Unique evidence for a RhIII to RhI reduction by deoxygenation of a carbonate moiety to CO2 by an out-of-sphere phosphane

Article abstract of DOI:10.1002/1099-0682(200107)2001:7<1801::AID-EJIC1801>3.0.CO;2-X

Rh^III carbonate generated from either a peroxocarbonate complex [RhCl(CO₄)(PR₃)₃] or [RhCl₃(PR₃)₃] and Na₂CO₃, is reduced to Rh^I by deoxygenation of the carbonate moiety to CO₂ by an out-of-sphere phosphane. The reaction takes place in mild conditions and is implied in the catalytic activity shown by Rh^I in the oxidation of styrene with O₂/CO₂ mixtures.

Full text of DOI:10.1002/1099-0682(200107)2001:7<1801::AID-EJIC1801>3.0.CO;2-X

FULL PAPER
Unique Evidence for a RhIII to RhI Reduction by Deoxygenation of a
Carbonate Moiety to CO2 by an Out-of-Sphere Phosphane
Michele Aresta,[a] Angela Dibenedetto,[a] and Immacolata Tommasi*[a][‡]
Keywords: Carbon dioxide / Deoxygenation / Redox chemistry / Rhodium / Phosphanes
RhIII carbonate generated from either a peroxocarbonate place in mild conditions and is implied in the catalytic activ-
complex [RhCl(CO4)(PR3)3] or [RhCl3(PR3)3] and Na2CO3, is
ity shown by RhI in the oxidation of styrene with O2/CO2
reduced to RhI by deoxygenation of the carbonate moiety to mixtures.
CO2 by an out-of-sphere phosphane. The reaction takes
Introduction
bond. This reaction has also been modelled using the Im-
pulse Oscillation Model, IOM,[6Ϫ8] which assumes that the
vibrations of the metal-dioxygen complex and CO2 must be
synchronised in order for the insertion reaction to occur.
IOM unequivocally shows that path A is preferred (up to
92% selectivity) over path B.[9,10]
Transition metal peroxo-carbonates have been known for
some time.[1,2] They are usually prepared by reaction of
metal dioxygen complexes with CO2. Accordingly, RhIII
peroxocarbonates are easily obtained according to Equa-
tion (1). Alternatively, transition metal carbon dioxide com-
plexes can react with dioxygen[3,4] to afford peroxocarbon-
ates.
Very interestingly, the reaction of carbon dioxide metal
complexes with dioxygen to afford peroxocarbonates pro-
ceeds through the substitution of coordinated CO2 by di-
oxygen and subsequent insertion of CO2 into the OϪO
bond of the newly formed dioxygen complex.[11]
(1)
We have also shown that peroxocarbonates may act as
selective one-oxygen transfer agents to an oxophile
(Scheme 2).[12Ϫ15] By using asymmetrically labelled species,
we have demonstrated that the oxygen bonded to the metal
is the atom of the peroxo group transferred to the oxoph-
ile.[5] The Rh peroxocarbonate is thus converted into a RhIII
carbonate complex.
By using labelled species, we have recently shown that
CO2 reacts with Rh(η2-O2) complexes by insertion into the
OϪO bond (Scheme 1, path A), rather than by insertion
into the MϪO bond (Scheme 1, path B).[5] The reaction
pathway has been demonstrated through coupling an IR
theoretical study (calculation of the vibrational frequencies
of selected bonds of the peroxocarbonate complexes) with
experimental vibrational frequencies. The latter match the
values calculated for the species obtained assuming CO2 in-
sertion into the OϪO bond (Ϯ2 cmϪ1 [5]
better than those
)
of products originating from CO2 insertion into the MϪO
Scheme 2. One-oxygen transfer from the peroxocarbonate moiety
to an oxophile
It should be noted that when the peroxocarbonate
[RhCl(CO4)(PEt2Ph)3] acts as a one-oxygen transfer agent,
the starting RhI complex [RhCl(PEt2Ph)3] is found along
with other compounds that undoubtedly originate from the
peroxocarbonate. In principle, the RhI complex could be
produced from the peroxocarbonate if the reaction in Equa-
tion (1) is reversible.
In an attempt to demonstrate if such a hypothesis holds
or not, we have carefully investigated the reaction and have
unequivocally shown that [RhCl(PEt2Ph)3] originates from
RhIII carbonate [RhCl(CO3)(PEt2Ph)3], through an unprec-
edented carbonate deoxygenation to carbon dioxide pro-
moted by an external phosphane. We describe in this paper
some properties of Rh peroxocarbonate 2 and the reduction
Scheme 1. Alternative pathways for CO2 reaction with a Rh dioxy-
gen complex
[a]
`
Dipartimento di Chimica, Universita degli Studi di Bari,
Campus Universitario,
Via Orabona, 4Ϫ70126 Bari, Italy
Fax: (internat.) ϩ39-080/544-2429
E-mail: miares@tin.it
[‡]
´
Present address: Universite de Bourgogne, L.S.E.O. Ϫ UMR
5632 Ϫ Equipe V, Sciences Mirande 9,
av. Alain Savary BP 47870Ϫ21078 Dijon Cedex, France
E-mail: ada.tommasi@u-bourgogne.fr
Eur. J. Inorg. Chem. 2001, 1801Ϫ1806
WILEY-VCH Verlag GmbH, 69451 Weinheim, 2001
1434Ϫ1948/01/0707Ϫ1801 $ 17.50ϩ.50/0
1801

M. Aresta, A. Dibenedetto, I. Tommasi
FULL PAPER
of carbonate 3 to [RhCl(PEt2Ph)3]. The latter reaction has
also been performed using an external carbonate source
(Na2CO3, Ag2CO3) and a RhIII chloride complex.
Results and Discussion
RhIII peroxocarbonate was prepared by reaction of
[RhCl(PEt2Ph)3(O2)] with CO2 at 240 K in toluene[4,5] ac-
cording to Equation (1). The pure compound is quite stable
at low temperature (below 240 K) in the solid state for sev-
eral days or for several minutes at a temperature as high as
340 K (see Experimental Section). Conversely, in solution
above 250 K, or in the solid state at 323 K in the presence
of an external phosphane, or any other species that may
interact with the Rh centre, it readily reacts and transfers
one oxygen atom to an oxophile to afford the carbonate
complex 3 (Scheme 2). It is worth recalling that olefins and
tetrahydrofuran can be oxidised in this way under mild con-
ditions by Rh peroxocarbonates.[13]
When pure 2 is dissolved in CH2Cl2 at 250 K, the 31P
NMR spectrum unequivocally shows the conversion of the
peroxocarbonate (Figure 1, spectrum A) into a number of
other species (Figure 1, spectrum B). These have all been
identified by comparison with the spectra of authentic
samples. Resonances aϪg have been assigned as follows: (a)
assigned to the OϭPEt2Ph species; (b) and (e) assigned to
the mer-RhCl(CO3)(PEt2Ph)3 species; (c) and (f) assigned
to the mer-RhCl(CO4)(PEt2Ph)3 species; (d and g) assigned
to the mer-RhCl3(PEt2Ph)3 species (see Experimental Sec-
tion).
Figure 1. Spectrum A: 31P NMR spectrum (213 K, CD2Cl2) of
[RhCl(CO4)(PEt2Ph)3]; spectrum B: 31P NMR (CD2Cl2) spectrum
of the solution from A warmed to 250 K
It is evident from these data that 2 behaves differently
whether it is pure or is in the presence of external ligands However, when solid Rh peroxocarbonate is warmed to
or a solvent. Moreover, the fact that a coordinated phos- 340 K with an equimolar amount of ethylene, phosphane is
phane is not oxidised (we recall that pure 2 is stable up to slowly released and then rapidly oxidised by the peroxocar-
345 K), demonstrates that the O-transfer process is inter- bonate to Et2PhPO (Scheme 3, path b). The selectivity of
molecular, not intramolecular.
phosphane oxidation with respect to ethylene is 100%, as
The reactivity in the solid state and in solution find a demonstrated by the gas-phase analysis which does not
common basis of interpretation. However, in solution the show the presence of ethylene oxidation products. The role
oxygen transfer is promoted by the solvent, which causes of ethylene is therefore that of favouring the phosphane re-
the phosphane dissociation (Scheme 3, path a).
lease from the Rh coordination sphere. Conversely, if more
The out-of-sphere phosphane formed in solution attacks reactive olefins like styrene are used, in excess with respect
the peroxo group affording carbonate and phosphane oxide to the stoichiometric ratio and under CO2/O2 atmosphere,
(Scheme 3, path a). The latter may be found free in solution they can be oxidised in solution by Rh peroxocarbonates
or coordinated to Rh, according to the nature of solvent. (Scheme 4).
Scheme 3. Reactivity of [RhCl(CO4)(PR3)3] in solution (path a) and in the solid state in the presence of ethylene (path b)
1802
Eur. J. Inorg. Chem. 2001, 1801Ϫ1806

RhIII to RhI Reduction by Deoxygenation of a Carbonate Moiety to CO2
FULL PAPER
The phosphane oxidation can go on in the presence of
an excess of olefin until complex 2 is totally converted into
other species that are not active catalysts. The use of bident-
ate phosphanes or chelating nitrogen ligands, like dipyridyl
or phenanthroline, stabilises the catalyst for up to two to
three hours,[14,16] after which it has been converted into in-
active species. The mechanism of deactivation of the cata-
lyst when monodentate or bidentate ligands are used is
quite different. In the case of monodentate ligands, as we
have reported above, the oxidation of the dissociated phos-
phane causes the destruction of the catalyst [Equation (3)],
whereas in the case of bidentate ligands, ligand scrambling
causes the destruction of the active catalytic species as
shown in Equation (4). In fact, the square-planar moiety
‘‘Rh(L-L)2’’ (where L-L is a bidentate phosphane) is not
active in catalysis as it does not promote the formation of
the peroxocarbonate because two cis-positions are not avail-
able. Similarly, the olefin complex does not coordinate
either dioxygen or CO2.
Scheme 4. Reaction of styrene with a CO2/O2 mixture promoted
by RhI
Styrene is converted into a mixture of phenylacetalde-
hyde, phenylmethylketone and styrene epoxide, as major
products, along with benzaldehyde which is produced by
the cleavage of the styrene CϪC double bond by O2. It is
worth noting that Rh peroxocarbonates and Rh dioxygen
complexes show a different behaviour towards styrene
oxidation.[5,13Ϫ18] While the dioxygen complex causes the
cleavage of the double bond [Equation (2)], the peroxo-car-
bonate promotes a one-oxygen transfer to the olefin, mim-
icking mono-oxygenase enzymes (Scheme 4).
(4)
Compound 5 has been isolated from the reaction solution
and its nature confirmed by comparison with an authentic
sample. As reported above, the oxygen transfer reaction as-
sumes a particular importance when olefins or ethers are
used,[13] affording useful products. Peroxocarbonates can
therefore be used to promote the transfer of a single oxygen
atom of the original dioxygen molecule to an oxophile. The
reaction path and products are reported in Scheme 5. In the
case of an olefin, carbocation 6 could be the intermediate[13]
that generates epoxides, aldehydes and ketones. The oxygen
transfer reaction has been also modelled using IOM.[17,18]
Very interestingly, if styrene is used (Scheme 4), the calcu-
lated distribution of products is very close to the experi-
mental one.
(2)
The formation of the ketone and phenylacetaldehyde can
be explained assuming a hydrogen shift promoted by the
metal centre (Scheme 5).
Scheme 6. Formula of compound 6
Scheme 5. Formation of ketone and terminal aldehyde by H-migra-
tion promoted by RhI
When peroxocarbonates react as oxidant towards an ex-
ternal phosphane or, in general when they act as one-oxy-
gen transfer agents, as reported above, the carbonate is
formed, as demonstrated by spectroscopic techniques (IR,
NMR). The 31P NMR spectrum shows a doublet of doub-
lets (dd) at δ ϭ 14.2 (JP-P ϭ 23.8 Hz, JP-Rh ϭ 89 Hz) and a
doublets of triplets (dt) at δ ϭ 22.7 (JP-Rh ϭ 125 Hz). These
signals are also found in the 31P spectrum when a solution
of 2 in CD2Cl2, prepared at 240 K, is heated to room tem-
perature. Interestingly, when the oxygen transfer occurs
from the peroxocarbonate to the external phosphane, the
In such an oxidation reaction, if Rh complexes stabilised
by monodentate P ligands are used, the catalyst has a life-
time of less than one hour. In fact, the reaction with the
olefin releases the phosphane, which is easily oxidised ac-
cording to Equation (3):
(3)
Eur. J. Inorg. Chem. 2001, 1801Ϫ1806
1803

M. Aresta, A. Dibenedetto, I. Tommasi
FULL PAPER
final reaction mixture shows the presence of a 31P signal
that can be attributed to [RhCl(PEt2Ph)3].[19]
The formation of this species might be explained in terms
of a reversible dissociation of CO2 from the peroxocarbon-
ate to afford the dioxygen complex that may give back
[RhCl(PEt2Ph)3] [Equation (1)]. We have tried to prove this
hypothesis by GC-MS analysis and found that dioxygen is
not detected. Conversely, CO2 is released and, if labelled O2
is used in the synthesis of the peroxocarbonate, a semi-la-
belled 18O species is formed. Moreover, if a phosphane is
added to the Rh carbonate formed from the peroxocarbon-
ate after oxygen-transfer to the external oxophile, phos-
phane oxide is formed together with [RhCl(PEt2Ph)3]
[Equation (5)].
In Equation (7), NaCl was isolated by extraction with
water, and the RhI complex was isolated by recrystallisation
from pentane. The latter has been characterised by ele-
mental analysis, and its spectroscopic properties compared
with those of an authentic sample. If PCy3 is added as ex-
ternal phosphane, OϭPCy3 is the most abundant phos-
phane oxide, with minor amounts of OϭPEt2Ph. A limited
degree of phosphane scrambling is observed (ca. 10%).
Equation (7) represents a quite unique and interesting reac-
tion, if one considers that the release of CO2 from Na2CO3
requires high temperatures (Ͼ1000 K).[21] Ag2CO3 behaves
in the same way. However, coordination to Rh lowers the
decomposition temperature of the carbonate by more than
600 K. This finding rationalizes the behaviour of peroxocar-
bonate 2 and elucidates the reaction pathway relevant to
Rh catalysis in styrene oxidation-carbonation,[12Ϫ15] al-
lowing us to propose the cycle represented in Figure 2 for
the RhIϪRhIII cyclic conversion.
(5)
This behaviour suggests that CO2 could originate from
a mechanism in which the carbonate is implicated in the
phosphane oxidation. As a matter of fact we have per-
formed a number of reactions which have clearly demon-
strated that the ‘‘RhIII carbonate ion’’ system can transfer
one oxygen atom to the phosphane with formation of phos-
phane oxide and CO2. The amount of CO2 has been evalu-
ated by a GC of the gas phase. The GC-MS spectrum of
the gas phase also shows that if the carbonate has one 18O
2Ϫ
oxygen atom that is used for binding the CO3 moiety to
the metal, the resulting CO2 has ca. 50% of 18O, as does
the phosphane oxide (Scheme 7).
Scheme 7. Carbonate deoxygenation by an external phosphane li-
gand
The conversion of the carbonate to CO2 with oxygen
transfer to a substrate is reminiscent of the reaction of inor-
ganic carbonates which form CO2 and the metal oxide when
heated at high temperature [Equation (6)].
Figure 2. Double one-oxygen transfer to two different substrates
from an O2 molecule mediated by a RhIϪRhIIIϪRhI shuttle with
the intermediacy of peroxocarbonate moiety; a ligand (P, L) scram-
bling is also observed
(6)
At the same time, the reduction of RhIII to RhI explains
the catalytic role of the metal. In fact, without reduction to
RhI, only a stoichiometric reaction is observed in the O-
transfer to the olefin from the peroxocarbonate complex.
The observed turnover number. which is higher than one,
can be interpreted only on the basis of a RhIII Ǟ RhI reduc-
tion and reactivation of the catalyst.
The release of CO2 from coordinated carbonate com-
plexes is not a trivial process. To the best of our knowledge,
it has not been documented in the literature with phos-
phane as the reducing agent. It is known that hydrogen at
high temperature and pressure is able to deoxygenate car-
bonates into carbon dioxide. In fact, CoCO3 has been used
for the synthesis of Co2(CO)8 with H2 at 673 K.[20] There-
fore, we decided to investigate if an external carbonate ion
could itself undergo the deoxygenation reaction under mild
conditions. We have thus reacted [RhCl3(PEt2Ph)3] with
Na2CO3 or Ag2CO3 in the presence of a free phosphane.
Conclusions
The unprecedented deoxygenation of a carbonate ion to
Interestingly, the reaction occurs at 353 K according to CO2 in the presence of RhIII has been observed. This find-
Equation (7) and (8), respectively.
ing represents an interesting new piece of chemistry demon-
1804
Eur. J. Inorg. Chem. 2001, 1801Ϫ1806

RhIII to RhI Reduction by Deoxygenation of a Carbonate Moiety to CO2
FULL PAPER
min. and the 31P NMR spectrum recorded at this temperature (Fig-
ure 1, spectrum B). Heating produced the oxygen transfer to the
phosphane. The following compounds were identified: trans-
strating that coordination of the carbonate anion allows the
lowering of its decomposition temperature by more than
600 °C. However, this finding allows us to rationalise the
catalytic role of RhI in the oxidation of styrene to styrene
epoxide and other one-oxygen transfer products in an
O2/CO2 atmosphere, with the intermediacy of
peroxocarbonateϪcarbonateϪcarbon dioxide complexes.
RhCl(CO)(PEt2Ph)2:
δ ϭ 26.0 (d, JP-Rh ϭ 128.7 Hz); mer-
RhCl3(PEt2Ph)3: δ ϭ 3.7 (dd, JP-Rh ϭ 84 Hz, JP-P ϭ 23.1 Hz), 16.9
(dt, JP-Rh ϭ 111.3 Hz); mer-RhCl(CO3)(PEt2Ph)3: δ ϭ 14.2 (dd, JP-
Rh ϭ 89 Hz, JP-P ϭ 23.8 Hz), 22.7 (dt, JP-Rh ϭ 125 Hz); OϭPEt2Ph:
δ ϭ 42.9.
Reaction of [RhCl(CO4)(PEt2Ph)3] with PEt2Ph: a) Solid 2 was
heated with an equimolar amount of PEt2Ph at 320 K under dini-
trogen. The IR spectrum of the mixture showed the appearance of
phosphane oxide (1150 cmϪ1) within 4 hours and the formation of
RhIII carbonate (IR band at 1660 cmϪ1), which was isolated from
the mother mixture.
Experimental Section
General: Unless otherwise stated, all reactions and manipulations
were conducted under a dinitrogen or CO2 atmosphere by using
vacuum line techniques. All solvents were dried as described in the
literature[22] and stored under dinitrogen. CO2 (Ն99.95%) and O2
(Ն99.98%) were purchased from Air Liquide Spa. The RhCl(PR3)3
complexes were prepared from [RhCl(C2H4)]2 (Aldrich) and the
corresponding phosphanes (Aldrich) according to previously re-
b) Compound 2 in CH2Cl2 solution was treated with an equimolar
amount of PEt2Ph at room temperature under dinitrogen. The so-
lution was monitored at intervals of time by GC and GC/MS. The
formation of phosphane oxide was clearly demonstrated to be a
product of the oxidation process.
ported
experimental
procedures.[4]
The
synthesis
of
RhCl[18O16OC(16O)18O](PEt2Ph)3 has been described previously.[5]
Reaction of Solid [RhCl(CO4)(PEt2Ph)3] with C2H4: Solid
[RhCl(CO4)(PEt2Ph)3] (0.05 g, 0.07 mmol) was stirred at room tem-
perature under a C2H4 atmosphere. The reaction was monitored by
GC analysis of the equilibrium gas. No ethylene oxidation products
were detected in the gas or solid phase, while in the solid phase
phosphane oxide [31P NMR: δ ϭ 42.9 (s)] appeared after a couple
of days. After a week the solid completely changed its appearance
and the reaction was stopped and the solid analyzed by 31P NMR
spectroscopy. The residual solid was a mixture of compounds which
had a 31P spectrum close to that shown in Figure 1B. If an inert
gas was used instead of ethylene, the complex lasted for a much
longer time under the same experimental conditions.
31P NMR spectra were obtained at 81 MHz with a Varian XL-200
instrument. 31P chemical shifts are referred to 85% H3PO4 using
the high-frequency-positive convention. FTIR spectra were re-
corded using a Bruker 113V Fourier transform interferometer. Fre-
quencies are accurate to Ϯ1 cmϪ1. Solid samples were studied as
Nujol mulls (Nujol was previously dried over sodium wire and de-
oxygenated with argon). GC analyses were performed with a HP
5890 Series II gas chromatograph equipped with an Alltech-Heli-
flex AT-1000 capillary column (30 m ϫ 0.00025 m, 0.2 µm thick-
ness). GC-MS analysis was carried out with a gas chromatograph
SHIMADZU 17 A-1000 (capillary column: 30 m ϫ 0.00025 m,
0.25 µm thickness) detector linked to a SHIMADZU GCMS-QP
5050 mass spectrometer. GC analysis of gaseous phases was carried
out with a gas chromatograph DANI equipped with a Carbowax
column 3 m ϫ 2.1 mm (ID) and TCD detector.
Reaction of (PEt2Ph)3RhCl(CO4) with Styrene: Styrene (0.1 mL) in
THF (2 mL) was added to 2 (0.100 g, 0.16 mmol) under a CO2/O2
atmosphere (10:1 v/v) and the solution stirred at room temperature.
At intervals of 30 min., a sample was withdrawn and analysed by
GC/MS. The presence of the following products was ascertained:
benzaldehyde, phenylacetaldehyde, phenyl methyl ketone and styr-
ene epoxide in a molar ratio 1:3:3:5. Benzoic acid and styrene car-
bonate were also present.
Study on the Thermal Stability of (PEt2Ph)3RhCl(CO4): a) Solid
[RhCl(CO4)(PEt2Ph)3] (0.050 g, 0.07 mmol) was heated at 338 K
for 10 min. under a dinitrogen atmosphere into Cell (a) (Figure 3)
that was then connected to a GC and the gas phase analysed with
a TCD. The GC analysis showed the absence of CO2. The IR ana-
lysis of the solid showed that no reaction had occurred.
b) Analogous results were obtained when C2H4 was admitted into
the cell and the system heated for 1Ϫ2 min. at 340 K. A longer
reaction time produced a conversion of the Rh complex (see be-
low).
Reaction of [RhCl3(PEt2Ph)3] with Na2CO3: [RhCl3(PEt2Ph)3]
(0.200 g, 0.3 mmol) was dissolved in toluene (5 mL) and a stoichi-
ometric amount of Na2CO3 (0.031 g, 0.3 mmol) and free phos-
phane (0.050 g, 0.3 mmol) added. The reaction mixture was heated
to 353 K. After 1 hour the GC analysis of the gas phase revealed
the presence of CO2 while in solution the free phosphane and phos-
phane oxide were detected. After 24 h, the amount of CO2 in the
gas phase increased to almost reach a CO2/Rh ratio of 0.5. CO2
was also present in solution. The total amount of released CO2
allowed us to estimate when the reaction had reached completion.
NaCl separated from the solution and was isolated by filtration.
Its mass was 95% of the expected stoichiometric amount. From
the mother solution 0.150 g of crude [RhCl(PEt2Ph)3] was isolated,
which was then crystallised from pentane. Ϫ C30H45ClP3Rh
(636.97): calcd. C 56.57, H 7.12, Cl 5.57, P 14.59; found C 56.03,
H 7.01, Cl 5.61, P 15.03.
Figure 3. Apparatus used for the analysis of the gas phase in equi-
librium with a solid
Reaction of (PEt2Ph)3RhCl3 with Ag2CO3: [RhCl3(PEt2Ph)3]
(0.140 g, 0.2 mmol) was dissolved in CH3CN (3 mL) and a
Conversion of [RhCl(CO4)(PEt2Ph)3] in CD2Cl2: A solution of 2 in stoichiometric amount of Ag2CO3 (0.050 g, 0.2 mmol) and PEt2Ph
CD2Cl2 was prepared at 213 K and the NMR spectrum recorded (0.030 g, 0.2 mmol) were added. The reaction mixture was heated
(Figure 1, spectrum A). The solution was heated to 250 K for 20
to 353 K. After 1 hour the GC analysis of the gas phase revealed
1805
Eur. J. Inorg. Chem. 2001, 1801Ϫ1806

M. Aresta, A. Dibenedetto, I. Tommasi
FULL PAPER
quille, M. Borowiak, in Advances in Chemical Conversion for
Mitigating Carbon Dioxide, Studies in Surface Science and
Catalysis (Eds.: T. Inui, M. Anpo, K. Izui, S. Yanagida, T. Ya-
maguchi), Elsevier Science B. V., 1998, 114, 677Ϫ680.
M. Aresta, E. Quaranta, I. Tommasi, Proceedings of the 215th
ACS National Meeting, Dallas, Texas, March 29ϪApril 2,
1998, COLL 086.
M. Aresta, M. A. Borowiak, M. G. H. Jamroz, I. Tommasi, E.
Quaranta, J. Mol. Catal. A 2000, in press.
M. Aresta, I. Tommasi, A. Dibenedetto, M. Fouassier, J. Mas-
cetti forthcoming paper.
M. Aresta, E. Quaranta, A. Ciccarese, J. Mol. Catal. 1987,
41, 355Ϫ359.
M. Aresta, C. Fragale, E. Quaranta, I. Tommasi, J. Chem. Soc.,
Chem. Commun. 1992, 315Ϫ317.
M. Aresta, A. Dibenedetto, Proceedings of the X Congresso
Brasileiro de Catalise, 22Ϫ24 September 1999, Salvador, Bahia,
Brazil, 410Ϫ413.
M. Aresta, A. Dibenedetto, Appl. Organomet. Chem., in press.
M. Aresta, A. Dibenedetto, I. Tommasi, Proceedings of the In-
ternational Symposium on CO2 conversion and utilisation in re-
finery and chemical processing, ACS Meeting, March 26Ϫ30,
2000, San Francisco, 108Ϫ109.
First workshop of the COST D9 Action, Advanced Computa-
tional Chemistry of Increasingly Complex Systems, July 2Ϫ4,
1999, Geneva, Switzerland.
J. C. Dobrowolski, M. H. Jamroz, J. K. Kazimirski, K. Bajdor,
M. Borowiak, R. Larsson, J. Mol. Struct. 1999, 483, 183Ϫ187.
According to the temperature and time of permanence in
chlorinated solvents, [RhCl3(PR3)3] may also be formed, M.
Aresta, M. Rossi, A. Sacco, Inorg. Chim. Acta 1969, 3,
227Ϫ231.
K. Yamamoto, K. Sato Japan Pat., (mar 9, 1953); Chem. Abstr.
1954, 48, 2996.
Y. A. Chang, H. Ahamad, Thermodynamic Data on Metal Car-
bonates and Related Oxides, Warrendale, Pa.: Metallurgical So-
ciety of AIME, 1982.
D. D. Perrin, W. L. Armarego, D. R. Perrin, Purification of
Laboratory Chemical Pergamon Press, Oxford, England, 1986.
Received September 6, 2000
the presence of CO2. After 24 h the gas phase showed an increased
amount of CO2 and from the solution a whitish-yellow solid separ-
ated, which was filtered, washed with pentane and dried in vacuo.
It was analyzed as AgCl(PEt2Ph). From the solution
[RhCl(PEt2Ph)2]2 was isolated together with OϭPEt2Ph. The dimer
was characterized by elemental analysis and its spectroscopic prop-
erties were compared to an authentic sample. Ϫ C18H33AgClP
(423.75): calcd. C 51.01, H 7.85, P 7.30, Cl 8.36; found C 50.85, H
7.54, P 7.10, Cl 8.16. Ϫ C40H60Cl2P4Rh2 (941.54): calcd. C 51.03,
H 6.42, Cl 7.53, P 13.16; found C 50.90, H 7.70, Cl 8.01, P 7.03.
[9]
[10]
[11]
[12]
[13]
[14]
In one run we used PCy3 as the external phosphane. This was es-
sentially totally converted into OϭPCy3. The isolated Rh com-
plexes showed that some phosphane scrambling also took place
and mixed phosphane species were obtained. The exchange did not
exceed 10Ϫ15%, as demonstrated by 31P NMR spectroscopy.
[15]
[16]
Acknowledgments
The MURST (Project no. 9803026360) is gratefully acknowledged.
The authors thank Dr. M. Miglietta for some experimental assist-
ance.
[17]
[1]
[18]
[19]
P. J. Hayward, D. M. Blake, G. Wilkinson, C. J. Nyman, J. Am.
Chem. Soc. 1970, 92, 5873Ϫ5878.
[2]
L. Malatesta, R. Ugo, J. Chem. Soc. 1963, 2080Ϫ2082.
[3]
M. Aresta, C. F. Nobile, J. Chem. Soc., Dalton Trans. 1977,
708Ϫ714.
M. Aresta, E. Quaranta, A. Ciccarese, C1 Mol. Chem. 1985,
1, 267Ϫ281.
[4]
[20]
[21]
[5]
M. Aresta, I. Tommasi, E. Quaranta, C. Fragale, J. Mascetti,
M. Tranquille, F. Galan, M. Fouassier, Inorg. Chem. 1996, 35,
4254Ϫ4260.
M. Borowiak, J. Mol, Catal., 2000, in press.
COST-D3/007/94 Project: ‘‘Modelling of selective energy trans-
[6]
[7]
[22]
fer, spatial and time coherence in catalytic reactions, tested on
carbon dioxide reactions’’, 1998 Report.
M. Aresta, E. Quaranta, I. Tommasi, J. Mascetti, M. Tran-
[8]
[I00334]
1806
Eur. J. Inorg. Chem. 2001, 1801Ϫ1806