[PtOTf(triphos)]OTf and [PtMe2(triphos-P,P′)] as versatile synthons of platinum(II)-triphos species

Article abstract of DOI:10.1016/S0020-1693(01)00372-3

The new complex [PtOTf(triphos)]OTf (triphos = bis(2-diphenylphosphinoethyl)phenyl-phosphine, OTf = CF₃SO₃) (1) can be most efficiently prepared by adding triflic acid to the known complex [PtMe₂(triphos-P,P′)] (5) where triphos acts as a bidentate ligand. The fluxional behaviours of 1 and 5 in solution and their reactivity have been investigated by NMR: [PtOTf(triphos)]OTf is a very electrophilic complex and its reactivity is dominated by the tendency of the labile ligand OTf to be replaced by a variety of nucleophiles, while the chemistry of [PtMe₂(triphos-P,P′)] is controlled by the proclivity of the third phosphorus to coordinate to platinum, as soon as a vacancy is created via Pt-Me protonolysis.

Full text of DOI:10.1016/S0020-1693(01)00372-3

www.elsevier.nl/locate/ica
Inorganica Chimica Acta 316 (2001) 25–32
[PtOTf(triphos)]OTf and [PtMe2(triphos-P,P%)] as versatile synthons
of platinum(II)-triphos species
G. Annibale a, P. Bergamini b,*, M. Cattabriga a
a Dipartimento di Chimica, Uni6ersita` di Venezia, Calle Larga S. Marta, 2137, 31023 Venice, Italy
b Dipartimento di Chimica dell’Uni6ersita` di Ferrara, 6ia L. Borsari 46, 44100 Ferrara, Italy
Received 22 September 2000; accepted 26 January 2001
Abstract
The new complex [PtOTf(triphos)]OTf (triphos=bis(2-diphenylphosphinoethyl)phenyl-phosphine, OTf=CF3SO3) (1) can be
most efficiently prepared by adding triflic acid to the known complex [PtMe2(triphos-P,P%)] (5) where triphos acts as a bidentate
ligand. The fluxional behaviours of 1 and 5 in solution and their reactivity have been investigated by NMR: [PtOTf(triphos)]OTf
is a very electrophilic complex and its reactivity is dominated by the tendency of the labile ligand OTf to be replaced by a variety
of nucleophiles, while the chemistry of [PtMe2(triphos-P,P%)] is controlled by the proclivity of the third phosphorus to coordinate
to platinum, as soon as a vacancy is created via Pt–Me protonolysis. © 2001 Elsevier Science B.V. All rights reserved.
Keywords: Platinum complexes; Triphosphine complexes; Triflate complexes
[PtOTf(triphos)]OTf is a very convenient example of
this class of compounds: platinum has a consolidated
catalytic value, much is known about the relative ligand
affinity to this metal, mainly in terms of hard-soft
reactivity, Pt(II) complexes are stable and finally the
coexistence of two NMR active nuclei (31P and 195Pt)
makes this technique extremely diagnostic for mecha-
nistic studies and for the identification of reaction
products.
1. Introduction
The coordination chemistry of linear and tripodal
triphosphines is still a topic of general interest because
of their ability to give complexes with a variety of
metals, to support metal–metal bonds and to stabilise
unusual coordination numbers. Moreover their practi-
cal value is recognised because of their applications in
homogeneous catalysis [1].
Because of the inductive effects of CF3 and SO3,
CF3SO3 is one of the strongest electron withdrawing
groups and therefore is an excellent leaving group.
Moreover, in [PtOTf(triphos)]OTf, the high tendency of
the OTf group to dissociate is enhanced by the strong
labilising effect of the trans phosphine. Consequently
the metal centre becomes very electrophilic and the
complex behaves as a Lewis acid, showing a remarkable
susceptibility to nucleophilic attack at platinum.
For all the above reasons, [PtOTf(triphos)]OTf can
be regarded as a stabilised form of the coordinatively
unsaturated species [Pt(triphos)]2+; in fact on the one
hand it can be readily isolated as a solid and does not
show any tendency to decompose in an inert solvent, on
the other hand the labile OTf can be replaced by a wide
number of electron donors. These observations suggest
Another lively field of coordination chemistry is the
study of metal complexes containing poorly coordi-
nated ligands. In particular, transition metal complexes
containing a coordinated triflate have assumed an in-
creasingly important role in mechanistic and synthetic
endeavours [2].
Therefore, we reasoned that transition metal com-
plexes where a tridentate phosphine is associated with
the presence of a coordinate triflate on the same metal-
lic centre should be powerful tools for mechanistic
investigations and versatile inorganic synthons. For
several reasons the mono-cationic Pt(II) complex
* Corresponding author. Tel.: +39-053-229 1129; fax: +39-053-
224 0709.
E-mail address: bgp@dns.unife.it (P. Bergamini).
0020-1693/01/$ - see front matter © 2001 Elsevier Science B.V. All rights reserved.
PII: S0020-1693(01)00372-3

26
G. Annibale et al. / Inorganica Chimica Acta 316 (2001) 25–32
a potential catalytic role for this complex in processes
where electron-donating substrates need to be activated
by the coordination to a metal-based Lewis acid [3].
Another source of several Pt-triphos complexes is
[PtMe2(triphos-P,P%)], where triphos acts as a bidentate
ligand. In this paper it will be also shown how the
tendency of the dangling phosphorus to coordinate to
platinum makes this complex very susceptible to Pt–C
protonolysis.
2.2. Synthesis of [PtOTf(triphos)]OTf (1) (Scheme 1)
2.2.1. Method a
[PtCl(triphos)]Cl (2) (200 mg, 0.25 mmol) was dis-
solved in 5 ml of CH2Cl2 and treated with 2 equiv. of
AgOTf: the reaction mixture was stirred at room tem-
perature (r.t.) for 15 min and the AgCl filtered off over
celite. The clear solution was taken to dryness under
reduced pressure and the product, obtained as a white
powder, dried over P2O5 (200 mg, 77%).
2. Experimental
2.2.2. Method b
We have previously reported this route [7].
2.1. General procedures
[PtCl(triphos)]Cl [4], [PtMe2(COD)] [5] and
[PtMe2(dppe)] [6] were prepared by literature methods.
All the other chemicals were purchased (reagent
grade) and solvents were distilled before using. Elemen-
tal analyses were performed using a Carlo Erba instru-
ment model EA1110. FT-IR spectra were recorded on a
Nicolet 510P FT-IR instrument in KBr. NMR spectra
were recorded on a Bruker AM spectrometer 200 MHz
for 1H NMR (TMS internal reference), 188.15 MHz for
19F{1H} NMR (neat CFCl3 external reference) and
81.15 MHz for 31P{1H} NMR (H3PO4 85% external
reference). MS-FAB (fast atomic bombardment) spec-
tra were acquired by a Hewlett–Packard MS engine
HP5989 A mass spectrometer using a p-nitrobenzylal-
cohol matrix.
2.2.3. Method c
[PtMe2(COD)] (200 mg, 0.6 mmol) was dissolved in 5
ml of anhydrous CH2Cl2 and the solution was kept
under nitrogen atmosphere; 1 equiv. of triphos dis-
solved in the same solvent was added dropwise and the
reaction mixture was stirred at r.t. for 15 min. After
this time, 2.1 equiv. of HOTf were added and the
solution stirred for a further 10 min. The solution was
then rapidly washed with H2O and the volume reduced
by half. The product was precipitated with diethyl ether
as a white powder, filtered, washed with diethyl ether
and dried over P2O5, (525 mg, 85%). Anal. Calc. for
C36H33F6O6P3PtS2: C, 42.07; H, 3.24; S, 6.24. Found:
C, 42.25; H, 3.29; S, 6.18%. IR (cm−1, KBr): wcoord-OTf
1339, wOTf 1269. MS-FAB+: m/z 878, [PtOTf(triphos)]+;
Scheme 1.

G. Annibale et al. / Inorganica Chimica Acta 316 (2001) 25–32
27
2.4. Generation and NMR obser6ation of sol6ento
complexes in solution (Scheme 2)
In a typical experiment, 1 (15 mg, 0.0145 mmol) was
dissolved in acetone or acetonitrile (0.5 ml) and the
31P{1H} NMR spectrum was observed after 10 min (see
Scheme 2 and Table 1 for data).
The identity of the solvento complexes was also
confirmed by the NMR observation of the same species
either when complex 2 was treated with an excess of
AgBF4 in acetone or acetonitrile or when an excess of
HBF4 was added to a solution of 5 in the same
solvents.
2.5. Preparation of bicationic complexes (Scheme 2)
2.5.1. Synthesis of [Pt(PPh3)(triphos)](OTf )2
Scheme 2.
A total of 25.5 mg (0.097 mmol) of PPh3 was dis-
solved in 2 ml of dichloromethane and the solution was
added dropwise to a second solution containing 100 mg
(0.097 mmol) of 1 in 2 ml of the same solvent. After
checking the completeness of the reaction by 31P{1H}
NMR, the volume was reduced in vacuo to about 0.5
ml and Et2O was added to precipitate the product as a
white solid which was filtered and washed several times
with diethyl ether (90.4 mg, 72%).
Table 1
31P NMR data of [PtL(triphos)]2+
L
lPa
1JPtP (Hz) lPb 1JPtP (Hz)
a b
Acetone
80.5 3482
84.7 3294
94.0 2147
91.9 2738
78.3 2886
91.6 2788
97.3 2674
49.9 2482
47.8 2370
44.5 2375
47.9 2412
50.1 2436
47.7 2418
48.3 2192
Acetonitrile
a
PPh3
SMe2
Pyridine
N-Acetyl-methionine
CO
Anal. Calc. for C54H48F6O6P4PtS2: C, 50.27; H, 3.72;
S, 4.96. Found: C, 49.98; H, 3.76; S, 4.69%. 31P{1H}
NMR (CDCl3): see Table 1.
1
2
a Pc (coordinated PPh3): lPc 12.7, JPtP =2570 Hz, JP P =298
2.5.2. Synthesis of [Pt(SMe2)(triphos)](OTf )2
c
c
a
2
Hz, JP P =22 Hz.
c
A total of 36 ml (0.5 mmol) of SMe2 was added to a
solution containing 100 mg (0.097 mmol) of 1 in 2 ml of
dichloromethane. After checking the completeness of
the reaction by 31P{1H} NMR, the work up of the
reaction was the same as the previous (85 mg, 80%).
Anal. Calc. for C38H39F6O6P3PtS3: C, 41.87; H, 3.58;
S, 8.81. Found: C, 42.40; H, 3.59; S, 8.35%. 31P{1H}
b
729, [Pt(triphos)]+
.
31P{1H} NMR (CDCl3): l 77.6 (s,
’
1JPtP =3314 Hz, Pa); l 46.5 (s, JPtP =2517 Hz, Pb).
1
a
b
19F {1H} NMR (CDCl3): l −78.3 [s, (CF3SO3)].
2.3. Synthesis of [PtMe2(triphos-P,P%)] (5)
1
NMR (CDCl3): see Table 1. H NMR (CDCl3), rele-
vant data: PtS(CH3)2: l 1.92 (d with satellites 3JHPt=35
4
To a solution of [PtMe2(COD)] (200 mg, 0.6 mmol)
in 5 ml of anhydrous CH2Cl2 a second solution con-
taining 1 equiv. of triphos dissolved in 5 ml of CH2Cl2
was added dropwise. The reaction mixture was stirred
at r.t. for 15 min and then taken to dryness under
reduced pressure. The product, obtained as a white
powder, was washed with n-hexane and dried over P2O5
(364 mg, 80%). Anal. Calc. for C36H39P3Pt: C, 56.99; H,
5.18. Found: C, 56.73; H, 4.98%. 31P{1H} NMR
Hz, JHP=3.5 Hz).
2.5.3. Synthesis of [Pt(py)(triphos)](OTf )2
The procedure was the same as in the previous
reaction, using 39 ml (0.5 mmol) of distilled pyridine.
The product was obtained as a white solid (78 mg,
72%).
Anal. Calc. for C41H38F6NO6P3PtS2: C, 44.40; H,
3.52; N, 1.26; S, 5.80. Found: C, 44.91; H, 3.46; N,
1.10; S, 5.52%. 31P{1H} NMR (CDCl3): see Table 1.
1
2
1
(CDCl3): l 47.5 (t, JPtP=1795 Hz, JPP=20 Hz). H
NMR (CDCl3): Pt(CH3)2 l 0.65 [dt (5 lines 1:2:2:2:1)
2
3
3
with satellites, JPtH=70 Hz, JP H=7.5 Hz, JP H=4
2.5.4. Synthesis of
a
b
Hz, 3H] and l 0.75 [dt (5 lines 1:2:2:2:1) with satellites,
[Pt(N-acetylmethionine)(triphos)](OTf )2
3
3
2JPtH=70 Hz, JP H=7.5 Hz, JP H=4 Hz, 3H], CH2P
A solution of 23 mg (0.12 mmol) of N-acetyl-me-
thionine in 2 ml of dichloromethane was added drop-
wise to a second solution containing 100 mg (0.097
a
l 1.8–2.5 [unresolved multiplet, b8H], PhP l 7.2–7.8
[unresolved multiplet, 25H].

28
G. Annibale et al. / Inorganica Chimica Acta 316 (2001) 25–32
mmol) of 1 in 2 ml of the same solvent. After checking
the completeness of the reaction by 31P{1H} NMR, the
volume was reduced in vacuo to about 0.5 ml and Et2O
was added to precipitate the product as a white solid
which was filtered off and washed several time with
ether (96 mg, 81%).
Anal. Calc. for C43H46F6NO9P3PtS3: C, 42.36; H,
3.80; N, 1.15; S, 7.89. Found: C, 42.51; H, 3.85; N,
1.10; S, 7.99%. 31P{1H} NMR (CDCl3): see Table 1.
ture after 24 h: 78% with N3−, 67% with SMe− and
62% with I−). The 31P{1H} NMR data of the products
are reported in Table 2.
2.7. Synthesis of mono-cationic complexes (Scheme 2)
2.7.1. Synthesis of [PtN3(triphos)]OTf
A total of 17 mg (0.06 mmol) of tetra-n-butylammo-
nium azide were added to a solution of [PtOTf-
(triphos)]OTf (0.05 mmol, 51 mg) in 5 ml of CH2Cl2.
The reaction mixture was stirred at r.t. for 15 min and
then the volume reduced to a half. The product was
precipitated as a white powder by adding diethyl ether
and the solid filtered and washed several times with
2.5.5. Synthesis of [Pt(CO)(triphos)](OTf )2
Carbon monoxide was bubbled for 30 min into a
solution containing 51 mg of [PtOTf(triphos)]OTf (0.05
mmol) in 5 ml of CH2Cl2; the volume was then reduced
to a half under CO atmosphere and the product precip-
itated as a white powder with diethyl ether. The solid
was filtered and washed several times with ether (40 mg,
76%). Anal. Calc. for C37H33F6O7P3PtS2: C, 42.09; H,
3.15; S, 6.07. Found: C, 42.83; H, 3.35; S, 5.85%. IR
(cm−1, KBr): wOTf 1269, wCO 2110. 31P{1H} NMR
(CDCl3): see Table 1.
ether
C35H33F3N3O3P3PtS: C, 45.66; H, 3.61; N, 4.56; S, 3.48.
Found: C, 45.67; H, 3.58; N, 4.62; S, 3.55%. IR (cm−1
(40
mg,
86%).
Anal.
Calc.
for
,
KBr): wOTf 1271, wN3 2043. MS-FAB+: m/z 771,
[PtN3(triphos)]+; 744, [PtN(triphos)]+
(CDCl3): see Table 2.
.
31P{1H} NMR
’
2.7.2. Synthesis of [PtCCPh(triphos)]OTf
A total of 0.3 mmol of phenylacetylene were added
to a solution of [PtOTf(triphos)]OTf (0.05 mmol, 51
mg) in 5 ml of CH2Cl2 and the mixture was stirred at
r.t. for 2 h. After this time the volume was reduced to
a half under reduced pressure and the product was
precipitated with diethyl ether. The white powder was
filtered, washed with diethyl ether and dried over P2O5,
(40 mg, 80%). Anal. Calc. for C43H38F3O3P3PtS: C,
52.71; H, 3.91; S, 3.27. Found: C, 52.73; H, 4.03; S,
3.10%. IR (cm−1, KBr): wOTf 1263, wCꢀC 2114. MS-
FAB+: m/z 830, [Pt(CꢀCPh)(triphos)]+; 729,
[Pt(triphos)]+. 31P{1H} NMR (CDCl3): see Table 2.
2.6. Estimate of the relati6e rates of substitution at
platinum in [PtOTf(triphos)]OTf (1) and
[PtCl(triphos)]Cl (2) with X− (X−=N3−, SMe− and
I−)
Two equimolar (3×10−2 M) solutions of [PtOTf-
(triphos)]OTf and [PtCl(triphos)]Cl in CDCl3 were pre-
pared and three couples of 0.5-ml aliquots were placed
in NMR tubes.
A total of 8.6 mg of NBu4N3 (2 equiv.) were added to
each tubes of the first pair, 5.2 mg (5 equiv.) of NaSMe
to the second, and 11 mg of NBu4I (2 equiv.) to the
third pair and the completeness of the reactions was
monitored by 31P{1H} NMR: in each case 1 was com-
pletely reacted at the first observation (after 2 min)
while in the solutions containing 2, beside the reaction
products, the reagent complex was still the main species
after 24 h (% of [PtCl(triphos)]Cl in the reaction mix-
2.8. Protonolysis of [PtMe2(triphos-P,P%)], (5)-NMR
experiments and comparison with [PtMe2(dppe)]
(Scheme 3)
Three 0.5-ml aliquots of a 3×10−2 M solution of
complex 5 in CH2Cl2 were placed in NMR tubes and a
large excess of HOTf (10 ml), CF3COOH (10 ml) and
HOTs·H2O (p-toluensulfonic acid) (10 mg) were added,
respectively. After 5 min, the 31P{1H} NMR spectra
showed the formation of the final products [Pt-
(triphos)X]X (X=OTf, OOCCF3, OTs).
Table 2
31P NMR data of [PtX(triphos)]+
X
lPa
1JPtP (Hz)
lPb
1JPtP (Hz)
a
b
In a group of analogue experiments a stoichiometric
amount (2.1 equiv.) of the above acids was used: it was
observed that step (i) was complete in 5 min in each
case. Step (ii) is slower and occurred at different times
depending on the acid: using HOTf step (ii) is complete
in 10 min, in the case of HOTs the reaction mixture
contained 50% of the final product after 18 h and only
10% in the case of CF3COOH, after the same time.
The same experiment performed with an excess of
weaker acids, PhOH, HF (48% aq.) and CH3-
COCH2COCH3, gave the complete conversion of
OTf
Cl
Br
77.6
88.2
89.8
92.3
89.7
81.8
76.2
91.7
91.8
77.5
74.8
3314
3065
3016
2902
2190
2780
2622
2216
1998
3458
3362
46.5
43.1
42.6
42.7
45.4
42.5
40.6
41.4
38.8
47.2
46.1
2517
2456
2449
2417
2639
2553
2619
2373
2483
2521
2494
I
SMe
N3
OH
CN
PhCC
OTs
CF3COO

G. Annibale et al. / Inorganica Chimica Acta 316 (2001) 25–32
29
Scheme 3.
[PtMe2(triphos)] to [Pt(triphos)Me]+ after about 10
min. No sign of the second step product was observed
after 24 h. In a set of comparison reactions it was found
that using the same weak acids in the same conditions,
[PtMe2(dppe)] did not react at an appreciable extent
after 24 h.
[PtMe2(COD)] is treated in sequence with 1 equiv. of
triphos to give the known complex [PtMe2(triphos-
P,P%)] (5), where the triphosphine acts as a bidentate
ligand [9]. The solution containing 5 is then treated
with 2 equiv. of triflic acid to give 1 by double Pt–C
protonolysis. The synthesis is performed through two
simple and fast steps in sequence (there is no need to
isolate the intermediate 5) and product 1 is recovered
very cleanly in good yield (85%).
3. Results and discussion
3.1. Synthesis of [PtOTf(triphos)]OTf (1)
3.2. Solution beha6iour of 1 by NMR
The preparation of complex 1, [PtOTf(triphos)]OTf,
has not been described until our recent preliminary
report of this work [7].
The metal bonding of a poorly coordinating ligand
such as OTf can be induced by creating a vacancy in
the coordination sphere of a suitable parent complex in
the presence of the weak donor ligand in a non coordi-
nating solvent. This can be obtained by two conven-
We characterised 1 in the solid state through its
X-rays crystal structure [7], while its characterisation in
solution is based on NMR observations. The 31P{1H}
NMR shows two singlets with satellites at 77.6 ppm
(1JPtP =3314 Hz) and 46.5 ppm (1JPtP =2517 Hz) due
a
to the P trans to OTf (Pa) and to twbo mutually trans
1
equivalent Pb, respectively. The large value of JPtP
a
reflects the low trans effect of OTf, when compared
tional approaches: (i) abstraction of
a
metal
with other triphos Pt complexes, e.g. in [Pt-
1
coordinated halide from a parent complex using silver
or thallium reagents; (ii) C-protonation of a coordi-
nated R alkyl group (CH3 in most cases) with release of
the corresponding hydrocarbon RH.
(triphos)Cl]Cl JPtP =3065 Hz (Table 2). These values
a
also show that all P–Pt coupling constants in
(triphos)Pt systems are smaller than the corresponding
constants in single five-membered ring analogues [10].
We have already underlined that the line width of the
Pb signal is strongly dependent on a few physical condi-
tions as medium polarity, concentration and tempera-
ture and we tentatively ascribed this to the fluxional
behaviour described in Scheme 4: the nucleophilic at-
tack of the external OTf on platinum induces the
detachment of Pb, which then rapidly exchanges with
Pb% , via the five coordinate trigonal bipyramidal inter-
mediate (a).
We found both the above routes successful for the
preparation of [PtOTf(triphos)]OTf: in fact it can be
prepared either treating the corresponding chloride
complex [PtCl(triphos)]Cl (2), with silver triflate, in the
same way reported before for the palladium analogue
[8] (Scheme 1, method a) or treating [PtMe-
(triphos)]OTf (4), with triflic acid (Scheme 1, method
b). Nevertheless both reactions show some disadvan-
tages: the first route requires the use of light-sensitive
hygroscopic silver salts while the second needs several
steps as we reported before [7].
This hypothesis is supported by the analysis of the
conditions inducing the broadening of the signal of Pb:
in a polar solvent like CD3NO2, Pb is sharp, because the
solvation of the two ions [Pt(triphos)OTf]+ and OTf−
We propose here a third very convenient synthetic
route to complex 1 (Scheme 1, method c): the precursor

30
G. Annibale et al. / Inorganica Chimica Acta 316 (2001) 25–32
favours the ion pair indicated as 1 (left side) in equi-
librium (i). The line broadening appears in low-polarity
solvents like CDCl3, where the favoured species is a,
whose fluxionality requires rapid exchange between
a–b–c. In the same solvent the line broadening disap-
pears at high concentration because the overall process
becomes very rapid. A further support to the hypothe-
sis that a fluxional behaviour is operating is offered by
a low temperature NMR experiment. At −55°C in
CDCl3, the situation described by equilibrium (i) is
frozen and two 31P{1H} NMR patterns are observed
[78.2 ppm (3579 Hz) and 47.6 ppm (2460 Hz) for the
first; 77.7 ppm (3121 Hz, broad) and 44.3 ppm (2501
Hz, broad) for the other] which we tentatively attribute
to the two species 1 and a, observed in a 1:1 ratio.
Increasing the temperature, the above patterns coalesce
at 0°C. This experiment supports the hypothesis that
the fluxionality is initiated by the nucleophilic attack of
the external OTf on platinum. The addition of external
OTf− (LiOTf or AsPh4OTf) to a CDCl3 solution does
not have any appreciable effect on the line width: this
observation indicates that the concentration of OTf
does not affect the rate of the exchange.
We dissolved 1 in some potentially coordinating sol-
vents (S=acetonitrile, acetone, water, methanol) with
the aim to check if they are able to replace OTf on
platinum, as depicted in Eq. (1):
[Pt–OTf]++S“[Pt–S]2+ +OTf−
(1)
The 31P{1H} NMR data observed for 1 in acetonitrile
and acetone are reported in Table 1 [12]: the recorded
data show small differences with the data of 1 in a non
coordinating solvent as CDCl3. In order to exclude that
these changes are due to solvent effects, we generated
the genuine solvento complexes by two different routes:
(i) by treating the chloride complex 2 with an excess of
AgBF4 in acetone or acetonitrile and (ii) by treating the
dimethyl complex 5 with an excess of HBF4 in the
solvent to be coordinated. In every case the obtained
solvento complex had 31P{1H} NMR data exactly coin-
cident with those of the species derived from the OTf
complex. Attempts to isolate these solvento complexes
were unsuccessful.
The addition of water or methanol to a solution of 1
in CDCl3 or CH3NO2 does not induce any significant
change in the 31P{1H} NMR spectrum, indicating that
OTf is not replaced by these solvents in these
conditions.
3.3. Substitution reactions of coordinated OTf in
[PtOTf(triphos)]OTf (1)
3.3.2. Neutral ligands
The labile OTf ligand in [PtOTf(triphos)]OTf is read-
ily substituted by other anions or electron-donating
molecules offering a convenient route to a large number
of Pt(triphos) complexes (Scheme 2) [2]. The 31P{1H}
NMR data of the substitution products are listed in
Tables 1 and 2.
Besides the above-mentioned coordinating solvents,
some other neutral s donors ‘L’ can also replace OTf
giving di-cationic complexes [PtL(triphos)](OTf)2. We
found that the exchange occurs with P (PPh3), N (py,)
and S (SMe2) donors (see Section 2 and Table 1 for
complete characterisation).
The OTf ligand can also be replaced by more hin-
dered molecules as N-acetyl-L-methionine, giving a
3.3.1. Sol6ents
product whose 31P{1H} NMR data are consistent with
those of the reported complex bearing the amino acid
as S coordinated monodentate ligand [13]. It has been
suggested that this adduct could have some biological
significance; it has been prepared before by adding the
ligand to [PtCl(triphos)]Cl in the presence of AgNO3.
Our preparation from [PtOTf(triphos)]OTf has the ad-
vantage to avoid the reported contamination with the
side-product [PtONO2(triphos)]NO3.
Pt solvento complexes are of great interest because of
their common use in mechanistic studies and because
they are involved as intermediates in many transition
metal catalysed processes [11].
After prolonged (30 min) bubbling of CO into a
solution of [PtOTf(triphos)]OTf in CDCl3, the carbonyl
Pt complex [Pt(CO)(triphos)](OTf)2 can be identified in
solution by 31P{1H} NMR (Pt–P coupling constant for
Pa 2674 Hz typical of phosphorus trans to CO [14]).
The starting complex 1 is recovered as the only Pt
containing species when the solution is taken to dryness
in vacuo as shown by the 31P{1H} NMR spectrum of
the residue redissolved in CDCl3. This indicates that the
equilibrium in Eq. (2) is present in solution, lying on its
right side when the CO pressure is high.
Scheme 4.

G. Annibale et al. / Inorganica Chimica Acta 316 (2001) 25–32
31
Scheme 5.
[(triphos)Pt–OTf]++CO
= [(triphos)Pt–CO]2+ +OTf−
[PtOTf(triphos)]+ +HCꢀCPh
1
= [PtCꢀCPh(triphos)]++HOTf
(3)
(2)
Complex [PtCCPh(triphos)]OTf has been isolated and
its identity was confirmed by elemental analysis, IR,
MS-FAB (see Section 2). The triflate complex 1 is
re-obtained when an excess of triflic acid is added to a
CDCl3 solution of [PtCCPh(triphos)]OTf.
When [PtCCPh(triphos)]OTf is generated in CDCl3
inside a NMR tube from 1 and HCꢀCPh, an ageing
effect is observed whereby the equilibrium in Eq. (3) is
completely shifted to the left after 3 days.
[Pt(CO)(triphos)](OTf)2 can be isolated as a solid by
precipitation with Et2O under CO pressure. The IR
spectrum shows the typical w(CO) at 2110 cm−1. When
the isolated compound is redissolved in CDCl3, the
pattern of [Pt(CO)(triphos)](OTf)2 is initially observed
with 1 as a secondary species (approximately 20% after
3 min) whose concentration reaches about 50% in 2 h;
finally, after 48 h complex 1 becomes the only de-
tectable species. The observed lability of the coordi-
nated CO might be ascribed to the scarce metal to CO
backdonation, the platinum being highly electrophilic.
The relevance of this ability of Pt(triphos) to bond and
to release CO in hydroformylation catalysis is at
present under investigation.
3.4. NMR characterisation of 5
We have previously mentioned complex [PtMe2-
(triphos-P,P%)] (5), as an intermediate in the synthesis of
1 (Scheme 1, method c). It originates from the substitu-
tion of the easy replaceable chelating diolefin COD
with triphos acting as a bidentate ligand, the third
phosphorus dangling out of the coordination sphere.
The preparation of 5 was reported before as well as a
31P{1H} NMR study of its fluxional, temperature-de-
pendent behaviour in solution [9]. The structure of 5
was described in terms of the equilibrium in Scheme 5,
fast on the NMR scale at r.t. In these conditions the
31P{1H} NMR spectrum appears as a triplet with satel-
lites due to the central phosphorus Pa coupled with the
two external ones (Pb and Pb%). They are equivalent
because of the fast exchange and for the same reason
their signal is so broad to disappear in the spectrum
baseline.
3.3.3. Anionic ligands
OTf on platinum is rapidly substituted by a large
number of anionic nucleophiles X−, giving a series of
monocationic Pt complexes [PtX(triphos)]OTf, Scheme
1
2, where the value of JPtP can be used as an indicator
a
of the trans influence of the ligand occupying the forth
position (Table 2) [15].
Besides the expected reactions of 1 with Br−, I−,
SMe− and N3−, which occur much faster than on
[PtCl(triphos)]Cl (see Section 2), the reaction with
LiOH and water in chloroform gives a species which we
tentatively attribute to complex [PtOH(triphos)]+. The
same species appears also as a secondary product in the
31P{1H} NMR spectra of solutions of 1 in the presence
of non-anhydrous basic nucleophiles as pyridine or
NEt3.
It is worth noticing that Meek and co-authors tenta-
tively proposed an exchange mechanism through a five-
coordinated intermediate. In order to verify this
1
hypothesis we analysed the H NMR of 5. In CDCl3 at
3.3.4. Organometallics
r.t. (rapid exchange conditions, as shown by 31P{1H}
NMR) two distinct and identical patterns for the Pt–
Me groups are observed [0.65 and 0.75 ppm (doublets
of triplets with satellites, see Section 2)], indicating that
the two methyl groups do not become equivalent as a
result of the fluxionality. These observations fit the
structure of the trigonal bipyramidal intermediate I. In
Coordinated OTf can be replaced also by anionic C
donor ligands such as CN, CCPh [8], as proved by the
1
small value of the JPtP of Pa in 31P{1H} NMR of the
products (Table 1). For CCPh the coordination occurs
via a spontaneous deprotonation without the assistance
of an external base (Eq. (3)).

32
G. Annibale et al. / Inorganica Chimica Acta 316 (2001) 25–32
fact in that arrangement the two Pt–Me groups are
chemically inequivalent (note that the phenyl group on
the central phosphorus is always cis to one Me and
trans to the other with respect to the P3Pt plane), but
they have the same geometrical relations with all the P
atoms: therefore two signals are observed with identical
coupling patterns.
The ability of the dangling phosphorus to coordinate
to metals giving polinuclear complexes is at present
under investigation.
Acknowledgements
We thank MURST (Project: ‘Pharmacological and
Diagnostic Properties of Metal Complexes’) for finan-
cial support and the Centro di Fotoreattivita` e Catalisi
del C.N.R. (Ferrara, Italy) for the availability of the
NMR instrument.
3.5. Reacti6ity of [PtMe2(triphos-P,P%)] (5)
The addition of 2 equiv. of strong acids as HOTf,
HOTs and CF3COOH, to
a solution of 5 in
dichloromethane gives [Pt(triphos)Me]X as a primary
product which then evolves to [Pt(triphos)X]X (X=
OTf, OTs, CF3COO) as shown by 31P{1H} NMR. The
first step is fast, while the second is slower [16] (see
Section 2).
References
[1] (a) H.A. Mayer, W.C. Kaska, Chem. Rev. (1994) 1239. (b) S.
Herold, A. Mezzetti, L.M. Venanzi, A. Albinati, F. Lianza, T.
Gerfin, V. Gramlich, Inorg. Chim. Acta 235 (1995) 215. (c) T.
Tanase, H. Toda, Y. Yamamoto, Inorg. Chem. 36 (1997) 1571.
(d) C. Bianchini, A. Marchi, L. Marvelli, M. Peruzzini, A.
Romerosa, R. Rossi, A. Vacca, Organometallics 14 (1995) 3203.
[2] (a) G.A. Lawrence, Chem. Rev. 86 (1986) 17. (b) P. Bergamini,
F. Fabrizi De Biani, L. Marvelli, N. Mascellani, M. Peruzzini,
R. Rossi, P. Zanello, New J. Chem. (1999) 207.
[3] B. Bosnich, Aldrichim. Acta 31 (1998) 76.
[4] R.B. King, P.N. Kapoor, R.N. Kapoor, Inorg. Chem. 10 (1971)
1841.
[5] R. Bassan, K.H. Bryars, L. Judd, A.W.G. Platt, P.G. Pringle,
Inorg. Chim. Acta 121 (1986) L41.
[6] T.G. Appleton, M.A. Bennet, I.B. Tomkins, J. Chem. Soc.,
Dalton Trans. (1976) 439.
[7] G. Annibale, P. Bergamini, V. Bertolasi, M. Cattabriga, V.
Ferretti, Inorg. Chem. Commun. 3 (2000) 303.
The protonolysis of 5 by a non coordinating acid
(HBF4) in the presence of acetone or acetonitrile gives
the above mentioned solvento complexes very cleanly.
The treatment of 5 with HBF4 in a non coordinating
solvent (CH2Cl2 or benzene) gives an unidentified spe-
cies with all three P atoms bonded to platinum (lPa
1
1
48.7, JPtP 2464 Hz and lPb 77.3, JPtP 3675 Hz),
a
b
except in CHCl3 where the chloride complex 2 was
observed as the only Pt containing species after 24 h.
In a series of NMR experiments we compared the
susceptibility of 5 and [PtMe2(dppe)] to Pt-methyl pro-
tonolysis (Scheme 3). We observed that 5 undergoes the
first step very quickly (less then 10 min) even with
extremely weak acids like PhOH, HF or
CH3COCH2COCH3, which do not show any tendency
to induce protonolysis in [PtMe2(dppe)] after 24 h (see
Section 2).
[8] C. Scheffknecht, P. Peringer, J. Organomet. Chem. 535 (1997)
77.
[9] (a) K.D. Tau, D.W. Meek, J. Organomet. Chem. 139 (1977)
C83. (b) K.D. Tau, R. Uriarte, T.J. Mazanec, D.W. Meek, J.
Am. Chem. Soc. 101 (1979) 6614.
The driving force of the Pt–Me protonolysis in both
complexes is the formation of CH4 but in our opinion
the remarkable susceptibility of 5 to this reaction can
be ascribed to the presence of an additional driving
force in this system, namely the great tendency of the
third phosphorus to coordinate, evolving to species
where triphos behaves as a tridentate ligand, thus be-
coming stabilised by the formation of two five-mem-
bered chelate rings. The same driving force is probably
responsible of the above-described fluxional behaviour
of 5 in solution, as well.
[10] R.G. Peters, S. White, D.M. Roddick, Organometallics 17 (1998)
4493.
[11] J.A. Davies, F.R. Hartley, Chem. Rev. 81 (1981) 79.
[12] (a) D.L. DuBois, A. Miedaner, J. Am. Chem. Soc. 109 (1987)
113. (b) J.R. Briggs, C. Crocker, B.L. Shaw, Inorg. Chim. Acta
40 (1980) 245.
[13] P. Sevillano, A. Habtemariam, S. Parsons, A. Castineiras, M.E.
Garcia, P.J. Sadler, J. Chem. Soc., Dalton Trans. (1999) 2861.
[14] G.K. Anderson, R.J. Cross, J. Chem. Soc., Dalton Trans. (1980)
1988.
[15] D.W. Meek, T.J. Mazanec, Acc. Chem. Res. 14 (1981) 266.
[16] B.L. Bennet, J.M. Hoerter, J.F. Houlis, D.M. Roddick,
Organometallics 19 (2000) 215.
.