EPR study of the oxidation reaction of nickel(0) phosphine complexes with Lewis and Br?nsted acids

Article abstract of DOI:10.1016/j.ica.2006.01.028

The oxidation of Ni(PPh₃)₄ with BF₃ · OEt₂, H₃CCOOH, and F₃CCOOH, and that of (PPh₃)₂Ni(C₂H₄) with BF₃ · OEt₂ is studied by EPR spectroscopy. The reaction of the Ni(0) complexes with BF₃ · OEt₂ gives Ni(II) complexes with which they react to form Ni(I) compounds with covalent Ni-F and Ni-B bonds that transform with excess BF₃ · OEt₂ into cationic paramagnetic Ni(I) complexes. Acetic acid also adds oxidatively to Ni(PPh₃)₄ to form a Ni(II) complex that reacts further to give Ni(I) hydride and carboxylate complexes. The Ni(I) hydride is transformed by the acid into the Ni(I) carboxylate with release of hydrogen, the amount of which depends on the rate of acid addition. The following Ni(I) complexes are identified in the reaction medium: [Ni(PPh₃)₃]BF₄, [(PPh₃)₂Ni(OEt₂)]BF₄, [(PPh₃)Ni(OEt₂)_n]BF₄, (PPh₃)₂NiBF₂, (PPh₃)₃NiOOCCH₃, and [(PPh₃)₂Ni(OEt₂)P(OEt)₃]BF₄. Oxidation schemes of Ni(0) complexes by Lewis and Br?nsted acids are given.

Full text of DOI:10.1016/j.ica.2006.01.028

Inorganica Chimica Acta 359 (2006) 2314–2320
www.elsevier.com/locate/ica
Note
EPR study of the oxidation reaction of nickel(0) phosphine
complexes with Lewis and Brønsted acids
a,
a
b
V.V. Saraev *, P.B. Kraikivskii a,b, D.A. Matveev , S.N. Zelinskii a,b, K. Lammertsma
a
Department of Chemistry, Irkutsk State University, Str. K. Marksa, 1, Irkutsk 664033, Russian Federation
Faculty of Sciences, Department of Chemistry, Vrije University, De Boelelaan 1083, 1081 HV, Amsterdam, The Netherlands
b
Received 10 November 2005; accepted 24 January 2006
Available online 2 March 2006
Abstract
The oxidation of Ni(PPh3)4 with BF3 Æ OEt2, H3CCOOH, and F3CCOOH, and that of (PPh3)2Ni(C2H4) with BF3 Æ OEt2 is studied by
EPR spectroscopy. The reaction of the Ni(0) complexes with BF3 Æ OEt2 gives Ni(II) complexes with which they react to form Ni(I) com-
pounds with covalent Ni–F and Ni–B bonds that transform with excess BF3 Æ OEt2 into cationic paramagnetic Ni(I) complexes. Acetic
acid also adds oxidatively to Ni(PPh3)4 to form a Ni(II) complex that reacts further to give Ni(I) hydride and carboxylate complexes. The
Ni(I) hydride is transformed by the acid into the Ni(I) carboxylate with release of hydrogen, the amount of which depends on the rate of
acid addition. The following Ni(I) complexes are identified in the reaction medium: [Ni(PPh3)3]BF4, [(PPh3)2Ni(OEt2)]BF4,
[(PPh3)Ni(OEt2)n]BF4, (PPh3)2NiBF2, (PPh3)3NiOOCCH3, and [(PPh3)2Ni(OEt2)P(OEt)3]BF4. Oxidation schemes of Ni(0) complexes
by Lewis and Brønsted acids are given.
ꢀ 2006 Elsevier B.V. All rights reserved.
Keywords: Oxidation; Ni(0) complexes; Lewis and Brønsted acids; Ni(I) complexes; EPR
1. Introduction
In Eq. (1), the Ni(I) complex results directly from a two-
electron oxidative addition of the Lewis acid, giving a
Ni(II) complex as byproduct. In Eq. (2), the Ni(I) complex
is formed by two sequential one-electron transfers. At-
tempts have been made to establish which oxidation pro-
cess is applicable with the Lewis acids Al(Et)Cl2 [7],
AlBr3 [8] BBr3, and B(Ph)2Br [9], but so far to no avail.
Also the mechanism and products resulting from the olig-
omerization with Ni(0) complexes and Brønsted acids re-
main elusive although it is believed that intermediate
Ni(II) hydride complexes are involved (Eq. (3)), based on
their NMR identification in model systems [10,11]. Assum-
ing a two-electron oxidative addition, the Ni(I) complexes
are then formed by comproportionation
Ni(0) phosphine complexes in combination with Lewis
(AXn) and Brønsted (HX) acids exhibit high activity in
the oligomerization of unsaturated hydrocarbons [1–3].
Lewis acids quantitatively oxidize Ni(0) to Ni(I) [4], so that
the coordinatively unsaturated cationic Ni(I) complexes
are, in fact, the active catalysts for the low molecular olefin
oligomerization [5]. According to the literature [6], this oxi-
dation can proceed by different directions that are summa-
rized in Eqs. (1) and (2); for simplicity the phosphine
ligands are not shown:
Nið0Þ
X
AXnꢀ1
Nið0Þ þ AXn ! Ni—
Nið0Þ þ AXn ! NiþAXꢀ ! Niþ þ A Xnꢀ1 þ Xꢀ
ꢀ! NiX þ NiAXnꢀ1
ð1Þ
ð2Þ
ꢁ
Â
Ã
H
X
Nið0Þ þ HX ! Ni—
ð3Þ
n
In the present study, we report EPR data on (1) the oxi-
dation of tetrakis(triphenylphosphine)nickel(0) with boron
trifluoride etherate, and acetic and trifluoroacetic acids,
*
Corresponding author. Fax: +7 3952 42 59 35.
E-mail address: saraev@chem.isu.ru (V.V. Saraev).
0020-1693/$ - see front matter ꢀ 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.ica.2006.01.028

V.V. Saraev et al. / Inorganica Chimica Acta 359 (2006) 2314–2320
2315
and (2) the oxidation of bis(triphenylphosphine)(p-ethyl-
ene)nickel(0) with boron trifluoride etherate to provide fur-
ther mechanistic insights.
tilled water [17], and also stored under argon in a polyeth-
ylene vessel and used within 2 h.
EPR studies were carried out using a PS-100X spectrom-
eter with an operating frequency of 9.6 GHz. Mn(II) in
MgO and diphenylpicrylhydrazyl were used as standards.
EPR spectra were recorded for samples frozen in a glass
ampoule at ꢀ170 ꢁC (liquid nitrogen vapors). These sam-
ples were collected with a syringe from the reaction vessel
(23 ꢁC) under argon and frozen in about 15 s. To estimate
the concentrations of paramagnetic Ni(I) complexes in the
samples, double integrated intensities of the EPR signals
of the samples were taken and compared to those of the
individual [Ni(PPh3)3]BF4 complex in diethyl ether. The
EPR spectra were simulated with our published program
[18], in which the hyperfine interaction (HFI) is limited to
the second-order term and where the main axes of the g-ten-
sor and the HFI tensors coincide. The EPR parameters of
all experimental signals shown in Fig. 1 were examined by
computer modeling and are summarized in Table 1.
19F NMR spectra were recorded at 25 ꢁC on a Varian
VXR-500S spectrometer using sealed ampoules and
F3CCOOH as standard.
2. Experimental
All operations were performed under moisture and oxy-
gen free argon using standard Schlenk-line technique at
t = 23 ꢁC under atmospheric pressure. Gas chromatogra-
phy grade argon was purified [12] and checked by the
AlEt3 test. Vacuum-calcined tubes were used as reaction
vessels.
Toluene, benzene, and hexane (Merck) were refluxed
over sodium for 10 h, distilled to glass vessels that were
coated with a sodium mirror, and stored at 20 kPa gauge
argon pressure. Acetylacetone and trifluoroacetic acid
(Merck) were distilled under argon before use. Acetic acid
(Merck) was additionally dried by the usual freezing tech-
nique. Boron trifluoride etherate (Merck) was distilled over
lithium hydride under argon less than 3 h before use and
contained less than 0.05% of acidic protons according to
NMR. Tetramethylpiperidinyloxy (TEMPO, Aldrich) has
a 37–38 ꢁC melting range.
Ni(PPh3)4 [13], (PPh3)2Ni(C2H4) [14], and [Ni(PPh3)3]-
BF4 [15] were synthesized by known methods. The initial
concentration of the Ni(0) complex in a toluene solution
was 10ꢀ2 mol/l. Reactions between the Ni(0) complex
and the Lewis and Brønsted acids were carried out at
23 ꢁC. Boron trifluoride etherate was introduced into the
reaction mixture in 0.2 mol aliquots with respect to nickel.
4. Results and discussion
4.1. Boron trifluoride etherate
[Ni(PPh3)3]BF4: Diamagnetic Ni(PPh3)4 is known to
partly dissociate in toluene [6], enabling association with
Lewis acids. Resonance absorption is not observed upon
addition of the first aliquots of boron trifluoride etherate
until the B:Ni molar ratio is over 2. The resulting EPR res-
onance (labeled 1 in Fig. 1) is due to the tricoordinate com-
plex [Ni(PPh3)3]BF4 [8,18], whose saturated toluene
solution is greenish-brown in color. This signal 1 reaches
a maximum at a B:Ni ratio of 4, whereupon its intensity
decreases. We note that the variation in intensity of this sig-
nal is within the integration error and P90% of that if all
the nickel were to become paramagnetic. The EPR spectra
are the same for all samples taken from the reaction mix-
ture at intervals of 30 s. Hence, the Ni(0) oxidation with
BF3 Æ OEt2 must proceed rather fast (<30 s).
3. Method
The reaction between the Ni(PPh3)4 complex and acetic
acid was conducted in a thermostated reaction vessel
(23 ꢁC), plugged by a rubber stopper pierced with a syringe
needle and thermally connected to a U-shaped gas measur-
ing burette. A toluene solution of acetic acid was injected
by a syringe (of 1 ml) into a vigorously stirred toluene solu-
tion of Ni(PPh3)4 in 0.01 ml portions (1 s) at intervals of
19 s, i.e., at an average volume rate of 5 · 10ꢀ4 ml/s. The
syringe was filled with acetic acid in a 1-, 2-, 6-, 20-, 200-
fold molar excess relative to Ni(0) and diluted with toluene
to the desired 1 ml of volume. Hence, the average relative
rates of acetic acid addition were 0.5, 1, 3, 10, and
The 19F NMR spectrum of the reaction mixture shows a
resonance at 25.58 ppm, which is assigned to F2BBF2 [19];
line broadening due to the presence of paramagnetic spe-
cies masked the fine structure and hence its coupling con-
stants could not be determined. The d (19F) intensity
peaked 1–2 min after the start of the reaction and then
diminished, indicating the depletion of F2BBF2.
Because HF and H2O are potential contaminants in
boron trifluoride etherate that might participate in the
oxidation of Ni(0) to Ni(I), we conducted independent
test on the interaction of Ni(PPh3)4 with mixtures of
BF3 Æ OEt2 + HBF4 and BF3 Æ OEt2 + HBF3OH in which
the contents of the Brønsted acid were varied from 0%
to 100%. However, as the content of HBF4 or HBF3OH
was increased the Ni(I) EPR signal intensity reduced.
100 ½CH3COOHꢂ ꢃ 10ꢀ3
s
ꢀ1, respectively.
½Niꢂ
For the acid HBF4, we used its toluene complex that
contained >48% of HBF4 (by chromatography) and was
prepared by the method reported in Ref. [16]. Acetylace-
tone (0.352 g, 3.52 mmol) was added dropwise (2 min) to
a vigorously stirred solution of 10 ml boron trifluoride eth-
erate (1 g, 7.05 mmol, toluene), after which the reaction
mixture was separated into two layers. The lower one, con-
sisting of the HBF4-toluene complex (P98% by NMR
spectroscopy), was stored under argon in a polyethylene
vessel and used within 2 h. HBF3OH was prepared from
equimolar amounts of boron trifluoride etherate and dis-

2316
V.V. Saraev et al. / Inorganica Chimica Acta 359 (2006) 2314–2320
Fig. 1. Experimental (a) and model (b) EPR spectra of the Ni(I) complexes: (1) [Ni(PPh3)3]BF4, (2) [(PPh3)2Ni(OEt2)]BF4, (3) [(PPh3)Ni(OEt2)n]BF4, (4)
(PPh3)2NiBF2, (5) mixture of complexes, (6) [(PPh3)2Ni(OEt2)P(OEt)3]BF4, and (7) (PPh3)3NiOOCCH3 in toluene at T = 77 K.
Table 1
Parameters of EPR spectra for Ni(I) complexes
Signal
1
Complex
gi
g^
Ai (mT)
A^ (mT)
Ref.
[8]
[Ni(PPh3)3]BF4
2.385z
2.12y
2.07x
6.1 (1P)z
<3.2 (2P)z
6.4 (1P)y
<3.2 (2P)y
8.1 (1P)x
<2.0 (2P)x
2
3
4
[(PPh3)2Ni(OEt2)]BF4
[(PPh3)Ni(OEt2)n]BF4
(PPh3)2NiBF2
2.45
2.03z
2.22
2.11
6.0 (1P)
3.7 (1P0)
7.6 (1P)
2.9 (1P0)
2.14y
2.30x
5.7 (1P)z
7.4 (1P)y
4.9 (1P)x
2.08
3.0 (2P)
3.0 (1B)
3.1 (2P)
3.1 (1B)
5
6
7
Mixture of complexes
[(PPh3)2Ni(OEt2)P(OEt)3]BF4
(PPh3)3NiOOCCH3
No EPR parameters determined
2.01
2.04
2.21
2.35
19.5 (1P)
17.0 (1P)
No Ni(I) EPR signals were even observed throughout the
molar ratio range 1 < B:Ni <80 with a 100% content of
HBF4 or HBF3OH. Consequently, the formation of
Ni(I) complexes in the system Ni(PPh3)4/BF3 Æ OEt2 can-
not be ascribed to the presence of HF and H2O as
contaminants.
[(PPh3)2Ni(OEt2)]BF4: With increasing the amount of
boron trifluoride etherate in the Ni(PPh3)4/BF3 Æ OEt2 sys-
tem, signal 1 decreases in intensity and another resonance
emerges (labeled 2 in Fig. 1) that has its maximum at a
B:Ni ratio of 50. The four equidistant components of the
hyperfine structure (HFS) resolved in signal 2 are due
either to two nonequivalent 31P (IP = 1/2) nuclei or to
one 11B (IB = 3/2) nucleus. To distinguish between these
two possibilities, EPR spectra were obtained for the com-
plexes formed in the reaction of BF3 Æ OEt2 with
[Ni(PPh3)3]BF4 in toluene to show the same signals. In this
case, the formation of a Ni–B valence bond is highly
improbable, so that the four-component HFS of signal 2
is attributed to two nonequivalent 31P nuclei. The ratio
between the g-factor components (gi > g^) and the non-
equivalence of the 31P nuclei is typical for tricoordinate
Ni(I) complexes; these have a ground pseudodegenerate
electronic state [18]. Consequently, signal 2 can be assigned
to [(PPh3)2Ni(OEt2)]BF4.
[(PPh3)Ni(OEt2)2]BF4: Increasing the amount of
boron trifluoride etherate in the Ni(PPh3)4/BF3 Æ OEt2 sys-
tem leads to the disappearance of signal 2 and to the emer-
gence of yet another resonance (labeled 3 in Fig. 1), which
has a maximum intensity at a B:Ni ratio of 80, while fur-

V.V. Saraev et al. / Inorganica Chimica Acta 359 (2006) 2314–2320
2317
ther addition of boron trifluoride etherate leads to discolor-
ation of the toluene solution, precipitation of finely divided
metallic nickel, and the appearance of a broad EPR signal
due to ferromagnetic nickel. Similar signal transformations
are observed if BF3 Æ OEt2 is added to a toluene solution of
Ni(PPh3)4 in portions corresponding to B:Ni molar ratios
of 3, 50, and 80. Signal 3 has a HFS with a well-resolved
31P nucleus that indicates the presence of one phosphine
ligand in the paramagnetic complex. Therefore, the succes-
sive change in EPR resonance absorptions from 1 ! 2 ! 3
that is induced by excess BF3 Æ OEt2 corresponds to the
sequential elimination of organophosphorus ligands from
the Ni(I) coordination sphere from 3 ! 2 ! 1. We note
that the activity of the Ni(PPh3)4/BF3 Æ OEt2 system in
low molecular ethylene oligomerization peaks within a
B:Ni ratio of 60–70 [15].
To establish whether free radicals are involved in the
formation of the cationic Ni(I) complexes, e.g., through
one-electron oxidation, reactions between Ni(PPh3)4 and
BF3 Æ OEt2 were performed in the presence of the free rad-
ical spin trap TEMPO (mole ratio TEMPO:Ni = 0.01).
However as virtually no effect was exerted on the signals
of both the Ni(I) complexes and the TEMPO radical, we
conclude that the oxidation of Ni(0) phosphine complexes
does not involve free radicals and hence consider the one-
electron mechanism for this oxidative system improbable
under the conditions used.
in the B:Ni ratio of 2–8 may be due to dimerization of the
paramagnetic complexes and their subsequent dissociation
caused by excess BF3 Æ OEt2. To verify this assumption, tri-
ethylphosphite was introduced in a P:Ni molar ratio of 1
after signal 4 had disappeared, because this ligand is capa-
ble of destructing the Ni(I) dimers. Indeed, a new EPR res-
onance resulted (signal 6 in Fig. 1) that is characteristic for
a tetrahedral Ni(I) complex containing one phosphite
ligand [20] and hence, signal 6 is assigned to [(PPh3)2-
Ni(OEt2)P(OEt)3]BF4. Unlike triethylphosphite, triphenyl-
phosphine did not destroy the Ni(I) dimers under the same
conditions, even when introduced in a 3-fold excess.
In contrast to the Ni(PPh)4 system, (PPh3)2Ni(C2H4) is
highly sensitive to the manner in which BF3 Æ OEt2 is intro-
duced. For instance, addition in portions corresponding to
a B:Ni ratio of 3, 6, and 9 gives subsequently a moderately
strong signal 2 that transforms into signal 3, whose
intensity remains virtually unchanged with the higher
BF3 Æ OEt2 content.
In summary, the experimental data provide evidence
that the Ni(0) complexes, Ni(PPh3)4 and (PPh3)2Ni(C2H4),
can be oxidized by addition of Lewis acids with kinetic fac-
tors influencing the composition of the redox products. The
sequential steps for the formation of Ni(I) complexes in the
Ni(PPh3)4/BF3 Æ OEt2 catalytic system are summarized in
Scheme 1.
In Scheme 1 the prospective Ni(II) and Ni(I) complexes
are shown in italics.
To further explore the oxidative reaction, we carried out
experiments with bis(triphenylphosphine)(p-ethylene)-
nickel(0) in which two phosphine ligands of Ni(PPh)4 have
been replaced by a p-complexing ethylene group. This sys-
tem behaves differently as the addition of even the first ali-
quots of boron trifluoride etherate (B:Ni ratio of 0.2) to the
toluene solution of (PPh3)2Ni(C2H4), which becomes yel-
low-brown, generates a new EPR resonance (labeled 4 in
Fig. 1) whose intensity increases with an increasing concen-
tration of BF3 Æ OEt2 to reach a maximum at a B:Ni ratio
of 1 (dark green; no precipitate). At this ratio, the intensity
of the signal does not exceed 30% of the value calculated
for the complete conversion of nickel to the paramagnetic
state. At a B:Ni ratio of 2, signal 4 has disappeared fully
and only at a B:Ni ratio of 8 does a strong new EPR reso-
nance reappear (labeled 5 in Fig 1), which is a superposi-
tion of several signals with only that of signal 3 identifiable.
(PPh3)2NiBF2: Signal 4 exhibits a comparatively com-
plicated HFS, which is attributed to one 11B and two equiv-
alent 31P nuclei as based on extensive comparisons with
theoretical spectra calculated with the program described
in Ref. [18]. By further taking into account the ratio between
the g-factor components (gi > g^), signal 4 is assigned to the
tricoordinate (PPh3)2NiBF2 structure with a Ni–B valence
bond. This interpretation is supported by the reported syn-
thesis of the Ni–B containing Ni(I) complex {(PPh3)2NiB-
Ph2 Æ 1/2OEt2}n (n P 2) that reportedly results from the
BBrPh2-mediated oxidation of (PPh3)2Ni(C2H4) [9].
The first step consists of the oxidative addition to form
the Ni(II) complex, which then reacts with the Ni(0) pre-
cursor to give Ni(I) complexes with covalent Ni–F and
Ni–B bonds. Such processes are known to underlie the syn-
thesis of, e.g., (PR3)nNiX (X = Cl, Br, J) [21]. The Ni(I)
borides in Scheme 1 enter the metathesis reaction as boron
trifluoride etherate to be converted to Ni(I) fluorides. In the
presence of excess BF3 Æ OEt2, these are transformed into
the monomeric paramagnetic cationic Ni(I) complexes that
contain several phosphine ligands. The dimerization of the
Ni(I) fluoro complexes is hampered, because the oxidation
occurs only in the presence of an excess of BF3 Æ OEt2 (B:
Ni > 2). The substitution of phosphine ligands for diethyl
ether molecules is an equilibrium process that is supported
by the significant excess of BF3 Æ OEt2 that is needed rela-
tive to Ni(I). The overall reaction of the phosphine com-
plex Ni(PPh3)4 with BF3 Æ OEt2 is given in Scheme 1.
Equilibrium signs are used to take into account both the
phosphine ligand substitution process and the potential
oxidative addition of F2BBF2 to the Ni(0) complex [22].
4.2. Acetic acid
(PPh3)3NiOOCCH3: The addition of acetic acid to a
toluene solution of diamagnetic Ni(PPh3)4 results in the
evolution of hydrogen, and an EPR signal (labeled 7 in
Fig. 1) whose intensity maximizes at a CH3COOH:Ni ratio
of 3. Further addition of acetic acid up to a CH3COOH:Ni
ratio of 150 does not cause any visible change in the EPR
[(PPh3)2Ni(OEt2)P(OEt)3]BF4: The disappearance of
signal 4 at a B:Ni ratio of 2 and the appearance of signal 5

2318
V.V. Saraev et al. / Inorganica Chimica Acta 359 (2006) 2314–2320
F
(PPh3 )4 Ni + 3BF ⋅OEt2 →(PPh3 )2 Ni
+ 2BF ⋅ PPh3 + 3OEt2
3
3
BF2
F
(PPh3 )2 Ni
+ (PPh3 )4 Ni → (PPh3 )3 NiF + (PPh3 )3 NiBF
2
BF2
(PPh3 )3 NiBF2 + BF3 ⋅ OEt2 → (PPh3 )3 NiF + F2 BBF + OEt2
2
+ (PPh3 )3 NiF → ((PPh3 )3 NiF)2
+ BF ⋅OEt2 →[(PPh3 )3 Ni]BF + OEt2
⎧
⎨
⎩
(PPh ) NiF
3
3
3
4
[(PPh3 )3 Ni]BF + nBF ⋅OEt2 ↔[(PPh3 )3−n Ni(OEt2 )n ]BF4 + nBF ⋅ PPh3
4
3
3
2(PPh3 )4 Ni + 2(3 + n)BF ⋅OEt2 ↔
3
↔ 2[(PPh3 )3−n Ni(OEt2 )n ]BF + 2(1+ n)BF ⋅ PPh3 + F2 BBF + 6OEt2
4
3
2
where n = 0, 1, 2
Scheme 1.
signal, indicating that the formed Ni(I) complex is stable in
an acidic medium. The observed EPR signal is typical for a
tetrahedral Ni(I) complex with trigonal symmetry [20] and
is therefore assigned to (PPh3)3NiOOCCH3. To verify this
assumption, we synthesized this complex by reacting
(PPh3)3NiCl with sodium acetate (Eq. (4)) and obtained
the same EPR signal 7, but could not determine the HFS
constants for the phosphorus nuclei due to line broadening,
which is likely the result of the dynamic character of the
Jahn–Teller effect in the tetracoordinate Ni(I) complex [20]
ðPPh3Þ3NiCl þ NaOOCCH3
! ðPPh3Þ3NiOOCCH3 þ NaCl
ð4Þ
The integral intensity of the EPR signal and the amount
of released hydrogen depend on the rate at which acetic
acid is added to the toluene solution of Ni(PPh3)4. The
slower the rate at which the acid is added, the greater the
amount of hydrogen is formed until a CH3COOH:Ni ratio
of 3 is reached after which its formation ceases.
Fig. 2. Plot of the amount of released hydrogen vs. the rate of acetic acid
addition to a toluene solution of the Ni(PPh3)4 complex at T = 298 K.
the rate is increased 100-fold. The amount of released
hydrogen is proportional to the Ni(I) acetate concentration.
Fig. 2 shows the relationship between the amount of
released hydrogen and the rate of acetic acid addition. On
adding a toluene solution of acetic acid to Ni(PPh3)4 at a
4.3. Trifluoroacetic acid
rate of ½CH3COOHꢂ ꢃ 10ꢀ3
s
ꢀ1, about 0.47 mol of molecular
(PPh3)3NiOOCCF3: To establish whether a stronger
Brønsted acid affects the formation of the Ni(I) complexes,
we oxidized Ni(PPh3)4 with trifluoroacetic acid. The EPR
½Niꢂ
hydrogen is released with respect to 1 mol of the starting
Ni(0) complex, whereas this is only 0.09 mol of H2 when

V.V. Saraev et al. / Inorganica Chimica Acta 359 (2006) 2314–2320
2319
H
(PPh3 )4 Ni + 3RCOOH →(PPh3 )2 Ni
+ 2RCOOH⋅ PPh3
OOCR
H
(PPh3 )2 Ni
+ (PPh3 )4 Ni → (PPh3 )3 NiH + (PPh3 )3 NiOOCR
OOCR
+ (PPh3 )3 NiH → 2(PPh3 )3 Ni + H2
⎧
⎨
+ RCOOH → (PPh 3 )3 NiOOCR + H2
⎩
(PPh ) NiH
3
3
2(PPh3 )4 Ni + 4RCOOH → 2(PPh3 )3 NiOOCR + 2RCOOH ⋅ PPh3 + H2
where R = CH3, CF3.
Scheme 2.
signal of the resulting Ni(I) complex does not differ signif-
icantly in shape from that of the Ni(I) acetate complex
obtained with acetic acid. The parameters of the EPR
signal of the trifluoroacetate complex (gi = 2.04,
g^ = 2.35) correspond to the tetracoordinated Ni-structure
(PPh3)3NiOOCCF3. We note that oxidative addition with
trifluoroacetic acid occurs 5–7 times more rapidly than
for acetic acid. Scheme 2 summarizes the sequence of steps
that likely govern the oxidative interaction of both acetic
and trifluoroacetic acids with Ni(PPh3)4.
(1) and (3)), (2) the Ni(I) complexes result from compropor-
tionation of Ni(0) and Ni(II) complexes, and (3) the relative
concentration of the Ni(I) complexes depends on kinetic
factors and increases as the rate of comproportionation
increases relative to that of the oxidative addition.
This process of oxidative addition of boron trifluoride
etherate to Ni(0) complexes may similarly govern the cor-
responding addition using organoaluminum compounds
[23].
In Scheme 2, the prospective Ni(II) and Ni(I) complexes
are shown in italics.
Acknowledgements
The first step is the well-established oxidative addition
of the acid to form the Ni(II) carboxylate hydride [6],
which subsequently comproportionates in the presence of
Ni(PPh3)4 to give Ni(I) hydride and carboxylate complexes.
The Ni(I) hydride, being unstable in an acidic medium, is
transformed by the acid into the Ni(I) carboxylate or
decomposes bimolecularly to reform the Ni(0) complex,
which reenters the oxidative addition cycle. Any decompo-
sition of Ni(I) hydride is accompanied by evolution of
hydrogen. The overall reaction summarizing the process
of the slow addition of an organic acid to phosphine com-
plex Ni(PPh3)4 is given in Scheme 2.
We assume that the formation of Ni(I) complexes is
influenced by the rate of acid addition, which, when slow,
enables the intermittently formed Ni(II) carboxylate
hydride to comproportionate in the presence of unreacted
Ni(PPh3)4.
This work was supported by the Netherlands Organiza-
tion for Scientific Research (NWO) and the Russian Foun-
dation for Basic Research (RFBR) (Project NWO – RFBR
No. 03-03-89007), and by the Ministry of education of
Russian Federation (Project No. E02-5.0-131).
References
[1] W. Keim, J. Mol. Catal. 52 (1989) 19.
[2] K. Nomura, M. Ishino, M. Hozama, G. Suzukamo, J. Mol. Catal.
Sect. A 126 (1997) L93.
[3] C. Carlini, M. Marchionna, A.R. Galletti, G. Sbrana, J. Mol. Catal.
Sect. A 169 (2001) 79.
[4] V.V. Saraev, F.K. Shmidt, J. Mol. Catal. Sect. A 158 (2000) 149.
[5] V.V. Saraev, P.B. Kraikivskii, S.N. Zelinskii, G.V. Ratovskii, V.S.
Tkach, F.K. Shmidt, Rus. J. Coord. Chem. 27 (2001) 757.
[6] J.P. Collman, L.S. Hegedus, J.R. Norton, R.G. Finke, Principles and
Applications of Organotransition Metal Chemistry, University Sci-
ence Books, Mill Valley, CA, 1987.
[7] F.K. Shmidt, L.V. Mironova, V.V. Saraev, V.A. Gruznykh, T.V.
Dmitrieva, G.V. Ratovskii, Kinet. Katal. 20 (1979) 622.
[8] P.B. Kraikivskii, V.V. Saraev, D.A. Matveev, S.N. Zelinskii, V.S.
Tkach, Rus. J. Coord. Chem. 29 (2003) 431.
[9] C.S. Cundy, H. Noth, J. Organomet. Chem. 30 (1971) 135.
[10] C.A. Tolman, J. Am. Chem. Soc. 92 (1970) 4217.
[11] W. Keim, Angew. Chem. 102 (1990) 251.
5. Conclusions
Our study of the oxidation of Ni(0) phosphine complexes
with Lewis and Brønsted acids (boron trifluoride etherate,
acetic and trifluoroacetic acids) shows that (1) the reaction
proceeds through a two-electron oxidative addition (Eqs.

2320
V.V. Saraev et al. / Inorganica Chimica Acta 359 (2006) 2314–2320
[12] D.F. Shriver, The Manipulation of Air-Sensitive Compounds, New
York, 1969.
[18] V.V. Saraev, P.B. Kraikivskii, P.G. Lazarev, G. Mjagmarsuren, V.S.
Tkach, F.K. Shmidt, Koord. Khim. 22 (1996) 648.
[13] G. Wilke, Angew. Chem. 73 (1961) 33.
[14] J. Ashley-Smith, M. Green, F.G.A. Stone, J. Chem. Soc. Sect. A 8
(1969) 3019.
[19] W. Haubold, P. Jacob, Z. Anorg. Allg. Chem. 507 (1983) 231.
[20] V.V. Saraev, P.B. Kraikivskii, P.G. Lazarev, G. Mjagmarsuren, V.S.
Tkach, F.K. Shmidt, Koord. Khim. 22 (1996) 655.
[15] V.S. Tkach, V.A. Gruznykh, M. Gomboogiin, L.B. Belykh, V.V.
Saraev, F.K. Shmidt, Koord. Khim. 20 (1994) 618.
[16] N.M.D. Brown, P. Blandon, J. Chem. Soc. A, Inorg. Phys. Theor.
(1969) 526.
[21] P. Heimbach, Angew. Chem. 76 (1964) 586.
[22] N. Lu, N.C. Norman, A.G. Orpen, M.J. Quayle, P.L. Timms, G.R.
Whittell, J. Chem. Soc., Dalton Trans. 22 (2000) 4032.
[23] J.I. Eisch, M. Xin Ma Singh, G. Wilke, J. Organomet. Chem. 527
(1997) 301.
[17] R.J. Gillespie, J.S. Hartman, Can. J. Chem. 45 (1967) 859.