The Effect of pH on the Reactions of Catalytically Important RhI Complexes in Aqueous Solution: Reaction of [RhCl(tppms)3] and trans-[RhCl(CO)(tppms)2] with Hydrogen (TPPMS = mono-sulfonated triphenylphosphine)

Article abstract of DOI:10.1002/1521-3765(20010105)7:1<193::AID-CHEM193>3.0.CO;2-Q

Hydrolysis and hydrogenation of [RhCl(tppms)3] (1) and trans-[RhCl(CO)(tppms)2] (2) was studied in aqueous solutions in a wide pH range (2 < pH < 11) in the presence of excess TPPMS (3-diphenylphosphinyl-benzenesulfonic acid sodium salt). In acidic solutions hydrogenation of 1 yields a mixture of cis-mer- and cis-fac-[RhClH2(tppms)3] (3a,b) while in strongly basic solutions [RhH(H2O)(tppms)3] (4) is obtained, the midpoint of the equilibrium between these hydride species being at pH 8.2. The paper gives the first successful 1H and 31P NMR spectroscopic characterization of a water soluble rhodium(I)-monohydride (4) bearing only monodentate phosphine ligands. Hydrolysis of 2 is negligible below pH 9 and its hydrogenation results in formation of [Rh(CO)H(tppms)3] (5), which is an analogue to the well known and industrially used hydroformylation catalyst [Rh(CO)H(tppts)3] (6) (TPPTS = 3,3',3"-phosphinetriyltris(benzenesulfonic acid) trisodium salt). It was shown by pH-potentiometric measurements that formation of 5 is strongly pH dependent in the pH5 - 9 range; this gives an explanation for the observed but previously unexplained pH dependence of several hydroformylation reactions. Conversely, the effect of pH on the rate of hydrogenation of maleic and fumaric acid catalyzed by 1 in the 2< pH < 7 range can be adequately described by considering solely the changes in the ionization state of these substrates. All these results warrant the use of buffered (pH-controlled) solutions for aqueous organometallic catalysis.

Full text of DOI:10.1002/1521-3765(20010105)7:1<193::AID-CHEM193>3.0.CO;2-Q

FULL PAPER
The Effect of pH on the Reactions of Catalytically Important
RhI Complexes in Aqueous Solution: Reaction of
[RhCl(tppms)3] and trans-[RhCl(CO)(tppms)2] with Hydrogen
(TPPMS ˆ mono-sulfonated triphenylphosphine)
Ferenc JooÂ,*[a, b] JoÂzsef KovaÂcs,[b] Attila Cs. BeÂnyei,[a]
Levente NaÂdasdi,[b, c] and GaÂbor Laurenczy[c]
Abstract: Hydrolysis and hydrogena-
tion of [RhCl(tppms)3] (1) and trans-
[RhCl(CO)(tppms)2] (2) was studied in
aqueous solutions in a wide pH range
(2 < pH < 11) in the presence of excess
TPPMS (3-diphenylphosphinyl-benze-
nesulfonic acid sodium salt). In acidic
solutions hydrogenation of 1 yields a
mixture of cis-mer- and cis-fac-[RhClH2-
(tppms)3] (3a,b) while in strongly basic
solutions [RhH(H2O)(tppms)3] (4) is
obtained, the midpoint of the equilibri-
um between these hydride species being
at pH 8.2. The paper gives the first
successful 1H and 31P NMR spectroscop-
ic characterization of a water soluble
rhodium(i)-monohydride (4) bearing on-
ly monodentate phosphine ligands. Hy-
drolysis of 2 is negligible below pH 9 and
its hydrogenation results in formation of
[Rh(CO)H(tppms)3] (5), which is an
analogue to the well known and indus-
trially used hydroformylation catalyst
that formation of 5 is strongly pH
dependent in the pH 5 ± 9 range; this
gives an explanation for the observed
but previously unexplained pH depend-
ence of several hydroformylation reac-
tions. Conversely, the effect of pH on the
rate of hydrogenation of maleic and
fumaric acid catalyzed by 1 in the 2 <
pH < 7 range can be adequately descri-
bed by considering solely the changes in
the ionization state of these substrates.
All these results warrant the use of
buffered (pH-controlled) solutions for
aqueous organometallic catalysis.
[Rh(CO)H(tppts)3]
(6)
(TPPTS ˆ
3,3',3''-phosphinetriyltris(benzenesul-
fonic acid) trisodium salt). It was shown
by pH-potentiometric measurements
Keywords: biphasic catalysis ´ hy-
drides ´ hydroformylation ´ hydro-
genation ´ rhodium
industrial processes using aqueous ± organic biphasic technol-
ogy.[5±7, 16, 17] Due to its elegant simplicity, the Ruhrchemie ±
Introduction
Ã
Catalysis by water soluble phosphine complexes in aqueous
solutions[1] and in aqueous ± organic two-phase systems[2] has
developed into a most successful way for product isolation
and catalyst recycling both on the laboratory and industrial
scale.[3±7] Of the several versions of liquid ± liquid biphasic
catalysis based on the use of immiscible organic[8] (among
them fluorous[9, 10]) or ionic[11, 12] liquids and supercriticial
fluids,[13, 14] only the Shell higher olefin process[15] (SHOP) is
currently in large scale operation while there are at least six
Rhone Poulenc (RCH-RP) propene hydroformylation proc-
ess[6, 17] is unquestionably the benchmark and clearest example
of the operational, environmental and economical advantages
made possible by conducting a large scale reaction in an
aqueous ± organic two-phase system. The key to the success of
this process was the use of a highly hydrophylic rhodium
catalyst, [Rh(CO)H(tppts)3] (6) containing the TPPTS 7
(3,3',3''-phosphinetriyltris(benzenesulfonic acid) trisodium
salt) ligand. This important achievement triggered an im-
mense effort of research into aqueous organometallic catal-
ysis which resulted in detailed mechanistic description of
several important processes such as olefin hydroformyla-
tion,[18] hydrogenation,[19] ring opening metathesis polymer-
ization[20] or vinylic/allylic substitutions.[21]
Â
Â
[a] Prof. F. Joo, Dr. A. Cs. Benyei
Institute of Physical Chemistry, University of Debrecen
4010 Debrecen, P.O. Box 7 (Hungary)
Fax : (‡ 36)52-512915
E-mail: jooferenc@tigris.klte.hu
For an excellent performance in homogeneous organic
reactions the catalysts have to be solubilized in aqueous
mixtures, which usually translates into ligand modification.
Two notable representatives of the ever-growing family of
water soluble phosphine ligands[16, 19] are the mono-sulfonated
triphenylphosphine, TPPMS, 8 (3-diphenylphosphinyl-benze-
Â
Â
Â
[b] Prof. F. Joo, Dr. J. Kovacs, Dr. L. Nadasdi
Research Group of Homogeneous Catalysis
Hungarian Academy of Sciences, 4010 Debrecen (Hungary)
Â
[c] Dr. L. Nadasdi, Dr. G. Laurenczy
Institute of Inorganic and Analytical Chemistry
University of Lausanne, 1015 Lausanne-Dorigny (Switzerland)
Chem. Eur. J. 2001, 7, No. 1
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0701-0193 $ 17.50+.50/0
193

Â
F. Joo et al.
FULL PAPER
nesulfonic acid, sodium salt[22]) and TPPTS, 7.[3, 23] It is often
assumed that in aqueous ± organic two-phase reactions water
simply gives a liquid phase which dissolves the catalyst and
that nothing else happens. Several observations show this is
not the case.[24] For example, Delmas et al. observed that the
biphasic hydroformylation of 1-octene catalyzed by
[RhCl(cod)]2‡7 (COD ˆ 1,5-cyclooctadiene) proceeded with
a higher rate at pH 10 than at pH 7.[25] Yan et al. have
determined a maximum in the rate and selectivity of 1-hexene
hydroformylation at pH 7 using a catalyst prepared in situ
from RhCl3 and P[C6H4-4-O(CH2CH2O)6H]3.[26] A similar
rate increase with increasing pH was found by Mieczynska
et al. in hydrogenation and hydroformylation of unsaturated
alcohols catalyzed by [Rh(acac)(CO)2]‡PNS (PNS ˆ
Ph2PCH2CH2CONHC(CH3)2CH2SO3Li).[27] We suggested
earlier that in hydrogenation of [RhCl(tppms)3] (1) in water
a [RhH(tppms)3or4] monohydrido-species was formed instead
of [RhClH2(tppms)3] (3) as the primary product of oxidative
addition.[28a] A similar effect was invoked also in case of
hydrogenations catalyzed by [RhCl(pta)3] in aqueous solu-
tions (PTA ˆ 1,3,5-triaza-7-phosphaadamantane).[28b, c] How-
ever, we were unsuccessful in characterizing the supposed
[RhHP3or4] species (P ˆ 8 or PTA) by NMR spectroscopy.
Formation of a similar rhodium(i) monohydride was proposed
by Leitner et al. in the [RhCl(tppts)3]-catalyzed hydrogena-
tion of CO2 in the presence of aminesÐagain without a direct
NMR evidence.[29] On the other hand, hydrogenation of
[RuCl2(tppms)2]2 (in the presence of excess TPPMS) at
controlled pH at several points in the 1 < pH < 12 range
the midpoint of the equilibrium shown in Equation (1) was
around pH 6 and it could be easily shifted each way by the
addition of acid or base.[30] This led to very pronounced
changes in the rate and selectivity of hydrogenation of
unsaturated aldehydes[30, 31] as well as of aqueous bicarbon-
ate.[32]
‡
[RuClH(tppms)3] ‡ H2 ‡ TPPMS > [RuH2(tppms)4] ‡ H ‡ Clꢀ
(1)
The purpose of this paper is to describe the effects of
varying pH on the hydridorhodium(i) complexes formed
in aqueous solutions from [RhCl(tppms)3] (1) and
[RhCl(CO)(tppms)2] (2) and to relate the changes in the
reactivity of these complexes to their catalytic performance.
Compound 1 is a water soluble analogue of the Wilkinsonꢁs
catalyst, [RhCl(PPh3)3] (9); properties of the latter in organic
solvents are known in very fine detail.[19, 33] Similarly, 2 is the
precursor of [Rh(CO)H(tppms)3] (5) analogous to the im-
portant hydroformylation catalyst [Rh(CO)H(tppts)3] (6).
Some information regarding the hydride species formed from
[RhCl(tppts)3] (10) in neutral and strongly acidic solutions is
available from the work of Larpent and Patin,[34] however,
despite several efforts rhodium(i) hydrides in basic aqueous
Â
solutions have not been characterized before. Also, Horvath
has shown that in water 6 is less prone to phosphine
dissociation than [Rh(CO)H(PPh3)3] (11) in toluene, suppos-
edly due to a network of hydrogen bonds and cation binding
between the nine sulfonate groups.[35] Such indirect effects of
the aqueous phase are outside the scope of the present study.
1
together with H and 31P NMR measurements revealed that
Results and Discussion
Â
Abstract in Hungarian: A [RhCl(tppms)3] (1) es transz-
Reaction of 1 and molecular hydrogen: When 1 was dissolved
under argon in water with an orange color in acidic and with a
red color in basic solutions (308C, 0.2m KCl was used to
provide sufficient ionic strength for pH measurements) a
Â
Â
Â
[RhCl(CO)(tppms)2] (2) komplexek hidrolíziset es hidroge-
nezeset tanulmanyoztuk vizes oldatban, szeles pH tartomany-
 Â
Â
Â
Â
ban (2 < pH < 11), TPPMS feleslegben (TPPMS ˆ 3-difenil-
Â
Â
foszfinil-benzolszulfonsav natrium so). Savas oldatban 1
‡
certain amount of H was produced, this amount being a
Â
Â
Â
hidrogenezesekor cisz-mer- es cisz-fac-[RhClH2(tppms)3],
3a, b kepzodik, míg erosen lugos oldatban [RhH(H2O)-
(tppms)3] (4) alakul ki. E ket, egymassal egyensulyban
kepzodo komplex koncentracioja pH 8.2 eseten azonos. Dol-
gozatunkban elsokent adjuk meg egy kizarolag egyfogu foszfin
ligandumot tartalmazo monohidridorodium(i) komplex 4 H
es P NMR jellemzoit. 2 hidrolízise pH 9 alatt elhanyagolhato
merteku, hidrogenezese soran pedig [Rh(CO)H(tppms)3] (5)
function of the actual pH of the solution. Upon admission of
H2 the color of the solutions turned yellow independent of the
pH and this was accompanied by a second stage of proton
liberation. We have followed these reactions by using a pH-
potentiometric apparatus for automatic compensation of any
acidification (see Experimental Section). The actual time
course of both proton producing processes at a constant pH 10
is shown on Figure 1. The first step can be ascribed to the
hydrolysis of the complex under an argon atmosphere,
yielding [Rh(OH)(tppms)3] (12) while the second one in-
dicates a heterolytic fission of H2 with concomitant proton
formation. Hydrogenation of 1 was investigated at several
constant pH values in the 2 < pH < 11 range; the results are
summarized on Figure 2. It can be seen, that hydrolysis does
not take place at all at pH 2, and its extent is less than 10% at
pH 4 (front set of bars). Interestingly, even in strongly basic
solutions (pH 11) the amount of protons formed is only about
35% relative to rhodium. Herrmann et al. have shown that 10
gives [Rh(OH)(tppts)3] (13) with 70 ± 90% isolated yield
upon standing of its aqueous solution at room temperature.[36]
In our case, incomplete hydrolysis of 1 can be explained
Â
Â
Â
Â
Â
Â
Â
Â
Â
 Â
Â
 Â
Â
Â
Â
Â
 Â
 Â
Â
1
Â
Â
31
Â
Â
Â
Â
Â
  Â
Â
Â
Â
Â
Â
Â
Â
kepzodik, ami a jol ismert [Rh(CO)H(tppts)3] (6) ipari
Â
Â
Â
Â
hidroformilezo katalizator analogja (TPPTS ˆ 3,3',3''-fosz-
Â
Â
fintriil-benzolszulfonsav natrium so). pH-potenciometrikus
 Â
Â
Â
meresekkel kimutattuk, hogy a pH 5 ± 9 tartomanyban 5
Â
Â
Â
 Â
Â
Â
Â
kepzodese erosen pH-függo, ami magyarazatot ad több
hidroformilezesi folyamat korabban is eszlelt de mindeddig
nem ertelmezett pH-függesere. Ugyanakkor, bar a maleinsav es
fumarsav 1 altal katalizalt hidrogenezesenek sebessege a
2 < pH < 7 tartomanyban szinten erosen függ a pH-tol, ezt a
jelenseget kelloen ertelmezhetjük kizarolag e szubsztratumok
ionizacios allapotanak megvaltozasaval. Mindezen eredme-
nyek arra figyelmeztetnek, hogy a vizes közegu femorganikus
Â
Â
Â
Â
 Â
Â
Â
Â
Â
Â
Â
Â
 Â
 Â
Â
Â
Â
Â
Â
Â
Â
Â
Â
Â
Â
Â
Â
Â
Â
 Â
Â
Â
Â
Â
Â
Â
Â
Â
katalízisben allando pH-ju (pufferelt) oldatokat kell hasznalni.
194
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0701-0194 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 1

RhI Complexes
193±199
reaction. One proton is produced in the hydrolysis of 1
[Eq. (2)] and in the dehydrochlorination of 3 [Eq. (4)], the
latter reaction yields the monohydridorhodium(i) species,
[RhH(H2O)(tppms)3] (4). The same monohydride is produced
in the reaction of [Rh(OH)(tppms)3] (12) with H2 [Eq. (5)];
however, this reaction is again electroneutral. For this reason
the front set of bars on Figure 2 directly gives the relative
amount of 12 at the indicated pH on the mol fraction scale,
and the back set shows the same for 4.
Reductive elimination of HCl from various transition metal
compounds is a well known process often used to obtain
highly reactive low valent complexes.[38] Compounds of the
general formula [RhClH2(PR3)3] readily undergo such dehy-
drochlorination upon addition of a base such as Et3N or alkali;
this was frequently applied for generation of active catalysts
for ketone reduction (PR3 ˆ PPh3 and various optically active
tertiary phosphines).[39] The procedure for the preparation of
[HRh(PPh3)4] also calls for the use of KOH.[40] However, one
should consider that the reductive elimination of HCl is a
reversible process and in a one-phase procedure it is retarded
Figure 1. Time course of proton production upon dissolution and hydro-
genation of [RhCl(tppms)3] (1) in aqueous solution at pH 10. [1] ˆ 1.8 Â
10ꢀ3 m, [8] ˆ 5.4  10ꢀ3 m, [KCl] ˆ 0.2m, 0.1 MPa Ar or H2, Tˆ 308C.
‡
by the H and Clꢀ produced. Extraction of HCl, for example
‡
as [Et3NH] Clꢀ into a separate aqueous phase,[41a] or using an
aqueous alkali such as 40% w/w NaOH[41b] shifts the reaction
completely towards dehydrochlorination. Even with no added
base water as solvent facilitates processes such as in Equa-
tions (2) and (4) due to the strong solvation of ions, especially
Figure 2. Proton production upon hydrolysis (Ar) and hydrogenation of
[RhCl(tppms)3] (1) at various constant pH in the 2 < pH < 11 range. [1] ˆ
1.8  10ꢀ3 m, [8] ˆ 5.4  10ꢀ3 m, [KCl] ˆ 0.2m, 0.1 MPa Ar or H2, Tˆ 308C.
‡
that of H .
Notably we have made several attempts to isolate
[HRh(tppms)4] in solid form. Both the direct synthesis[40]
and the generally applicable TPPMS/PPh3 metathesis[22, 42]
methods failed and at best yielded a mixture of rhodium
hydrides based on the several absorptions in the hydride
stretching region of the infrared spectrum. In a similar
attempt [HRh(tppts)4] could also not be prepared.[36a]
considering the presence of a 1000 times excess of Clꢀ over
Rh. However, the presence of 12 in these solutions was
confirmed by 31P NMR data[37] although the spectra are not
well resolved probably due to the exchange with the excess of
8. Due to the ligand excess ([8]:[1] ˆ 3) formation of
[Rh(OH)(tppms)2]2 need not be considered.
The strength of our novel potentiometric approach is its
capabilility of indicating the pH boundaries within which a
certain species exists as an appreciable or even major
component of the reaction mixture. With a few excep-
tions[30, 31] metal ion hydrolysis in homogeneous or biphasic
aqueous organometallic catalysis has been neglected so far
and the formation of hydroxo-complexes was addressed only
by preparative methods.[36] Only a few studies describe the
effect of pH variation on the kinetics of hydrogenation[28, 43, 44]
and hydroformylation[25±27] reactions. Quantitative measure-
ments of proton production (or consumption) at several
constant pH values in the widest possible pH range, as shown
above, provide important additional information on the ªpH
windowº for the formation of catalytically important com-
plexes. On the other hand these pH-static hydrogenations can
only give an estimate regarding the chemical composition of
such species, not to mention structural information.
Replacement of the argon atmosphere by hydrogen results
in further proton production in the 4 < pH < 11 range, on top
of that observed in the hydrolysis process. The middle set of
bars on Figure 2 represents this additional proton production,
while the back set shows the combined amount of protons
produced upon dissolution and hydrogenation. It should be
emphasized that 1 does react with H2 at pH 4 but this reaction
is not accompanied by proton production. On the other
‡
extreme, a stoichiometric amount of H is formed in the
reaction of 1 and H2 at pH 10. Based on these results
combined with spectroscopic evidence (see below), the
processes taking place in aqueous solutions of 1 under argon
and hydrogen can be represented by Equations (2) ± (5):
‡
[RhCl(tppms)3] (1)‡H2O
[RhCl(tppms)3] (1)‡H2
[RhClH2(tppms)3] (3a,b)‡H2O
>
[Rh(OH)(tppms)3] (12)‡Clꢀ‡H
(2)
(3)
>
[RhClH2(tppms)3] (3a, b)
>
Characterization of the hydride species at various pH values
(4)
(5)
[RhH(H2O)(tppms)3] (4)‡Clꢀ‡H
‡
1
by H and 31P NMR spectroscopy: When 1 and a two-fold
excess of 8 was dissolved in water containing HClO4 (0.1m)
and pressurized with 6 MPa of H2, the originally deep orange
solution turned light yellow. With multinuclear NMR spec-
troscopy under medium gas pressure we observed the
quantitative formation of two rhodium-dihydride species,
[Rh(OH)(tppms)3] (12)‡H2
> [RhH(H2O)(tppms)3] (4)
Oxidative addition of H2 to 1 as in Equation (3) gives the
dihydridorhodium(iii) complex, [RhClH2(tppms)3] (3) in an
ªelectroneutralº reaction, as no ions are formed during the
Chem. Eur. J. 2001, 7, No. 1
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0701-0195 $ 17.50+.50/0
195

Â
F. Joo et al.
FULL PAPER
cis-mer-[RhClH2(tppms)3] (3a) (23%) and cis-fac-
[RhH2X(tppms)3] (3b) (77%, X ˆ H2O or Clꢀ) as shown in
Scheme 1.[45] Addition of three equivalents NaCl to the
reaction mixture resulted in an increase of the proportion of
the cis-mer-dihydride (up to 50%), while a ten-fold excess of
chloride shifted significantly the equilibrium further towards
the formation of 3a (88%). Although solely on the basis of 1H
and 31P NMR spectra it is not possible to discriminate
between X ˆ H2O or X ˆ Clꢀ as the ligand in axial position in
these complexes, such an effect of excess chloride suggests the
coordination of a chloride ion in 3a, and a water molecule in
3b. These data are in good agreement with those observed by
Larpent et al.[34] for the analogous Rh-TPPTS hydrides under
similar conditions in strongly acidic solutions.
evidence for the formation of rhodium(i)-monohydride in
aqueous solutions containing only monodentate phosphine
ligands. Part of the difficulties of obtaining well resolved
spectra in basic aqueous solutions is caused by a hydride
exchange with the solvent; indeed, we have observed a fast
H ± D exchange in solutions of 1 in D2O under moderate H2
pressure (2 MPa) which is markedly accelerated upon in-
creasing the pH.[49]
The acidity of hydrido transition metal complexes is an
important question of chemical bonding and homogeneous
catalysis, however, only a few acid dissociation constants
(pKa) are known for rhodium complexes in aqueous solu-
tions.[50] 3a, b yields 4 with concomitant HCl loss, that is not in
a simple acid deprotonation process; therefore it cannot be
characterized by a single pKa. Nevertheless, our pH-potentio-
metric measurements establish that under the particular
experimental conditions used the pH at which 3 and 4 are
present in 50:50 molar ratio is equal to pH 8.2, showing the
thermodynamic acidity of 3a,b being close that of
[Rh13(CO)24H2]3ꢀ.[50]
H1
Pd
-
H
Pc
Pc
Pa
Cl
H2
Pa
Cl
Rh
Rh
H
H2O
Pb
3b
3a
H2O/HClO4
6 MPa H2
H2O/HClO4
6 MPa H2
Hydrogenation of maleic and fumaric acid as a function of pH
with 1 as a catalyst precursor: In order to study possible
effects of the change in the catalystꢁs composition brought
about by changes in the pH we studied the hydrogenation of
maleic and fumaric acids in a wide pH range. Initial rates of
hydrogen uptake at 608C were determined by gas volumetry;
results are shown on Figures 3 and 4. The Figures also show
the molar distribution a) of the undissociated (H2A) and
deprotonated (HAꢀ and A2ꢀ) forms of the substrate acids as a
function of pH, calculated with acid dissociation constants
taken from the literature.[51]
P
Cl
P
Rh
0.1 M NaOH
20 % MeOH
6 MPa H2
P
1
0.1 M NaOH
Ar
H
Pf
OH
Pf
20 % MeOH
6 MPa H2
Pg
Pg
Rh
Pg
Rh
Pe
12
H2O
4
Scheme 1. Hydrolysis and hydrogenation equilibria in the aqueous sol-
ution of [RhCl(tppms)3] (1) under argon or hydrogen. P ˆ TPPMS (8).
In contrast to the above findings, completely different
NMR spectra were recorded using a strongly basic (0.1m
NaOH) aqueous solution of 1, although a similar color change
to light yellow was observed upon addition of 6 MPa H2. Both
in the 1H and 31P NMR spectra only broad signals were
observed in the temperature range of 2 ± 908C which may be
due to fast exchange with the solvent and excess TPPMS.
However, in solutions containing 20% v/v methanol the
signals sharpened and well resolved spectra could be ob-
tained. At pH 13.0 the only species, formed in quantitative
yield was the monohydride: [RhH(H2O)(tppms)3] (4) con-
taining the hydride, aqua, and the three phosphine ligands in a
trigonal bipyramidal geometry (Scheme 1). Oxidation of
TPPMS to phosphine oxide was always observed in strongly
basic solutions, however its extent usually did not exceed 10%
of all phosphorus which means that there was enough free
TPPMS left if needed for a tetrakis-phosphine species.[40, 47, 48]
Figure 3. The effect of pH on the initial rate of hydrogenation of maleic
acid (MA) catalyzed by [RhCl(tppms)3] (1) in aqueous solution. [1] ˆ 1.0 Â
10ꢀ3 m, [MA] ˆ 5.0  10ꢀ2 m, Tˆ 608C, ptotal ˆ 0.1 MPa. The calculated
distribution (a%) of nondissociated (H2A) and dissociated (HAꢀ, A2ꢀ
)
maleic acid is also shown.
1
However, the H and 31P NMR spectra[46] clearly show that
such a compound is not formed; the instability of the putative
[HRh(tppms)4] complex is also indicated by the failed
attempts of its preparation and can be ascribed to the large
steric demand of the meta-sulfonated triphenylphosphine
ligand.
As mentioned earlier, the existence of a species, such as 4,
has been inferred before from various observations, however,
well resolved 1H NMR spectra of such compounds have never
been obtained.[28, 29] Thus this paper gives the first direct
Figure 4. The effect of pH on the initial rate of hydrogenation of fumaric
acid (FA) catalyzed by [RhCl(tppms)3] (1) in aqueous solution. [1] ˆ 5.2 Â
10ꢀ4 m, [FA] ˆ 5.0  10ꢀ2 m, Tˆ 608C, ptotal ˆ 0.1 MPa. The calculated dis-
tribution (a%) of nondissociated (H2A) and dissociated (HAꢀ, A2ꢀ
)
fumaric acid is also shown.
196
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0701-0196 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 1

RhI Complexes
193±199
The most striking feature of these graphs is in that while the
rate of maleic acid hydrogenation goes through a deep
minimum with increasing pH, the case of fumaric acid is just
the opposite and a sharp maximum is seen. Both extrema
coincide with the highest concentration of the HAꢀ form of
the acids. It is an intriguing question why the monoanion of
maleic acid reacts sluggishly in contrast to the monoanion of
fumaric acid which was found an extremely reactive substrate;
presently we do not have the answer. However, it can be also
seen from the Figures, that most of the changes in the rate of
these hydrogenations occur below pH 6 ± 7 where the majority
of rhodium is present as dihydride 3, and that there is no
significant further change in the hydrogenation rate upon
increasing the relative amount of monohydride 4 by raising
the pH to 7 ± 10. It can be concluded therefore that in case of
these substrates the dihydride > monohydride equilibrium
has little or no influence on the overall rate of the hydro-
genation. Interestingly in line with the known higher reac-
tivity of Wilkinsonꢁs catalyst towards cis-olefins, maleate
esters are reduced faster by 9 in organic solvents than the
corresponding fumarates.[33] This general trend is also ob-
served with 1 in strongly acidic and strongly basic aqueous
solutions, that is with the H2A and A2ꢀ forms of the substrates.
In sharp contrast to this, the monoanion, HAꢀ formed from
the trans-olefinic acid is reduced about seven times faster than
the corresponding cis-isomer. In this case the effect of pH can
be linked to the changes caused in the ionization state of the
substrates. This conclusion agrees with that of Andersson
et al. who studied the hydrogenation of acetamidoacrylic acid
Figure 5. Proton production upon hydrolysis and hydrogenation of trans-
[RhCl(CO)(tppms)2] (2) at various constant pH in the 4 < pH < 10 range.
[2] ˆ 2.4  10ꢀ3 m, [8] ˆ 7.2  10ꢀ3 m, 0.1 MPa Ar or H2, Tˆ 358C.
5%. Therefore practically the total amount of proton
production can be ascribed to the formation of 5 in a direct
reaction of 2 with hydrogen. The second, even more important
observation, is in that monohydride 5 is not formed in
appreciable quantities at pH ꢁ5 but is formed in yields >90%
at pH ꢁ9. These reactions are summarized in Scheme 2.
OC
P
H2, P
P
no reaction
Rh
pH <5
Cl
2
H2, P
H+, Cl-
OH-
- Cl-
pH >5
H
OC
P
H2, P
P
Rh
15
P
Rh
P
catalyzed by a water soluble cationic bisphosphine complex,
P
OH
CO
‡
[Rh(bdppts)(nbd)]
(BDPPTS ˆ tetrasulfonated 1,4-bis-
5
(diphenylphosphinyl)butane, NBD ˆ norbornadiene). They
have found that the reaction proceeded with the same
mechanism both in organic and aqueous solvents and the
pH effects on the rate in aqueous solution were ascribed solely
to the dissociation/protonation of the substrate.[42]
Scheme 2. Hydrolysis and hydrogenation equilibria in aqueous solution of
trans-[RhCl(CO)(tppms)2] (2) under argon or hydrogen. P ˆ TPPMS (8).
It was not possible to record useful NMR spectra of
[Rh(CO)(OH)(tppms)2] (15) due to its low solubility even in
strongly basic solutions. However, at pH 10 finely resolved
spectra were obtained for 5 which was fully characterized by
1H, 31P, and 13C NMR spectroscopy.[54] We note that 13C-
enriched 2 was not formed in measurable ratios when an
aqueous solution of 2 was placed under 13CO at room
temperatureÐobviously the exchange is very slowÐbut could
be obtained in quantitative yield in the fast reaction of 1 and
13CO in methanol. The NMR data obtained by us for 5 are
very close to the corresponding parameters[35, 36, 48, 52d] deter-
mined for 6 (or 11) and are consistent with a trigonal-
bipyramidal structure of 5 with the three TPPMS ligands in
the equatorial plane. This structural correspondence gives
further evidence to the general observation that there are only
minor differences in the chemical properties of 7 and 8 and
complexes thereof.
The observed effect of pH on the formation of 5 (together
with the similarity of complexes of 7 and 8) explains the
results of Delmas et al.[25] who found a substantial increase in
the rate of hydroformylation of 1-octene with 6 on an increase
in the pH from 7 to 10, despite proton is not involved in the
catalytic reaction. Such a rise in the pH should indeed result in
a higher concentration of 6 and consequently in that of the
Reaction of 2 with molecular hydrogen: In striking contrast to
Vaskaꢁs compound, trans-[IrCl(CO)(PPh3)2], the analogous
rhodium complex trans-[RhCl(CO)(PPh3)2] (14) does not
form the corresponding dihydride in benzene or toluene
solutions at 208C and 0.1 MPa hydrogen, although it catalyzes
slow hydrogenation of olefins and aldehydes under more
harsh conditions.[33] Compound 2,[52a,b] the water soluble
analogue of 14,[52c,d] has been known for quite some time,
although it has only now been characterized by NMR
spectroscopy; the data confirm its trans-geometry.[53] In order
to obtain detailed data on the possible effects of a base on
reactions of this carbonyl complex similar to 1, 2 was
hydrogenated in several solutions of constant pH using a
ligand excess ([8]:[2] ˆ 3). The golden yellow solutions of 2 did
not show any change when stirred under H2 at pH 3 but
turned significantly to deeper yellow at pH 10.5 and in
sufficiently basic solutions proton production was observed
by pH potentiometry. The detailed results concerning proton
formation in these solutions are presented on Figure 5.
Two features are apparent on Figure 5 especially in
comparison with Figure 2. First, there is virtually no hydroly-
sis of 2 below pH 9 and even at pH 10 its extent is only about
Chem. Eur. J. 2001, 7, No. 1
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0701-0197 $ 17.50+.50/0
197

Â
F. Joo et al.
FULL PAPER
(Hungary). Rhodium(iii) chloride was purchased from Pressure Chemicals.
Maleic and fumaric acids were supplied by Aldrich and were recrystallized
from aqueous ethanol. Other reagents (Na2HPO4, NaH2PO4, HClO4)
obtained from Fluka and Aldrich were used as received.
real catalytic species, [Rh(CO)H(tppts)2], formed from 6 by
phosphine dissociation. It is apparent from the NMR data that
the effects of pH on hydrogenation of 1 and 2 bear no
connection to the presence of sulfonate substituent(s) in the
ligands. Therefore it seems reasonable to assume that similar
pH effects are responsible for the rate increase in the
hydroformylation and hydrogenation experiments of Yan
Acknowledgement
et al.[26] and Mieczynska et al.[27] where the tertiary phosphine
Â
Â
This work was supported by the Office Federal de lꢁEducation et de la
Science, Suisse (OFES C98.0011) and by the National Research Founda-
tion of Hungary (OTKA T029934 and F023159). N.L. is grateful for an
OFES fellowship. This research is part of the collaboration within the
COST Action D10/0001 Working Group.
Â
ligands in the [Rh(CO)HP3] catalyst carry poly(ethene
glycol)- and alkylsulfonato-type substituents, respectively.
Conclusion
Â
Â
Â
[1] a) F. Joo, M. T. Beck, Magy. Kem. Foly. 1973, 79, 189 ± 191; b) F. Joo,
M. T. Beck, React. Kinet. Catal. Lett. 1975, 2, 257 ± 263.
[2] Y. Dror, J. Manassen, J. Mol. Catal. 1976/1977, 2, 219 ± 222.
We have characterized some important protic equilibria of
water soluble hydroxo- and hydridorhodium complexes
including the first NMR characterization of a long anticipated
Rh(i)-monohydride, [RhH(H2O)(tppms)3], in basic aqueous
solution. It is shown that applying water in homogeneous or
aqueous ± organic biphasic systems for organometallic catal-
ysis can be by no means regarded as using just another inert
(innocuous) solvent. The molecular state of catalytically
important transition metal complexes may be strongly influ-
enced by the solution pH and this effect explains seemingly
unreasonable kinetic features of such important reactions as
hydrogenation and hydroformylation. A systematic study of
the effect of pH and the use of buffered aqueous solutions of
the appropriate pH in case of synthetically useful processes
may bring gratifying results and, in fact, is a must in aqueous
organometallic catalysis.
Ã
[3] E. Kuntz (Rhone-Poulenc Ind. S.A.), DE 2627354A1, 1976 [Chem.
Abstr. 1977, 87, 101944n]; DE 2700904, 1977 [Chem. Abstr. 1977, 87,
117602k]; DE 2733516, 1978 [Chem. Abstr. 1978, 88, 152026t].
[4] E. Kuntz, CHEMTECH 1987, 17, 570 ± 575.
Â
Â
Â
[5] F. Joo, A. Katho, J. Mol. Catal. A. 1997, 116, 3 ± 26.
[6] a) Aqueous-Phase Organometallic CatalysisÐConcepts and Applica-
tions (Eds.: B. Cornils, W. A. Herrmann), Wiley-VCH, Weinheim,
1998; b) NATO ASI Series 3/5ÐAqueous Organometallic Chemistry
Â
Â
and Catalysis (Eds.: I. T. Horvath, F. Joo), Kluwer, Dordrecht, 1995;
c) Applied Homogeneous Catalysis with Organometallic Compounds
(Eds.: B. Cornils, W. A. Herrmann), VCH, Weinheim, 1996.
[7] W. A. Herrmann, C. W. Kohlpaintner, Angew. Chem. 1993, 105,
1588 ± 1609; Angew. Chem. Int. Ed. Engl. 1993, 32, 1524 ± 1544.
[8] a) A. Durocher, W. Keim, P. Voncken, Compendium 75/76 (Ergän-
zungsband ªErdöl und Kohleº) 1976, 347 ± 354; b) S. Steines, U.
Englert, B. Drieûen-Hölscher, Chem. Commun. 2000, 217 ± 218.
Â
Â
Â
[9] a) I. T. Horvath, J. Rabai, Science 1994, 266, 72 ± 75; b) I. T. Horvath,
Acc. Chem. Res. 1998, 31, 641 ± 650.
[10] a) L. P. Barthel-Rosa, J. A. Gladysz, Coord. Chem. Rev. 1999, 190 ±
192, 587 ± 605; b) W. Keim, M. Vogt, P. Wasserscheid, B. Drieûen-
Hölscher, J. Mol. Catal. A. 1999, 139, 171 ± 175; c) B. Richter, A. L.
Spek, G. van Koten, B.-J. Deelman, J. Am. Chem. Soc. 2000, 122,
3945 ± 3951.
[11] a) K. R. Seddon, Ionic Liquids for Clean Technology: An Update,
2000, to be found under http://www.ch.qub.ac.uk/krs/krs.html; b) Y.
Chauvin, H. Olivier-Bourbigou, CHEMTECH 1995, 25, 26 ± 30; c) T.
Welton, Chem. Rev. 1999, 99, 2071 ± 2083.
[12] A. J. Carmichael, M. J. Earle, J. D. Holbrey, P. B. McCormac, K. R.
Seddon, Org. Lett. 1999, 1, 997 ± 1000; b) V. P. W. Böhm, W. A.
Herrmann, Chem. Eur. J. 2000, 6, 1017 ± 1025; c) J. E. L. Dullius,
P. A. Z. Suarez, S. Einloft, R. F. de Souza, J. Dupont, J. Fischer, A.
De Cian, Organometallics 1998, 17, 815 ± 819.
[13] a) Chemical Synthesis Using Supercritical Fluids (Eds.: P. G. Jessop, W.
Leitner), Wiley-VCH, Weinheim, 1999; b) W. Leitner, Top. Curr.
Chem. 1999, 206, 108 ± 132; c) Chemistry under Extreme or Non-
Classical Conditions (Eds.: R. van Eldik, C. D. Hubbard), Wiley, New
York, 1996.
Experimental Section
All manipulations were done under an inert atmosphere (argon, N2 or H2)
using conventional Schlenk techniques.
Instrumentation: Hydrogenations at controlled pH were carried out in a
magnetically stirred, jacketed reactor equipped with a RadelkisOP-0808P
combined glass-Ag/AgCl electrode, gas inlet/outlet, a capillary inlet for
base delivery and a sampling port closed by a rubber septum.[30] If required,
an ABU91 autoburette (Radiometer) supplied 0.2m KOH into the reactor
to keep the pH constant. The autoburette was controlled by a PC used also
‡
for data collection. The amount of H produced in the reaction was
calculated from the known volume and exact concentration of base.
Medium pressure (p < 12 MPa) NMR measurements were carried out
1
using sapphire NMR tubes[55] and the H, 13C, and 31P NMR spectra were
collected by Bruker AC200, AM360 and DRX400 NMR spectrometers.
1H, 13C, and 31P NMR spectra are referenced to 3-(trimethylsilyl)-1-
propanesulfonic acid sodium salt (TSPSA, Fluka), to internal or external
solvent peaks and 85% phosphoric acid, respectively. The spectra were
fitted with WINNMR, GNMR4.0, and NMRICMA/MATLAB programs
on a PC. Infrared spectra were recorded on a PE Paragon1000 PC FTIR
spectrometer in KBr discs. For hydrogen uptake measurements a constant
pressure gas-volumetric apparatus was used temperature controlled to
Æ0.18C by a Julabo F25 ultrathermostat/circulator.
[14] a) P. G. Jessop, T. Ikariya, R. Noyori, Chem. Rev. 1999, 99, 475 ± 493;
b) P. E. Savage, Chem. Rev. 1999, 99, 603 ± 621.
[15] a) W. Keim, T. M. Shryne, R. S. Bauer, H. Chung, P. W. Glockner, H.
van Zwet (Shell Int. Res.) DE-P 2054009, 1969 [Chem. Abstr. 1968,
69, 87167t]; b) W. Keim in Fundamental Research in Homogeneous
Catalysis, Vol. 4 (Eds.: M. Graziani, M. Giongo), Plenum, New York,
1984, pp. 131 ± 144.
Reagents: All solvents were purified by distillation and carefully deaerated
before use. Doubly distilled water was used throughout. TPPMS (8) was
prepared by sulfonation of PPh3, and compounds 1 and 2 by phosphine
metathesis in 9 and in trans-[RhCl(CO)(PPh3)2] using tetrahydrofuran
solutions as described earlier.[22, 42] The purity of the products was routinly
[16] Industrial scale aqueous ± organic biphasic processes: hydroformyla-
tion of propene (RCH-RP), butenes (Celanese), and higher olefins
(Union Carbide); telomerization of butadiene (Kuraray); production
of geranylacetone and hydrogenation of a,b-unsaturated aldehydes
Ã
(Rhone-Poulenc). a) G. Papadogianakis, R. A. Sheldon in Catalysis, A
checked by NMR[53] and by FTIR spectroscopy [KBr, 2: n ˆ 1980 cmꢀ1
Specialist Periodical Report, Vol. 13 (Senior reporter: J. J. Spivey),
Royal Society of Chemistry, 1997, pp. 114 ± 193; b) B. Cornils, W. A.
Herrmann, R. W. Eckl, J. Mol. Catal. A. 1997, 116, 27 ± 33; c) B.
Drieûen-Hölscher, Adv. Catal. 1998, 42, 473 ± 505.
ꢀ1
ˆ
ˆ
(C O, vs) and absence of TPPMS oxide, n ˆ 1120 cm (P O, s) in both 1
and 2]. D2O (99.9%) was purchased from Cambridge Isotope Laboratories.
H2, N2, and Ar were acquired from Carbagas-CH or from Messer
198
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0701-0198 $ 17.50+.50/0
Chem. Eur. J. 2001, 7, No. 1

RhI Complexes
193±199
[17] B. Cornils, Top. Curr. Chem. 1999, 206, 133 ± 152.
[18] M. Beller, B. Cornils, C. D. Frohning, C. W. Kohlpaintner, J. Mol.
Catal. A 1995, 104, 17 ± 85.
[40] N. Ahmad, J. J. Levason, S. D. Robinson, M. F. Uttley, Inorg. Synth.
1974, 15, 45 ± 64.
Â
Â
Â
[41] a) F. Joo, E. Trocsanyi, J. Organomet. Chem. 1982, 231, 63 ± 70; b) V. V.
Grushin, H. Alper, Organometallics 1991, 10, 1620 ± 1622.
[42] A. Borowski, D. J. Cole-Hamilton, G. Wilkinson, Nouv. J. Chim. 1978,
2, 137 ± 144.
Â
[19] P. A. Chaloner, M. A. Esteruelas, F. Joo, L. A. Oro, Homogeneous
Hydrogenation, Kluwer, Dordrecht, 1994.
[20] R. H. Grubbs, D. M. Lynn in Aqueous-Phase Organometallic Catal-
ysisÐConcepts and Applications (Eds: B. Cornils, W. A. Herrmann),
Wiley-VCH, Weinheim, 1998, pp. 466 ± 476.
[43] T. Malström, O. F. Wendt, C. Andersson, J. Chem. Soc. Dalton Trans.
1999, 2871 ± 2875.
[21] J. P. Genet, M. Savignac, J. Organomet. Chem. 1999, 576, 305 ± 317.
[22] a) S. Ahrland, J. Chatt, N. R. Davies, A. A. Williams, J. Chem. Soc.
[44] D. P. Paterniti, P. J. Roman, Jr., J. D. Atwood, Organometallics 1997,
16, 3371 ± 3376.
Â
Â
Â
Â
Â
1958, 276 ± 288; b) F. Joo, J. Kovacs, A. Katho, A. C. Benyei, T. Decuir,
[45] NMR data for 3a and 3b. All measurements in 80% H2O/20% D2O at
358C. 3a: 1H (400.03 MHz): d ˆ ꢀ9.85 (d, 2J(H1,Pb) ˆ 154.2 Hz),
ꢀ18.05 (s); 1H{31P}: d ˆ ꢀ9.85 (s), ꢀ18.05 (s); 31P (161.94 MHz):
d ˆ 39.1 (d, 1J(Pa,Rh) ˆ 111.8 Hz), 21.1 (m); 31P{1H}: d ˆ 39.1 (dd,
2J(Pa,Pb) ˆ 18.7 Hz), 21.1 (dt, 1J(Pb,Rh) ˆ 91.8 Hz). 3b: 1H
(400.03 MHz): d ˆ ꢀ9.85 (brs); 1H{31P}: d ˆ ꢀ9.85 (s); 31P
(161.94 MHz): d ˆ 37.3 (d, 1J(Pc,Rh) ˆ 111.2 Hz), 24.6 (m); 31P{1H}:
D. Darensbourg, Inorg. Synth. 1998, 32, 1 ± 8.
[23] W. A. Herrmann, C. W. Kohlpaintner, Inorg. Synth. 1998, 32, 8 ± 25.
[24] E. Fache, C. Santini, F. Senocq, J. M. Basset, J. Mol. Catal. 1992, 72,
337 ± 350.
[25] R. M. Deshpande, Purwanto, H. Delmas, R. V. Chaudhari, J. Mol.
Catal. A 1997, 126, 133 ± 140.
[26] a) Y.-Y. Yan, H.-P. Zhuo, B. Yang, Z.-L. Jin, J. Nat. Gas Chem. 1994,
436 ± 442; b) Y.-Y. Yan, H.-P. Zhuo, Z.-L. Jin, J. Mol. Catal. (China)
1994, 8, 147 ± 150; c) Y. Zhang, Y.-Z. Yuan, X.-L. Liao, J.-L. Ye, C.-X.
Yao, K. Tsai, Chem. J. Chin. Univ. 1999, 20, 1589 ± 1594.
2
1
d ˆ 37.3 (dd, J(Pc,Pd) ˆ 17.7 Hz), 24.6 (dt, J(Pd,Rh) ˆ 90.6 Hz).
[46] Complex 4: 1H (200.13 MHz, 66% H2O/17% D2O/17% MeOH,
258C): d ˆ ꢀ9.93 (q, J(H,Pg) ˆ 13.4 Hz); 31P (81.01 MHz, 66% H2O/
2
17% D2O/17% MeOH, 258C): d ˆ 40.8 (d, 1J(Pg,Rh) ˆ 156.3 Hz, 3P);
[27] E. Mieczynska, A. M. Trzeciak, J. J. Zioøkowski, J. Mol. Catal. A 1999,
148, 59 ± 68.
31P{1H}: d ˆ 40.8 (d, J(Pg,Rh) ˆ 156.3 Hz).
1
Â
Â
[47] B. Ilmayer, R. S. Nyholm, Naturwiss. 1969, 56, 415 ± 416.
Â
Â
[28] a) F. Joo, A. Benyei, P. Csiba, Chem. Commun. 1993, 1602 ± 1604; b) F.
[48] a) R. V. Kastrup, J. S. Merola, A. A. Oswald in Adv. Chem. Ser. 196:
Catalytic Aspects of Metal Complexes (Eds.: E. C. Alyea, D. W.
Meek), A.C.S., Washington, D.C., 1982, pp. 43 ± 64; b) C. Bianchini,
H. M. Lee, A. Meli, F. Vizza, Organometallics 2000, 19, 849 ± 853.
Â
Â
Â
Joo, L. Nadasdi, A. Cs. Benyei, D. J. Darensbourg, J. Organomet.
Chem. 1996, 512, 45 ± 50; c) L. Nadasdi, F. Joo, Inorg. Chim. Acta 1999,
293, 218 ± 222.
Â
Â
  Â
[49] a) G. Kovacs, L. Nadasdi, F. Joo, G. Laurenczy, Comp. Rend. Acad.
[29] W. Leitner, E. Dinjus, F. Gaûner in Aqueous-Phase Organometallic
CatalysisÐConcepts and Applications (Eds: B. Cornils, W. A. Herr-
mann), Wiley-VCH, Weinheim, 1998, pp. 486 ± 498.
Â
Â
Â
Sci., Series IIC, Chemistry, in press; b) G. Kovacs, L. Nadasdi, F. Joo,
G. Laurenczy, Proc. 12th J. Cartier Meeting, Lyon, France, 1999, Poster
6/9.
Â
Â
Â
Â
Â
[30] F. Joo, J. Kovacs, A. Cs. Benyei, A. Katho, Catal. Today 1998, 42, 441 ±
Â
Â
448.
[50] S. S. Kristjansdottir, J. R. Norton in Transition Metal Hydrides
Â
Â
Â
Â
Â
[31] F. Joo, J. Kovacs, A. Cs. Benyei, A. Katho, Angew. Chem. 1998, 110,
1024 ± 1026; Angew. Chem. Int. Ed. 1998, 37, 969 ± 970.
[32] G. Laurenczy, F. Joo, L. Nadasdi, unpublished results.
(Ed.: A. Dedieu), VCH, New York, 1992, Chapter 9.
[51] CRC Handbook of Chemistry and Physics, 78th ed., (Ed.: D. R. Lide),
CRC Press, Boca Raton, FL, 1997 ± 1998, pp. 8 ± 46.
Â
Â
 Â
[52] a) F. Joo, Z. Toth, M. T. Beck, Inorg. Chim. Acta L 1977, 25, 61 ± 62;
[33] a) J. A. Osborn, F. H. Jardine, J. F. Young, G. Wilkinson, J. Chem. Soc.
A 1966, 1711 ± 1732; b) B. R. James, Homogeneous Hydrogenation,
Wiley, New York, 1973, pp. 204 ± 248; c) R. S. Dickson, Homogeneous
Catalysis with Compounds of Rhodium and Iridium, Reidel, Dord-
recht, 1985; d) see ref. [19], pp. 8 ± 15.
Â
Â
b) F. Joo, Z. Toth, J. Mol. Catal. 1980, 8, 369; c) J. A. McCleverty, G.
Wilkinson, Inorg. Synth. 1966, 8, 214 ± 217; d) H. Schumann, G.
Cielusek, S. Jurgis, E. Hahn, J. Pickardt, J. Blum, Y. Sasson, A. Zoran,
Chem. Ber. 1984, 117, 2825 ± 2838.
[34] a) C. Larpent, H. Patin, J. Organomet. Chem. C 1987, 335, 13 ± 16;
b) C. Larpent, R. Dabard, H. Patin, Inorg. Chem. 1987, 26, 2922 ± 2924.
[53] Complex 2: 31P NMR (146 MHz, 80% H2O/20% D2O, 258C): d ˆ 30.5
(d, 1J(P,Rh) ˆ 127 Hz). For 13C enrichment, 1 was reacted with 13CO
(0.1 MPa in MeOH followed by evaporation to dryness. The residue
was taken up in 80% H2O/20% D2O. 13C (50.32 MHz): d ˆ 186.4 (d,
1J(C,Rh) ˆ 68 Hz).
Â
[35] I. T. Horvath, R. V. Kastrup, A. A. Oswald, E. J. Mozeleski, Catal.
Lett. 1989, 2, 85 ± 90.
[36] a) W. A. Herrmann, J. A. Kulpe, W. Konkol, H. Bahrmann, J.
Organomet. Chem. 1990, 389, 85 ± 101; b) W. A. Herrmann, J. A.
Kulpe, J. Kellner, H. Riepl, H. Bahrmann, W. Konkol, Angew. Chem.
1990, 102, 408 ± 414; Angew. Chem. Int. Ed. Engl. 1990, 29, 391 ± 393;
c) W. A. Herrmann, J. Kellner, H. Riepl, J. Organometal. Chem. 1990,
389, 103 ± 128.
[37] Complex 12: 31P NMR (161.93 MHz, 80% H2O/20% D2O, 408C):
dˆ35.9 (brd, 1J(Pf,Rh) ˆ 169.9 Hz) and 56.6 (brd, 1J(Pe,Rh) ˆ
187.4 Hz).
[54] Complex 5: all NMR measurements in 80% H2O/20% D2O at 408C.
1H (200.13 MHz): d ˆ ꢀ9.49 (dq, 2J(H,P) ˆ 13.8 Hz, 2J(H,C) ˆ
38.2 Hz); 1H{31P} (400.03 MHz): d ˆ ꢀ9.49 (d, 2J(H,C) ˆ 38.2 Hz;
2
1H{13C}: d ˆ ꢀ9.49 (q, J(P,H) ˆ 13.8 Hz); 13C (50.32 MHz): d ˆ 206.0
(ddq, 1J(C,Rh) ˆ 53.9 Hz, 2J(C,P) ˆ 10.4 Hz, 2J(H,C) ˆ 38.2 Hz);
13C{1H}: d ˆ 206.0 (dq); 31P(161.93 MHz): d ˆ 41.2 (brd, 1J(P,Rh) ˆ
155.0 Hz); 31P{1H} d ˆ 41.2 (dd, 1J(P,Rh) ˆ 155.0 Hz, 2J(P,C) ˆ
10.4 Hz).
Â
[38] V. V. Grushin, Acc. Chem. Res. 1993, 26, 279 ± 286.
[55] a) I. T. Horvath, J. M. Millar, Chem. Rev. 1991, 91, 1339; b) A.
Â
Â
[39] a) B. Heil, S. Torös, J. Bakos, L. Marko, J. Organomet. Chem. 1979,
175, 229 ± 232; b) M. Gargano, P. Giannoccaro, M. Rossi, J. Organo-
met. Chem. 1977, 129, 239 ± 242.
Cusanelli, U. Frey, D. T. Richens, A. E. Merbach, J. Am. Chem. Soc.
1996, 118, 5265 ± 5271.
Received: April 14, 2000 [F2428]
Chem. Eur. J. 2001, 7, No. 1
ꢀ WILEY-VCH Verlag GmbH, D-69451 Weinheim, 2001
0947-6539/01/0701-0199 $ 17.50+.50/0
199