Modification of Wilkinson's catalyst with triphenyl phosphite: Synthesis, structure, 31P NMR and DFT study of trans-[RhCl(P(OPh) 3)(PPh3)2]

Article abstract of DOI:10.1016/j.poly.2012.06.086

The complex trans-[RhCl(P(OPh)₃)(PPh₃)₂] (1) has been prepared and characterized by ³¹P NMR spectroscopy and single crystal X-ray crystallography. It was found that in the solid state there are two forms of complex 1 in the unit cell forming a cocrystal. DFT theoretical computations have confirmed the existence of the two forms and have provided evidence for the greater stability of 1 compared with Wilkinson's catalyst, [RhCl(PPh₃)₃] (2), in terms of the dissociation energy of the Rh-P(PPh₃) bonds. On the basis of the phosphorus chemical shifts, δ P(PPh₃), and the results of the theoretical computations, it is suggested that the Rh-P(PPh₃) bonding interactions are slightly enhanced in 1 compared with 2. A distinct difference between complexes 1 and 2, was found to be the catalytic activity of 1 in the alkylation of allyl acetate with sodium diethylmalonate, while 2 is almost catalytically inefficient.

Full text of DOI:10.1016/j.poly.2012.06.086

Polyhedron 45 (2012) 255–261
Contents lists available at SciVerse ScienceDirect
Polyhedron
journal homepage: www.elsevier.com/locate/poly
Modification of Wilkinson’s catalyst with triphenyl phosphite: Synthesis, structure,
31P NMR and DFT study of trans-[RhCl(P(OPh)3)(PPh3)2]
Ioannis Choinopoulos a, Ioannis Papageorgiou a, Silverio Coco b, Emmanuel Simandiras c, Spyros Koinis a,
⇑
a Laboratory of Inorganic Chemistry, Faculty of Chemistry, National and Kapodistrian University of Athens, Panepistimiopolis, 15771 Athens, Greece
b IU/CINQUIMA, Química Inorgánica, Facultad de Ciencias, Universitad de Valladolid, 47071 Valladolid, Spain
c Theoretical and Physical Chemistry Institute, National Hellenic Research Foundation, 48 Vassileos Constantinou Avenue, 11635 Athens, Greece
a r t i c l e i n f o
a b s t r a c t
The complex trans-[RhCl(P(OPh)3)(PPh3)2] (1) has been prepared and characterized by 31P NMR spectros-
copy and single crystal X-ray crystallography. It was found that in the solid state there are two forms of
complex 1 in the unit cell forming a cocrystal. DFT theoretical computations have confirmed the existence
of the two forms and have provided evidence for the greater stability of 1 compared with Wilkinson’s cat-
alyst, [RhCl(PPh3)3] (2), in terms of the dissociation energy of the Rh-P(PPh3) bonds. On the basis of the
Article history:
Received 30 April 2012
Accepted 29 June 2012
Available online 20 July 2012
Keywords:
Rhodium
Phosphane
Phosphite
DFT
phosphorus chemical shifts, d(PPPh ), and the results of the theoretical computations, it is suggested that
3
the Rh-P(PPh3) bonding interactions are slightly enhanced in 1 compared with 2. A distinct difference
between complexes 1 and 2, was found to be the catalytic activity of 1 in the alkylation of allyl acetate
with sodium diethylmalonate, while 2 is almost catalytically inefficient.
Ó 2012 Elsevier Ltd. All rights reserved.
Alkylation
1. Introduction
phosphite ligands [6]. This was revised as soon as hydridophosph-
itorhodium(I) clusters and mononuclear complexes were found to
The importance of [RhCl(PPh3)3], Wilkinson’s catalyst, in the ad-
vance of organometallic chemistry and catalysis is universally rec-
ognized [1]. Wilkinson’s catalyst has been shown to be an efficient
catalyst not only for the homogeneous hydrogenation of unsatu-
rated substrates but also for a large variety of organic reactions
[1b,2]. Furthermore, the start of the development of catalysts for
asymmetric hydrogenation was the concept of replacing the tri-
phenylphosphine ligand of Wilkinson’s catalyst with a chiral ligand
[3].
Trivalent phosphorus ligands are ubiquitous in organometallic
chemistry and homogeneous catalysis and their primary role is
the fine-tuning of the electron density, and consequently of the
reactivity, of the metal center in the course of catalytic reactions
[4]. This is accomplished through steric and electronic effects first
described in detail by Tolman [5]. Thus in the context of the
development of new catalysts it is of major importance the under-
standing of the nature of the effects by which trivalent phosphorus
ligands modulate the reactivity of the metal center.
be efficient hydrogenation catalysts [6,7].
Phosphite ligands have been used widely for the modification
of rhodium catalysts bearing phosphine ligands, usually in situ.
These modifications can be classified in the following three broad
categories.
(i) Complexes of the form [(L-L)Rh(P-OP)]+ (L-L = acac, cod, nbd
and P-OP = chiral bidentate phosphine-phosphite ligand, e.g.
BINAPHOS), which were introduced by Takaya et al. as
modifications of the corresponding complexes with chiral
biphosphines and have been used as catalyst precursors for
homogeneous hydrogenations and hydroformylations [8].
(ii) The in situ modification of Wilkinson’s catalyst with phos-
phites, which resulted in highly efficient catalytic systems
for allylic alkylation reactions. These were introduced by
Tsuji et al. [9] and their applications were extended by Evans
et al. [10].
(iii) Complexes of the form [(L-L)RhPaPb]+ (L-L = cod, Pa = mono-
dentate phosphine, Pb = monodentate phosphite) which were
introduced by Reetz et al. and were used in the context of the
Combinatorial Enantioselective Transition Metal Catalysis
with S-BINOL derived monodentate P-ligands [11].
Phosphites (phosphite esters) are extremely attractive ligands
for transition metal catalysts. Their available synthesis allows the
modification of electronic and steric properties. Metal complexes
of phosphites were generally supposed to be inefficient for hydro-
genation catalysts, by reason of the putative high
p-acidity of the
Although the effect of the phosphite, in the cases of the in situ
modifications of Wilkinson’s catalyst, has been studied from the
point of view of the catalytic performance [10,12], the role of the
⇑
Corresponding author.
E-mail address: koinis@chem.uoa.gr (S. Koinis).
0277-5387/$ - see front matter Ó 2012 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.poly.2012.06.086

256
I. Choinopoulos et al. / Polyhedron 45 (2012) 255–261
(40 mL). The mixture was stirred and in a few minutes a bright yel-
[RhCl(P(OPh3)(PPh3)2] isomers
Cl
Rh PPh3
P(OPh)3
low solution resulted. The stirring was continued for 24 h and re-
sulted to the formation of a bright yellow solid. The mixture was
condensed to dryness and hexane (20 mL) was added to the solid
residue. The mixture was stirred vigorously for 10 min and the light
yellow liquid phase was carefully suctioned. Hexane (20 mL) was
added and after 10 min of vigorous stirring the yellow solid was col-
lected by filtration and dried in vacuo over P2O5. (Yield 0.48 g, 95%).
Anal. Calcd. for C54H45ClO3P3Rh: C, 66.64; H, 4.66; N. Found: C,
66.47; H, 4.38%.
Cl
Ph3P
Ph3P
Rh P(OPh)3
PPh3
[RhCl(P(OPh3)2(PPh3)] isomers
Cl
Rh P(OPh)3
P(OPh)3
2.2.2. 2nd method
In a Schlenk tube P(OPh)3 (0.0310 g, 0.10 mmol) and PPh3
(0.0525 g, 0.20 mmol) were introduced with CH3CN (10 mL). To
the resulting solution [Rh2Cl2(cod)2] (0.0247 g, 0.05 mmol) was
added with CH3CN (10 mL). The reaction mixture was stirred and
a bright yellow solution resulted. The stirring was continued for
3 h and then it was condensed to dryness. Hexane (20 mL) was
added to the solid residue, the mixture was stirred vigorously for
10 min and the light yellow liquid phase was carefully suctioned.
Hexane (20 mL) was added and after 10 min of vigorous stirring
the yellow solid was collected by filtration and dried in vacuum
over P2O5. (Yield 0.024 g, 95%).
Cl
Ph3P
(PhO)3P
Rh P(OPh)3
PPh3
Fig. 1. Possible forms of the triphenyl phosphite modified Wilkinson’s catalyst.
phosphite remains unidentified, since nothing is known about the
nature of the catalyst precursors and the catalytically active spe-
cies involved. In most cases the ratio [RhCl(PPh3)3]/phosphite used
for the modification of Wilkinson’s catalyst was in the range 1ꢀ3 to
1ꢀ4. Most probably among the catalyst precursors, the mononu-
clear Rh(I) complexes with mixed trivalent phosphorus ligands,
PPh3 and phosphite, are an obvious choice. Therefore we began a
systematic work towards the synthesis, the characterization and
the study of the catalytic activity of Rh(I) complexes with mixed
monodentate trivalent phosphorus ligands, phosphines and phos-
phites, starting with PPh3 and P(OPh)3. The only report, to our
knowledge, of a Rh(I) complex, in which both PPh3 and P(OPh)3 li-
gands are coordinated is [RhCl(CO)(PPh3)(P(OPh)3)] [13].
2.3. Alkylation of allyl acetate
In a Schlenk tube NaH (0.0348 g, 1.45 mmol) and THF (4 mL)
were introduced and diethyl malonate (228 lL, 1.50 mmol) was
added dropwise, resulting in the evolution of H2 gas. This solution
was transferred and added slowly by syringe to another Schlenk
tube containing trans-[RhCl(P(OPh)3)(PPh3)2] (0.0487 g, 0.050
mmol), allyl acetate (54 lL, 0.50 mmol) and THF (4 mL). The reac-
tion mixture was stirred for 19 h at room temperature and an or-
ange solution with a white precipitate resulted. The mixture was
condensed until the liquid phase became oily. Diethylether
(15 mL) was added and the white solid was filtered and washed
with diethylether (10 mL). The filtrate was condensed until an oily
liquid remains. The conversion was determined by 1H NMR
spectroscopy.
The possible forms of the modified forms of Wilkinson’s catalyst
with triphenyl phosphite are, in principle, four; two mono-phosph-
ito and two di-phosphito isomers, Fig. 1.
We, herein, shall describe the synthesis of trans-[RhCl(P(OPh)3)
(PPh3)2], which is the first structurally characterized modified form
of Wilkinson’s catalyst with mixed phosphine and phosphite li-
gands. We shall also discuss the effects induced to the ‘‘RhCl(PPh3)2’’
fragment as a result of the substitution of PPh3 by P(OPh)3 in terms
of structural and 31P NMR spectroscopic data, supported by DFT the-
oretical computations results. Finally we shall present the catalytic
performance of trans-[RhCl(P(OPh)3)(PPh3)2] in the alkylation of al-
lyl acetate with sodium diethylmalonate.
2.4. Crystallographic analysis
X-ray diffraction measurements were made using a Bruker
SMART CCD area-detector diffractometer with Mo K
(k = 0.71073 Å) [16]. Suitable single crystal of the complex was
a radiation
mounted in glass fiber. Intensities were integrated from several
2. Experimental
series of exposures, each exposure covering 0.3° in
x, the total data
set being a hemisphere [17]. Absorption corrections were applied
based on multiple and symmetry-equivalent measurements [18].
The structures were solved by direct methods and refined by least
squares on weighted F2 values for all reflections [19]. All non-
hydrogen atoms were assigned anisotropic displacement parame-
ters and refined without positional constraints. Hydrogen atoms
were taken into account at calculated positions and refined as rid-
ing atoms. Complex neutral-atom scattering factors were used
[20].
2.1. Materials and methods
Starting materials and solvents were purchased from Sigma–Al-
drich. [RhCl(PPh3)3] [14] and [Rh2Cl2(cod)2] [15] were synthesized
according to literature methods. THF and diethyl ether were dis-
tilled over Na/Ph2CO prior to use. The solvents were degassed by
bubbling nitrogen for 30 min. All operations were performed under
a pure nitrogen atmosphere, using Schlenk and syringe techniques
on an inert gas/vacuum manifold. NMR spectra were recorded on a
Varian Unity Plus 300 spectrometer. For the 1H NMR spectra the
solvent peak was used as an internal reference and the 31P NMR
spectra were referenced with an external standard of H3PO4 85%.
2.5. Computational studies
All calculations were done with the GAUSSIAN 09 suite of pro-
grams [21]. The SVP basis is a split-valence plus polarization [22]
and the TZVPP is a triple zeta valence plus double polarization
including an effective core potential [23,24]. The popular B3LYP
functional is due to Becke [25] and Lee et al. [26] and contains
empirical data, whereas the fully ab initio PBE functional used is
due to Perdew [27]. The TZVPP computations were made faster
2.2. Synthesis of trans-[RhCl(P(OPh)3)(PPh3)2]
2.2.1. 1st method
In a Schlenk tube [RhCl(PPh3)3] (0.4626 g, 0.50 mmol) and
P(OPh)3 (0.1551 g, 0.50 mmol) were introduced with CH3CN

I. Choinopoulos et al. / Polyhedron 45 (2012) 255–261
257
using the Density Fitting approximation [28]. The structures were
fully optimized and stationary points characterized by second
derivative calculations to confirm that they are true minima. The
frequencies calculated were used for the estimation of the
vel energy differences.
m = 0 le-
3. Results and discussion
3.1. The system: [RhCl(PPh3)3]+n P(OPh)3
It is impressive that for almost 50 years after the synthesis of
Wilkinson’s catalyst, the consequent syntheses of a variety of rho-
dium complexes using Wilkinson’s catalyst as starting material
and the use of in situ modifications of Wilkinson’s catalyst with
phosphites in homogeneous catalysis, the title complex has not
been reported yet. In order to gain some insight into the Rh(I) com-
plexes with mixed PPh3–P(OPh)3 ligands we have measured the 31
P
NMR spectra of [RhCl(PPh3)3]+nP(OPh)3 systems, shown in Fig. 2.
From the spectra it comes out that for n = 1, complex 1 is the
only rhodium complex present, while the systems for n = 2 and 3
are multicomponent systems containing rhodium complexes with
mixed PPh3–P(OPh)3 ligands. Thus, in order to prepare complex 1,
accurate quantities of the reagents should be used and special pre-
caution should be taken regarding the purity of the easily oxidized
triphenyl phosphite.
3.2. Synthesis
The complex trans-[RhCl(P(OPh)3)(PPh3)2] (1) was prepared
quantitatively either by reacting [RhCl(PPh3)3] (2) with P(OPh)3
(Eq. (1)) or by reacting [Rh2Cl2(cod)2] with a stoichiometric mix-
ture of PPh3 and P(OPh)3, (Eq. (2)):
½RhClðPPh3Þ3ꢁ þ PðOPhÞ3 ! trans-½RhClðPðOPhÞ3ÞðPPh3Þ2ꢁ þ PPh3
ð1Þ
½Rh2Cl2ðcodÞ2ꢁ þ 4PPh3 þ 2PðOPhÞ3
! 2trans-½RhClðPðOPhÞ3ÞðPPh3Þ2ꢁ þ 2cod
ð2Þ
Both reactions can be carried out in acetonitrile, acetone or
dichloromethane. Complex 1 can be stored under vacuum in a des-
iccator, but samples can be weighed in the atmosphere. It can be
heated without decomposition (toluene, reflux, 2 h), while Wilkin-
son’s catalyst, under the same conditions is dimerized completely
giving [(PPh3)2Rh(l-Cl)2Rh(PPh3)2] [[1b]]. In solution, in the pres-
ence of oxygen, it decomposes and a brown solid is formed. It re-
acts with CO in CH2Cl2 solution, yielding labile species even at
low temperature as deduced from 31P NMR spectroscopy. Addition
of acetone to the solution under CO bubbling results in the precip-
itation of trans-[RhCl(CO)(PPh3)2]. The nature of the hydrido spe-
cies formed in the presence of dihydrogen is under investigation.
The structure of 1 was assigned on the basis of analytical and
spectroscopic data. The formation of this species as unique product
was demonstrated by measuring the 31P NMR spectra of the reac-
tion mixtures (see Table 1).
Fig. 2. 31P NMR spectra of [RhCl(PPh3)3]+nP(OPh)3 systems. A: n = 1; B: n = 2; C:
n = 3 (upper: P(OPh)3 region; lower: PPh3 region).
presented in Table 2. The molecular structures of the two forms
of 1, shown in Figs. 3 and 4, can be described in terms of square
planar coordination to a first approximation.
3.3. X-ray crystal and molecular structure of 1
The RhClP3 fragment for both forms is planar within experimen-
tal error. In contrast to this, for both forms of Wilkinson’s catalyst
there is a distortion toward tetrahedral geometry, which is not due
to steric crowding, since, as stated elsewhere, the trans Rh–P bonds
are similar to those in the less congested trans-[RhCl(CO)(PPh3)2],
suggesting that steric crowding is not responsible for this distor-
tion, which is also found in [RhCl(PMe3)3] [30].
The differences of the Rh–Cl and the Rh–P(PPh3) bond lengths of
the two forms of 1 with the respective bond lengths of Wilkinson’s
catalyst forms are not significant.
Single crystals of 1 suitable for X-ray diffraction measurements,
were obtained by slow diffusion of hexane into a solution of the
complex in CH3CN. The crystal examined was found to be a cocrys-
tal of two chemically identical forms of complex 1 in 1ꢀ1 ratio,
Z = 8. These forms which shall be referred to as 1a and 1b, are anal-
ogous to the red and orange forms of Wilkinson’s catalyst, which
crystallize independently depending upon the conditions of syn-
thesis [29]. Selected interatomic distances and bond angles are

258
I. Choinopoulos et al. / Polyhedron 45 (2012) 255–261
Table 1
Table 2
Crystal data and structure refinement data for complex 1.
Bond lengths [Å] and bond angles [°] of the two forms of trans-[RhCl(P(OPh)3)(PPh3)2]
(experimental and computed with the B3LYP and PBE functionals) and of Wilkinson’s
catalyst (experimental [29] and computed with the BP86 functional [34]).
Complex
1
C
Formula
Mr
54H45ClO3P3Rh
973.17
T (K)
298(2)
k (Å)
0.71073
orthorhombic
Pca2(1)
25.318(8)
19.138(6)
19.097(6)
90
Crystal system
Space group
a (Å)
b (Å)
c (Å)
Parameter
[RhCl(P(OPh)3)(PPh3)2]
[RhCl(PPh3)3]
1a
1b
Red
Orange
a
(°)
Rh–Cl (Å)
X-ray
BP86
B3LYP/SVP
PBE/TZVPP
Rh–Pa trans to Pc (Å)
X-ray
b (°)
90
90
9253(5)
8
1.397
2.3814(13)
2.3857(13)
2.376(4)
2.42
2.437
2.404(4)
2.42
2.443
c
(°)
V (Å3)
2.426
2.376
2.431
2.386
Z
qcalc (gmꢂ3
)
l
(mmꢂ1
)
0.574
4000
2.3220(12)
2.3146(13)
2.322(4)
2.35
2.392
2.304(4)
2.35
2.376
F000
Crystal size (mm)
hmax (°)
No. reflections collected
Rint
Data
BP86
0.26 ꢃ 0.24 ꢃ 0.14
B3LYP/SVP
PBE/TZVPP
Rh–Pc trans to Pa (Å)
X-ray
2.387
2.335
2.386
2.331
28.20
89378
0.0638
22357
1117
0.0477
0.0840
R1 = 0.0840, wR2 = 0.1089
1.248/ꢂ0.547
2.3236(13)
2.3321(13)
2.334(3)
2.37
2.424
2.338(4)
2.38
2.434
BP86
Parameters
B3LYP/SVP
PBE/TZVPP
Rh–Pb trans to Cl (Å)
X-ray
2.401
2.341
2.390
2.340
Ra [I > 2
r
(I)]
axb (all data)
R indices (all data)
Residuals (eÅꢂ3
R
2.1454(12)
2.1406(12)
2.214(4)
2.27
2.301
2.225(4)
2.27
2.309
)
BP86
B3LYP/SVP
PBE/TZVPP
Pa–Rh–Cl (°)
Pc–Rh–Cl (°)
Pa–Rh–Pb (°)
Pc–Rh–Pb (°)
PBE/TZVPP
2.192
2.153
2.189
2.150
83.64(5)
87.41(4)
94.89(5)
94.06(4)
85.2, 87.6,
96.1, 98.1
171.03(4)
155.9
85.61(5)
84.59(5)
96.08(4)
93.48(4)
87.4, 84.7,
95.1, 95.6
169.38(4)
163.7
86.1(2)
85.2(2)
97.9(2)
100.4(1)
85.3(1)
84.5(1)
97.7(1)
96.4(2)
The Rh–P(P(OPh)3) bond lengths are the same within experi-
mental error and significantly shorter compared to the respective
Rh–P(PPh3) bonds of Wilkinson’s catalyst indicative of a higher
bond order, as a result of the
P(OPh)3.
The Rh–P(P(OPh)3) bond lengths of the two forms of 1,
2.1454(12) and 2.1406(12), are among the shortest so far reported:
[(cod)Rh(l-Cl)2Rh(P(OPh)3)2] 2.1378(7), 2.1475(8) Å [31], [(acac)
r-donor and p-acceptor nature of
Pa–Rh–Pc (°)
PBE/TZVPP
Cl–Rh–Pb (°)
PBE/TZVPP
152.8(1)
156.2(2)
159.1(2)
166.7(2)
172.82(5)
161.4
176.41(5)
165.9
Rh(P(OPh)3)2] 2.147(2), 2.156(2) and 2.142(2), 2.150(2) Å [32]
and [(CF3)COCH2C(CH3)O)Rh(P(OPh)3)2] 2.148(4), 2.136(6) and
2.145(6), 2.130(6) Å [33].
A probable explanation of the self-assembling of the two forms
of 1 in a cocrystal compared to Wilkinson’s catalyst (and the anal-
ogous iridium complex [34]) for which the two forms crystallize
separately forming different crystals, could be that the two forms
of 1 are more closely related both structurally and/or energetically.
expected since we are comparing a similar size basis set, for the
SVP case, but different functionals. The B3LYP/SVP results, of this
work in Table 2 for both molecules clearly show that, overall, the
Rh–P(PPh3) bond lengths are shorter for 1 than for 2. In particular
for 2 we find that one of the Rh–P(PPh3) bonds, the Rh–Pc is consis-
tently the longest compared to all other calculated and measured
bond lengths for both compounds. The PBE/TZVPP results, which
are of much higher quality confirm the consistently shorter Rh–
P(PPh3) bonds. Summarizing, the DFT calculations show that 1 ex-
ists, even in the gas phase, in two very similar forms which have a
common feature, namely that all Rh–P(PPh3) bonds are shorter and
hence stronger compared to the corresponding ones in 2.
The energy differences between the two forms for various levels
of theory are given in Table 3. In all cases the energy differences are
very small and certainly within the possible range of error for the
methods used. However, the consistency observed, and the agree-
ment with a previous study [34], support the findings. It can there-
fore be concluded that the two forms of 1 are closer in energy
compared to the two forms of 2. This may be the reason for the for-
mation of a cocrystal and not two different crystal allotropes.
A measure of the stability of the compound is the dissociation
energy of the two Rh–P(PPh3) bonds. Although it is solvent depen-
dent, a gas phase theoretical calculation can in any case provide a
measure of the relative Rh–P(PPh3) bond strengths. This was calcu-
lated in Ref. [34] for 2 and found to be 13.9 kcal/mol. For 1, a
phosphine was removed and the two fragments were reoptimized
at the PBE/TZVPP level. The dissociation energy was found to be
3.4. DFT calculations
The structures of 1 and 2 were fully optimized using Density
Functional Theory combined with split-valence (SVP) and larger
triple zeta (TZVPP) basis sets using B3LYP and PBE functionals
respectively. The data are reported in Table 2.
Firstly it is noted that theoretical computations clearly confirm
the existence of two chemically equivalent forms of 1. These two
forms are stable minima of the potential energy surface, which
was confirmed by force constant calculations.
The second observation worth noting from Table 2 is that there
exists a very good broad agreement between the theoretical and X-
ray data, especially for the PBE/TZVPP calculations. Differences of
0
less than 0.005 ÅA for the Rh–Cl bond lengths are seen, and less than
0
0.015 ÅA for the Rh–P lengths, the correct ordering between 1a and
1b being well reproduced. The angles are also in good broad agree-
ment; in fact that comparison is between a closely packed cocrys-
tal and a computation for the isolated zero point gas phase
structure should be taken into account.
26.2 kcal/mol (or 25.2 between
double that of 2.
m = 0 energy levels), approximately
Further, it is seen that our results for 2 are in good broad agree-
ment with the previously reported theoretical results [34]; this is

I. Choinopoulos et al. / Polyhedron 45 (2012) 255–261
259
Table 3
Computed energy differences (in kcal/mol) for the two
forms of 1 and 2.
1
2
B3LYP/SVP
eq
0.8
1.2
1.9
1.7
m
= 0
PBE/TZVPP
eq
1.0
1.4
m
= 0
Table 4
31P NMR spectral parameters of the complexes [RhCl(X)(PPh3)2] (X=P(OPh)3, PPh3).
Complex d(P)/ppm d(P)/ppm 1J(Rh–P)/Hz 1J(Rh–P)/Hz
2J(P–P)/Hz
PPh3
P(OPh)3
Rh–PPh3
Rh–P(OPh)3
1
2
36.57
31.07dd
48.05dt
114.81
134.4
143.7
192.1
310.8
44.6
36.9
36.9
The variation of the 31P NMR parameters of the, cis to the chlo-
rine, PPh3 can be used to probe the cis-electronic effect induced as
a result of the substitution of the, trans to the chlorine, PPh3 (class
Fig. 3. ORTEP drawing of complex 1a, projected onto the mean molecular plane
RhClP3. Hydrogen atoms have been omitted for clarity. Selected bond lengths and
bond angles are reported in Table 1.
II phosphorus ligand,
III phosphorus ligand,
r
r
-donor) of [RhCl(PPh3)3] by P(OPh)3 (class
-donor and -acceptor) [35].
p
These are in accordance with the 31P NMR spectrum of trans-
[RhCl(Ppyrl3)(PPh3)2] (Ppyrl3: N-pyrrolylphosphine; class III
phosphorus ligand), which shows two distinct sets of phosphorus
resonances: a dd due to the pair of mutually trans PPh3 ligands (d
(P) 34.3 ppm, 1J(Rh–P) 133 Hz, 2J(P–P) 42 Hz) and a dt assignable
to the unique P(pyrl)3 ligand (d (P) 101.1 ppm, 1J(Rh–P) 278 Hz,
2J(P–P) 42 Hz) [36]. The variation of the 31P NMR parameters, of
trans-[RhCl(Ppyrl3)(PPh3)2] compared with 2, are in the same
direction with those found for 1. Thus it is seen that upon substi-
tution of the, trans to the chlorine, PPh3, of Wilkinson’s catalyst
by
r-donors/p-acceptors phosphorus ligands the chemical shift, d
(PPPh ), tends to be shifted to higher frequencies.
3
3.6. Comments on the variation of the Rh–P(PPh3) bonding
An interpretation of the observed variation of d (PPPh3), for 1
compared to 2, in terms of respective variation in Rh–P(PPh3)
bonding interactions can thus be given taking into account the
following:
(a) In the case of Wilkinson’s catalyst the formation of the dimer
Fig. 4. ORTEP drawing of complex 1b, projected onto the main molecular plane
RhClP3. Hydrogen atoms have been omitted for clarity. Selected bond lengths and
bond angles are reported in Table 1.
[(PPh3)2Rh(l-Cl)2Rh(PPh3)2] has been correlated with the
dissociation of triphenylphosphine according to the follow-
ing reactions [1b,37]:
½RhClðPPh3Þ3ꢁ ꢀ ½RhClðPPh3Þ2ꢁ þ PPh3
ð3Þ
ð4Þ
3.5. 31P NMR
The 31P NMR spectral parameters of complex 1 are presented in
Table 4. The only signals observed in the spectrum of 1, are a dd
(2P, PPh3 region) and a dt (1P, P(OPh)3 region), which are assigned
to the coordinated phosphorus ligands.
2½RhClðPPh3Þ2ꢁ ꢀ ½ðPPh3Þ2Rhð
l-ClÞ2RhðPPh3Þ2ꢁ
Cl
Cl
Although the X-ray molecular structures do not provide any sig-
nificant differentiation for 1 compared with 2 in terms of the Rh–
P(PPh3) bond lengths of their common fragments, RhCl(PPh3)2,
the 31P NMR spectra provide such a differentiation, namely the
spectral parameters of the, cis to the chlorine, PPh3. As shown in
Table 4, for 1 compared with 2, the chemical shift, d(PPPh3), is
shifted to higher frequencies (downfield) and the spin–spin cou-
Ph3P
Rh
PPh3
Ph3P
Rh
P(OPh)3
C
O
C
O
ν(CO)=1983 cm-1
ν (CO)=1978 cm-1
pling constant 1J(Rh–PPPh ) is reduced.
3
Fig. 5. P(OPh)3 cis-effect in mixed PPh3–P(OPh)3 rhodium carbonyl species.

260
I. Choinopoulos et al. / Polyhedron 45 (2012) 255–261
COOEt
+ NaCH(COOEt) 2
OCOMe
COOEt
[RhCl(P(OPh)3)(PPh3)2]
100%
Scheme 1. Alkylation of allyl acetate catalyzed by 1.
As mentioned above DFT computations provide evidence of
enhancement of the Rh–P(PPh3) bonds for 1 compared with 2,
which are based on the calculated bond dissociation energies
and bond lengths. Thus the aforementioned stability of 1 with
respect to dimerization could be attributed to an enhancement
of Rh–P(PPh3) bonding interactions of 1 relative to 2.
that trans-[RhCl(P(OPh)3)(PPh3)2] is the first well defined Rh com-
plex with mixed trivalent phosphorus ligands that catalyzes effec-
tively an allylic alkylation reaction.
4. Conclusions
In summary, the synthesis of the modified form of Wilkinson’s
catalyst with triphenyl phosphite, trans-[RhCl(P(OPh)3)(PPh3)2],
was accomplished using as starting materials either [RhCl(PPh3)3]
or [Rh2Cl2(cod)2]. The X-ray crystal structure showed the presence
of two forms of 1 in the unit cell, forming a cocrystal. Based on
experimental, spectroscopic and computational data it is suggested
that the result of the substitution of the trans, to the chlorine, PPh3
of Wilkinson’s catalyst by P(OPh)3 is the fine enhancement of the
Rh–P(PPh3) bonding interactions in 1 compared with 2, in agree-
ment with the superior thermal stability of 1 in solution compared
with Wilkinson’s catalyst.
(b) A cis-effect induced by P(OPh)3 can be recognized by com-
paring of the carbonyl vibrational frequencies of the com-
plexes trans-[RhCl(CO)(PPh3)2] and [RhCl(CO)(P(OPh)3)
(PPh3)] [13], Fig. 5. The slight decrease of
m(CO) could be
attributed, to a crude approximation, to the fine-tuning of
the Rh–C bond order towards enhancement [38].
(c) The interpretation of the 31P NMR coordination chemical
shifts,
shift upon coordination, d(P)c, and the chemical shift of the
free phosphorus ligand, d(P)f, in terms of the -donor and
-acceptor abilities of these ligands has been pioneered by
Dd(P), equal to the difference between the chemical
r
p
This low cost modification of Wilkinson’s catalyst resulted in an
efficient catalyst precursor for the alkylation of allyl acetate with
sodium diethylmalonate, while [RhCl(PPh3)3] is almost catalyti-
cally inactive.
Alyea et al. [39]. According to a comprehensive discussion
they have proposed that: (i) predominantly electronic,
rather than steric, factors are active in determining d (P)
and both the sign and the magnitude of the coordination
chemical shift
and (ii) both
D
d(P) in terms of
r- and p-electronic effects
Acknowledgements
r
(M P) and (M ? P) components should
p
bring 31P chemical shift to high frequency. These proposals
were based on the theoretical treatment of the chemical
shift of the free phosphorus ligands according to which the
31P NMR chemical shifts are thought to arise primarily from
variations in the paramagnetic contribution from electrons
in valence orbitals [40]. Ziegler et al., have conducted an
elaborate theoretical study (DFT-GLIAO) of the 31P NMR
chemical shifts for complexes of the type [M(CO)5PR3]
(M@Cr, Mo; R@H, CH3, C6H5, F and Cl) which has confirmed
the domination of the paramagnetic shielding term in deter-
mining the phosphorus chemical shift [41].
We thank the Special Account for Research Grants, National and
Kapodistrian University of Athens, for financial support (70/4/
7561). Ioannis Choinopoulos is deeply indebted to Professor Emer-
itus of Cardiology Pavlos Toutouzas for his succor in health and
financial issues. This contribution was taken in part from the
Ph.D. Thesis of I.C.
Appendix A. Supplementary data
CCDC 699645 contains the supplementary crystallographic data
for 1. These data can be obtained free of charge via http://
www.ccdc.cam.ac.uk/conts/retrieving.html, or from the Cambridge
Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ,
UK; fax: +44 1223-336-033; or e-mail: deposit@ccdc.cam.ac.uk.
It can thus be suggested that in complex 1 the withdrawal of
electron density, from rhodium to P(OPh)3 through
interaction (backbonding) together with the diminished
p
-bonding
electron
r
donor capacity of P(OPh)3 as compared with PPh3, is compensated
by the enhancement of the Rh–P(PPh3), bonding interactions. This
variation is not reflected explicitly in the solid state Rh–P(PPh3)
bond lengths and should be considered as a fine tuning, towards
enhancement.
References
[1] (a) M.A. Bennett, P.A. Longstaff, Chem. Ind. (London) (1965) 846;
(b) J.A. Osborn, F.H. Jardine, J.F. Young, G. Wilkinson, J. Chem. Soc. A (1966)
1711.
[2] For a review, see: F.H. Jardine, Prog. Inorg. Chem. 28 (1981) 63.
[3] W.S. Knowles (Nobel Lecture), Adv. Synth. Catal. 345 (2003) 3.
[4] (a) R.P. Hughes, in: G. Wilkinson, F.G.A. Stone, E.W. Abel (Eds.),
Comprehensive Organometallic Chemistry, Pergamon Press, Oxford, UK,
1982. Chapter 35;
3.7. Alkylation of allyl acetate
(b) P.R. Sharp, in: G. Wilkinson, F.G.A. Stone, E.W. Abel (Eds.), Comprehensive
Organometallic Chemistry II, vol. 8, Pergamon Press, Oxford, UK, 1995. Chapter
2;
(c) L.H. Pignolet (Ed.), Homogeneous Catalysis with Metal Phosphine
Complexes, Plenum, New York, 1983.
trans-[RhCl(P(OPh)3)(PPh3)2] was found to catalyze the alkyl-
ation of allyl acetate with sodium diethylmalonate in THF. See
Scheme. 1.
Tsuji et al. have reported that although [RhCl(PPh3)3] showed
almost no activity in the allylic alkylation, highly efficient catalytic
systems have resulted by the addition of PnBu3 or P(OEt)3 and,
Evans et al. have reported that efficient ‘‘modified’’ catalysts were
formed in situ, by using mixtures of [RhCl(PPh3)3] and phosphites
(P(OPh)3, P(OMe)3) in a 1ꢀ3 molar ratio [9,10]. Our finding showed
[5] C.A. Tolman, Chem. Rev. 77 (1977) 313.
[6] H.K.A.C. Coolen, R.J.M. Nolte, P.W.N.M. van Leeuwen, J. Organomet. Chem. 496
(1995) 159.
[7] A.J. Sivak, E.L. Muetterties, J. Am. Chem. Soc. 101 (1979) 4878.
[8] (a) N. Sakai, S. Mano, K. Nozaki, H. Takaya, J. Am. Chem. Soc. 115 (1993) 7033;
(b) H. Fernández-Pérez, S.M.A. Donald, I.J. Munslow, J. Benet- Buchholz, F.
Maseras, A. Vidal-Ferran, Chem. Eur. J. 16 (2010) 6495 (and references therein).

I. Choinopoulos et al. / Polyhedron 45 (2012) 255–261
261
[9] J. Tsuji, I. Minami, I. Shimizu, Tetrahedron Lett. 25 (1984) 5157.
[10] (a) P.A. Evans, J.D. Nelson, Tetrahedron Lett. 39 (1998) 1725;
(b) P.A. Evans, J.D. Nelson, J. Am. Chem. Soc. 120 (1998) 5581.
[11] (a) M.T. Reetz, G. Mehler, Tetrahedron Lett. 44 (2003) 4593;
(b) M.T. Reetz, G. Mehler, A. Meiswinkel, Tetrahedron: Asymmetry 15 (2004)
2165;
J.V. Ortiz, J. Cioslowski, D.J. Fox, Gaussian 09, Revision A.1, Gaussian Inc.,
Wallingford CT, 1996.
[22] R. Ahlrichs, K. May, Phys. Chem. Chem. Phys. 2 (2000) 943.
[23] F. Weigend, R. Ahlrichs, Phys. Chem. Chem. Phys. 7 (2005) 3297.
[24] D. Andrae, U. Haussermann, M. Dolg, H. Stoll, H. Preuss, Theor. Chim. Acta 77
(1990) 123.
(c) M.T. Reetz, Angew. Chem., Int. Ed. 47 (2008) 2556.
[12] U. Kazmaier, D. Stolz, Angew. Chem., Int. Ed. 45 (2006) 3072.
[13] D.M. Barlex, M.J. Hacker, R.D.W. Kemmit, J. Organomet. Chem. 43 (1972) 425.
[14] J.A. Osborn, G. Wilkinson, Inorg. Synth. 28 (1990) 77.
[15] G. Giordano, R.H. Crabtree, Inorg. Synth. 28 (1990) 88.
[16] SMART V5.051 Diffractometer Control Software, Bruker Analytical X-ray
Instruments Inc., Madison, WI, 1998.
[25] A.D. Becke, Phys. Rev. A 38 (1988) 3098.
[26] C.T. Lee, W.T. Yang, R.G. Parr, Phys. Rev. B 37 (1988) 785.
[27] (a) J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 78 (1997) 1396;
(b) J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77 (1996) 3865.
[28] B.I. Dunlap, J. Mol. Struct. THEOCHEM 529 (2000) 37.
[29] M.J. Bennett, P.B. Donaldson, Inorg. Chem. 16 (1977) 655.
[30] S.A. Cotton, The Chemistry of Precious Metals, Blackie Academic
Professional, London, 1997. pp. 90.
[31] J. Coetzer, G. Gaffner, Acta Crystallogr., Sect B: Struct. Crystallogr. Cryst. Chem.
26 (1970) 985.
&
[17] SAINT V6.02 Integration Software, Bruker Analytical X-ray Instruments Inc.,
Madison, WI, 1999.
[18] G.M. Sheldrick, SADABS:
A Program for Absorption Correction with the
Siemens SMART System, University of Göttingen, Göttingen, Germany, 1996.
[19] SHELXTL Program System Version 5.1, Bruker Analytical X-ray Instruments
Inc., Madison, WI, 1998.
[32] G.L. Lamprecht, J.G. Leipoldt, G.J. Van Zyl, Inorg. Chim. Acta 96 (1985) L31.
[33] G.J. Van Zyl, G.L. Lamprecht, J.G. Leipoldt, Inorg. Chim. Acta 102 (1985) L1.
[34] J. Goodman, V.V. Grushin, R.B. Larichev, S.A. Macgregor, W.J. Marshall, D.C. Roe,
J. Am. Chem. Soc. 132 (2010) 12013.
[35] M.N. Golovin, M.M. Rahman, J.E. Belmonte, W.P. Giering, Organometallics 4
(1985) 1981.
[36] C.R. Landis, S. Feldgus, J. Uddin, C.E. Wozniak, K.G. Moloy, Organometallics 19
(2000) 4878.
[37] D.A. Wink, P.C. Ford, J. Am. Chem. Soc. 109 (1987) 436.
[38] F.A. Cotton, G. Wilkinson, Advanced Inorganic Chemistry, fifth ed., Wiley, New
York, 1988. pp. 58.
[20] International Tables for Crystallography, vol. C, Kluwer, Dordrecht, 1992.
[21] G.W.T.M.J. Frisch, H.B. Schlegel, G.E. Scuseria, M.A. Robb, J.R. Cheeseman, G.
Scalmani, V. Barone, B. Mennucci, G.A. Petersson, H. Nakatsuji, M. Caricato, X.
Li, H.P. Hratchian, A.F. Izmaylov, J. Bloino, G. Zheng, J.L. Sonnenberg, M. Hada,
M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda,
O. Kitao, H. Nakai, T. Vreven, J.A. Montgomery Jr., J.E. Peralta, F. Ogliaro, M.
Bearpark, J.J. Heyd, E. Brothers, K.N. Kudin, V.N. Staroverov, R. Kobayashi, J.
Normand, K. Raghavachari, A. Rendell, J.C. Burant, S.S. Iyengar, J. Tomasi, M.
Cossi, N. Rega, J.M. Millam, M. Klene, J.E. Knox, J.B. Cross, V. Bakken, C. Adamo,
J. Jaramillo, R. Gomperts, R.E. Stratmann, O. Yazyev, A.J. Austin, R. Cammi, C.
Pomelli, J.W. Ochterski, R.L. Martin, K. Morokuma, V.G. Zakrzewski, G.A. Voth,
P. Salvador, J.J. Dannenberg, S. Dapprich, A.D. Daniels, Ö. Farkas, J.B. Foresman,
[39] E.C. Alyea, S. Song, Inorg. Chem. 34 (1995) 3864.
[40] J.H. Letcher, J.R. Van Wazer, J. Chem. Phys. 44 (1966) 815.
[41] Y. Ruiz-Morales, T. Ziegler, J. Phys. Chem. A 102 (1998) 3970.