Synthesis, characterization and use in enantioselective hydroformylation of (BINAPO)PtCl2 (BINAPO = 2-diphenylphosphino-2'-diphenylphosphinyl-1,1'- binaphthalene), the first chiral catalyst with an atropisomeric hemilabile P,O-heterodonor ligand

Article abstract of DOI:10.1016/S1381-1169(98)00378-1

Platinum(II) complexes with the axially chiral phosphinyl phosphine (S)- BINAPO (1) have been prepared and their behaviour in solution has been studied by ³¹P-NMR spectroscopy. Reaction of PtCl₂(PhCN)₂ with 1 in benzene leads to the isolation of a neutral complex, 4, which maintains the P,O-chelate coordination of the ligand even in solvents of low polarity. The hemilabile character of the ligand is apparent from the reactions with DMSO and with carbon monoxide which promote the cleavage of the chelate ring of 4 through displacement of the oxygenated arm. Insertion of tin(II) chloride into the Pt-Cl bond takes readily place at room temperature affording only one of the possible trichlorostannato derivatives (6) with complete selectivity. In the presence of SnCl₂, the platinum complex 4 originates a catalyst of remarkable regioselectivity which, in the hydroformylation of styrene, is able to produce in up to 30% e.e. the branched aldehyde as the prevalent product.

Full text of DOI:10.1016/S1381-1169(98)00378-1

Ž
.
Journal of Molecular Catalysis A: Chemical 143 1999 155–162
www.elsevier.comrlocatermolcata
Synthesis, characterization and use in enantioselective
ž
/
hydroformylation of BINAPO PtCl2
ž
BINAPOs2-diphenylphosphino-
2 -diphenylphosphinyl-1,1-binaphthalene , the first
chiral catalyst with an atropisomeric hemilabile
P,O-heterodonor ligand
X
X
/
a
a
b
Serafino Gladiali a,), Elisabetta Alberico , Sonia Pulacchini , Laszlo Kollar
` `
`
a
Dipartimento di Chimica, UniÕersita di Sassari, Via Vienna 2, 07100 Sassari, Italy
Department of Inorganic Chemistry, Janus Pannonius UniÕersity, P.O. Box 266, H-7601 Pecs, Hungary
`
b
´
Received 15 May 1998; accepted 30 July 1998
Abstract
Ž .
Ž .
Ž .
Platinum II complexes with the axially chiral phosphinyl phosphine S -BINAPO 1 have been prepared and their
behaviour in solution has been studied by 31P-NMR spectroscopy. Reaction of PtCl2 PhCN with 1 in benzene leads to the
Ž
.
2
isolation of a neutral complex, 4, which maintains the P,O-chelate coordination of the ligand even in solvents of low
polarity. The hemilabile character of the ligand is apparent from the reactions with DMSO and with carbon monoxide which
Ž .
promote the cleavage of the chelate ring of 4 through displacement of the oxygenated arm. Insertion of tin II chloride into
Ž .
the Pt–Cl bond takes readily place at room temperature affording only one of the possible trichlorostannato derivatives 6
with complete selectivity. In the presence of SnCl2, the platinum complex 4 originates a catalyst of remarkable
regioselectivity which, in the hydroformylation of styrene, is able to produce in up to 30% e.e. the branched aldehyde as the
prevalent product. q 1999 Elsevier Science B.V. All rights reserved.
Keywords: Platinum complexes; Pt–Sn hydroformylation catalysts; Hemilabile coordination; Atropisomeric heterodonor ligands
1. Introduction
providing a free coordination site. Due to this
peculiar aptitude, these hemilabile ligands have
found several useful applications in the design
of those homogenous catalysts where dissocia-
tion–recombination processes are presumed to
Bidentate heterodonor ligands, in which a
soft phosphorus is flanked by a hard oxygen,
albeit able of chelate binding to the metal, may
easily undergo displacement of the oxygen donor
Ž
play an important role in the stabilization or
.
activation of the catalysts.
)
This feature appears of particular significance
in some metal-catalyzed carbonylation reac-
Corresponding author. Fax: q39-079212069; E-mail:
gladiali@ssmain.uniss.it
1381-1169r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved.
Ž
.
PII: S1381-1169 98 00378-1

(
)
156
S. Gladiali et al.rJournal of Molecular Catalysis A: Chemical 143 1999 155–162
Ž
.
tions. Several patents claim the activating prop-
erties of simple phosphine oxides in rhodium-
lary column J&W Scientific . The elemental
analyses were performed on a 1108 Carlo Erba
apparatus. The optical rotation of 2-phenylpro-
panal was measured in benzene solution on a
Perkin Elmer 241 polarimeter.
w
x
catalyzed hydroformylation 1–5 . Rhodium
w x
complexes with diphosphine hemioxides 6 and
w x
hemisulfides 7 have been found more active
than the classical Monsato catalyst in the car-
bonylation of methanol. Aminoalkylphosphine
oxides are better activators than the relevant
phosphines in the rhodium-catalyzed hydro-
2.2. Reagents
Commercial chemical reagents were used as
received and solvents were dried by standard
procedures and stored over molecular sieves
under inert atmosphere. Styrene was freshly dis-
w x
formylation of styrene 8 .
In a recent paper, we have reported the
preparation of 2-diphenylphosphino-2X-diphen-
X
Ž .
Ž . w x
tilled before use. S -BINAPO
1 9
and
10 were synthesized as previ-
Ž
.
ylphosphinyl-1,1-binapthalene BINAPO, 1 ,
w
Ž
.
x w
x
PtCl2 PhCN
the first example of a potentially hemilabile
2
ously reported.
w x
ligand with axial chirality 9 . This compound
X
Ž
.
has been obtained from 1,1-binaphtol BINOL
[
( )]
2.3. Synthesis of PtCl2 1
through a four-step procedure involving the Pd-
catalyzed sequential substitution of two homo-
topic triflate groups by diphenylphosphine oxide
and is an effective chiral inducer in the Pd-
catalyzed asymmetric hydrosilylation of styrene
Ž
.
w
Ž
.2 x
Some 0.25 mmol 118 mg of PtCl2 PhCN
was dissolved in 12 ml refluxing benzene, and a
Ž
.
solution of 0.25 mmol 159.7 mg of 1 in 3 ml
benzene was added. The mixture was refluxed
for 2.5 h. A pale yellow powder-like solid was
formed, which after cooling the mixture to room
temperature, was filtered off. The product was
washed with cold benzene and dried in vacuum.
Analysis calculated for C44H32OCl2 P2 Pt
Ž
.
e.e.)70% . It acts as a bidentate ligand to-
Ž .
wards Pd II centres, affording P,O-chelate com-
plexes which have been isolated in the solid
state and are fairly stable even in solution.
These promising figures stimulated us to in-
vestigate the catalytic activity of the correspond-
Ž
.
MWs904.67 : C, 58.42; H, 3.57; Cl, 7.84.
Ž .
ing trichlorostannato platinum II complex in
Found: C, 58.30; H, 3.24; Cl, 8.08. Yield: 89%.
the hydrofomylation of styrene. In this paper,
we report the results of this investigation and
some aspects of the coordination chemistry of
The product was characterized by NMR, IR and
Raman spectroscopy Table 1 .
Ž
.
Ž .
ligand 1 towards platinum II which are signifi-
Table 1
cant as to the catalytic process.
Selected NMR and IR data of 1 and its platinum complexes
1
Ž .
d P
Ž
.
x
Ž
.
Ž
cm
.
Ž
cm
.
d PO
w
J Pt,P
x
n P5O
n Pt–Cl
y1
y1
w
x
w
w
x
w
x
ppm
y14.7
9.2d
ppm
Hz
2. Experimental
1
3
27.8
26.5
52.6
26.4
26.5
48.3
26.7
25.3
1201
ca. 3600
3883
2640
4
7.5
27.1
26.0
16.7
13.0d
22.7
1161
351a, 283b
2.1. General methods
5a
5b
6
7
8
2625
1
The H- and 31P-NMR spectra were recorded
3812c
ca. 3230
3750
in CDCl3 on a Varian Unity 300 spectrometer
at 300 and 121.4 MHz, respectively. The 31P
chemical shifts were reported relative to H3PO4.
The samples from the catalytic runs were anal-
ysed with a Hewlett Packard 5890A gas chro-
matograph fitted with a 30 m cyclodex-b capil-
aCl trans to P5O.
bCl trans to P.
c P cis to SnCl3
J
Ž2 Ž117,119
.
Sn,31P s185 Hz, 117Sn and 119Sn
.
satellites coincide .
d Broad signal.

(
)
S. Gladiali et al.rJournal of Molecular Catalysis A: Chemical 143 1999 155–162
157
Ž
.
R -2-phenylpropanoic acid are 15.9 and 16.2
2.4. Hydroformylation experiments
min, respectively.
In a typical experiment, a solution of 0.025
w
Ž .x
mmol of PtCl2 1 and 0.05 mmol of SnCl2 in
30 ml toluene containing 50 mmol of styrene
was transferred under argon into a 150 ml stain-
less steel autoclave. The reaction vessel was
3. Results and discussion
w
Ž .x
The synthesis of the PtCl2 1 complex was
Ž
pressurised to 80 bars total pressure COrH2 s
1r1 and placed in an oil bath and the mixture
was stirred with a magnetic stirrer. The pressure
was monitored throughout the reaction. After
cooling and venting, the solution was removed
and conversion and composition of the reaction
products were determined by GC 1108C . The
enantiomer excess was determined in the same
way on the corresponding carboxylic acids ob-
tained by KMnO4 oxidation 11 of a sample of
the crude reaction mixture. Retention times
carried out by reacting the appropriate amount
.
w
Ž
.2 x2
of PtCl2 PhCN
racemic or enantiopure S -configuration lig-
with one equivalent of the
ŽŽ .
.
Ž
.
and 1 in refluxing benzene Scheme 1 . It is
clearly apparent from the IR and 31P-NMR
spectra of the isolated complex 4 that both
ancillary benzonitrile ligands have been dis-
placed and that P,O-bidentate coordination of
the ligand 1 to the metal affording the neutral
P,O-chelate complex 4 has taken place. The
most significant spectroscopic data of complex
4 are collected in Table 1.
Ž
.
w
x
Ž
.
Ž .
from 1008 to 1708 at 58Crmin of S - and
Scheme 1.

(
)
158
S. Gladiali et al.rJournal of Molecular Catalysis A: Chemical 143 1999 155–162
The 22.2 and 24.8 ppm downfield shifts ob-
served in the 31P-NMR for the PPh2 and
disappears completely, being mainly converted
into the chelate complex 4. At the same time
another species is built up in solution at a lower
rate and this one, at the end of the transforma-
tion, accounts for about the 30% of the whole
product. In the 31P-NMR, this complex shows
two peaks at 27.1 and 26.4 ppm. It is readily
apparent from these chemical shifts that only
the PPh2 group of the ligand 1 is coordinated to
the metal, while the phosphinyl group is not
Ž .
P O Ph2 of complex 4 with respect to the free
ligand provides a strong indication that both
donor arms of the ligand 1 are bound to the
metal. Further support comes from presence of
1
31
Ž195
.
J
Pt, P s3883 Hz, which indicates that
the PPh2 group is directly bound to platinum
and occupies a position trans to a chloro ligand.
On the contrary, the phosphinyl phosphorus does
not exhibit any coupling with the metal. This
fact is in keeping with previous observations
that magnetization exchange between phospho-
rus and platinum is suppressed when an oxygen
1
31
Ž195
.
bound. The value of the J Pt, P s2640 Hz
indicates that the phosphorus of the PPh2 group
is opposed by a donor of strong trans influence.
From these data this product can be attributed
the structure 5a, where two moles of ligand
w
x
is interposed 12 .
Ž .
IR and Raman spectra are consistent with the
chelate coordination of the ligand. Upon com-
plexation the P5O stretching band undergoes a
S -1 are bound to the platinum with a trans
geometry.
This attribution is confirmed by the result
1
shift from 1201 to 1161 cmy as a consequence
observed when the NMR experiment was per-
formed on the same Pt-precursor using the lig-
and 1 in racemic form. In this case, one more
species with chemical shifts and coupling con-
stants quite similar to 5a is present in solution
w x
of the slight weakening of the P5O bond 6 . In
the far IR region, the two separate bands at 351
1
and 283 cmy can be attributed to the Pt–Cl
trans to oxygen and phosphorus, respectively.
Ž
.
The reaction of 2 with the enantiopure ligand
Ž .
Table 1 . This is due to the formation of the
S -1 in CDCl3 solution, monitored by 31P-
meso-complex 5b, which contains two moles of
ligand of opposite chiral notation. Notably, the
coordination to platinum of the second unit of
BINAPO takes place with a pronounced chiral
recognition which leads to the preferential for-
mation of the meso-compound 5b with 70%
diastereoselectivity.
NMR, showed that in the early stages of the
reaction, the concentration of the chelate com-
plex 4 is quite low and that the prevailing
products are two different species which contain
either one or two moles of ligand 1 bound to the
metal through the PPh2 donor in a monodentate
fashion.
In view of its use as hydroformylation cata-
At the very beginning, the main product
lyst, the reactivity of the chelate complex 4
Ž .
shows a broad peak at 9.2 ppm with Pt-satellites
towards tin II chloride and carbon monoxide in
1–4–1 pattern; J Pt, P f3.600 Hz and a
sharp singlet at 26.5 ppm, attributable to a
CDCl3 solution has been followed by 31P-NMR
spectroscopy. At 258C, complex 4 undergoes a
smooth carbene-like insertion of SnCl2, afford-
ing a poorly soluble derivative. The presence of
the trichlorostannato ligand in complex 6 is
evidenced by the characteristic tin satellites of
the peak of the bound PPh2 phosphorus. The
value of these coupling constants is about 200
1
31
Ž
Ž195
.
.
Ž .
bound PPh2 and to a free P O Ph2 groups,
respectively. This NMR pattern suggests for this
product the structure 3, corresponding to the
substitution of one ancillary PhCN by one
equivalent of monodentate ligand 1. The broad-
ening of the first signal is most probably the
consequence of the fast exchange experienced
by the benzonitrile ligand of 3.
2Jcis
Ž
.
Hz and, as usual for
Sn,P , they are coinci-
dent for the 117Sn and Sn nuclides 13 . This
provides convincing evidence that the insertion
of SnCl2 in 4 has occurred with complete posi-
119
w
x
The concentration of compound 3 decreases
steadily at room temperature and within 1 h, it

(
)
S. Gladiali et al.rJournal of Molecular Catalysis A: Chemical 143 1999 155–162
159
tional selectivity into the Pt–Cl bond trans to
the oxygen donor, leading exclusively to 6
DMSO. This results in the formation of a new
Ž
.
Ž
complex Table 1 , most probably 8 Scheme
Ž
.
.
Scheme 1 . The preference of the trichlorostan-
1 , where one DMSO has displaced the oxygen
nato ligand for the position trans to the harder
donor is not unprecedented and seems to config-
ure a general trend for the insertion of tin
donor taking up the position cis to the PPh2
group. Taken together with the previous results
with carbon monoxide, this fact gives a measure
of the lability of the P5O donor in ligand 1.
w
x
chloride into inequivalent Pt–Cl bonds 14 .
When complex 4 is treated with carbon
Ž
.
monoxide under pressure 25 bars at 258C in a
high-pressure NMR tube, a complete and fast
displacement of the P5O fragment takes place
3.1. Catalytic hydroformylation of styrene pro-
(
)
moted by PtCl2 BINAPO complex
Ž
.
and some ligand about 10% is completely
The preformed complex 4 has been inspected
for its catalytic activity in the hydroformylation
released from the metal. The 31P-NMR shows a
1
31
Ž195
.
Ž
.
new peak at 13.0 ppm, J Pt, P s3230 Hz,
attributable to the still bound PPh2 arm. Under
25 bars of CO, this is rather broad Dn1r2 f220
Hz , probably due to the fast exchange of a
carbonyl ligand in cis position, but becomes
sharper when the pressure is reduced to 1 bar
Dn1r2 f120 Hz . Even in these mild condi-
tions, the phosphinyl oxygen is almost com-
pletely displaced, whereas no release of free
ligand 1 is noticed. These NMR data and the
presence in the IR spectrum of carbonyl stretch-
ing bands at 2108, 2022 and 1998 cmy1, sug-
gest that under CO pressure, the carbonyl com-
plex 7 and some undefined dinuclear carbonyl
complexes should be contextually present in
of styrene Scheme 2 .
The catalytic reactions were run in toluene at
a substrate-to-metal ratio 2000:1 in the presence
of two equivalents of SnCl2 as a promoter, at
70–1008C under 80–90 bars of an equimolar
mixture of CO and H2. The reaction product
consisted of a mixture of branched and linear
aldehydes, 9 and 10, respectively, and some
ethylbenzene 11 arising from hydrogenation of
the substrate. The most significant results are
collected in Table 2.
Ž
.
Ž
.
The catalytic activity of complex 4, either
alone or in the presence of additional free ligand
1, is fairly low as compared to similar
platinum-derived catalysts containing chelating
Ž
.
w x
solution Scheme 1 .
diphosphines, such as BINAP 15 , as chiral
modifiers. Several hours of reaction were re-
quired to attain moderate conversions. The ac-
tivity of the catalyst improved slightly after
addition of one more equivalent of free ligand,
but remained modest.
A similar set of experiments were run on
complex 6, but they led to inconclusive results,
probably due to the low solubility of this deriva-
tive.
Easy displacement of the oxygenated arm of
the ligand 1 resulting in the opening of the
chelate ring is observed also when complex 4 is
dissolved in solvent of high donicity such as
The chemoselectivity of the reaction was
surely more satisfactory since the amount of
ethylbenzene ranged between 2 and 15%, which
Scheme 2.

(
)
160
S. Gladiali et al.rJournal of Molecular Catalysis A: Chemical 143 1999 155–162
Table 2
Hydroformylation of styrene in the presence of 4 and SnCl2 as in situ catalysta
Run
Catalyst
Reaction
w x
Temp.
Conv.b
RCc
RRd
w x
%
w
8C
x
w x
%
w
x
%
time h
1
2
3
4
5
6
4qSnCl2
22
21
43
21
7
90
6
82
86
85
90
93
94
59
53
52
53
41
41
4q1qSnCl2
4q1qSnCl2
4qSnCle2
100
100
100
100
100
34
57
9
21
51
Ž
.
.
PtCl2 binap qSnCl2
Ž
PtCl2 PPh3 2 qSnCl2
23
a
Ž
.
Ž
.
Reaction conditions: 0.025 mmol Pt-complex; 0.05 mmol SnCl2; 50 mmol styrene; solvent: toluene; p CO sp H2 s40 bar.
bmmol productrmmol initial substrate=100.
c
Ž
. Ž
.
Chemoselectivity towards hydroformylation: 9q10 r 9q10q11 =100.
d
e
Ž
.
Regioselectivity towards branched aldehyde: 9r 9q10 =100.
Ž
.
Ž
.
p CO s80 bar; p H2 s40 bar.
is quite acceptable and even better than usual
for a platinum-catalyzed hydroformylation reac-
tion. The most interesting feature of this cata-
lyst, however, is related to the regioselectivity
of the reaction. The branched isomer in this
case is formed in an amount definitely higher
than with the usual mono- and bidentate phos-
phorus ligands. In all but one of the experi-
ments, the aldehyde 9 was, albeit for a neck, the
main product of the reaction and its share was
as high as 60% in the best run. This result is
notable since a prevalence of the branched over
the linear aldehyde has been never noticed in
the hydroformylation of styrene with plati-
reaction, the experiments were run at a styrene-
to-metal ratio as low as 200:1.
This resulted in a substantial reduction of the
reaction times necessary for attaining a reason-
able conversion and in improved chemoselectiv-
ities as the amount of ethylbenzene was limited
to 2–4%. The branched isomer accounted for up
to 60% of the aldehydes in the experiment run
at 608C, whereas, only 40% was obtained at
858C. This latter value is definitely lower than
the one observed under comparable conditions
Ž
but at a lower catalyst concentration Table 2,
.
entry 2 . It is our feeling that this decrease can
be hardly ascribed solely to the difference of
metal concentration, but that it may probably
arise from the presence of two different di-
astereomeric catalytic species, each one featur-
w
x
num–phosphine catalysts 16,17 . This is appar-
ently the first time that a reversal of the usual
selectivity is obtained in this reaction without
making use of dibenzophosphole or structurally
related derivatives as supporting ligands for the
Table 3
w
x
Ž .
Pt-catalyst 18 .
Asymmetric hydroformylation of styrene in the presence of S -4
and SnCl2 as in situ catalysta
The enantioselective hydroformylation exper-
iments have been carried out using the pre-
formed complex 4 containing the enantiopure
Run Catalyst
T
Conv.b RCc
RRd
e.e.e
w x
%
w
8C
x
w x
%
w
x
%
w
x
%
Ž .
1
2
S -4q1qSnCl2 85
100
59
96
98
40
60
30
28
Ž .
S -ligand in the presence of an additional
equivalent of free S -1 and two equivalents of
SnCl2 Table 3 . In a comparative experiment,
run in the conditions of entry 3 of Table 1 with
the enantiopure catalyst, a slight decrease of the
Ž .
S -4q1qSnCl2 60
Ž .
a
Ž .
Ž
.
Reaction conditions: 0.025 mmol Pt-complex; 0.025 mmol S -4;
0.05 mmol SnCl2; 5 mmol styrene; solvent: benzene 6 ml;
Ž
.
Ž
.
p CO s p H2 s45 bar; reaction time: 24 h.
bmmol productrmmol initial substrate=100.
cChemoselectivity towards hydroformylation: 9q10 r 9q10q
Ž
.
.
Ž
. Ž
branched selectivity was noticed 47% vs. 52%
.
11 =100.
Ž
and the e.e. was as low as 8% R-configuration .
In the aim to minimize the extent of racemiza-
tion of the chiral aldehyde 9 in the course of the
d
Ž
.
Regioselectivity towards branched aldehyde: 9r 9q10 =100.
e In both cases, the configuration of the prevailing enantiomer is
Ž
.
R .

(
)
S. Gladiali et al.rJournal of Molecular Catalysis A: Chemical 143 1999 155–162
161
ing its own regoiselectivity, when using the
racemic ligand 1. This may provide a hint into
the way to 4 or by an additional external phos-
phino donor on the way to 5 can be expected to
occur at the intermediate stage on compound 3.
The intramolecular process should be favoured
when the second benzonitrile to be displaced is
Ž .
the real structure of the catalyst s active under
our conditions.
The prevailing enantiomer of the branched
Ž
.
Ž
.
aldehyde is R -configured and its e.e. is mod-
in cis position Scheme 1 , but the intermolecu-
lar substitution should prevail if the leaving
ligand is trans to 1. We may then speculate that
5 can originate from a species such as 3 but
with a trans geometry. This should be highly
reactive towards substitution due to the labiliz-
ing effect of the phosphino donor on the trans
PhCN leaving group. The lability of this species
may well account for the fact that we were not
able to collect any NMR evidence of its pres-
ence in the reacting mixture.
Ž
.
est not higher than 30% and poorly dependent
on the reaction temperature. This means that the
direction of the asymmetric induction is oppo-
site to the one observed in the asymmetric
hydroformylation of styrene with BINAP-based
w
x
platinum catalysts under similar conditions 15
and that our catalytic system is significantly less
stereoselective. Both these facts provide support
to the assumption that the catalytic species orig-
Ž .
Ž .
inated from S -BINAPO and S -BINAP may
involve a different type of binding of the chiral
ligand at the metal centre.
Complex 5 readily insert SnCl2 affording the
trichlorostannato derivative 6 in quantitative
yield. The reaction is remarkable for its selectiv-
ity and this is a positive feature in view of the
fact that 6 is the putative catalyst in the hydro-
formylation of styrene. We do not know, how-
ever, if the oxygen of the trichlorostannato
species is displaced by CO under hydroformyla-
tion conditions as it occurs in the case of the
dichloro complex 4. Although we have no direct
evidence of this process, this possibility finds
some support from the results observed in the
enantioselective reactions which are more com-
fortably assessed if one assumes that two cat-
alytic species, each one containing two mono-
hapto BINAPO units around the Pt-centre, are
present during the reaction. The modest stereo-
selections, the significant differences of regiose-
lectivity recorded going from the racemic to the
enantiopure BINAPO-based catalyst and the re-
4. Conclusions
It has been already shown previously that the
binaphthyl phosphinyl phosphine 1 shows a
sharp preference for P,O-chelate coordination to
Ž .
w x
palladium II centres 9 . The same apparently
Ž .
occurs in the case of platinum II derivatives
which, by reaction with 1, produce the chelate
species 4 as the main or the exclusive product.
The hemilabile character of the ligand in com-
plex 4 is, however, evident from the reactions
with carbon monoxide and dmso, which readily
afford products 7 and 8 where the entering
ligand has taken up the position cis to the
phosphorus vacated by the oxygen.
Monodentate coordination of two equivalents
Ž .
of 1 through the phosphino donor is noticed in
versal of handness observed with S -BINAPO
w
Ž
.
2 x
Ž .
the reaction of PtCl2 PhCN
with excess 1.
as compared to S -BINAP are all factors which
This constitutes a minor reaction path leading to
5. This compound, however, shows a trans
geometry and most probably its formation does
not involve the intermediacy of 4, but should
take place directly from 3 via a stepwise substi-
tution of the ancillary benzonitrile ligands.
Competition for the substitution of the second
PhCN ligand by the internal oxygen donor on
better support this view.
Acknowledgements
Financial supports from the Italian Ministero
dell’Universita e della Ricerca Scientifica e
Tecnologica and from the Hungarian Research
`

(
)
162
S. Gladiali et al.rJournal of Molecular Catalysis A: Chemical 143 1999 155–162
w x
7
M. J. Baker, M. F. Giles, A. G. Orpen, M. J. Taylor, R. J.
Ž
.
Fund OTKA 23525 and FKFP 0818 are grate-
fully acknowledged. The authors would like to
Ž
.
Watt, J. Chem. Soc., Chem. Commun. 1995 197.
w x
8
Ž
.
C. Abu-Gnim, I. Amer, J. Organomet. Chem. 516 1996
235.
Ž
.
thank Dr. Antonio Zucca Universita di Sassari
`
w x
9
Ž
S. Gladiali, S. Pulacchini, D. Fabbri, M. Manassero, M.
and Gabor Szalontai NMR Laboratory, Univer-
`
Ž
.
Sansoni, Tetrahedron Asymmetry 9 1998 391.
.
sity of Veszprem for the NMR measurements
w
w
x
Ž
.
10 F.R. Hartley, Organomet. Chem. Rev. A 6 1970 119.
Ž
.
and Otto Berkessy Jozsef Attila University for
IR measurements.
`
`
x
11 A. Abiko, J.C. Roberts, T. Takemasa, S. Masamune, Tetrahe-
Ž .
dron Lett. 27 1986 4537.
w
w
w
w
w
w
w
x
12 S.J. Higgins, R. Taylor, B.L. Shaw, J. Organomet. Chem.
Ž
.
325 1987 285.
x
13 E. Farkas, L. Kollar, A. Sironi, M. Moret, Organometallics
`
References
Ž
.
15 1996 1345.
x
14 L. Kollar, S. Gladiali, M.J. Tenorio, W. Weissensteiner, J.
w x
Ž
.
1
w x
2
T. Onoda, Chemtech 1993 34.
R. M. Gipson, US Patent 3954877 1976 CA 85, 77653c .
A. G. Abatjoglou, EP Patent 73961 1983 CA 99, 53124u .
S. Arimitsu, K. Shikakura, JP Patent 02174740 1990 CA
Ž
.
Clust. Sci. 9 1998 321.
Ž
Ž
. w
. w
x
x
x .
15 L. Kollar, P. Sandor, G. Szalontai, J. Mol. Catal. 67 1991
`
Ž
w x
3
191.
w x
Ž
. w
4
x
16 S. Gladiali, J.C. Bayon, C. Claver, Tetrahedron Asymmetry 6
`
x
113, 39942f .
Ž
.
1995 1453.
w x
5
C. Myazawa, H. Mikami, S. Orita, Y Omori, JP Patent
0242040 1990 CA113, 190750s .
R. Wegman, A. G. Abatjoglu, A. M. Harrison, J. Chem.
Soc., Chem. Commun. 1987 1891.
x
17 F. Agbossou, J.F. Carpentier, A. Mortreux, Chem. Rev. 95
Ž
. w
x
Ž
.
1995 2485.
w x
6
x
18 S. Gladiali, D. Fabbri, L. Kollar, J. Organomet. Chem. 491
`
Ž
.
Ž
.
1995 91, and references cited therein.