The construction of P-O/P-N ligands on platinum and palladium

Article abstract of DOI:10.1002/ejic.200300642

A variety of diphenyldiphosphinites have been prepared in a very stable form as Pt^II or Pd^II complexes by adding diols to a solution containing cis-[MCl₂(PPh₂Cl)₂] (M = Pt or Pd) preformed in situ from cis-[MCl₂(1,5-COD)] (M = Pt; M = Pd) and PPh₂Cl. This synthetic pathway can be used for diphosphinites derived from aliphatic 1,2 and 1,3 diols and also from the aromatic catechol. Besides the P-O bond, it is also shown here that the P-N bond can also be formed via nucleophilic attack of an amine group on coordinated PPh₂Cl in the presence of a base resulting in some AMPP and BAMP complexes of Pt ^II and Pd^II. The X-ray crystal structures of the [cis-1,2-cyclohexanediyl bis(diphenylphosphinite)]platinum dichloride, of the AMPP palladium complex from (1R,2S)-(-)-norephedrine and of the BAMP palladium complex [P,P′-diphenyl-1,2-cyclohexanediylbis(imino)bisphosphane]- palladium dichloride have been determined. The chemical and optical stabilities of many complexes in solution have been monitored by ³¹P NMR spectroscopy. Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim, Germany, 2004.

Full text of DOI:10.1002/ejic.200300642

FULL PAPER
The Construction of P؊O/P؊N Ligands on Platinum and Palladium[‡]
Paola Bergamini,*[a] Valerio Bertolasi,[a] and Federico Milani[a]
Keywords: Template synthesis / Ligand design / P ligands / Platinum / Palladium
A variety of diphenyldiphosphinites have been prepared in
a very stable form as PtII or PdII complexes by adding diols
to a solution containing cis-[MCl2(PPh2Cl)2] (M = Pt or Pd)
preformed in situ from cis-[MCl2(1,5-COD)] (M = Pt; M = Pd)
and PPh2Cl. This synthetic pathway can be used for diphos-
phinites derived from aliphatic 1,2 and 1,3 diols and also from
the aromatic catechol. Besides the P−O bond, it is also shown
here that the P−N bond can also be formed via nucleophilic
attack of an amine group on coordinated PPh2Cl in the pres-
ence of a base resulting in some AMPP and BAMP complexes
of PtII and PdII. The X-ray crystal structures of the [cis-1,2-
cyclohexanediyl bis(diphenylphosphinite)]platinum dichlor-
ide, of the AMPP palladium complex from (1R,2S)-(−)-nore-
phedrine and of the BAMP palladium complex [P,PЈ-di-
phenyl-1,2-cyclohexanediylbis(imino)bisphosphane]-
palladium dichloride have been determined. The chemical
and optical stabilities of many complexes in solution have
been monitored by 31P NMR spectroscopy.
( Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
Germany, 2004)
Introduction
phinites,[5] AMPP (‘‘aminophosphanes-phosphinites’’) and
BAMP [‘‘bis(amino)phosphanes’’][6] have been recently con-
sidered particularly for their synthetic accessibility based on
the reaction of chlorophosphanes with diols, amino-ols and
diamines (Scheme 1).
P-donor chiral ligands are widely used in metal-based
asymmetric catalysis, because of the intrinsic ability of
phosphorus to favour the substrate coordination, to stabil-
ise intermediates of various processes and to provide selec-
tivity through steric effects.[2] So far the most significant
success is still represented by the asymmetric hydrogenation
of prochiral olefins using RhI complexes of chiral diphos-
phanes.[3]
Research in this field has been developed since the early
seventies and a large variety of chiral diphosphanes have
been prepared and tested. Although great success has been
achieved in the enantioselective conversion of some sub-
strates, a few problems still remain unsolved:
Scheme 1
i) Rh or Ru/chiral diphosphane catalysts are very efficient
but their activity is restricted to a narrow group of trans-
formations because only highly functionalised substrates
(e.g. amino acids precursors) undergo coordination, the
primary step of the process.
ii) The synthesis of chiral diphosphanes is often expensive
and time consuming and requires delicate and awkward
reactants.[4]
For these reasons the design and preparation of new li-
gands for chiral catalysts is still a lively topic and groups
of phosphorated compounds other than diphosphanes have
been taken into consideration. In particular chiral diphos-
The wide variety of asymmetric precursors belonging to
these groups, especially from the natural chiral pool, enor-
mously extends the number of easily accessible structural
variations on the basic structure of these ligands.
Although the classical preparation of 1,2-(diphenyl)di-
phosphinites from diols and chlorodiphenylphosphane has
been described in some papers as an easy clean process,[7]
some problems have been raised[8] and in our hands this
reaction gave poor and variable results because of the tend-
ency of both the reagent chlorodiphenylphosphane and of
the reaction products to undergo oxidative processes giving
a variety of PV species. On the other hand we noticed that
1,2-diphenylphosphinites are greatly stabilised by coordi-
nation to metal ions and therefore we have recently pro-
posed an alternative synthetic pathway where a few ex-
amples of diphosphinite ligands were built up in the coordi-
nation sphere of PtII or PdII acting as templates.[1]
[‡]
Part 2. Part 1: Ref.[1]
.
[a]
Dipartimento di Chimica
e
Centro di Strutturistica
`
Diffrattometrica dell’Universita di Ferrara,
Via L. Borsari 46, 44100 Ferrara, Italy
Fax: (internat.) ϩ 39-0532-240709
E-mail: bgp@unife.it
Eur. J. Inorg. Chem. 2004, 1277Ϫ1284
DOI: 10.1002/ejic.200300642
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1277

P. Bergamini, V. Bertolasi, F. Milani
FULL PAPER
With the aim to demonstrate the generality of the above broadness of the signal is due to the fast equilibrium be-
metal template method, in this paper we extend its appli- tween two conformational forms where one P atom occu-
cation to the preparation of different kinds of coordinated pies the equatorial position while the other one is axial and
diphosphinites, and also show that the PϪN bond can be vice versa.
formed by nucleophilic substitution on coordinated chloro-
phosphane by an amine group leading to aminophosphane-
phosphinites (AMPP) or bis(amino)phosphanes (BAMP).
A variable-temperature experiment showed, at low tem-
perature (Ϫ 55 °C), two distinct signals with satellites
1
(δPA ϭ 74.8 ppm, JPt,P ϭ 3834 Hz, δP ϭ 98.7 ppm
1JPt,P ϭ 4149 Hz), which coalesce at 25 °CBinto the above
described broad singlet, which becomes sharper at ϩ55 °C.
A
B
Results and Discussion
Solutions of 2a in pure chloroform or in a CHCl3/acetone
mixture, examined by 31P NMR, did not show any sign of
decomposition even in the presence of oxygen and/or water
after five days.
i) Preparation of Coordinated Diphosphinites
In our previous paper we showed that 1,2- and 2,3-bis(di-
phenylphosphinito)butanediols and isopropyl bis(diphenyl-
phosphinito)tartrate can be prepared as coordinated ligands
on Pt and Pd by a two-step process, starting from
[MCl2(1,5-COD)] (M ϭ Pt, Pd), whose reaction with
chlorodiphenylphosphane gave the intermediate complexes
cis-[MCl2(PPh2Cl)2] (1a, M ϭ Pt; 1b, M ϭ Pd), character-
ised in situ by 31P NMR spectroscopy and converted into
the relative diphenylphosphinite complex by addition of the
corresponding diols in the presence of a base (Scheme 2).[1]
A recrystallisation of 2a at room temperature from di-
chloromethane and diethyl ether gave crystals for X-ray
crystal structure determination.
In order to demonstrate also that aromatic bis(diphenyl-
phosphinites) can be formed on Pt, complex 3a derived
from catechol was prepared. The functionalisation of
catechol can be proposed as a simple model for more
complicated and more relevant aromatic systems like calix-
arenes.[10]
The 31P spectrum of 3a in CHCl3 is a sharp singlet with
satellites at δ ϭ 93.4 ppm (1JPt,P ϭ 4043 Hz); when the solu-
tion was treated with an excess of soluble CNϪ, the spec-
trum changed into a singlet at δ ϭ 114.1 ppm, due to the
free ligand 3. The following addition of [PdCl2(1,5-COD)]
to this solution gave the Pd analogue 3b in situ, showing a
singlet shifted to δ ϭ 122 ppm. This chemical shift trend,
with the Pd complex downfield and the Pt complex upfield,
relative to the free ligand, has been found in general for the
diphosphinites we have considered here and for the pre-
viously reported analogues.[1]
Scheme 2
In this paper we show that this template synthesis can
be extended to other ligands: in particular we prepared the
diphenylphosphinite from cis-1,2-cyclohexanediol in the
platinum-coordinated form. This ligand is the optically in-
active form derived from (1R,2S)-cyclohexanediol, and is
the diastereomeric form of the optically active (1R,2R)
or (1S,2S)-cyclohexane-1,2-diyl bis(diphenylphosphinite),
which have already been tested in asymmetric catalysis.[9]
Using the optically active (R,R)-(Ϫ)-2,4-pentanediol,
which gave the optically active Pt and Pd complexes 4a and
4b, we succeeded in demonstrating that it is possible to ob-
tain eight-membered rings by this route.
The platinum complex 4a has been obtained from
[PtMe2(1,5-COD] by treatment with PPh2Cl to give the di-
methyl intermediate, which is then treated with (R,R)-(Ϫ)-
2,4-pentanediol, giving the immediate precipitation of 4a.
Complex 2a co-precipitated with py·HCl from the reac-
The reaction occurs via nucleophilic attack of both the
tion mixture in THF and was purified by extraction with OH groups on the phosphorus atoms. The outgoing HCl
water and dichloromethane. causes the protonolysis of the two PtϪMe groups giving
Due to the equivalence of the two phosphorus atoms, the rise to the final dichloro complex. The use of the dimethyl
31P NMR of 2a shows a broad single signal at δ ϭ intermediate is very convenient because it avoids the use of
1
87.9 ppm, with satellites and JPt,P ϭ 3995 Hz, a typical the base and produces volatile CH4 as the only side prod-
value for a phosphorus atom trans to a chloride atom. The uct.
1278
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 1277Ϫ1284

Construction of PϪO/PϪN Ligands on Platinum and Palladium
FULL PAPER
Because of the instability of the corresponding precursor
[PdMe2(1,5-COD)], the corresponding palladium complex doublets with satellites at δ ϭ 43.5 ppm (PA, JPt,P
4b was obtained from [PdCl2(1,5-COD)].
The complete characterisation of 4a and 4b is reported 4075 Hz; JP ϭ 12 Hz) in the ranges expected for amino-
The 31P NMR spectrum of 5a shows two doublets of
1
ϭ
ϭ
A
1
3887 Hz; 2JP
ϭ 12 Hz) and 94.7 ppm (PB, JPt,P
BPA
B
2
APB
in the Exp. Sect. It is worth noting that for 4a the value phosphane and phosphinites respectively.
1
of JPt,P ϭ 4176 Hz, is more than 100 Hz larger than the
corresponding coupling constant in seven-membered ring
complexes.
The palladium complex 5b presents two doublets at δ ϭ
2
2
67.8 ppm (PA, JP ϭ 32 Hz) and 124.9 ppm (PB, JP
ϭ
BPA
APB
32 Hz) both at lower fields with respect to the correspond-
ing peaks in 5a.
Another example of optically active AMPP complexes,
6a and 6b, have been obtained via template synthesis using
the optically active amino alcohol (1R,2S)-(Ϫ)-norephed-
rine.
In 31P NMR, 6a is characterised by two doublets with
satellites, due to the aminophosphinic moiety 45.9 ppm (PA,
1JPt,P ϭ 3917 Hz and 2JP ϭ 11 Hz) and the phosphinitic
BPA
A
1
2
group at 93.8 (PB, JPt,P ϭ 4104 Hz and JP ϭ 11 Hz)
APB
B
ii) Preparation of Coordinated AMPP
ppm. 1H NMR spectroscopy is very useful for checking the
identity of the product and its purity (see Exp. Sect. for
complete assignments): in particular at δ ϭ 2.0 ppm a sing-
let is visible with satellites (3JPtH ϭ 55 Hz) belonging to the
nitrogen bonded proton.
After the preparation of a series of bis(diphenylphosphi-
nites) we reasoned that XϪP bonds other than OϪP could
be conveniently built by nucleophilic substitution of chlo-
ride on the coordinated phosphorus centre.
Due to the interest in the catalysis of both aminophos-
phane-phosphinites (AMPP) and bis(amino)phosphanes
(BAMP),[6] we tried to use amino groups (NH2 or NHR)
as nucleophiles in order to form NϪP bonds.
The palladium complex 6b, obtained by treatment of 1b
with (1R,2S)-(Ϫ)-norephedrine and NEt3 in benzene, shows
two doublets at δ ϭ 70.5 ppm (PA, 2JP ϭ 35 Hz) and
BPA
2
123.6 ppm (PB, JP ϭ 35 Hz). Its X-ray crystal structure
APB
By adding 1,2-amino-ols to a solution containing the pre-
formed intermediates 1a or 1b, we succeeded in obtaining
some examples of coordinated AMPP (Scheme 3).
has been determined (see further).
In these reactions the use of NEt3 as a base affects the
purity of the products, always contaminated by NEt3HCl,
which therefore require purification procedures.
In order to avoid salt contamination we reasoned that
the use of a heterogeneous base could be convenient. With
this aim, we tried to add solid NaHCO3 to the reaction
mixture in THF, together with a water sequestrating agent
(e.g. anhydrous Na2SO4). Under these conditions, the reac-
tion between 1b and (1R,2S)-(Ϫ)-norephedrine gave com-
plex 7b, where the OϪP bond is formed while the second-
coordinated PPh2Cl has been hydrolysed to PPh2OH by the
equivalent of water resulting from the reaction of NaHCO3
with HCl. This indicates that the dehydrating agents
(Na2SO4 or molecular sieves) that we tried, in order to
avoid this hydrolytic process, are not efficient in comparison
with the competing hydrolysis of the homogeneous coordi-
nated PPh2Cl.
Scheme 3
Complex 7b contains a coordinated monodentate amino-
phosphinite (AMP), a type of ligand mentioned previously
for its ability to give P/N six-membered chelate rings of
interest in catalysis.[6]
By following the reactions using 31P NMR spectroscopy,
we noticed that the formation of the OϪP bond is the first
step occurring in a few minutes, followed by the slow mak-
ing of NϪP, with ring closure.
In this case it is necessary to assist the reaction using
NEt3, which is a stronger base than the ϪNH2 group, in
order to avoid the protonation of the amine group, which
will make it unreactive.
We chose the methyl ester of the natural occurring amino
acid -serine, bearing an OH group in the β position, for
this reaction and we obtained the Pt complex 5a and the
Pd analogue 5b.
The reaction of preformed 1a with the aminoalcohol -
prolinol bearing a secondary amino group gave complex 8a.
Eur. J. Inorg. Chem. 2004, 1277Ϫ1284
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1279

P. Bergamini, V. Bertolasi, F. Milani
FULL PAPER
The NMR spectroscopic data of 8a are consistent with the iv) X-ray Crystal Structures of 2a, 6b and 9b
data previously reported for the same complex obtained by
By the slow diffusion of Et2O into CH2Cl2 solutions,
the classical synthetic procedure from the preformed li-
crystals suitable for structure diffractometric analysis were
obtained for one member of each group: diphosphinites
(2a), AMPP (6b) and BAMP (9b).
gand.[11]
An ORTEP[13] view of each structure (2a, 6b and 9b) is
given in Figure 1Ϫ3 and a selection of bond lengths and
angles is reported in Table 1. All the complexes present a
slightly distorted square-planar geometry with the metal
centre bonded to two chloride and two phosphorus atoms
belonging to the ligand. The seven-membered chelate rings
resemble twisted half-chair conformations.
iii) Preparation of Coordinated BAMP
Finally we have considered the possibility of extending
the metal template method to the preparation of BAMP li-
gands.
The most interesting member of this group is the
N,NЈ-bis(diphenylphosphanyl)-1,2-diaminocyclohexane
(DACHP2), whose palladium complex has recently been de-
scribed as an efficient tool for the resolution of racemic
amino acids.[12]
The platinum complex has been obtained from
[PtMe2(1,5-COD] by treatment with PPh2Cl to give the di-
methyl intermediate, which is then treated with (1R,2R)-
trans-diaminocyclohexane, resulting in the immediate pre-
cipitation of 9a. The corresponding palladium complex 9b
was obtained starting from [PdCl2(1,5-COD)] (see Exp.
Sect.). We determined the X-ray crystal structure of 9b,
recrystallised from dichloromethane and diethyl ether.
Figure 2. ORTEP view of compound 6b showing the thermal
ellipsoids at 30% level of probability
Figure 3. ORTEP view of compound 9b showing the thermal
ellipsoids at 30% level of probability
The molecules of complex 2a, in the crystals, display C1
and C6 atoms in configurations (R) and (S), respectively.
Since the solution does not show optical activity, it is evi-
dent that this compound crystallizes with separation of the
Figure 1. ORTEP view of compound 2a showing the thermal
ellipsoids at 30% level of probability
1280
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 1277Ϫ1284

Construction of PϪO/PϪN Ligands on Platinum and Palladium
FULL PAPER
˚
Table 1. Selected bond lengths (A) and angles (°) (* values of the four independent molecules of the asymmetric unit)
Compound
2a
6b*
9b
Pt1/Pd1ϪCl1
Pt1/Pd1ϪCl2
Pt1/Pd1ϪP1
Pt1/Pd1ϪP2
P1ϪO1
2.364(2)
2.359(1)
2.223(1)
2.218(1)
1.605(3)
1.597(4)
2.358(3), 2.359(3), 2.358(3), 2.358(3)
2.359(3), 2.345(2), 2.345(3), 2.357(3)
2.286(2), 2.261(2), 2.248(5), 2.239(2)
2.230(2), 2.252(2), 2.244(2), 2.222(2)
2.383(1)
2.367(1)
2.238(1)
2.263(1)
P2ϪO2
P2ϪO1
P1ϪN1
P2ϪN2
1.609(6), 1.598(6), 1.595(7), 1.614(7)
1.694(8), 1.663(8), 1.675(8), 1.670(8)
1.663(3)
1.656(3)
O1ϪC1
O1ϪC2
O2ϪC6
N1ϪC1
N2ϪC6
C1ϪC2
1.452(4)
1.456(5)
1.442(9), 1.463(9), 1.422(12), 1.440(10)
1.463(12), 1.485(12), 1.499(13), 1.450(12)
1.521(12), 1.546(13), 1.531(17), 1.536(12)
1.462(6)
1.462(5)
C1ϪC6
1.511(7)
88.19(4)
171.67(4)
88.82(4)
85.59(4)
177.01(3)
97.41(4)
118.5(1)
1.521(6)
91.13(3)
174.34(3)
86.24(3)
86.35(4)
176.87(3)
96.12(3)
Cl1ϪPt1/Pd1ϪCl2
Cl1ϪPt1/Pd1ϪP1
Cl1ϪPt1/Pd1ϪP2
Cl2ϪPt1/Pd1ϪP1
Cl2ϪPt1/Pd1ϪP2
P1ϪPt1/Pd1ϪP2
Pt1/Pd1ϪP1ϪO1
Pt1/Pd1ϪP2ϪO1
Pt1/Pd1ϪP2ϪO2
Pt1/Pd1ϪP1ϪN1
Pt1/Pd1ϪP2ϪN2
91.0(1), 91.0(1), 91.4(1), 89.6(1)
169.6(1), 169.6(1), 169.6(1), 163.5(1)
86.9(1), 88.5(1), 89.1(1), 88.5(1)
88.1(1), 86.1(1), 85.2(1), 90.1(1)
176.4(1), 176.4(1), 175.2(1), 173.6(1)
93.5(1), 95.0(1), 95.1(1), 93.5(1)
116.9(2), 114.6(2), 115.3(3), 119.5(3)
116.9(3), 116.1(3), 115.7(3), 118.1(3)
118.1(1)
120.0(1)
123.2(1)
cial suppliers. 1H and 31P NMR spectra (always 1H decoupled)
were recorded with a Bruker AC200 spectrometer operating at
200.13 MHz (1H) and 81.01 MHz (31P). Peak positions are given
in ppm relative to tetramethylsilane (1H) and to external 85%
H3PO4 (31P). Elemental analyses (C,H,N) were performed using a
Carlo Erba model 1106 elemental analyser. Optical rotations were
obtained with a PerkinϪElmer 241 at 25 °C.
enantiomers. The C1 and C2 atoms in complex 6b are in
configurations (S) and (R), respectively. In complex 9b both
C1 and C6 atoms display the (R) configuration. The mol-
ecules in the crystals are coupled by means of two
N2ϪH···Cl1 and N1ϪH···Cl2 hydrogen bonds: N2···Cl1(x
˚
ϩ 1, y, z) ϭ 3.236(3) A and N1···Cl2(xϩ 1, y, z) ϭ
˚
3.348(3) A.
Literature methods were used for the preparation of [PtCl2(1,5-
COD)],[14] [PtMe2(1,5-COD)][15] and [PdCl2(1,5-COD)].[16]
Conclusions
Synthesis of 2a: PPh2Cl (144 µL, 0.802 mmol) was added dropwise,
whilst stirring at room temperature, to a suspension of [PtCl2(1,5-
COD)] (0.150 g, 0.401 mmol) in THF (25 mL) placed in a Schlenk
flask. The immediate formation of the dichloro diphosphane com-
plex 1a was unambiguously confirmed by 31P NMR spectroscopy
The preparation of a variety of PtII and PdII complexes
through the metal template procedure proves that this new
method can be proposed as an easy and accessible general
synthetic pathway to PϪO and PϪN containing ligands,
which are obtained in a convenient metal-stabilized form.
1
(δ ϭ73.4 ppm, JPt,P ϭ 4069 Hz [1]).
The addition of pyridine (65 µL, 0.802 mmol) and cis-1,2-cyclo-
hexanediol (0.047 g, 0.401 mmol) gave a cloudy mixture, which per-
sisted whilst stirring at room temperature for 18 h. The white pre-
cipitate, composed of 2a and pyridinium chloride, was then filtered,
re-dissolved in CH2Cl2 and the solution was washed with 10 mL
of water three times, dried with Na2SO4 and then taken to dryness
under vacuum to give pure 2a as a white solid. Yield 0.170 g (56%).
C30H30Cl2O2P2Pt (750.5): calcd. C 48.01, H 4.03; found C 47.95,
H 4.31. 1H NMR (200 MHz, CDCl3, 25 °C): δ ϭ 1.2Ϫ1.5 (m, 4
H, CH), 1.6Ϫ1.9 (m, 4 H, CH), 4.5 (m, 2 H, CHϪO), 7.3Ϫ7.9 (m,
20 H, Ph) ppm. 31P NMR (81.01 MHz, CDCl3, 25 °C): δ ϭ
Experimental Section
All reactions and manipulations were routinely performed under
dry nitrogen by using standard Schlenk-tube techniques. Tetra-
hydrofuran was bi-distilled over LiAlH4, while diethyl ether and
benzene were purified by distillation over sodium/benzophenone.
Pyridine and triethylamine were distilled and stored over KOH.
CDCl3, used as solvent for NMR spectra, was dried with molecular
1
˚
87.9 ppm (br. s, JPt,P ϭ 3995 Hz). Recrystallisation of crude 2a
sieves (4 A).
from dichloromethane/ethanol gave crystals suitable for dif-
fractometric study.
PPh2Cl (d ϭ 1.23 g·mLϪ1) was purified by distillation under nitro-
gen, stored at Ϫ15 °C and checked by 31P NMR spectroscopy be-
fore use. All the other solvents and chemicals were reagent grade
and, unless otherwise stated, were used as received from commer-
Synthesis of 3a: Complex 3a was prepared in the same way as above
using catechol (0.044 g, 0.401 mmol). After 18 h a small amount of
Eur. J. Inorg. Chem. 2004, 1277Ϫ1284
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1281

P. Bergamini, V. Bertolasi, F. Milani
FULL PAPER
pyridinium chloride was filtered off, while the product remained in
solution. The product was confirmed by 31P NMR spectroscopic
inspection. The solution was taken to dryness under vacuum and
the oily residue was redissolved in CH2Cl2 (1 mL). Pure 3a, as a
white solid, was precipitated by addition of Et2O (15 mL) and sepa-
Synthesis of 6a: This complex was prepared in the same manner as
5a, but by replacing -serin methyl ester hydrochloride with
(1R,2S)-(Ϫ)-di(norephedrine) (0.041 g, 0.267 mmol). The reaction
gave 6a as a white solid. Yield 0.096 g (46%). C33H31Cl2NOP2Pt
(785.5): calcd. C 50.46, H 3.98, N 1.78; found C 49.96, H 4.03, N
3
rated by filtration; yield 0.270 g (90%). C30H24Cl2O2P2Pt (744.4): 1.98. 1H NMR (200 MHz, CDCl3, 25 °C): δ ϭ 0.95 (d, JH,H
ϭ
calcd. C 48.40, H 3.25; found C 47.91, H 3.35. 1H NMR (200 MHz,
7.3 Hz, 3 H, CH3), 2.0 (br. s, JPtH ϭ 55 Hz, 1 H, NH), 4.0 (br. s,
3
CDCl3, 25 °C): δ ϭ 6.1 (m, 2 H, CH), 6.85 (m, 2 H, CH), 7.4Ϫ8.1 1 H, CHϪN), 5.35 (m, 1 H, CHϪO), 6.8Ϫ8.2 (m, 25 H, CH arom.)
(m, 20 H, Ph) ppm. 31P NMR (81.01 MHz, CDCl3, 25 °C): δ ϭ
93.7 (s, JPt,P ϭ 4043 Hz) ppm.
ppm. 31P NMR (81.01 MHz, CDCl3, 25 °C): δ ϭ 45.9 (d, PA,
1
2
1
1JPt,P ϭ 3917, JP P ϭ 11.3 Hz), 93.8 (d, PB, JPt,P ϭ 4104,
B A
A
2JP P ϭ 11.5 Hz) ppm. [α]2D5ϭ Ϫ42.8 (c ϭ 0.54, CH2Cl2B).
A
B
Synthesis of 4a: [PtMe2(1,5-COD)] (0.100 g, 0.300 mmol) was dis-
solved in bi-distilled THF (20 mL) and PPh2Cl (108 µL,
0.600 mmol) was added. After checking the complete formation
Synthesis of 6b: [PdCl2(1,5-COD)] (0.150 g, 0.525 mmol) was sus-
pended in distilled benzene (25 mL) and PPh2Cl (189 µL,
of the Pt containing intermediate [PtMe2(PPh2Cl)2] by 31P NMR 1.050 mmol) was added to give the immediate formation of the
1
spectroscopy (δ ϭ100.3 ppm, JPt,P ϭ 1915 Hz), the solution was
yellow soluble intermediate 1b. Triethylamine (147 µL, 1.050 mmol)
treated with (2R,4R)-(Ϫ)-pentanediol (0.031 g, 0.300 mmol). The and (1R,2S)-(Ϫ)-norephedrine (0.080 g, 0.525 mmol) were added in
initially clear solution was kept at room temperature for three days
and during this time the progressive precipitation of a white solid
sequence, followed by the precipitation of NEt3HCl. After 18 hours
of stirring, the solution containing 6b was separated by filtration,
was observed. Later the pure solid product was filtered away and and taken to dryness. The yellow-orange solid product was purified
dried under vacuum over P2O5. Yield 0.140 g (63%).
C29H30Cl2O2P2Pt (738.5): calcd. C 47.17, H 4.09; found C 46.88,
by recrystallisation from CH2Cl2 and diethyl ether Yield 0.215 g
(59%). C33H31Cl2NOP2Pd (696.9): calcd. C 56.88, H 4.48, N 2.01;
1
3
H 4.24. H NMR (200 MHz, CDCl3, 25 °C): δ ϭ 0.8 (d, JH,H
ϭ
found C 56.54, H 4.48, N 1.91. 1H NMR (200 MHz, CDCl3, 25
3
3
7.3 Hz, 6 H, CH3), 1.8 (t, JH,H ϭ 7.3 Hz, 2 H, CH2), 4.35 (m, 2 °C): δ ϭ 0.95 (d, JH,H ϭ 7.3 Hz, 3 H, CH3), 2.0 (br. s, 1 H, NH),
H, CH), 7.3Ϫ8.1 (m, 20 H, Ph) ppm. 31P NMR (81.01 MHz,
3.9 (br. s, 1 H, CHϪN), 5.2 (m, 1 H, CHϪO), 6.8Ϫ8.2 (m, 25 H,
CDCl3, 25 °C): δ ϭ 91.8 ppm (s, JPt,P ϭ 4176 Hz). [α]2D5ϭ Ϫ42.6 Ph) ppm. 31P NMR (81.01 MHz, CDCl3, 25 °C): δ ϭ 70.5 (d, PA,
1
2JP P ϭ 35 Hz), 123.6 (d, PB, JP P ϭ 35 Hz) ppm. [α]2D5ϭ Ϫ13.2
2
(c ϭ 0.55, CH2Cl2).
B
A
A B
(c ϭ 0.53, CH2Cl2). Recrystallisation of crude 6b from dichloro-
methane/diethyl ether gave crystals suitable for a diffractometric
study.
Synthesis of 4b: PPh2Cl (126 µL, 0.700 mmol) was added to
[PdCl2(1,5-COD)] (0.100 g, 0.350 mmol) suspended in 25 mL of
dry THF: the solution turned yellow and the presence of 1b was
checked by 31P NMR spectroscopy (δ ϭ96 ppm, broad singlet) .[1]
Synthesis of 7b: PPh2Cl (189 µL, 1.050 mmol) was added to a sus-
At this point pyridine (57 µL) and (2R,4R)-(Ϫ)-2,4-pentanediol pension containing [PdCl2(1,5-COD)] (0.150 g, 0.525 mmol) in
(0.036 g, 0.350 mmol) were added to the solution which became
cloudy. After stirring for 24 h the solid (4b and pyHCl) was filtered,
25 mL of bi-distilled THF: the mixture became clear and yellow
and the complete formation of 1b was checked by 31P NMR spec-
redissolved in 1 mL of CH2Cl2 and extracted three times with water troscopy. A large excess (0.500 g each) of NaHCO3 and Na2SO4,
(10 mL). The organic phase was taken to dryness, resulting in 4b previously dried at 120 °C for 10 h, was suspended in this solution,
as a yellow powder. Yield 0.100 g (44%). C29H30Cl2O2P2Pd (649.8): followed by the addition of (1R,2S)-(Ϫ)-norephedrine (0.079 g,
calcd. C 53.60, H 4.65; found C 53.80, H 4.73. 1H NMR (200 MHz, 0.525 mmol). After stirring vigorously for 24 h, the salts were elim-
3
CDCl3, 25 °C): δ ϭ 0.75 (d, JH,H ϭ 7.3 Hz, 6 H, CH3), 1.8 (t, inated by filtration and the solution was taken to dryness, giving a
3JH,H ϭ 7.3 Hz, 2 H, CH2), 4.35 (m, 2 H, CH), 7.5Ϫ8.2 (m, 20 H,
solid residue of 7b; yield 0.220 g (59%), which was recrystallised
Ph) ppm. 31P NMR (81.01 MHz, CDCl3, 25 °C): δ ϭ 121.8 (s) with CH2Cl2 and diethyl ether. C33H33Cl2NO2P2Pd (714.9): calcd.
1
ppm. [α]2D5ϭ Ϫ61.4 (c ϭ 0.5, CH2Cl2).
C 55.44, H 4.65, N 1.96; found C 55.72, H 4.90, N 2.01. H NMR
3
(200 MHz, CDCl3, 25 °C): δ ϭ 1.1 (d, JH,H ϭ 7.3 Hz, 3 H, CH3),
3.0 (m, 2 H, NH2), 3.2 (br. s, 1 H, CHϪN), 4.6 (m, 3JH,H ϭ 8.2 Hz,
1 H, CHϪO), 7.0Ϫ8.0 (m, 25 H, Ph) ppm. 31P NMR (81.01 MHz,
Synthesis of 5a: PPh2Cl (96 µL, 0.535 mmol) was added dropwise
whilst stirring at room temperature to a suspension of [PtCl2(1,5-
COD)] (0.100 g, 0.267 mmol) in THF (25 mL), resulting in 1a. The
addition of triethylamine (0.802 mmol) and -serin methyl ester hy-
drochloride (0.042 g, 0.267 mmol) gave a cloudy mixture due to the
formation of partially insoluble NEt3 hydrochloride.
After 2 hours, another 50 mL of THF was added and the solution
was stirred for a further two hours. The white precipitate was then
filtered away and the solution, containing the impure product was
taken to dryness. The residue was dissolved in CH2Cl2 and water
was added in order to eliminate NEt3HCl. The organic layer was
finally dried with Na2SO4 and then taken to dryness under vacuum
2
CDCl3, 25 °C): δ ϭ 63.2 (d, PA, JP P ϭ 15 Hz), 115.4 (d, PB,
B
A
2JP P ϭ 15 Hz) ppm.
A
B
Synthesis of 8a: PPh2Cl (144 µL, 0.802 mmol) was added dropwise,
whilst stirring at room temperature, to a suspension of [PtCl2(1,5-
COD)] (0.150 g, 0.401 mmol) in THF (25 mL), resulting in 1a. The
addition of triethylamine (112 µL, 0.802 mmol) and 40 µL of -
prolinol (0.401 mmol, d ϭ 1.025 g·mLϪ1) gave a cloudy mixture
due to the formation of partially insoluble triethylamine hydrochlo-
ride. After stirring for 24 h at room temperature the salt was filtered
away and the solution treated as above (see preparation of 5a) to
obtain 8a as a white solid; yield 0.153 g (52%). C29H29Cl2NOP2Pt
to give pure 5a as
a white solid. Yield 0.090 g (45%)
C28H27Cl2NO3P2Pt (753.5): calcd. C 44.64, H 3.61, N 1.86; found
1
C 44.60, H 3.80, N 1.89. H NMR (200 MHz, CDCl3, 25 °C): δ ϭ (735.5): calcd. C 47.36, H 3.97, N 1.90; found C 46.96, H 4.12, N
3
1
3.4 (br. s, JPtH ϭ 45 Hz, 1 H, NH), 3.75 (s, 3 H, CH3O), 4.0 (m, 1.89. H NMR (200 MHz, CDCl3, 25 °C): δ ϭ 1.5Ϫ4.0 (m, 6 H,
1 H, CHϪN), 4.4 (m, 2 H, C-CH2O), 7.3Ϫ8.0 (m, 20 H, Ph) ppm. CH2), 3.3Ϫ3.8 (m, 2 H, CH2ϪO), 4.9 (br. s, 1 H, CHϪN), 7.0Ϫ8.0
31P NMR (81.01 MHz, CDCl3, 25 °C): δ ϭ 43.6 (d, PA, JPt,P
ϭ
(m, 25 H, Ph) ppm. 31P NMR (81.01 MHz, CDCl3, 25 °C): δ ϭ
1 2 1
1
3885, 2JP P ϭ 12 Hz), 94.9 (d, PB, 1JPt,P ϭ 4074, 2JP P ϭ 12AHz)
51.3 (d, PA, JPt,P ϭ 4030, JP P ϭ 14 Hz), 80.7 (d, PB, JPt,P ϭ
B A
A B
B
A
A
B
B
3941, JP P ϭ 14 Hz) ppm. [α]2D5ϭ ϩ17.8 (c ϭ 0.53, CH2Cl2).
2
ppm. [α]2D5ϭ Ϫ20.8 (c ϭ 0.58, CH2Cl2).
A B
1282
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 1277Ϫ1284

Construction of PϪO/PϪN Ligands on Platinum and Palladium
Synthesis of 9a: The complex was prepared from [PtMe2(1,5-COD)]
FULL PAPER
above. The optical rotations were redetermined for the same solu-
(0.100 g, 0.300 mmol), following the same procedure described tions after 3 days and each value was found unchanged.
above for 4a, replacing the diol with (1R,2R)-trans-diaminocy-
X-ray Crystal Structure Determinations: X-ray diffraction data for
clohexane (0.034 g, 0.300 mmol). The precipitation of 9a occurred
compounds 2a, 6b and 9b were collected on a Nonius Kappa CCD
within five minutes as a white solid, which was filtered and dried
diffractometer, at room temperature (T ϭ 295 K), with graphite
with P2O5; yield 0.170 g (77%). C30H32Cl2N2P2Pt (748.5): calcd. C
˚
monochromated Mo-Kα radiation (λ ϭ 0.7107 A) and corrected
48.14, H 4.31, N 3.74; found C 47.88, H 4.38, N 3.67. 1H NMR
for Lorentz polarization and absorption (SORTAV)[17] effects. The
(200 MHz, CD2Cl2, 25 °C): δ ϭ 0.6Ϫ1.0 (m, 4 H, CH), 1.5 (m, 2
structures were solved by direct methods (SIR97)[18] and refined
H, CH), 1.7 (d, 2 H, CH), 2.4 (s, 3JPtH ϭ 32 Hz, 2 H, NH), 3.3 (br.
(SHELXL-97)[19] by full-matrix least-squares with anisotropic non-
s, 2 H, CHϪN), 7.3Ϫ8.0 (m, 20 H, Ph) ppm. 31P NMR
hydrogen and hydrogen atoms on calculated positions, riding on
their carrier atoms.
1
(81.01 MHz, CD2Cl2, 25 °C): δ ϭ 40.6 (s, JPt,P ϭ 4003 Hz) ppm.
Synthesis of 9b: This palladium complex was prepared as described
above for 6b, using THF as solvent and replacing the amino-ol
Crystal Data. 2a: C30H30Cl2O2P2Pt·CH2Cl2; trigonal, space group
3
˚
˚
P32, a ϭ 9.8264(1), c ϭ 29.2991(2) A, V ϭ 2450.04(4) A , Z ϭ 3,
Dcalcd. ϭ 1.699 g·cmϪ3. Intensity data collected with θ Յ 30°; 9447
independent reflections measured; 8644 reflections observed [I Ͼ
with (1R,2R)-trans-diaminocyclohexane (0.060 g, 0.525 mmol). The
product co-precipitated with NEt3HCl in two hours and after fil-
tration was re-dissolved in CH2Cl2 and extracted three times with
water. The organic phase, taken to dryness, gave pure 9b as a yellow
solid; yield 0.160 g (46%). C30H32Cl2N2P2Pd (659.9): calcd. C
54.61, H 4.89, N 4.25; found C 54.68, H 4.70, N 4.55. 1H NMR
(200 MHz, CD2Cl2, 25 °C): δ ϭ 0.6Ϫ1.0 (m, 4 H, CH), 1.5 (m, 2
H, CH), 1.7 (d, 2 H, CH), 2.3 (s, 2 H, NH), 3.3 (br. s, 2 H, CHϪN),
7.3Ϫ8.0 (m, 20 H, Ph) ppm. 31P NMR (81.01 MHz, CD2Cl2, 25
°C): δ ϭ 65.9 ppm (s). [α]2D5ϭ ϩ32.8 (c ϭ 0.54, CH2Cl2). Recrystal-
lisation of crude 9b from dichloromethane/diethyl ether gave crys-
tals suitable for a diffractometric study.
2σ(I)]. Final R (observed reflections) ϭ 0.028 and Rw (all
[20]
reflections) ϭ 0.063, S ϭ 1.026, Flack parameter
ϭ Ϫ0.012(3).
6b: C33H31Cl2NOP2Pd·1/2CH2Cl2; triclinic, space group P1, a ϭ
˚
11.6343(2), b ϭ 14.3753(3), c ϭ 20.3486(4) A, α ϭ 93.0527(7), β ϭ
3
˚
92.4847(7), γ ϭ 91.6164(9)°, V ϭ 3393.6(1) A , Z ϭ 4, Dcalcd.
ϭ
1.447 g·cmϪ3. Intensity data collected with θ Յ 27°; 26342 indepen-
dent reflections measured; 20677 reflections observed [I Ͼ 2σ(I)].
Final R ϭ 0.098 (observed reflections) and Rw ϭ 0.281 (all reflec-
tions), S ϭ 1.029, Flack parameter ϭ 0.03(4). The asymmetric unit
contains four independent molecules of the complex and two mol-
ecules of dichloromethane.
Removal of Coordinated Ligands from Platinum and Their Transfer
to Palladium: This experiment was performed with the platinum
complexes 4a, 5a, 6a and 8a. In a typical experiment, an excess of
tetrabutylammonium cyanide (5 equiv.) was added to a solution
containing the Pt complex (0.010 g) in CDCl3 (0.8 mL). The pro-
gress of the reaction was monitored by 31P NMR spectroscopy.
After 15 min the spectrum showed the signals of the free ligand:
9b: C30H32Cl2N2P2Pd; monoclinic, space group P21, a ϭ 7.5164(1),
˚
b ϭ 16.2585(3), c ϭ 12.4538(2) A, β ϭ 96.4409(7)°, V ϭ 1512.32(4)
3
A , Z ϭ 2, Dcalcd. ϭ 1.449 g·cmϪ3. Intensity data collected with
˚
θ Յ 30°, 7940 independent reflections measured; 7416 reflections
observed [I Ͼ 2σ(I)]. Final R ϭ 0.039 (observed reflections), Rw ϭ
0.101 (all reflections), S ϭ 1.043, Flack parameter ϭ 0.00(2).
4: δ ϭ 105.8 ppm (s)
CCDC-219076, 219077 and CCDC-219078 contain the supplemen-
tary crystallographic data for this paper. These data can be ob-
tained free of charge at www.ccdc.cam.ac.uk/conts/retrieving.html
[or from the Cambridge Crystallographic Data Centre, 12 Union
Road, Cambridge CB2 1EZ, UK; Fax: (internat.) ϩ44-1223-336-
033; E-mail: deposit@ccdc.cam.ac.uk].
5: δ ϭ 116.9 ppm (s)
6: δ ϭ 112.8 ppm (s) (PO), 38.8 ppm (s) (PN)
8: δ ϭ 114.1 ppm (s) (PO), 46.4 ppm (s) (PN).
At this point an excess (2 equiv.) of [PdCl2(1,5-COD)] was added to
the solution and after ten minutes the signals of the corresponding
palladium dichloride complex were visible in the 31P NMR spec-
trum.
In an analogous manner the treatment of the palladium complex
9b with tetrabutylammonium cyanide in CDCl3, followed by the
addition of [PtCl2(1,5-COD)], allowed the observation of the spec-
trum of ligand 9 (δ ϭ 33.6 ppm, s) and then of 9a, assigned by
comparison with a genuine sample.
The palladium complex 7b, containing the monodentate amino-
phosphinite ligand 7, after addition of CNϪ, also gave the free
ligand [δ ϭ 111.5 ppm (s), PO] together with another unidentified
phosphorus containing product [δ ϭ 62.7 ppm (s)].
[1]
P. Bergamini, V. Bertolasi, M. Cattabriga, V. Ferretti, U. Lopri-
eno, N. Mantovani, L. Marvelli, Eur. J. Inorg. Chem. 2003,
918Ϫ925.
[2]
S. J. Berners-Price, P. J. Sadler, Structure and Bonding 1988,
70, 26Ϫ102.
[3]
W. S. Knowles, Angew. Chem. Int. Ed. 2002, 41, 1998Ϫ2007.
[4]
´
´
M. Nogradi, Stereoselective Synthesis, VCH, Weinheim, Ger-
many, 1987.
[5] [5a]
T. V. RajanBabu, S. Shin, Org. Lett. 1999, 1, 1229Ϫ1232.
N. Derrien, C. B. Dousson, S.M. Roberts, U. Berens, M. J.
[5b]
Burk, M. Ohff, Tetrahedron: Asymmetry 1999, 10, 3341Ϫ3352.
[5c]
W. Hu, M. Yan, C. Lau, S. M. Yang, A. S. C. Chan, Tetra-
hedron Lett. 1999, 40, 973Ϫ976.
Chemical Stability Tests: In a typical experiment, 10Ϫ2  solutions
of the complex in CDCl3, CH2Cl2, THF and benzene were pre-
pared without exclusion of air and the 31P NMR spectrum moni-
tored after 30 min, 5 h and after 3 days. No changes were observed.
A few drops of water were then added to each solution and the
spectrum checked after 3 min and then after 24 h. Under these con-
ditions all of the spectra were also found to be unchanged.
[6] [6a]
F. Agbossou, J. Carpentier, F. Hapiot, I. Suisse, A. Mor-
[6b]
treux, Coord. Chem. Reviews 1998, 178Ϫ180, 1615Ϫ1645.
F. Agbossou-Niedercorn, I. Suisse, Coord. Chem. Reviews 2003,
242, 145Ϫ158.
[7]
[8]
A. S. C. Chan, W. Hu, C. C. Pai, C. P. Lau, J. Am. Chem. Soc.
1997, 119, 9570.
L. D. Quin, A guide to Organophosphorus Chemistry, John
Wiley & Sons, New York, 2000, p. 59.
Optical Stability Tests: The values of [α]2D5 were determined for
solutions of complexes 4a, 4b, 5a, 6a, 6b, 8a and 9b as reported
[9] [9a]
V. I. Tararov, R. Kadyrov, T. H. Reirmeier, J. Holz, A.
[9b]
Börner, Tetrahedron: Asymmetry 1999, 10, 4009Ϫ4015.
V.
Eur. J. Inorg. Chem. 2004, 1277Ϫ1284
www.eurjic.org
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
1283

P. Bergamini, V. Bertolasi, F. Milani
FULL PAPER
[15]
I. Taranov, R. Kadyrov, T. H. Reirmeier, J. Holz, A. Börner,
R. Bassan, K. H. Bryars, L. Judd, A. G. Platt, P. G. Pringle,
Inorg. Chim. Acta 1986, L41ϪL42.
Tetrahedron Lett. 2000, 41, 2351Ϫ2355.
[10] [10a]
[16]
[17]
[18]
`
M. Stolmar, C. Floriani, A. Chiesi-Villa, C. Rizzoli, Inorg.
D. Drew, J. R. Doyle, Inorg. Synth. 1972, 13, 47.
R. H. Blessing, Acta Crystallogr., Sect. A 1995, 51, 33Ϫ38.
A. Altomare, M. C. Burla, M. Camalli, G. L. Cascarano, C.
Giacovazzo, A. Guagliardi, A. G. G. Moliterni, G. Polidori, R.
Spagna, J. Appl. Crystallogr. 1999, 32, 115Ϫ116.
G. M. Sheldrick, SHELXL-97, Program for Crystal Structures
Refinement, University of Göttingen, Germany, 1997.
H. D. Flack, Acta Crystallogr., Sect. A 1983, 39, 876Ϫ881.
Received September 11, 2003
[10b]
Chem. 1997, 36, 1694Ϫ1701.
C. Jeunesse, C. Dieleman, S.
Steyer, D. Matt, J. Chem. Soc., Dalton Trans. 2001, 881Ϫ892.
S. Na¨ıli, J. Carpentier, F. Agbossou, A. Montreux, Organomet-
allics 1995, 14, 401Ϫ406.
J. Bravo, C. Cativiela, J. E. Chaves, R. Navarro, E. P. Urriolab-
eitia, Inorg. Chem. 2003, 42, 1006Ϫ1013.
M. N. Burnett, C. K. Johnson, ORTEPIII, Report ORNL-
6895, Oak Ridge National Laboratory, Oak Ridge, TN, 1996.
J. X. Mc Dermott, J. F. White, G. M. Whitesides, J. Am. Chem.
Soc. 1976, 98, 6521.
[11]
[12]
[13]
[14]
[19]
[20]
Early View Article
Published Online February 16, 2004
1284
 2004 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjic.org
Eur. J. Inorg. Chem. 2004, 1277Ϫ1284