Synthesis of [Ni(η2-CH2C6H 4R-4){PPh(CH2CH2PPh2) 2}]+ (R = H, Me or MeO) and protonation reactions with HCl

Article abstract of DOI:10.1016/j.jorganchem.2005.01.034

The complexes [Ni(η²-CH₂C₆H ₄R-4)(triphos)]BPh₄ {R = H, Me or MeO; triphos = PhP(CH₂CH₂PPh₂)₂} have been prepared and characterised by spectroscopy and X-ray crystallography. In all cases the coordination geometry of the nickel is best described as square-planar with an η²-benzyl ligand occupying one of the positions. The orientation of the η²-benzyl ligand is dictated by the steric restrictions imposed by the phenyl groups on the triphos ligand, so that the phenyl group on the unique secondary phosphorus and the aromatic group of the benzyl ligand (which are trans to one another) are oriented in the same direction. [Ni(η²-CH₂C₆H₄R-4)(triphos)] ⁺ react with an excess of anhydrous HCl in MeCN to form [NiCl(triphos)]⁺ (characterised as the [BPh₄]^- salt by X-ray crystallography) and the corresponding substituted toluene. The kinetics of the reaction of all [Ni(η²-CH₂C ₆H₄R-4)(triphos)]⁺ and HCl in the presence of Cl^- have been determined using stopped-flow spectrophotometry. All reactions exhibit a first-order dependence on the concentration of complex and a first-order dependence on the ratio [HCl]/[Cl^-]. Varying the 4-R-substituent on the benzyl ligand shows that electron-withdrawing substituents facilitate the rate of the reaction. It is proposed that the mechanism involves initial rapid protonation at the nickel to form [NiH(η²-CH₂C₆H₄R-4)(triphos)] ²⁺, followed by intramolecular proton migration from nickel to carbon to yield the products.

Full text of DOI:10.1016/j.jorganchem.2005.01.034

Journal of Organometallic Chemistry 690 (2005) 1763–1771
www.elsevier.com/locate/jorganchem
Synthesis of [Ni(g2-CH2C6H4R-4){PPh(CH2CH2PPh2)2}]+
(R = H, Me or MeO) and protonation reactions with HCl
Valerie Autissier, Ellie Brockman, William Clegg, Ross W. Harrington,
*
Richard A. Henderson
Department of Chemistry, School of Natural Sciences, University of Newcastle, Newcastle upon Tyne NE1 7RU, UK
Received 7 December 2004; accepted 14 January 2005
Available online 8 March 2005
Abstract
The complexes [Ni(g2-CH2C6H4R-4)(triphos)]BPh4 {R = H, Me or MeO; triphos = PhP(CH2CH2PPh2)2} have been prepared
and characterised by spectroscopy and X-ray crystallography. In all cases the coordination geometry of the nickel is best described
as square-planar with an g2-benzyl ligand occupying one of the positions. The orientation of the g2-benzyl ligand is dictated by the
steric restrictions imposed by the phenyl groups on the triphos ligand, so that the phenyl group on the unique secondary phosphorus
and the aromatic group of the benzyl ligand (which are trans to one another) are oriented in the same direction. [Ni(g2-CH2C6H4R-
4)(triphos)]+ react with an excess of anhydrous HCl in MeCN to form [NiCl(triphos)]+ (characterised as the [BPh4]ꢀ salt by X-ray
crystallography) and the corresponding substituted toluene. The kinetics of the reaction of all [Ni(g2-CH2C6H4R-4)(triphos)]+ and
HCl in the presence of Clꢀ have been determined using stopped-flow spectrophotometry. All reactions exhibit a first-order depen-
dence on the concentration of complex and a first-order dependence on the ratio [HCl]/[Clꢀ]. Varying the 4-R-substituent on the
benzyl ligand shows that electron-withdrawing substituents facilitate the rate of the reaction. It is proposed that the mechanism
involves initial rapid protonation at the nickel to form [NiH(g2-CH2C6H4R-4)(triphos)]2+, followed by intramolecular proton
migration from nickel to carbon to yield the products.
ꢀ 2005 Elsevier B.V. All rights reserved.
Keywords: Nickel; Benzyl; Protonation; Mechanism
1. Introduction
which control the rate of protonation of metal com-
plexes. Understanding these factors will allow us to
predict the sites of protonation in new reactions.
(ii) Elucidation of the kinetically favoured site of pro-
tonation of the complex, in particular, investigating
whether a proton binds directly to the site identified in
the product, or whether the proton binds initially to
some other site (kinetically controlled product) and then
moves to the ultimate site (thermodynamically con-
trolled product). (iii) Probing the pathways by which
protons move between the kinetically favoured and the
thermodynamically favoured sites. These pathways can
be intramolecular or intermolecular.
Studies on the kinetics of protonation of transition
metal complexes are of fundamental interest in under-
standing the mechanisms of proton transfer to inorganic
complexes as well as being relevant to the action of cer-
tain industrially important catalysis, such as hydrocya-
nation [1]. Whilst there have been significant advances
in understanding the mechanisms of protonation of
transition metal complexes, the following aspects are
still unresolved [2]. (i) The electronic and steric factors
Earlier work from our laboratories focussed on the
protonation of complexes based on the substitutionally
robust {M(Ph2PCH2CH2PPh2)2} (M = Mo or W) core,
*
Corresponding author. Tel.: +44 191 222 6636; fax: +44 191 222
6929.
E-mail address: r.a.henderson@ncl.ac.nk (R.A. Henderson).
0022-328X/$ - see front matter ꢀ 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.jorganchem.2005.01.034

1764
V. Autissier et al. / Journal of Organometallic Chemistry 690 (2005) 1763–1771
containing a variety of different ligands such as N2,
RCN, RNC, alkyne or alkene [2,3]. More recently we
have been studying the protonation reactions of com-
plexes based on the {Ni(triphos)}+ {triphos = PhP
(CH2CH2PPh2)2} site, containing methyl [4], g3-allyl
[5] or thiolate [6] ligands. In particular, a study of the
kinetics of the the reaction between [NiMe(triphos)]+
and HCl to form methane {Eq. (1)}, indicates that
although initial protonation occurs at the nickel, forma-
tion of methane has to await the subsequent direct pro-
tonation of the methyl ligand [4]. Intramolecular
migration of hydrogen from nickel to methyl appears
to be slow.
4-MeOC6H4CH2Cl and 4-MeC6H4CH2Cl. The com-
pounds, RC6H4CH2MgCl (R = Me or MeO) [7] and
[NiCl(triphos)]BPh4 [5] were prepared by methods re-
ported in the literature. [NEt4]Cl Æ H2O was purchased
from Aldrich and dried by heating to 80 ꢁC in vacuo
overnight.
The compounds were characterised by 1H NMR
spectroscopy, 31P NMR spectroscopy, elemental analy-
1
sis and X-ray crystal structure determination. H NMR
and 31P NMR spectroscopy was carried out on a
300 MHz Bruker NMR spectrometer fitted with a wide
¨
bore to hold larger NMR tubes, operating at
121.5 MHz. The samples were dissolved in deuterated,
degassed solvents. Due to the low solubility and yield
of the products, 31P NMR spectroscopy needed to be
performed overnight in order to increase the signal to
noise ratio. The elemental analysis was carried out
using a Carlo Erba elemental analyzer, model 1106.
All X-ray crystal structures were obtained by the New-
castle University X-ray Crystallography Research
Group.
½NiMeðtriphosÞꢁþ þ HCl ! ½NiClðtriphosÞꢁþ þ CH4
ð1Þ
In order to more fully understand the factors affect-
ing the protonation of nickel alkyl complexes we have
prepared and structurally characterized [Ni(g2-CH2C6-
H4R-4)(triphos)]+ (R = H, Me or MeO) and studied
the kinetics of the reactions of these complexes with
HCl in MeCN.
2.1. Preparation of [Ni(g2-CH2C6H4R-4)(triphos)]-
BPh4 (R = H, Me or MeO)
2. Experimental
The series of complexes [Ni(g2-CH2C6H4R-4)(tri-
phos))]BPh4 were all prepared by the same method.
The complexes were characterised by elemental and
spectroscopic analysis as shown in Table 1, and by X-
ray crystallography. A typical preparation is described
below for [Ni(g2-CH2C6H4OMe-4)(triphos)]BPh4.
A 1.5 M solution of 4-MeOC6H4CH2MgCl was pre-
pared [7] in a two-neck, round-bottomed flask fitted
with an efficient condenser and a dropping funnel by
the slow addition of 4-MeOC6H4CH2Cl (25 g,
0.16 mol) in THF (50 mL) to a stirred mixture of magne-
sium turnings (4.2 g, 0.17 mol) in THF (50 mL).
All preparations and manipulations were routinely
performed under an atmosphere of dinitrogen using
Schlenk or syringe techniques as appropriate. All sol-
vents were dried and distilled under dinitrogen immedi-
ately prior to their use. Tetrahydrofuran (THF) was
distilled from Na/benzophenone, diethyl ether (Et2O),
from Na, acetonitrile (MeCN) from CaH2 and dichloro-
methane (CH2Cl2) from P2O5.
The following chemicals were purchased from Al-
drich and used as received: triphos; NiCl2 Æ 6H2O;
NaBPh4, SiMe3Cl, PhCH2MgCl (1 M solution),
Table 1
Elemental and spectroscopic characterisation of [Ni(g2-CH2C6H4R-4)(triphos)]BPh4
b
R
H
Elemental analysisa/%
NMR spectroscopy
1H
C
H
31P
77.8
(77.8)
5.3
(5.9)
7.2–8.3 (complex m, Ph groups)
6.2 (d, JHH = 8.0 Hz, CH2Ph)
2.1–2.7 (broad, CH2CH2)
87.7 (t, JPP = 45 Hz, Pc)
47.5 (d, JPP = 45 Hz, Pt)
Me
77.5
(77.9)
6.2
(6.1)
6.6-8.0 (complex m, Ph groups)
6.2 (d, JHH = 9.6 Hz, CH2Ar)
2.1–2.8 (broad, CH2CH2)
1.2 (s, CH3)
88.8 (t, JPP = 42 Hz, Pc)
47.7 (d, JPP = 42 Hz, Pt)
MeO
76.4
(76.7)
6.00
(6.00)
6.8–7.8 (complex m, Ph groups)
6.5 (d, JHH = 10.0 Hz, CH2Ar)
6.15 (d, JHH = 10.0 Hz, CH2Ar)
3.65 (s, OCH3)
86.3 (t, JPP = 44.5 Hz, Pc)
45.3 (d, JPP = 44.5 Hz, Pt)
2.1–2.8 (broad, CH2CH2)
a
Calculated values shown in parentheses.
Abbreviations: m = multiplet; t = triplet; d = doublet; s = singlet; Pc = central P in triphos; Pt = terminal P in triphos.
b

V. Autissier et al. / Journal of Organometallic Chemistry 690 (2005) 1763–1771
1765
To
a
suspension of [NiCl(triphos)]BPh4 (1 g,
fidence; although atomic sites could be assigned and re-
fined as a mixture of carbon and oxygen, the geometry is
not sensible for any recognisable solvent molecule, and
so this has not been included in the formula of the com-
pound. Solvent in the structure of the complex with
R = OMe is even more disordered and could not be
modeled with discrete atoms; instead, it was accounted
for by the SQUEEZE procedure in PLATON (A.L.
Spek, University of Utrecht, The Netherlands). The
structure of [NiCl(triphos)]BPh4 has two phenyl rings
disordered over two orientations each in the cation.
Crystal data and other experimental information are
given in Table 2. Selected bond lengths and angles for
the three benzyl complexes are reported in Table 3.
Crystallographic data for the structural analyses has
been deposited with the Cambridge Crystallographic
Data Centre: CCDC No. 249866 for [Ni(g2-CH2C6H5)-
(triphos))]BPh4; CCDC No. 249867 for [Ni(g2-CH2C6-
H4Me-4)(triphos))]BPh4; CCDC No. 249868 for
[Ni(g2-CH2C6H4OMe-4)(triphos))]BPh4 and CCDC
No. 249869 for [NiCl(triphos))]BPh4.
1.95 mmol) in THF (ca. 30 mL) at ꢀ75 ꢁC, an excess
of 4-MeOC6H4CH2MgCl (10 mL, 15 mmol) was added
slowly. The solution changed rapidly from orange to
dark red and was stirred overnight. The next day, etha-
nol was added dropwise to the solution until there was
no more effervescence. All volatiles were now removed
in vacuo, to produce a red microcrystalline solid. The
dry solid was extracted with the minimum amount of
CH2Cl2 to form a red solution and a white solid. The
magnesium salts were removed by filtration, washing
with more CH2Cl2 to ensure complete extraction of
the desired red product. Layering EtOH on the red solu-
tion produced (after several days of slow diffusion) red
crystals of [Ni(g2-CH2C6H4OMe-4)(triphos)]BPh4. The
crystals were removed by filtration and dried in vacuo.
The best yield obtained by this method was 35%. How-
ever, if the addition of the Grignard reagent to the
[NiCl(triphos)]BPh4 was performed at room tempera-
ture the maximum yield obtained was 11%.
On several occasions when we have prepared these
complexes a mixture has been obtained consisting of or-
ange and dark red crystals. Using NMR spectroscopy
and X-ray crystallography it was shown evident that
the red crystals are [Ni(g2-CH2C6H4OMe-4)(triphos)]-
BPh4 and the orange crystals are [NiCl(triphos)]BPh4.
The mixture can be separated as follows. Place the mix-
ture in a sintered Schlenk flask and wash with 4 · 5 mL
of MeCN whilst gently shaking. The [NiCl(tri-
phos)]BPh4 is much more soluble in MeCN than
[Ni(g2-CH2C6H4OMe-4)(triphos)]BPh4. When it was
evident that no orange crystals were left, the red crystals
were washed with a further 5 mL and dried in vacuo.
Using this procedure the isolated yield of [Ni(g2-
CH2C6H4OMe-4)(triphos)]BPh4 is only ca. 3%.
2.3. Kinetic studies
All kinetic studies were performed using an Applied
Photophysics SX.18MV stopped-flow spectrophotome-
ter, modified to handle air-sensitive solutions. The tem-
perature was maintained at 25.0 0.1 ꢁC using a Grant
LT D6G thermostated recirculating pump.
All solutions were prepared under an atmosphere of
dinitrogen and transferred by gas-tight, all-glass syrin-
ges into the stopped-flow spectrophotometer. Stock
solutions of anhydrous HCl in MeCN were prepared
by mixing equimolar amounts of SiMe3Cl and MeOH
in the solvent. Dilute solutions containing mixtures of
HCl and [NEt4]Cl were prepared from the stock solu-
tions of the two components. All solutions were pre-
pared and used within 1 h.
2.2. X-ray crystallography
Data were measured on Bruker AXS SMART 1K
and Nonius KappaCCD diffractometers at 150 K, using
Kinetics were studied under pseudo first-order condi-
tions [8] with all reagents in a large excess (>10-fold)
over the concentration of [Ni(g2-CH2C6H4R-4)(tri-
phos)]BPh4. Under all conditions the absorbance-time
curve for the reactions is an excellent fit to a single expo-
nential for at least 4 half-lives, indicating a first-order
dependence on the concentration of complex. The entire
curve was fitted using the Applied Photophysics com-
puter program to obtain the observed rate constants
(kobs). The rate laws were determined by graphical anal-
ysis as described in Section 3.
˚
Mo Ka radiation (k = 0.71073 A), and semi-empirical
absorption corrections were applied, based on repeated
and symmetry-equivalent reflections. The structures
were solved by direct and heavy-atom methods, and
were refined on all unique F2 values, with programs of
the SHELX family (G.M. Sheldrick, University of Go¨ttin-
gen, Germany); anisotropic displacement parameters
were refined for non-H atoms, and H atoms were refined
as riding on their parent atoms in ideal geometry. For
the benzyl complex with R = H, there are two indepen-
dent cations and two anions in the asymmetric unit; the
final difference map contains a significant peak close to
one Ni atom, but no sensible disorder model was found
to account for this. The structure of the benzyl complex
with R = Me contains disordered solvent molecules, the
exact nature of which could not be determined with con-
2.4. Characterisation of the product of protonation
The reaction of an excess of anhydrous HCl with all
[Ni(g2-CH2C6H4R-4)(triphos)]+ produced [NiCl(tri-
phos)]+. The product was characterised in situ by 31P
NMR spectroscopy {d110.2 (t, JPP = 49.6 Hz, Pc);

1766
V. Autissier et al. / Journal of Organometallic Chemistry 690 (2005) 1763–1771
Table 2
Crystal data and structure refinement parameters for [Ni(CH2C6H4R-4)(triphos)]BPh4 (R = H, Me or MeO) and [NiCl(triphos)]BPh4
[Ni(CH2C6H5)(triphos)]-
BPh4
[Ni(CH2C6H4Me-4)(triphos)]-
BPh4
[Ni(CH2C6H4OMe-4)(triphos)]-
BPh4
[NiCl(triphos)]-
BPh4
Chemical formula
Fw
C65H60BNiP3
1003.6
Triclinic
C66H62BNiP3
1017.6
C66H62BNiOP3
1033.6
C58H53BClNiP3
947.9
Triclinic
Crystal system
Space group
T (K)
Monoclinic
P21/n
150
Monoclinic
P21/n
150
ꢀ
ꢀ
P1
150
P1
150
˚
a (A)
18.4171(8)
18.6374(8)
19.6767(9)
65.798(2)
69.175(2)
61.155(2)
5289.8(4)
4
13.6232(16)
19.860(2)
20.622(3)
90
13.6171(6)
19.9305(9)
20.5807(9)
90
11.4983(5)
14.3108(6)
17.0594(7)
98.102(2)
105.120(2)
112.220(2)
2416.9(2)
2
˚
b (A)
˚
c (A)
a (ꢁ)
b (ꢁ)
c (ꢁ)
104.888(9)
90
103.266(2)
90
3
˚
V (A )
5392.2(11)
4
32,359
5436.5(4)
4
31,534
Z
Reflections measured
Unique data, Rint
Parameters
38,266
17,734
18,369, 0.048
1261
0.047
6970, 0.085
661
0.046
7108, 0.062
651
0.041
8461, 0.027
670
0.035
R [F2 > 2r]
Rw (F2, all data)
Goodness-of-fit [F2]
Maximum, minimum
0.118
0.997
0.094
1.083
0.103
1.021
0.089
1.050
1.80, ꢀ0.79
0.33, ꢀ0.31
0.39, ꢀ0.29
0.42, ꢀ0.40
3
˚
electron density (e/A )
Table 3
2
˚
Selected bond lengths (A) and angles (ꢁ) for [Ni(g -CH2C6H4R-4)(triphos)]BPh4
R
Bond lengths
Ni–Pt
Bond angles
Ni–Pc
Ni–CH2
1.970(3)
Ni–Cipso
2.484(3)
CH2–Cipso
1.448(5)
Ni–CH2–Cipso
91.9(2)
CH2–Ni–Cipso
Pt–Ni–Pt
138.19(4)
H
2.1813(9)
2.2040(9)
2.1694(9)
2.2104(10)
2.1844(10)
35.6(2)
38.3(2)
2.2112(9)
2.1899(11)
2.1896(9)
1.991(3)
1.974(3)
1.966(3)
2.335(3)
2.263(4)
2.250(3)
1.458(5)
1.466(5)
1.459(4)
83.7(2)
137.38(4)
129.75(4)
129.93(4)
Me
2.2425(10)
2.1897(10)
80.8(2)
39.74(14)
39.78(12)
MeO
2.2280(10)
2.1967(10)
80.65(18)
d48.7 (d, JPP = 49.6 Hz, Pt)} and visible absorption
3. Results and discussion
spectroscopy (kmax = 330 nm,
e
330 = 1.34 · 103 dm3
molꢀ1 cmꢀ1; kmax = 450 nm, e450 = 500 dm3 molꢀ1 cmꢀ1).
The spectra were identical to those of an authentic sam-
ple of [NiCl(triphos)]BPh4 prepared by the method de-
scribed in the literature [5].
3.1. Synthesis of [Ni(g2-CH2C6H4R-4)(triphos)]BPh4
The complexes [Ni(g2-CH2C6H4R-4)(triphos)]BPh4
were prepared by the reaction shown in Eq. (2). At-
tempts to improve the poor yields by: (i) using various
proportions of the Grignard to [NiCl(triphos)]BPh4,
from equimolar amounts to a large excess of the Grig-
nard; (ii) performing the reaction in thf or diethyl ether;
(iii) performing the reaction at ꢀ78 ꢁC or (iv) using a
variety of alcohols (PriOH, EtOH or MeOH) to destroy
the excess of the Grignard reagent, resulted in the best
isolated yield of ca. 35%. In addition, leaving the solu-
tion in the freezer, or adding a greater excess of MeOH,
to facilitate crystallization had no significant effect on
the yield.
The X-ray crystal structure of [NiCl(triphos)]BPh4
has also been determined. The structure of the cation
is shown in Fig. 1 and is as anticipated. The geometry
about the nickel is best described as distorted square-
planar. The bond distance of the phosphorus trans to
˚
the chloride ligand, Ni–P(2) (2.1191 (6) A), is shorter
˚
than the Ni–P(1) (2.2096(7) A) and Ni–P(3)
˚
(2.1956(7) A) bond distances, possibly due to the elec-
tronegative chloride ligand donating electron density
to the nickel, which in turn increases the p-back bonding
to the trans phosphorus, shortening the Ni–P(2) bond.

V. Autissier et al. / Journal of Organometallic Chemistry 690 (2005) 1763–1771
1767
Fig. 1. Structure of the cation in [NiCl(triphos)]BPh4 with 50%
probability displacement ellipsoids. Hydrogen atoms and minor
disorder components are omitted. Selected bond lengths are: Ni–
˚
P(1) = 2.2096(7); Ni–P(3) = 2.1956(7); Ni–P(2) = 2.1191(6) A. Selected
Fig. 2. Structure of [Ni(g2-CH2C6H4Me)(triphos)]+ with 50% proba-
bility displacement ellipsoids. Hydrogen atoms are omitted.
bond angles are: P(1)–Ni–P(2) = 85.75(2); P(1)–Ni–P(3) = 162.74(3);
P(2)–Ni–P(3) = 85.60(2); Cl–Ni–P(1) = 97.28(3); Cl–Ni–P(2) = 171.66(3);
Cl–Ni–P(3) = 93.37(3)ꢁ.
phorus is orientated in the same direction as the phenyl
group of the benzyl ligand. This orientation minimises
unfavourable steric interactions of the benzyl group with
the phenyl groups on the terminal phosphorus atoms.
Whilst the majority of benzyl ligands coordinate in an
end-on configuration, other binding modes are known,
such as g2- and g3-coordination [9]. The dimensions of
the Ni-benzyl group in [Ni(g2-CH2C6H4R-4)(triphos)]+
demonstrate that the benzyl ligand is bound in the g2-
configuration. The unusually short distance between
nickel and the ipso-carbon on the benzyl ligand
{R = H: Ni–Cipso = 2.484(3) and 2.335(3); R = Me:
Ni–Cipso = 2.263(4); R = MeO: Ni–Cipso = 2.250(3)}
strongly indicates an interaction in all cases, as do the
Ni–CH2–Cipso angles, which are small {R = H: an-
gle = 91.9(2) and 83.7(2)ꢁ; R = Me: angle = 80.8(2);
R = MeO: angle = 80.65(18)ꢁ}. In transition metal ben-
zyl complexes showing no evidence of a M–Cipso interac-
tion the M–CH2–Cipso angle is closer to the tetrahedral
angle of 109.5ꢁ. The nickel-Cortho distance in all
the complexes is too large for any interaction, thus pre-
cluding an g3-benzyl coordination. Further evidence that
the g2-benzyl description is correct comes from consider-
ations of the bond lengths within the benzyl ligand.
A typical CH3–C6H4R single bond length [10] is
½NiCðtriphosÞꢁBPh4
4-RC6H4CH2MgCl
½Niðg2-CH C H R-4ÞðtriphosÞꢁBPh
ƒƒƒƒƒƒƒƒƒ!
2
6
4
4
ð2Þ
We have been unable to extend the range of [Ni(g2-
CH2C6H4R-4)(triphos)]BPh4 to include benzyl deriva-
tives where R is an electron-withdrawing substituent
(e.g. 4-O2NC6H4CH2- or 4-ClC6H4CH2–) because of
the inability to prepare the necessary Grignard reagents.
All [Ni(g2-CH2C6H4R-4)(triphos)]BPh4 (R = H, Me
or MeO) were characterised by H and 31P NMR spec-
1
troscopy, elemental analysis and X-ray crystallography.
Most notably, the X-ray crystal structures unambigu-
ously demonstrate that the substituted benzyl ligands
bind to the nickel in what is best described as an g2-
coordination mode. We will discuss the crystallography
of these complexes first and then present the NMR spec-
troscopic information.
3.2. Structures of [Ni(g2-CH2C6H4R-4)(triphos)]BPh4
X-ray crystallographic investigation of [Ni(g2-
CH2C6H4R-4)(triphos)]BPh4 (R = H, Me or MeO)
shows that all the cations have similar structures typified
by that of [Ni(g2-CH2C6H4Me-4)(triphos)]+ shown in
Fig. 2. The geometry at the nickel is best described as dis-
torted square-planar. Three of the four coordination sites
are occupied by the phosphorus atoms of the triphos
ligand and the fourth site corresponds to the centre of
the CH2–Cipso bond. The CH2–Cipso bond is orientated
perpendicular to the coordination plane of the complex.
In all the structures the phenyl group on the central phos-
˚
1.53 0.01 A. The CH2–Cipso bond lengths in the
[Ni(g2-CH2C6H4R-4)(triphos)]+ complexes are 0.05–
˚
0.08 A shorter, suggesting partial double bond character
(R = H: CH2–Cipso = 1.448(5) and 1.458(5); R = Me:
˚
CH2–Cipso = 1.466(5); R = MeO: CH2–Cipso = 1.459(4) A).
The phenyl ring is coplanar with the CH2 carbon
atom, and the benzyl ligand lies approximately perpen-
dicular to the Ni–CH2–Cipso triangle.

1768
V. Autissier et al. / Journal of Organometallic Chemistry 690 (2005) 1763–1771
3.3. Spectroscopic characterisation
CH2C6H4R-4)(triphos)]+ (R = H, Me or MeO) can be
accommodated by the rate law shown in Eq. (4),
ðR ¼ H : kH2 ¼ 0:71 ꢂ 0:04 sꢀ1; R ¼ Me : kM2 e ¼ 0:20 ꢂ
0:03 sꢀ1; R ¼ MeO : k2MeO ¼ ð2:1 ꢂ 0:5Þ ꢃ 10ꢀ3 sꢀ1Þ.
1
The H NMR spectra of all the complexes [Ni(g2-
CH2C6H4R-4)(triphos)]BPh4 are similar. The aromatic
region is very complicated because of the large number
of various phenyl groups on the triphos ligand, the
[BPh4]ꢀ and the benzyl ligand. However, the 1H
NMR spectra of [Ni(g2-CH2C6H4Me-4)(triphos)]BPh4
and [Ni(g2-CH2C6H5)(triphos)]BPh4 both show a dou-
blet at d 6.2. We tentatively suggest that this peak is
assigned to the methylene protons of Cipso coupled to
the central P of the triphos ligand. The central P is
essentially trans to the benzyl ligand. If this assignment
is correct, we must conclude that coupling to the two
terminal P (which are effectively cis to the benzyl
ligand) is too small to be detected. [Ni(g2-
CH2C6H4OMe-4)(triphos)]BPh4 shows two doublets
(d 6.15 and d 6.50). The singlet at d 3.65 in the spec-
trum of [Ni(g2-CH2C6H4OMe-4)(triphos)]BPh4 is as-
signed to the three protons on the methoxy group,
and the singlet at d 1.2 in the spectrum of [Ni(g2-
CH2C6H4Me-4)(triphos)]BPh4 is assigned to the methyl
group.
½Niðg2-CH2C6H4R-4ÞðtriphosÞꢁþ þ HCl
! ½NiClðtriphosÞꢁþ þ 4-CH3C6H4R
ð3Þ
ꢀd½NiðCH2C6H4R-4ÞðtriphosÞþꢁ
dt
½HClꢁ
¼ kR2
½NiðCH2C6H4R-4ÞðtriphosÞꢁþ
ð4Þ
½Clꢁꢀ
The analysis of the kinetic data needs some further
explanation. When studying the reaction of compounds
with HCl in acetonitrile there is the problem of whether
the reacting acid is HCl or solvated H+ {Eq. (5)}.
KHa Cl
HCl ꢀ Hþ þ Clꢀ
ð5Þ
Clearly, the addition of Clꢀ to solutions of HCl in
MeCN will perturb the concentration of solvated H+
ðKHa ClÞ according to Eq. (5). However, the addition of
Clꢀ to solutions of HCl in MeCN also perturbs the con-
centration of HCl because of the formation of [HCl2]ꢀ
according to the homoconjugation equilibrium (KH)
shown in Eq. (6). Using the literature values [12,13] of
KH = 158.5 dm3 molꢀ1 with KHa Cl ¼ 1:3 ꢃ 10ꢀ9 dm3
molꢀ1 we have calculated the concentrations of HCl,
Clꢀ and H+ as shown in Table 4. The values of [HCl]/
[Clꢀ] shown in Fig. 3 are those corrected for the homocon-
jugation equilibrium.
Further evidence for the Ni–Cipso interaction comes
from the noticeable downfield shift of the CH2–Cipso
1
peaks in the H NMR spectrum. Formation of a metal-
locycle results in additional ring current which opposes
the external field, deshielding the proton. An agostic
interaction with the aromatic residue of the type C–
H. . .M can be excluded. This type of interaction would
result in a high-field shift [11].
All the spectroscopic data is consistent with the struc-
tures of [Ni(g2-CH2C6H4MR-4)(triphos)]+ having the
same structure in solution as established in the X-ray
crystal structures: the benzyl ligand lies approximately
perpendicular to the Ni–CH2–Cipso triangle.
KH
HCl þ Clꢀ ꢀ HClꢀ2
ð6Þ
The kinetics shown in Eq. (4) could represent, not a
dependence on [HCl]/[Clꢀ], but actually a dependence
on [H+] as shown in Eq. (7), where kRH ¼ k2R=KHa Cl. If Eq.
(7) were the true form of the rate law, and the kinetics ex-
hibit a dependence on the concentration of H+,
3.4. Kinetics and mechanism of the reaction between
[Ni(g2-CH2C6H4R-4)(triphos)]+ and HCl
kHH ¼ 5:5 ꢂ 0:3 ꢃ 108 dm3 molꢀ1 sꢀ1
,
kMH e ¼ 1:5 ꢂ 0:2 ꢃ
The reaction of an excess of anhydrous HCl with
[Ni(g2-CH2C6H4R-4)(triphos)]+ (R = H, Me or MeO)
in MeCN produces [NiCl(triphos)]+ and 4-RC6H4Me,
as described by Eq. (3). The kinetics have been moni-
tored by stopped-flow spectrophotometry, with both
HCl and Clꢀ present in a large excess over the concen-
tration of the complex. MeCN was chosen as the solvent
because of the extensive data concerning acid strengths
and homoconjugation equilibria which are known in
this solvent [12]. Under all conditions, the absorbance–
time curve can be fitted to a single exponential, with
the initial absorbance corresponding to that of [Ni(g2-
CH2C6H4R-4)(triphos)]+ and the final absorbance to
that of [NiCl(triphos)]+. The observed rate constant
(kobs) depends on the ratio [HCl]/[Clꢀ] as shown in
Fig. 3. The kinetics of the reactions of all [Ni(g2-
108 dm3 molꢀ1 sꢀ1 and kHMeO ¼ 1:6 ꢂ 0:3 ꢃ 106 dm3 molꢀ1
s
ꢀ1. Whilst these second order rate constants are fast they
are slower than the diffusion-controlled limit (k = 1 · 1010
dm3 molꢀ1 sꢀ1), and hence not unreasonable even for a
mechanism involving a rate-limiting protonation step.
Based on the kinetics, we cannot establish whether the
true (mechanistically significant) form of the rate law is
Eq. (4) or (7).
ꢀd½NiðCH2C6H4R-4ÞðtriphosÞþꢁ
dt
¼ kRH½Hþꢁ½NiðCH2C6H4R-4ÞðtriphosÞþꢁ
ð7Þ
In the case of [Ni(g2-CH2C6H4OMe-4)(triphos)]+ the
quality of the kinetic data is poor. Because of the scatter
of the data, the non-zero intercept shown in Fig. 3

V. Autissier et al. / Journal of Organometallic Chemistry 690 (2005) 1763–1771
1769
Fig. 3. Kinetic data for the reactions between [Ni(g2-CH2C6H4R)(triphos)]+ (R = H, Me or MeO) and HCl in MeCN at 25.0 ꢁC. Main. Dependence
of kobs on the ratio, [HCl]/[Clꢀ] for the reactions of R = H (full symbols) and R = Me (open symbols). The data points correspond to:
[HCl] = 20 mmol dmꢀ3, [Clꢀ] = 5–25 mmol dmꢀ3 (d); [HCl] = 10 mmol dmꢀ3, [Clꢀ] = 5–25 mmol dmꢀ3 (m); [HCl] = 20 mmol dmꢀ3, [Clꢀ] = 5–
25 mmol dmꢀ3 (s); [HCl] = 15 mmol dmꢀ3, [Clꢀ] = 5–25 mmol dmꢀ3 (n); [HCl] = 10 mmol dmꢀ3, [Clꢀ] = 5–25 mmol dmꢀ3 (h). Insert Dependence
of kobs on the ratio, [HCl]/[Clꢀ] for the R = MeO derivative. The data points correspond to: [HCl] = 50 mmol dmꢀ3, [Clꢀ] = 2–50 mmol dmꢀ3 (d). In
all cases the concentrations of HCl and Clꢀ have been corrected for the homoconjugation equilibrium shown in Eq. (6). The lines drawn are those
defined by Eq. (4) and the values presented in the text.
(insert) may be meaningless and we conclude that the
rate law, even for this derivative, is that shown in Eq.
(4). The reason for the poor quality of the kinetic data
with the R = MeO derivative is not clear.
in Fig. 4. The rate law associated with this mechanism is
shown in Eq. (8), and is derived assuming that [NiH(g2-
CH2C6H4R-4)(triphos)]2+ is a steady state intermediate.
If kRꢀ3 > kR4 then Eq. (8) simplifies to Eq. (9), which is
identical in form to that observed experimentally {Eq.
(4)}. Comparison of Eqs. (4) and (9) yields the values
of kR3 kR4 =kRꢀ3 ¼ kR2 shown in Table 5.
3.5. The mechanism of the reactions of [Ni(g2-
CH2C6H4R-4)(triphos)]+ with HCl
ꢀd½Niðg2-CH2C6H4R-4ÞðtriphosÞþꢁ
The simplest interpretation of the rate law shown in
Eq. (4) is that the mechanism involves protonation of
the benzyl carbon followed by dissociation of the substi-
tuted toluene. However, it is difficult to reconcile such a
simple protonation mechanism with the effect that the
4-R-substituent has on the rates of the reactions: as
the 4-R-substituent becomes more electron-releasing,
the reaction becomes slower (R = H > Me > MeO). It
is not easy to see how the rate of a simple protonation
of a carbon site would be facilitated by 4-R-substituents
which make the carbon site less basic.
dt
kR4 kR3 ½HClꢁ
¼
½Niðg2-CH2C6H4R-4ÞðtriphosÞþꢁ
kR4 þ kRꢀ3½Clꢀꢁ
ð8Þ
ð9Þ
ꢀd½Niðg2-CH2C6H4R-4ÞðtriphosÞþꢁ
dt
kR4 kR3 ½HClꢁ
¼
½Niðg2-CH2C6H4R-4ÞðtriphosÞþꢁ
kRꢀ3½Clꢀꢁ
It is proposed that the mechanism could involve ini-
tial rapid protonation of the nickel. Subsequent product
formation could then occur by migration of the hydro-
gen from the nickel to the coordinated benzyl as shown
The overall effect that the 4-R-substituents have on
the rate of the reaction depends on the effect that the
4-R-substituent has on the individual elementary rate
constants kR3 , k4R and kꢀR3. It seems intuitively reasonable

1770
V. Autissier et al. / Journal of Organometallic Chemistry 690 (2005) 1763–1771
Table 4
Kinetic data for the reactions of HCl with [Ni(g2-CH2C6H4R-4)(triphos)]BPh4 (R = H, Me or MeO) in the presence of Clꢀ in MeCN at 25.0 ꢁC
a
R
H
[HCl] (mmol dmꢀ3
)
[Clꢀ] (mmol dmꢀ3
)
[HCl]e (mmol dmꢀ3
)
[Clꢀ]ea(mmol dmꢀ3
)
[H+] (109 mol dmꢀ3
)
kobs (sꢀ1
)
10.0
5.0
10.0
15.0
20.0
25.0
7.31
5.39
4.09
3.23
2.67
2.31
5.39
42.5
3.0
1.26
0.62
0.32
0.44
0.19
9.09
13.23
17.67
0.567
0.307
0.190
20.0
10.0
15.0
20.0
50.0
5.0
10.0
15.0
20.0
25.0
16.39
13.2
10.6
8.46
6.9
1.39
3.2
14.9
5.16
2.38
1.26
0.73
8.6
2.20
1.55
0.73
0.49
5.64
8.46
11.9
Me
5.0
10.0
15.0
20.0
25.0
7.31
5.39
4.09
3.23
2.67
2.31
5.39
42.5
0.81
0.72
0.62
0.39
0.21
1.26
9.09
13.23
17.67
0.567
0.307
0.190
5.0
10.0
15.0
20.0
25.0
11.74
9.39
7.09
5.64
4.5
1.74
4.39
8.51
2.69
0.86
0.25
0.10
0.09
0.06
7.09
10.64
14.5
1.26
0.668
0.391
Me
5.0
10.0
15.0
20.0
25.0
16.39
13.2
10.6
8.46
6.9
1.39
3.22
5.64
8.46
11.9
14.9
5.16
2.38
1.26
0.73
2.46
1.18
0.79
0.51
0.31
MeO
2.0
5.0
48.25
45.6
0.25
0.60
1.35
3.2
243
0.50
0.45
0.25
0.14
0.17
95.8
38.6
13.1
1.26
10.0
20.0
50.0
41.35
33.2
14.8
14.8
a
Concentrations calculated allowing for homoconjugation equilibrium.
+
+
2+
R
k3
P
P
P
R
CH2
k4
HCl
Cl-
H
P
Ni
P
Ni
P
Ni Cl + R
P
CH3
CH2
R
P
P
k-3
R
R
Fig. 4. The mechanism of the reactions between [Ni(g2-CH2C6H4R)(triphos)]+ (R = H, Me or MeO) and HCl in MeCN. The mechanism is only
shown for HCl being the acid. An analogous mechanism could be drawn for H+ being the acid.
that the protonation step kR3 is facilitated by electron-
releasing 4-R-substituents, and deprotonation kRꢀ3 would
be slower with electron-releasing R, since such 4-R-sub-
stituents would increase the electron-richness of the
nickel site. The observation that electron-withdrawing
substituents facilitate the reaction must be attributable
to the effect of the 4-R-substituent on the unimolecular
kR4 step. If the kR4 step involved intramolecular hydride
transfer, electron-withdrawing 4-R-substituents would
stabilize the transition state by neutralizing the build-
up of negative charge on the methylene group. A similar
argument has been presented for the reactions involving
intramolecular hydrogen migration reactions of exo-
and endo-[Nb(g5-C5H5)2H(CH2 = CHC6H4R-4)]. It has

V. Autissier et al. / Journal of Organometallic Chemistry 690 (2005) 1763–1771
1771
Table 5
Summary of the elementary rate for the reactions of [Ni(g2-CH2C6H4-
References
R-4)(triphos)]+ with HCl in MeCN at 25.0 ꢁC
[1] (a) C. Masters, Homogeneous Transiton Metal Catalysis, Chap-
man and Hall, London, 1981, pp. 135–158;
R
kR3 kR4 =kRꢀ3 ðdm3 molꢀ1 sꢀ1
Þ
0.70 0.10
(b) P.W. Jolly, G. WilkeOrganic Chemistry of Nickel, vol. 2,
Academic Press, New York, 1975;
H
Me
MeO
0.21 0.05
(2.2 0.8) · 10ꢀ3
(c) R.A. Henderson, J. Chem. Res. (S) (2002) 407.
[2] (a) K.W. Kramarz, J.R. Norton, Prog. Inorg. Chem. 42
(1994) 1;
(b) R.A. Henderson, Angew. Chemie, Int. Ed. 35 (1996) 946;
(c) S.S. Kristjansdottir, J.R. Norton, in: A. Dedieu (Ed.),
Transition Metal Hydrides: Recent Advances in Theory and
Experiment, 1992 (chapter 9);
been observed that electron-releasing 4-R-substituents
accelerate migration [14], presumably by stabilizing the
modest accumulation of positive charge at Cb in the
transition state for the migration.
(d) E.T. Papish, M.P. Magee, J.R. Norton, in: M. Peruzzini, R.
Poli (Eds.), Recent Advances in Hydride Chemistry, Elsevier,
Amsterdam, 2001 (chapter 2).
3.6. Mechanisms of the reactions of [NiR(triphos)]+
with HCl
[3] R.A. Henderson, J. Chem. Soc. Dalton Trans. (1995) 503.
[4] R.A. Henderson, K.E. Oglieve, Chem. Commun. (1999)
2271.
The kinetics and mechanism of the reactions of
[Ni(g2-CH2C6H4R-4)(triphos)]+ with HCl described in
this paper are markedly different to those observed in
the earlier study on [NiMe(triphos)]+. Both systems ap-
pear to protonate preferentially at the nickel. However,
in the benzyl complexes the subsequent formation of
products involves migration of the hydride from nickel
to benzyl ligand, but in [Ni(H)Me(triphos)]2+ product
formation has to await the addition of a second proton
(presumably at the methyl ligand) [4]. What controls
how [Ni(H)R(triphos)]2+ converts into products (migra-
tion or additional protonation) is currently not clear.
However, it seems likely that the additional protonation
pathway is observed with more basic alkyl ligands (e.g.
Me ligand), whereas with less basic ligands (e.g. benzyl)
the migration pathway operates.
[5] W. Clegg, G. Cropper, R.A. Henderson, C. Strong, B. Parkinson,
Organometallics 20 (2001) 2579.
[6] (a) W. Clegg, R.A. Henderson, Inorg. Chem. 41 (2002) 1128;
(b) V. Autissier, P.M. Zarza, A. Petrou, R.A. Henderson, R.W.
Harrington, W. Clegg, Inorg. Chem. 43 (2004) 3106.
[7] A.I. Vogel, Textbook of Practical Organic Chemistry, third ed.,
Longmans, London, 1967, 812 pp.
[8] J.H. Espenson, Chemical Kinetics and Reaction Mechanisms,
McGraw-Hill, New York, 1981, pp. 30–32.
[9] (a) Selected examples of structures of benzyl-complexes are: N.H.
Dryden, P. Legzdins, J. Trotter, V.C. Yee, Organometallics 10
(1991) 2857;
(b) C. Pellecchia, A. Grassi, A. Immirzi, J. Am. Chem. Soc. 115
(1993) 1160;
(c) W. Clegg, M.R.J. Elsegood, P.W. Dyer, V.C. Gibson, E.L.
Marshall, Acta Cryst. 55 (1999) 916;
(d) W. Lesueur, E. Solari, C. Floriani, Inorg. Chem. 36 (1997)
3354.
[10] See Ref. [9a].
[11] M. Brookhart, M.L.H. Green, L. Wong, Prog. Inorg. Chem. 36
(1988) 1.
[12] K. Izutsu, Acid–Base Dissociation Constants in Dipolar Aprotic
Solvents, Blackwell Scientific, Oxford, 1990.
Acknowledgement
[13] J.F. Coetzee, Prog. Phys. Org. Chem. 4 (1967) 45.
[14] B.J. Burger, B.D. Santarsiero, M.S. Trimmer, J.E. Bercaw, J.
Am. Chem. Soc. 110 (1988) 3134.
One of us (V.A.) thanks the University of Newcastle
for a studentship.