = 5b–d.
them, and a decrease in electron density around the Pd–P bonds
[42] – a similar process in 3a could result in the JPP value falling
from oxidative-addition reactions, are generated via one-electron
processes involving halogen-atom transfer [45].
2
below those of the PNx-containing species.
NiðPN3Þ4 þ MeI ! trans-NiðMeÞðIÞðPN3Þ2ð9aÞ þ 2PN3
ð5Þ
3.4.2. Reactivity of the Ni0 complexes
NiðP—PÞ2 þ MeI ! trans-NiðMeÞðIÞðP—PÞ2; P—P
¼ dðpyÞpeð9bÞ; dðpyÞpcpð9cÞ
Studies on the reactivity of Ni0 complexes with small molecules
are myriad, as exemplified by reviews from the 1970s and 1980s
[43,44] that mention, for example, reactivity toward CO, NO,
alkenes, CS2, N2, O2, SO2, and CO2; oxidative addition reactions of
HCl and RX (X = halogen, CN) to give, respectively, NiII-hydrido
[43],and -alkyl species [43,45] were discovered in the same era.
Thus, the reactivity of Ni[d(py)pcp]2 (6a) and Ni(PN3)4 (8d) toward
small molecules was tested by introducing C2H4, O2, H2O, or Cl2
into a C6D6 solution of the complexes in an N2-flushed, J-Young
NMR tube. Ethylene was unreactive with 18-electron species 6a
ð6Þ
We attempted syntheses of the unreported trans-
Ni(CH3)(I)(dppe)2 (9d) and trans-Ni(CH3)(I)(PPh3)2 (9e), the phenyl
analogues of 9a and 9b, from Ni(dppe)2 or Ni(PPh3)4, respectively,
but isolated only the off-white solids [(CH3)(dppe)]I and
[(CH3)(PPh3)]I; the phosphonium iodides detected in this and
related reactions (see below) are discussed in Section 3.5. The
filtrates from these reactions revealed the presence of the Ni0
precursor, free phosphine, and sometimes phosphine oxides. No
phosphonium salts were detected in the syntheses of 9a–c, and
electronic differences between pyridyl and phenyl groups must
be responsible for the success and failure within the respective
syntheses. Four months after the attempted synthesis of 9d, the
Schlenk tube used was found to contain purple crystals of
NiI2(dppe) [1,7]. Of interest, oxidative addition of aryl and acyl
halides (R0X) to Ni(PR3)4 species containing more basic phosphines
(R = Et or Cy) does give trans-Ni(R0)(X)(PR3)2 complexes [45,50]. In
a patent that describes the use of mixtures of NiCl2(PPh3)2 and an
aluminum alkyl as olefin dimerization catalysts, one species is
listed as Ni(CH3)(Cl)(PPh3)2, but this is uncharacterized [51].
Species such as trans-Ni(R0)(Cl)(PMe3)2 have been made from
NiCl2(PMe3)2 and the appropriate Grignard reaction Mg(R0)Cl (e.g.
R0 = CH2CMe2Ph) [52].
A hexane solution of Ni(CH3)(I)(PN3)2 (9a) was treated with
1 atm CO in the hope of forming an acyl derivative, but only partial
conversion to CH3I, [(CH3)(PN3)]I, PN3, and OPN3 was observed.
Related to this, an attempted iodomethylation of Ni(CO)2
(PPh3)2 in CDCl3 generated only PPh3, OPPh3, and [(CH3)(PPh3)]I.
Synthesis of the Ni(COR0)(X)(PMe3)2, where R0 = CH2CMe2Ph, and
X = halide, has been reported via carbonylation of the alkyl halo
precursor [52]; the synthesis of acyl NiII phosphine species thus
appears to require a more basic phosphine ligand.
and 8d, unlike the 16-electron Pt(PN1)3 that forms Pt(g2
C2H4)(PN1)2 under similar conditions [8]. The Ni(
-
g
2-C2H4)(PPh3)2
complex has been made by reduction of an ether solution of
Ni(acac)2 and two equivalents of PPh3 with AlEt(OEt)2 under
C2H4 at 0 °C [46], or by addition of C2H4 to the 16-electron
Ni(PPh3)3 that forms upon dissociation of 8a in solution [47].
The reactions with O2, H2O, and Cl2 resulted in considerable
decomposition of 6a and 8d with concomitant release of the free
ligands and formation of their mono- and bis-oxides (see
Section 2.9). Such decomposition upon exposure to O2 or H2O is
typical of Ni0 complexes [43], and the aqueous solution chemistry
of 3–8 was thus not examined. Nickel(0) species are typically unre-
active toward Cl2 [43], although both Ni(CO)2[o-C6H4(PEt2)2]2 and
Ni[o-C6H4(PEt2)2]4 are reported to react to yield NiCl2[o-
C6H4(PEt2)2]2 [48]. The ligand oxides formed in our reactions likely
result from O2 impurities in the gas which was not of analytically
pure grade.
Reactions of Ni(PN3)4, Ni[d(py)pe]2, and Ni[d(py)pcp]2 with
CH3I give the d8 square planar, tan (9a) or octahedral, yellow
(9b–9c) trans oxidative addition products as shown in Eqs. (5)
and (6). The 31P{1H}spectrum of 9a in CDCl3 shows a sharp singlet
at d 31.5, while the 31P{1H}spectrum of the C6D6 filtrate from its
synthesis displayed the PN3 singlet at d ꢀ0.1 [1], of intensity of
ꢂ2 equivalents consistent with Eq. (5). The corresponding
31P{1H} singlets of 9b and 9c (d 29.0 and 36.5, respectively) were
‘noisy’ and typical of paramagnetic species, and the synthetic fil-
trates showed the absence of the pyridyl-diphosphines, consistent
with Eq. (6). The1H NMR spectra show for the Me group a triplet at
d 0.6 for 9a, and quintets at d 0.9 and 0.7 for 9b and c, respectively,
3.5. The phosphonium salts, [(CH3)(phosphine)]I and
[(CH3)2(diphosphine)]I2
Some mono- and bis-(methyl)phosphonium iodide salts
(Charts 1c and 6) were synthesized in order to establish their
formation in some reactions of Ni0 complexes with MeI, and in
the reactivity of iodo(methyl)-NiII species toward CO: 10e [9],
11a [10] and the PN1 analogue of 10a [8] are known.
3
with JPH values from 6.2 to 7.3 Hz, the signal intensities integrat-
ing to ꢂ3 H; for free CH3I, dH = 2.1. A similarly observed triplet (d
3
0.05, JPH = 5.7 Hz) is seen in trans-Pt(CH3)(I)(PN3)2 [8]. The well-
resolved 1H spectra result from the favorable electron relaxation
time of paramagnetic octahedral NiII complexes [43,49]. The for-
mulations of 9a–c were confirmed by elemental analysis. The trans
products, as opposed to the more common cis products derived
In our work, the monomethyl salts were first made in situ by
addition of excess CH3I to the selected pyridylphosphine in C6D6:
10a and e display one 31P{1H} NMR singlet, whereas 10b–d show
3
the AX pattern of two doublets with JPP coupling between the

M.D. Le Page et al. / Inorganica Chimica Acta 431 (2015) 276–288
287
Table 6
Ligand
Monomethyl Salt Bismethyl Salt
2I -
NMR Data for the monomethylphosphonium salts 10aꢀe in CDCl3.a
PN3
10a
10b
10c
10d
10e
−
dppe
11a
11b
11c
−
Salt of:
dP P+(3JPP
)
dP P (3JPP
)
dCH3 (2JPH
)
d(py)pe
d(py)pcp
PPh3
R2(Me)P
P(Me)R2
b
PN3
PPh3
10a
10e
10.0 s
21.8 s
–
–
3.29 d (15.0)
3.46 d (13.2)
b
11a−c
dppe
d(py)pe
d(py)pcp
10b
10c
10d
27.4 d (44.0)
21.7 d (38.5)
23.6 d (15.4)
ꢀ11.1 d (43.7)
ꢀ8.0 d (38.6)
ꢀ0.9 d (15.6)
3.27 d (13.6)
3.26 d (13.7)
3.18 d (13.5)
Chart 6. Labelling scheme for the mono and bis(methyl) phosphonium ligand salts.
a
J values in Hz.
b
two P-atoms (Chart 6), the assignments (Table 5) being based on
comparison with the dP singlet values for the appropriate bis-
methylated species and the free pyridylphosphines. No bismethyl
salts were detected or isolated from these in situ syntheses in
C6H6 even though excess CH3I was used, because the solutions of
all five systems generated precipitates within 1 h. The solids were
collected and identified as 10a–e by impurity-free 1H and
31P{1H}NMR spectra in CDCl3 (Table 6); the 1H doublet of the Me
Lit. dP (CDCl3) for 10e = 22.2, Ref. [9], and for [CH3(PN1)]I = 17.5, Ref. [8].
Table 7
Total downfield shift of 1H NMR Signals of PꢀP Ligands upon bismethylation.
Ligand
Da dH3
D
dH4
D
dH5
D dH6
d(py)pe
d(py)pcp
0.75
0.62
0.76
1.91
0.70
1.07
0.33
0.24
2
results from JPH coupling. This precipitation behavior is known
a
D
= [dH (bismethylated compound) ꢀdH (compound)]. Ha of d(py)pe and Hc of
for ditertiary diphosphines, and has been exploited for selective
formation of the diphosphine monoxides (versus bisoxides) by
alkaline hydrolysis of the insoluble monophosphonium salt – see
Eq. (7) [53]; under these same conditions [54], the bisphospho-
nium cation [Ph3P(CH2)2PPh3]2+ decomposes further to a complex
mixture containing OPPh3, PPh3, C2H4, and C6H6.
d(py)pcp shift downfield by 0.58 and 1.68 ppm, respectively, upon bismethylation.
to 11b–I, 11c–I, and 11c–2I; IR data that show m(P-C) stretches,
and masym and msym C–H stretches of the Me group, in the reported
ranges [22]; and finally, sharp melting points in the 221–234 °C
range, indicative of high purity.
ð7Þ
4. Conclusions
Reaction of dppe with 5 equivalents of CH3I in CDCl3 precipi-
tated a 3:1 mixture of the mono (10b) and bismethylated (11a)
phosphonium salts as determined by 31P{1H} NMR in CD3OD solu-
tion. A similar reaction of d(py)pe with 15 equivalents of CH3I gave
a 1:1 mixture of 10c and 11b, whereas the bismethylated d(py)pcp
(11c) was the sole-product on addition of 10 equivalents of CH3I to
d(pyp)cp in CDCl3; 11a and b were obtained as the sole-products
after heating a mixture of 20 equivalents of CH3I and the diphos-
phine in MeOH at 55 °C for 2 h. NMR data for 11a–c are reported
in Table S7. The conversion of mono- to bis-methylated d(py)pcp
was monitored over ꢂ4 h by observing the decrease in intensity
of the P-CH3 doublet at dH 3.18 with concomitant increase in the
P-CH3 doublet at dH 3.40 (Fig. S7); the doublets become singlets
in the 1H{31P} spectra.
The chemistry of Ni(CO)2L2 complexes, where L is a monoden-
tate (2-pyridyl)phosphine, or L2 is an analogous, bidentate
bis(2-pyridyl)diphosphine, has been studied in detail, following
our preliminary, earlier work on such systems 20 years ago; the
interest in such systems stems from their potential for homoge-
neous catalysis in aqueous media. The carbonyls are now shown
to dissolve in water through protonation of the o-pyridyl N-atoms
with formation of isolable dicationic species. Twenty-two com-
plexes of the type NiL4, are also investigated, L4 being either four
monodentate (2-pyridyl)phosphine ligands, two of the bidentate
(2-pyridyl)diphosphines, or various combinations of these ligands.
All the complexes discussed in the paper have the pyridyl phos-
phine ligands solely P-bonded, and several trends in the 31P{1H}
data depend on the number of pyridyl substituents in the phos-
phine ligands. The only clean reactivity toward small molecules
was oxidative addition of MeI to the NiL4 type complexes, which
yields trans-Ni(Me)(I)L2; this chemistry led us to characterize some
methylphosphonium and bis(methyl)phosphonium iodides.
The pyridyl 1H resonances within the d(py)pe and d(py)pcp sys-
tems shift downfield (by 0.24–1.91 ppm) on going from the neutral
to mono to bismethylated compounds, consistent with decreased
shielding of the 1H nuclei with increasing charge: e.g., the H3–H6
pyridyl proton shifts for the non- and bis-methylated d(py)pe
and d(py)pcp species are listed in Table 7, and the relevant 1H
spectra for all three d(py)pcp species are shown in Fig. S8. 1H,1H
COSY spectra associated with the d(py)pcp and d(py)pe systems
(Figs. S9ꢀS10) were needed to facilitate assignment of the pyridyl
protons.
Acknowledgements
We thank the Natural Sciences and Engineering Research
Council of Canada for financial support via a Discovery Grant.
Other data support the formulations of the isolated bismethyl
salts, 11b and c: good elemental analyses; conductivity values in
MeOH that are consistent with 2:1 electrolytes [17]; EIMS frag-
mentation peaks with isotopic splitting patterns that correspond
Appendix A. Supplementary material
CCDC 1044832 contains the supplementary crystallographic
data for Ni(CO)2[d(py)pcp]. These data can be obtained free of
charge from The Cambridge Crystallographic Data Centre via
www.ccdc.cam.ac.uk/data_request/cif. Supplementary data associ-
ated with this article can be found, in the online version, at http://
dx.doi.org/10.1016/j.ica.2015.04.003.
Table 5
31P{1H} NMR data for monomethylphosphonium iodides 10a–e in C6D6.a
Salt of:
dP P+(3JPP
)
dP P (3JPP
)
PN3
PPh3
10a
10e
10.3
21.6
–
–
References
dppe
d(py)pe
d(py)pcp
10b
10c
10d
26.1 (41.2)
21.7 (38.5)
23.4 (15.3)
ꢀ12.4 (41.3)
ꢀ7.9 (38.5)
ꢀ1.1 (15.2)
[1] M.D. Le Page, B.O. Patrick, S.J. Rettig, B.R. James, Inorg. Chim. Acta 425 (2015)
198.
a
Singlets for 10a and 10e; doublets for 10bꢀd; J values in Hz.
[2] Y. Xie, B.R. James, J. Organomet. Chem. 417 (1991) 277.

288
M.D. Le Page et al. / Inorganica Chimica Acta 431 (2015) 276–288
[3] I.R. Baird, M.B. Smith, B.R. James, Inorg. Chim. Acta 235 (1995) 291.
[4] A. Sacco, P. Mastrorilli, J. Chem. Soc., Dalton Trans. 19 (1994) 2761.
[5] S.D. Ittel, Inorg. Synth. 17 (1977) 117.
[6] (a) S.J. Berners-Price, R.E. Norman, P.J. Sadler, J. Inorg. Biochem. 31 (1987) 197;
(b) M.B. Smith, James, B. R. Unpublished work. University of British Columbia,
1994.
[31] Y. Chen, S. Sakaki, Inorg. Chem. 52 (2013) 13146. and references therein.
[32] (a) M.T. Atlay, M. Preece, G. Strukul, B.R. James, Can. J. Chem. 61 (1983) 1332;
(b) N. Decharin, B.V. Popp, S.S. Stahl, J. Am. Chem. Soc. 133 (2011) 13268;
(c) L. Boisvert, K.L. Goldberg, Acc. Chem. Res. 45 (2012) 899.
[33] P. Giannoccaro, E. Pannacciulli, G. Vasapollo, Inorg. Chim. Acta 96 (1985) 179.
[34] D.J. Darensbourg, J.B. Robertson, D.L. Larkins, J.H. Reibenspies, Inorg. Chem. 38
(1999) 2473.
[35] (a) L. Gonsalvi, M. Peruzzini, in: C. Bianchini, D.J. Cole-Hamilton, P.W.N.M. van
Leeuwen (Eds.), Catalysis by Metal Complexes, vol. 37, Springer, Dordrecht,
2011. pp. 183–212;
(b) L. Gonsalvi, M. Peruzzini, Pure Appl. Chem. 85 (2013) 385.
[36] D. Miksa, T.B. Brill, Ind. Eng. Chem. Res. 41 (2002) 5151.
[37] G.R. van Hecke, W.D. Horrocks Jr., Inorg. Chem. 5 (1966) 1968.
[38] (a) R. Mynott, A. Mollbach, G. Wilke, J. Organomet. Chem. 199 (1980) 107;
(b) K.J. Fisher, E.C. Alyea, Polyhedron 8 (1989) 13.
[7] (a) G. Booth, J. Chatt, J. Chem. Soc. (1965) 3238;
(b) G.R. van Hecke, W.D. Horrocks, Inorg. Chem. 5 (1966) 1968;
(c) P.S. Jarrett, P.J. Sadler, Inorg. Chem. 30 (1991) 2098.
[8] Y. Xie, Ph.D. Dissertation University of British Columbia, 1990.
[9] D. Hellwinkel, Chem. Ber. 98 (1965) 576.
[10] V. Mark, C.H. Dungan, J. Letcher, M.M. Crutchfield, J.R. Van Wazer, 31P Nuclear
Magnetic Resonance, vol. 5, Interscience, New York, 1967.
[11] W. Honggen, W. Ruji, Z. Zhengzhi, W. Xukun, Z. Weijun, Y. Limei, Chem. J. Chin.
Univ. 10 (1989) 809. through G.R. Newkome, Chem. Rev. 93 (1993) 2067 [Ref.
165].
[39] N.D. Jones, K.S. MacFarlane, M.B. Smith, R.P. Schutte, S.J. Rettig, B.R. James,
Inorg. Chem. 38 (1999) 3956.
[40] A. Sen, J. Halpern, Inorg. Chem. 19 (1980) 1073.
[12] Y. Xie, B.R. James, Inorg. Chim. Acta 217 (1994) 209.
[13] P.E. Garrou, Chem. Rev. 81 (1981) 229.
[14] Y. Xie, C.L. Lee, Y.P. Yang, S.J. Rettig, B.R. James, Can. J. Chem. 70 (1992) 751.
[15] A.J. Deeming, M.B. Smith, J. Chem. Soc., Dalton Trans. 22 (1993) 3383.
[16] (a) H.J. Jakobsen, J. Mol. Spectrosc. 34 (1970) 245;
(b) S.L. Schiavo, E. Rotondo, G. Bruno, F. Faraone, Organometallics 10 (1991)
1613.
[41] G. Rastar, M.Sc. Thesis. University of British Columbia, 1989.
[42] G.R. Newkome, D.W. Evans, F.R. Fronczek, Inorg. Chem. 26 (1987) 3500.
[43] L. Sacconi, F. Mani, A. Bencini, Nickel, in: G. Wilkinson, R. Gillard, J. McCleverty
(Eds.), Comprehensive Coordination Chemistry, vol. 5, Pergamon Press, Oxford,
1987. Ch. 50.
[17] W.J. Geary, Coord. Chem. Rev. 7 (1971) 81.
[44] (a) L. Malatesta, S. Cenini, Zerovalent Compounds of Metals, Academic Press,
London, 1974;
[18] C. Krüger, Y. Tsay, Cryst. Struct. Commun. 3 (1974) 455.
[19] G.L. Miessler, D.A. Tarr, Inorganic Chemistry, Prentice Hall, Englewood Cliffs,
1991. p. 335.
(b) J.H. Enemark, R.D. Feltman, Coord. Chem. Rev. 13 (1974) 339;
(c) K.G. Caulton, Coord. Chem. Rev. 14 (1975) 317;
[20] S. Sakaki, K. Kitaura, K. Morokuma, K. Ohkubo, Inorg. Chem. 22 (1983) 104. and
refs. therein.
(d) L. Vaska, Acc. Chem. Res. 9 (1976) 175;
(e) P.V. Yaneff, Coord. Chem. Rev. 23 (1977) 183;
[21] P. Giannoccaro, A. Sacco, G. Vasapollo, Inorg. Chim. Acta 37 (1979) L455.
[22] R.M. Silverstein, G.C. Bassler, T.C. Morrill, Spectrometric Identification of
Organic Compounds, fifth ed., Wiley & Sons, Inc., New York, 1991.
[23] L.D. Quin, A Guide to Organophosphorus Chemistry, Ch. 6, Wiley-Interscience,
New York, 2000.
[24] P. Giannoccaro, G. Vasapollo, A. Sacco, J. Chem. Soc., Chem. Commun. (1980)
1136.
[25] K. Nakamoto, IR Spectra of Inorganic & Coordination Compounds, second ed.,
Wiley Interscience, New York, 1970.
(f) J. Chatt, J.R. Kilworth, R.L. Richards, Chem. Rev. 78 (1978) 589;
(g) D.A. Palmer, R. Van Eldik, Chem. Rev. 83 (1983) 651.
[45] (a) M. Hidai, T. Kashiwagi, T. Ikeuchi, Y. Uchida, J. Organomet. Chem. 30 (1971)
279;
(b) T.T. Tsou, J.K. Kochi, J. Am. Chem. Soc. 101 (1979) 6319.
[46] J. Ashley-Smith, M. Green, F.G.A. Stone, J. Chem. Soc. A. (1969) 3019.
[47] C.A. Tolman, W.C. Seidel, D.H. Gerlach, J. Am. Chem. Soc. 94 (1972) 2669.
[48] J. Chatt, F.A. Hart, J. Am. Chem. Soc. 94 (1960) 1378.
[49] W.D. Horrocks Jr., C.R. Taylor, G.N. LaMar, J. Am. Chem. Soc. 86 (1964) 3031.
[50] S. Otsuka, M. Naruto, T. Yoshida, A. Nakamura, J. Chem. Soc., Chem. Commun.
(1972) 396.
[26] C.J. Pouchert (Ed.), The Aldrich Library of Infrared Spectra, third ed., The
Aldrich Chemical Company, Inc, Milwaukee, 1981.
[27] (a) C. Larpent, R. Dabard, H. Patin, Inorg. Chem. 26 (1987) 2922;
(b) A.M. Bond, J.C. Eklund, V. Tedesco, T. Vu, A.G. Wedd, Inorg. Chem. 37 (1998)
2366.
[28] R. Giannandrea, P. Mastrorilli, C.F. Nobile, M. Dilonardo, Inorg. Chim. Acta 249
(1996) 237.
[29] (a) E.G. Kuntz, O.M. Vittori, J. Mol. Catal. A 129 (1998) 159;
(b) R. Giannandrea, P. Mastrorilli, C.F. Nobile, Inorg. Chim. Acta 284 (1999)
116;
(c) L.B. Belykh, T.V. Dmitrieva, F.K. Shmidt, Russ. J. Coord. Chem. 25 (1999) 494.
[30] B.R. James, Adv. Chem. Ser. 191 (1980) 253.
[51] E.A. Zuech, U.S. Pat. 3,485, 881 (1969), cited in Chem. Abstr. 72 (1970)
P99989a.
[52] E. Carmona, F. Gonzales, M.L. Poveda, J.L. Atwood, R.D. Rogers, J. Chem. Soc.,
Dalton Trans. (1980) 2108.
[53] (a) A.G. Abatjoglou, L.A. Kapicak, Eur. Pat. Appl. EP 72,560 (1983), cited in
Chem. Abstr. 98 (1983) P198452p.;
(b) N.A. Bonderenko, M.V. Rudomino, E.N. Tsvetkov, Synthesis 125 (1991);
(c) I. Brassat, U. Englert, W. Keim, D.P. Keitel, S. Killat, G.-P. Suranna, R. Wang,
Inorg. Chim. Acta 280 (1998) 150.
[54] J.J. Brophy, M.J. Gallagher, Aust. J. Chem. 22 (1969) 1385.