Self-replication of tris(cyanoethyl)phosphine catalysed by platinum group metal complexes

Article abstract of DOI:10.1039/a704655c

The platinum(0) complex [Pt(tcep)₃], tcep = P(CH₂CH₂CN)₃, catalyses the formation of tcep from PH₃ and CH₂=CHCN. The complexes [M(tcep)₃] (M = Pt, Pd or Ni) and [MCl(tcep)₃] (M = Rh or Ir) are compared for their catalysis of the reaction of PH(CH₂CH₂CN)₂ with CH₂=CHCN to give tcep and it is shown that the platinum(0) complex is the most efficient. The platinum(0) catalysis has been studied in detail, monitoring the kinetics by ³¹P-{¹H} NMR spectroscopy. It is revealed that the kinetics are a complex function of the concentration of product tcep. Qualitatively, the rates also depend on [CH₂=CHCN] and [catalyst]. Both ³¹P-{¹H} and ¹⁹⁵Pt-{¹H} NMR spectroscopy suggests that addition of CH₂=CHCN to [Pt(tcep)₃] gives the complex [Pt(tcep)₂(η²-CH₂=CHCN)] which undergoes phosphine exchange on the NMR time-scale. The binuclear complex [Pt₂H₂(tcep)₂{η-P(CH₂CH ₂CN)₂}₂], formed upon addition of PH(CH₂CH₂CN)₂ to trans-[PtHCl(tcep)₂] in the presence of base, is shown to be a catalyst precursor for the reaction of PH(CH₂CH₂CN)₂ with CH₂=CHCN. Two parallel mechanisms involving mononuclear and binuclear intermediates are discussed to rationalise these observations.

Full text of DOI:10.1039/a704655c

View Article Online / Journal Homepage / Table of Contents for this issue
Self-replication of tris(cyanoethyl)phosphine catalysed by platinum
group metal complexes
a
,a
b
a
Emiliana Costa, Paul G. Pringle,* Martin B. Smith and Kerry Worboys
a
School of Chemistry, University of Bristol, Cantock’s Close, Bristol, UK BS8 1TS
Department of Chemistry, Loughborough University, Loughborough, UK LE11 3TU
b
The platinum(0) complex [Pt(tcep) ], tcep = P(CH CH CN) , catalyses the formation of tcep from PH and
3
2
2
3
3
CH ᎐CHCN. The complexes [M(tcep) ] (M = Pt, Pd or Ni) and [MCl(tcep) ] (M = Rh or Ir) are compared for
2
3
3
their catalysis of the reaction of PH(CH CH CN) with CH ᎐CHCN to give tcep and it is shown that the
2
2
2
2
platinum(0) complex is the most efficient. The platinum(0) catalysis has been studied in detail, monitoring
31
1
the kinetics by P-{ H} NMR spectroscopy. It is revealed that the kinetics are a complex function of the
concentration of product tcep. Qualitatively, the rates also depend on [CH ᎐CHCN] and [catalyst]. Both
2
31
1
195
1
P-{ H} and Pt-{ H} NMR spectroscopy suggests that addition of CH ᎐CHCN to [Pt(tcep) ] gives the
2
᎐
3
2
complex [Pt(tcep) (η -CH ᎐CHCN)] which undergoes phosphine exchange on the NMR time-scale. The binuclear
2
2
complex [Pt H (tcep) {µ-P(CH CH CN) } ], formed upon addition of PH(CH CH CN) to trans-[PtHCl(tcep) ]
2
2
2
2
2
2
2
2
2
2
2
in the presence of base, is shown to be a catalyst precursor for the reaction of PH(CH CH CN) with
2
2
2
CH ᎐CHCN. Two parallel mechanisms involving mononuclear and binuclear intermediates are discussed to
2
rationalise these observations.
Hydrophosphination of an alkene, the addition of a P᎐H bond
of R PH (n = 0–2) to a C᎐C bond, is a useful route to tertiary
amounts; thus we established that 1 catalyses the reaction
shown in equation (1). These observations encouraged us to
n
3Ϫn
1
phosphines. Alkene hydrophosphinations can be acid- or base-
2
3
catalysed, and they can also be promoted by radical initiators.
[
cat]
PH3 + 3
P(CH2CH2CN)3
(1)
Metal–phosphine complexes have been used as templates for
CN
the construction of polydentate phosphines via stoichiometric
4
5
P᎐H additions to alkenes but previous to our work, there were
no reports of metal–phosphine catalysed hydrophosphinations.
Tertiary phosphine complexes of late transition metals are
excellent catalysts for E᎐H (e.g. E = H, BR or SiR ) additions
investigate the hydrophosphination of acrylonitrile in detail.
The PH reaction shown in equation (1) would have presented
3
difficulties for study because of the dangers inherent in hand-
2
3
ling PH and furthermore, since it involves three consecutive
3
6
to C᎐C bonds and we were intrigued by the possibility that
PR3 complexes may catalyse the formation of PR3 from
R PH3Ϫn; such a reaction constitutes a self-replication process
hydrophosphinations, the kinetics would be very complicated to
interpret.
It was reasoned that the reaction shown in equation (2)
᎐
n
at a metal centre.
Tris(cyanoethyl)phosphine (tcep) is commercially important
[
cat]
7
(2)
PH(CH2CH2CN)2
P(CH2CH2CN)3
and its co-ordination chemistry has been well studied. It can be
8
CN
conveniently made by the base-catalysed reaction of CH ᎐
2
9
CHCN with [P(CH OH) ]Cl. In 1961, it was reported in a
2
4
would be convenient for detailed study since the secondary
phosphine PH(CH CH CN) (dcep) is an easily handled, com-
mercially available liquid and the reaction involves only one
hydrophosphination step. The conversion was readily moni-
patent that the addition of PH to acrylonitrile was catalysed by
3
2
2
2
PtCl , and other transition-metal compounds, though it was
4
not claimed that the catalysts were homogeneous. In this
paper we report an investigation of the hydrophosphination
of acrylonitrile catalysed by late-transition-metal complexes of
31
1
tored by P-{ H} NMR spectroscopy (see Fig. 1 for a typical
run); to obtain quantitative information about the course of the
31
1
tris(cyanoethyl)phosphine (tcep). Some of these results were
hydrophosphination, the P-{ H} NMR spectra were meas-
ured with 2 s pulse delay and inverse-gated decoupling so that
the relative intensities of signals for the secondary and tertiary
phosphines corresponded to their relative proportions. Under
the reaction conditions we used (see Experimental section), in
the absence of metal complex, it took 8 d for the P᎐H addition
to go to completion. Under the same conditions, but in the
presence of 0.04 equivalent of 1 (see Experimental section) the
reaction was 50% complete after 2 h and no secondary phos-
phine could be detected after 6 h.
5
previously described in
a preliminary communication.
10
Recently Glueck and co-workers have reported platinum(0)-
catalysed P ᎐H additions to acrylonitrile and Han and
Tanaka have reported palladium(0)-catalysed P ᎐H additions
III
11
V
to acetylenes.
Results and Discussion
We began our search for metal-complex catalysts for the hydro-
phosphination of acrylonitrile with the platinum(0) complex,
The complexes 2–5 were screened for hydrophosphination
Pt(tcep) ] 1 because we had previously shown that platinum(0)
catalytic activity and compared with 1. The synthesis of each of
these complexes has been previously reported with the excep-
3
7
complexes were efficient catalysts for the hydrophosphination
12
of formaldehyde. When PH is passed through an acetonitrile
tion of [Ni(tcep) ] 3. Complex 3 was generated in situ from
3
3
solution of acrylonitrile, no reaction was observed after 6 h and
[Ni(cod) ] (cod = cycloocta-1,5-diene) and 3 equivalents of tcep
2
31
1
31
1
only PH was observed to be present by P-{ H} NMR spec-
as evidenced by the P-{ H} NMR spectrum at Ϫ44 ЊC in
MeCN which showed a sharp signal at δ ϩ16.7. Though solu-
tions of 3 were apparently stable, an attempt to isolate the com-
plex by evaporation of the MeCN led to formation of a nickel
3
troscopy. Under similar conditions but in the presence of 1 (mol
ratio of 1 to acrylonitrile, ca. 1:1500) after 6 h, the phosphines
PH (CH CH CN) (n = 0–3) were observed in roughly equal
n
2
2
3Ϫn
J. Chem. Soc., Dalton Trans., 1997, Pages 4277–4282
4277

View Article Online
Table 1 Kinetic data
a
Entry
Catalyst
Conditions/solvent
Standard
t(50%)/min
1
2
3
4
5
6
7
8
9
None
6400
120
240
1290
b
250
120
80
1
2
3
4
5
1
1
1
1
1
1
MeCN
Me CO
2
Me SO
70
120
65
2
1
1
1
0
1
2
2 × [1]
1/30 [CH ᎐CHCN]
4250
2
᎐
a
See experimental for the standard conditions used in these reactions
b
and how the t(50%) values were obtained. The reaction did not reach
5
0% completion even after several days.
CH3
CN
(NCCH2CH2)2P
(NCCH2CH2)2P
CN
CN
n
X
Y
Comparison of the catalysts
The hydrophosphination shown in equation (2) was followed by
31
1
P-{ H} NMR spectroscopy in the presence of complexes 1–5
and t(50%), the average time for the reaction to be 50% com-
plete, was measured in order to compare the relative efficiency.
It can be seen from entries 2–6 in Table 1 that the order of
activity of the complexes was: [Pt(tcep) ] > [Pd(tcep) ] ≈ [IrCl-
3
3
(
tcep) ] > [Ni(tcep) ]. In the presence of the rhodium complex
3
3
4
, catalysis was observed in the early stages but the reaction
failed to reach the 50% point even after several days. In the runs
with 4, a pale yellow precipitate formed during the reaction
which was insoluble in all common solvents and was most likely
a phosphido-bridged rhodium species which is not only inactive
catalytically but apparently slows the hydrophosphination reac-
tion. The reaction in the absence of metal complex is likely to
be a radical process and the rhodium–phosphido species may
be acting as a radical scavenger and thereby inhibiting the
reaction.
31
1
Fig. 1 Variation of P-{ H} NMR spectra with time upon addition of
dcep to CH ᎐CHCN in the presence of [Pt(tcep) ] under the standard
2
᎐
3
conditions (see Experimental section) after (a) 10, (b) 120, (c) 240 and
d) 720 min
(
Effect of solvent
(NCCH2CH2)3P
The solvent has a small effect on the rate of the reaction
M
P(CH2CH2CN)3
(
entries 7–9, Table 1): under similar conditions the rate
(
NCCH2CH2)3P
increases in the order MeCN < Me CO ≈ Me SO. More sig-
2
2
1
M = Pt
M = Pd
M = Ni
nificantly the solvent influences the course of the reaction. In
MeCN, the product is mainly tcep [δ(P) Ϫ24.0] but a small
amount (<5%) of a second product with similar δ(P) of Ϫ24.6,
is observed. Two possibilities were considered for the struc-
ture of the minor species: (i) the branched species X derived
from Markovnikov addition of the P᎐H to acrylonitrile; (ii)
the telomeric species Y which would be the product of mul-
tiple insertions of acrylonitrile into the P᎐H bond. The pro-
2
3
(
NCCH2CH2)3P
NCCH2CH2)3P
P(CH2CH2CN)3
M
(
Cl
4
5
M = Rh
M = Ir
portion of the second product is greater (ca. 20%) in Me CO
2
and is the main product (ca. 60%) in Me SO. Therefore we
2
investigated the products of hydrophosphination in these sol-
1
3
mirror. Addition of a fourth equivalent of tcep gave two broad
signals in the ratio of 3:1 at δ ϩ16.9 and Ϫ25.9 (the latter
assigned to free tcep) which had coalesced at ϩ28 ЊC to a very
broad singlet centred at ca. δ ϩ5, consistent with rapid
exchange of co-ordinated and non-co-ordinated tcep on the
NMR time-scale. Thus it appears that nickel(0) forms a co-
ordinatively unsaturated tcep complex 3 analogous to the
palladium(0) and platinum(0) complexes (1 and 2). Exchange of
tcep is observed for each of the species 1–3 and from the broad-
vents by C distortionless enhancements by polarisation trans-
fer (DEPT) NMR spectroscopy. In both solvents, the spectra
obtained were complex with several signals in the alkyl region
(δ 10–45) and CN region (δ 115–135) in addition to the tcep
signals. For X we would expect one new CH signal and one
new CH signal but significantly, there were at least five CH/
3
CH signals in the range δ 25–30 which is more consistent
3
with the formation of Y. This assignment is further supported
by the observation that the proportion of this product in
MeCN increased when the initial concentration of CH2᎐
CHCN was increased.
31
1
ening of the P-{ H} NMR signals, it is clear that the rate of
the exchange is greatest for nickel and slowest for platinum.
4
278
J. Chem. Soc., Dalton Trans., 1997, Pages 4277–4282

View Article Online
Fig. 3 The variation of time, t(50%), to equal proportions of dcep and
tcep against initial concentration of tcep, [tcep]
The course of the hydrophosphination was monitored as a
31
1
function of time by integration of the P-{ H} NMR spectra as
described in the Experimental section. A plot of a typical run
of this type is shown in Fig. 2(a) and from the attempts to fit
these data to first order [Fig. 2(b)] or second order [Fig. 2(c)]
behaviour, it is clear that the kinetics are not simple since the
data do not fit with any single value of α. It was suspected that
the kinetics may be complicated by the presence of increasing
amounts of product tcep as the reaction ensues. To test this, the
time taken for conversion of half the PH(CH CH CN) to
2
2
2
P(CH CH CN) was measured as a function of initial [tcep]
2
2
3
(
0–50 equivalents relative to Pt catalyst) and the unusual curve
plotted in Fig. 3 was observed. At low initial concentrations of
tcep, increasing [tcep] reduces the rate of reaction while at high
initial concentrations of tcep, increasing [tcep] increases the
rate of reaction. This behaviour can be explained if the rate law
has the form shown in equation (5). At high [tcep] the k term
1
δ
ε
rate = k [tcep] ϩ k /[tcep]
(5)
1
2
would dominate and at low [tcep] the k term would dominate.
2
This expression is consistent with parallel mechanistic path-
ways operating (see below).
The complexity of the kinetics of product formation [e.g. Fig.
2
(a)] is a reflection of the complexity of the dependence of this
rate on [tcep] concentration (Fig. 3); indeed since the ratio of
tcep] to [Pt catalyst] rises from 0 to 50:1 during a typical run,
[
the entire curve of Fig. 3 is traversed in each experiment. It is
likely that for similar reasons, the kinetics will also be a complex
function of the secondary phosphine (dcep) concentration. The
conclusion is that this reaction defies simple kinetic analysis!
However we have attempted to determine the sensitivity of the
Fig. 2 (a) Plot of [dcep] [mole fraction of PH(CH CH CN) ], deter-
2
2
2
mined from the integrals of spectra such as those in Fig. 1, against time,
t. (b) A plot of 1n[dcep] against time, t, illustrating that the kinetics are
not first order and (c) a plot of 1/[dcep] against time, t, illustrating that
the kinetics are not second order
rate on [Pt catalyst] and [CH ᎐CHCN] by measuring the effect
2
on t(50%) of varying their initial concentration while keeping
the other conditions constant. Doubling [Pt catalyst] halved
t(50%) while reducing [CH ᎐CHCN] by a factor of 30
2
increased the t(50%) by a factor of 30 (entries 10–12, Table 1).
It is tempting to draw quantitative significance from these num-
bers but in view of the above discussion of the complexity of
the reaction we will resist this and put these simple numerical
relationships down to a coincidence of many factors and con-
Kinetic investigation of the platinum(0)-catalysed reaction
A more detailed investigation of the hydrophosphination reac-
tion was carried out using the most effective of the catalysts,
[
Pt(tcep) ]. We initially made the assumption that the rate
3
equation would be of the form given in equation (3). Assuming
clude that [CH ᎐CHCN] and, not surprisingly, [Pt catalyst] are
2
factors in the rate equation.
α
β
γ
rate = k [PHR ] [CH ᎐CHCN] [Pt catalyst]
(3)
2
2
᎐
Some properties of platinum–tcep systems
the [Pt catalyst] term remains constant throughout (and there is
no catalyst degradation) and a large excess of CH ᎐CHCN (15
mol equivalents were used) is present, equation (3) reduces to
expression (4).
Insight into likely elementary steps in the hydrophosphination
catalysis [equation (2)] has been obtained from our study of the
2
᎐
properties of [Pt(tcep)
The co-ordinatively unsaturated 1 does not form a new
] and related systems.
3
α
31
1
rate = kЈ [PHR2]
(4)
species in the presence of a large excess of tcep but the P-{ H}
J. Chem. Soc., Dalton Trans., 1997, Pages 4277–4282
4279

View Article Online
(
ii )
R3P
R3P
R3P
+
–
PHR2
PR3
Pt
Pt
CN
R2P
H
CN
6
(i )
C
+
CN
–
PR3
(iii )
CN
R2
P
R3P
R3P
(vi )
R2P
R3P
(v )
R2P
R3P
CN
(iv )
H
PR3
H
Pt PR3
Pt
Pt
1/2 Pt
R P
3
Pt
+
PR3
+
PR3
PR3
H
P
R2
1
E
D
7
(viii )
(vii )
+
PR3
R2P
CN
Pt
R3P
H
F
Scheme 1 Mononuclear mechanism proposed for the hydrophosphination at high [tcep] (R = CH CH CN)
2
2
1
NMR signals are broadened indicating that tcep exchange takes
place on the NMR time-scale. This exchange very likely pro-
acteristic of the presence of a Pt᎐H bond. The small J(PtH) of
7
Ϫ1
948 Hz and low frequency ν(PtH) of 1967 cm reflect the high
ceeds via the 18-electron intermediate A shown in equation (6).
trans influence of the µ-PR2 ligands. Significantly, we have
found that binuclear 7 also catalyses the hydrophosphination of
acrylonitrile [equation (2)].
R3P
R3P
PR3
Pt
Pt PR3
+
PR3
(6)
PR3
PR3
R3P
Mechanistic hypothesis
A
The conclusions that can be drawn from our kinetic measure-
ments are complicated by our observation that the product tcep
strongly influences the rate in a way that suggests that more
than one pathway is operating. The mechanisms shown in
Schemes 1 and 2 form the basis of the discussion below.
Addition of 1 equivalent of acrylonitrile to 1 in Me SO
equation (7)] gave a new platinum-containing species quantit-
2
[
R3P
R3P
In Scheme 1, the simplest sequence of reactions is presented.
We have shown [equation (7)] that displacement of tcep from 1
by acrylonitrile to give 6 [step (i)] occurs readily and since
acrylonitrile was in large excess, 6 is likely to be the main Pt
species present. Exchange of tcep in 6 occurs on the NMR
time-scale and therefore displacement of tcep by PH(CH2-
CH CN) [step (ii)] to give C is reasonable. Oxidative addition
R3P
R3P
Pt PR3
+
(7)
Pt
+ PR3
CN
CN
6
CN
2
2
of the P᎐H bond to platinum(0), which is known with other
14
Pt
low-valent metals, is proposed [step (iii)] to give mononuclear
intermediate D (and its geometric isomers). The displacement
PR3
PR3
R3P
of acrylonitrile from D [step (iv)] would give the µ-PR complex
B
2
7
.
atively, characterised as 6 in solution from the AB pattern of its
The completion of the catalytic cycle from D could be
3
1
1
1
P-{ H} NMR spectrum and the doublet of doublets in the
achieved by β-hydrogen migration to the co-ordinated acrylo-
nitrile [step (v)] to give E followed by reductive P᎐C elimination
[step (vi)]; though literature precedent for neither of these reac-
tions is available for Pt compounds, insertion of acrylonitrile
95
1
Pt-{ H} NMR spectrum. The same complex is formed by
reduction of trans-[PtCl (tcep) ] with NaBH in the presence of
2
2
4
acrylonitrile (see Experimental section). While the NMR sig-
nals for 6 are sharp in the absence of tcep, the addition of 10
equivalents of tcep to solutions of 6 resulted in broad signals
15
16
into other M᎐H bonds and P᎐C reductive elimination have
been reported. Alternatively a Michael-type attack by the PR
2
(
w
₁
= 25 Hz) but no evidence for the re-formation of 1. We can
on the co-ordinated acrylonitrile [step (vii)] would yield the
metallacycle F which upon reductive formation of the C᎐H
bond [step (viii)] would regenerate 1; a similar metallacycle is
proposed in the mechanism for iridium()-catalysed hydro-
₂
�
conclude that while phosphine exchange in 6 takes place on the
NMR time-scale, presumably via B, the equilibrium shown in
equation (7) lies to the right with K >> 10.
17
The diplatinum complex 7 [equation (8)] was made in high
amination. Insertion of another CH ᎐CHCN into the Pt᎐C
2
bond in F would relieve the strain and may explain the form-
ation of telomeric species Y under certain conditions (see
above).
The terminal phosphido–platinum() complexes proposed
in the mechanism in Scheme 1 should be stable with respect to
R2
H
PR3
Cl
H
P
PR3
NEt3
(8)
Pt
+
PHR2
Pt
Pt
MeCN
R3P
R P
P
H
3
R2
µ-PR complexes only at high [tcep]. We suggest that in low
7
2
[
tcep] regimes, the mechanism shown in Scheme 2 is more
yield by treatment of trans-[PtHCl(tcep) ] with PH(CH -
likely. In this mechanism steps (a)–(c) are the same as steps (i)–
(iii) in Scheme 1 but then D dissociates a tcep [step (d)] to
give the binuclear species G which could also be formed from 7
[step (e)]. From G the cycle can be completed by β-hydrogen
2
2
CH CN) and has been fully characterised (see Experimental
2
2
31
1
section). The P-{ H} NMR signal at δ Ϫ160.9 is characteristic
of µ-PR ligands and the H NMR signal at δ Ϫ4.84 is char-
1
3
1
2
4
280
J. Chem. Soc., Dalton Trans., 1997, Pages 4277–4282

View Article Online
(
b )
R2
P
R3P
R3P
3
R P
+
–
PHR2
(c )
R2P
R3P
H
PR3
H
CN
Pt
Pt
Pt
Pt
Pt
PR3
2
R P
H
R P
3
P
CN
CN
CN
R2
6
H
C
D
7
(
e )
(
d ) – PR3
+
+
(a )
CN
– PR3
–
PR3
CN
R2
P
R2
R3P
R3P
R3P
H
P
(
g )
(f )
+ PR3
CN
Pt PR3
Pt
Pt
Pt
Pt
+
PR3
NC
P
PR3
P
H
R2
R2
NC
PR3
G
H
(j )
+
(h )
CN
Pt
R3P R2P
Pt
H
CN
H
R2
R P
2
R3P
P
CN
PR2 PR3
H
H
Pt
Pt
Pt
Pt
NC
P
I
P
H
PR3
R2
R2
NC
CN
H'
I'
Scheme 2 Binuclear mechanism proposed for the hydrophosphination at low [tcep] (R = CH CH CN)
2
2
migration [step ( f )] to give H (or more likely H’) followed by
reductive P᎐C elimination [step (g)] which are directly analo-
gous to steps (v) and (vi) in Scheme 1. Intramolecular Michael
addition in the binuclear G [step (h)] may give the species I (or
more likely I’) from which reductive formation of the C᎐H
bond [step ( j )] would regenerate 1.
presence of small amounts of polyacrylonitrile) but the
phosphorus-containing product was identified as 6 from the
18
NMR data which are very similar to those reported for
2
31
1
[Pt(PRЈ ) (η -CH ᎐CHCN)].
P-{ H} NMR [36.2 MHz,
3
2
2
1
(CD ) SO]: δ 18.1 [d, J(PP) 44, J(PtP) 3784], 16.4 [d, J(PP) 44,
J(PtP) 3345 Hz]. Pt-{ H} NMR [19.2 MHz, (CD ) SO]:
3
2
1
195
1
3
2
If we propose that at high [tcep] a mononuclear pathway
Ϫ550 (d,d). The same species 6 was identified in the following
(
Scheme 1), while at low [tcep] a binuclear pathway (Scheme 2),
experiment carried out in a 5 mm NMR tube. To a solution of
3
operates then we can rationalise the complicated dependence of
the rate on [tcep].
[Pt(tcep) ] (30 mg, 0.039 mmol) in (CD ) SO (0.4 cm ) was
3
3 2
3
added CH ᎐CHCN (0.015 cm , 0.228 mmol, 5.85 equivalents)
2
31 1
to give a pale yellow solution, the P-{ H} NMR spectrum of
which showed the presence of 6 and tcep only.
Conclusion
It has been demonstrated that the formation of P(CH2-
Preparation of [Pt H (tcep) {ì-P(CH CH CN) }] 7
2
2
2
2
2
2
CH CN) from PH(CH CH CN) and CH ᎐CHCN is catalysed
2
3
2
1
2
2
2
7
0
8
To a solution of trans-[PtHCl(tcep) ] (233 mg, 0.38 mmol) in
2
3 3
by tcep complexes of d and d metals and platinum(0) is the
most efficient. The mechanism of this self-replication probably
involves parallel mononuclear and binuclear processes. The
application of metal-complex-catalysed hydrophosphinations
to the synthesis of other phosphines is in progress.
MeCN (5 cm ) was added PH(CH CH CN) (0.066 cm , 0.38
2
3
2
2
mmol) followed by NEt (0.130 cm , 0.933 mmol) and the mix-
3
ture was stirred. After 10 min, the white solid product was fil-
3
tered off, washed with acetonitrile (5 cm ) and then diethyl ether
3
(
(
20 cm ) and dried (150 mg, 75%) [Found (Calc): C, 34.15
34.10); H, 4.15 (4.00); N, 12.00 (12.45) %]. IR (Nujol mull,
Experimental
Ϫ1
1
cm ): 2247m (νCN), 1967m (νPtH). H NMR [270 MHz,
(CD ) SO]: 2.77 (m, 10 H, CH ), 2.50 (m, 10 H, CH ), Ϫ4.84
[d × m, J(PH) 146, J(PtH) 948 Hz]. P-{ H} NMR [36.2
MHz, (CD ) SO]: 28.1 [m, J(PtP) 2051, J(PP) 293], Ϫ160.9
[m, J(PtP) 1777 Hz].
The general methods used were as described in recent papers
3
2
2
2
7
1
31
1
from this laboratory. The phosphine tcep and its complexes 1,
7
1
2
, 4 and 5 were made by routes previously reported. The
3
2
1
compounds PH(CH CH CN) and [Ni(cod) ] were purchased
2
2
2
2
from Strem.
Platinum(0)-catalysed reaction of PH with acrylonitrile
3
in situ Preparation of [Ni(tcep) ] 3
3
CAUTION: Phosphine gas is extremely toxic and should only
be handled in a well ventilated fume cupboard; the exit gases
were passed through a solution of commercial bleach to oxidise
In a glovebox under nitrogen solid [Ni(cod) ] (10 mg, 0.036
2
mmol) and tcep (21 mg, 0.109 mmol) were put in a 5 mm NMR
3
31
1
tube and dry CD CN (0.4 cm ) was added. The P-{ H} NMR
3
unreacted PH .
3
spectrum of the resulting amber solution showed at Ϫ44 ЊC a
sharp peak at δ ϩ16.7.
3
A 250 cm , three-necked flask equipped with a gas inlet and
3
outlet was charged with acetonitrile (100 cm ). The solution was
purged with nitrogen for 30 min and then [Pt(tcep) ] (77 mg,
2
3
Preparation of [Pt(tcep) (ç -CH ᎐CHCN)] 6
2
2
0
.10 mmol) was added. After purging for a further 10 min,
7
3
To a solution of trans-[PtCl (tcep) ] (100 mg, 0.153 mmol) in
MeCN (10 cm ) was added CH ᎐CHCN (0.50 cm , 0.760
mmol, 50 equivalents) followed by NaBH (30 mg, 0.793 mmol,
.18 equivalents) and the mixture was stirred. After 1 h, the
solvent was evaporated to dryness and the residue washed with
water (2 × 10 cm ). Yield 70 mg of crude product. Satisfactory
acrylonitrile (10 cm , 8.06 g, 152 mmol) was added and then
2
2
3
3
PH was admitted at a slow rate for ca. 6 h. The solution was
2
3
3
then purged with nitrogen for 20 min and an aliquot (0.4 cm )
4
31
1
5
removed for analysis by P-{ H} NMR spectroscopy which
showed the presence of four signals in approximately equal
intensity corresponding to the species PH (CH CH CN)
3
n
2
2
3Ϫn
elemental analyses were not obtained (possibly because of the
(n = 0–3).
J. Chem. Soc., Dalton Trans., 1997, Pages 4277–4282
4281

View Article Online
Platinum(0)-catalysed reaction of PH(CH CH CN) with
V. P. Wystrach, J. Org. Chem., 1961, 26, 5138; I. Macleod,
2
2
2
L. Manojlovic-Muir, D. Millington, K. W. Muir, D. W. A. Sharp
and R. Walker, J. Organomet. Chem., 1975, 97, C7; G. Muhlbach,
B. Rausch and D. Rehder, J. Organomet. Chem., 1981, 205, 343;
M. Antberg, C. Prengel and L. Dahlenburg, Inorg. Chem., 1984, 23,
acrylonitrile
3
Ϫ3
Acrylonitrile (1.62 cm , 24.5 mmol, 2.65 mol dm ) was added
to a solution of [Pt(tcep) ] (72 mg, 0.093 mmol, 0.01 mmol
3
Ϫ3
3
3
dm ) in acetonitrile (7.2 cm ). The phosphine dcep (0.43 cm ,
4
170; E. Arpac and L. Dahlenburg, Angew. Chem., Int. Ed. Engl.,
Ϫ3
2
.45 mmol, 0.265 mol dm ) was then added and the reaction
1982, 21, 931; L. D. Field and I. P. Thomas, Inorg. Chem., 1996,
35, 2546; D. M. Schubert, P. F. Brandt and A. D. Norman,
Inorg. Chem., 1996, 35, 6204; P. F. Brandt, D. M. Schubert and
A. D. Norman, Inorg. Chem., 1997, 36, 1728; M. L. J. Hackney,
D. M. Schubert, P. F. Brandt, R. C. Haltiwanger and A. D. Norman,
Inorg. Chem., 1997, 36, 1867; D. Semenzin, G. Etemad-Moghadam,
D. Albouy, O. Diallo and M. Koenig, J. Org. Chem., 1997, 62, 2414.
4 R. L. Keiter, Y. Y. Sun, J. W. Brodack and L. W. Cary, J. Am. Chem.
Soc., 1979, 101, 2638; J. A. Iggo and B. L. Shaw, J. Chem. Soc.,
Dalton Trans., 1985, 1009; P. M. Treichel and W. K. Wong, Inorg.
Chim. Acta, 1979, 33, 171; S. J. Coles, P. G. Edwards, J. S. Fleming
and M. B. Hursthouse, J. Chem. Soc., Dalton Trans., 1995, 1139;
P. G. Edwards, J. S. Fleming and S. S. Liyanage, Inorg. Chem., 1996,
35, 4563; P. G. Edwards, J. S. Fleming and S. S. Liyanage, J. Chem.
Soc., Dalton Trans., 1997, 193.
31
1
monitored by P-{ H} NMR spectroscopy and the time,
t(50%), required for the signals for dcep and tcep to be
approximately equal was measured (see Table 1). The same
conditions (referred to as standard conditions in Table 1) were
used to compare the other metal catalysts 2–5; the reaction with
the rhodium complex 4 was carried out in Me CO because it is
2
insoluble in MeCN.
31
1
Kinetic investigations of the platinum(0) catalysis by P-{ H}
NMR spectroscopy
3
In a typical experiment, acrylonitrile (4.86 cm , 73.4 mmol, 5.87
Ϫ3
mol dm ) was added to a solution of [Pt(tcep) ] (81 mg,
0
3
3
5
P. G. Pringle and M. B. Smith, J. Chem. Soc., Chem. Commun., 1990,
701.
Ϫ3
.105 mmol, 8.4 mmol dm ) in acetonitrile (6.77 cm ). Second-
3 Ϫ3
1
ary phosphine, dcep (0.86 cm , 4.90 mmol, 0.392 mol dm ) was
6 Homogenous Catalysis with Metal-Phosphine Complexes, ed.
L. H. Pignolet, Plenum Press, New York, 1983; Catalytic Aspects of
Metal Phosphine Complexes, eds. E. C. Alyea and D. W. Meek, John
Wiley & Sons, New York, 1980; Comprehensive Organometallic
Chemistry, eds. E. W. Abel, F. G. A. Stone and G. Wilkinson,
Elsevier, Amsterdam, 1995.
3
then added and 0.4 cm samples were removed at regular inter-
vals and frozen in liquid nitrogen. The P-{ H} NMR spectra
were later recorded and the intensity of the signals measured
31
1
(
see Fig. 2 for a typical plot). Similar runs were carried out at
Ϫ3
different [dcep] (0.392, 0.489 and 0.588 mol dm ) and
7
A. G. Orpen, P. G. Pringle, M. B. Smith and K. Worboys,
J. Organomet. Chem., 1997, in the press and refs. therein.
Ϫ3
[
CH ᎐CHCN] (0.20 mol dm ). The conditions were different
2
for the runs with different initial [tcep] (0, 8.4, 17.0, 41.8, 84.0
8 E. Wolf and M. Reuter, Chem. Abstr., 1961, 55, 16422c; W. J. Vullo,
Ϫ3
Ind. Eng. Chem., Prod. Res. Dev., 1966, 6, 346.
and 420 mmol dm ) and runs where [catalyst] was varied
Ϫ3
3
9 M. Reuter and E. Wolf, Chem. Abstr., 1961, 55, 16427c.
10 D. K. Wicht, I. V. Kourkine, B. M. Lew, J. M. Nthenge and
D. S. Glueck, J. Am. Chem. Soc., 1997, 119, 5039.
(
3.4, 5.6 and 11.1 mol dm ): more dcep (0.86 cm , 0.394 mol
Ϫ3 Ϫ3
dm ) and complex 1 (105 mg, 10.1 mmol dm ) were used and
the NMR signal intensities for tcep corrected for the amount of
tcep added initially.
1
1 L. B. Han and M. Tanaka, J. Am. Chem. Soc., 1996, 118, 1571.
12 P. A. T. Hoye, P. G. Pringle, M. B. Smith and K. Worboys, J. Chem.
Soc., Dalton Trans., 1993, 269.
1
3 J. B. Brandon and K. R. Dixon, Can. J. Chem., 1981, 59, 1188 and
Acknowledgements
We would like to thank EPSRC and Albright and Wilson
for CASE studentships and Johnson-Matthey for a loan of
platinum salts.
refs. therein.
1
4 R. A. Schunn, Inorg. Chem., 1973, 12, 1573; E. A. V. Ebsworth,
R. O. Gould, R. A. Mayo and M. Walkinshaw, J. Chem. Soc.,
Dalton Trans., 1987, 2831; E. A. V. Ebsworth and R. A. Mayo,
J. Chem. Soc., Dalton Trans., 1988, 477.
1
5 W. H. Baddley and M. S. Fraser, J. Am. Chem. Soc., 1969, 91, 3661;
S. S. Deshpande, S. Gopinathan and C. Gopinathan, J. Organomet.
Chem., 1989, 378, 103.
6 G. L. Geoffroy, S. Rosenberg, P. M. Shulman and R. R. Whittle,
J. Am. Chem. Soc., 1984, 106, 1519; M. D. Fryzuk, K. Joshi,
R. K. Chadha and S. J. Rettig, J. Am. Chem. Soc., 1991, 113, 8724.
7 A. L. Casalnuovo, J. C. Calabrese and D. Milstein, J. Am. Chem.
Soc., 1988, 110, 6738.
References
1
1
W. Wolfsberger, Chim.-Z., 1988, 112, 53 and refs. therein; L. Maier,
in Organic Phosphorus Compounds, eds. G. M. Kosolopoff and
L. Maier, Wiley-Interscience, New York, 1972, vol. 1, ch. 1 and refs.
therein.
1
1
2
For examples, see, B. D. Dombek, J. Org. Chem., 1978, 43, 3408;
M. M. Rauhut, I. Hechenbleikner, H. A. Currier, F. C. Schaefer and
V. P. Wystrach, J. Am. Chem. Soc., 1959, 81, 1103; P. N. Kapoor,
D. D. Pathak, G. Gaur and M. Kutty, J. Organomet. Chem., 1984,
8 Y. Xie and B. R. James, J. Organomet. Chem., 1991, 417, 277;
P. A. Chaloner, S. E. Davies and P. B. Hitchcock, Polyhedron, 1997,
1
6, 765.
2
76, 167; R. B. King, Adv. Chem. Ser., 1982, 196, 313.
3
For examples, see, M. M. Rauhut, H. A. Currier, A. M. Semsel and
Received 2nd July 1997; Paper 7/04655C
4
282
J. Chem. Soc., Dalton Trans., 1997, Pages 4277–4282