Synthesis, characterization, and catalytic activity of Rh(I) complexes with (S)-BINAPO, an axially chiral inducer capable of hemilabile P,O-heterobidentate coordination

Article abstract of DOI:10.1007/s007060070015

The reaction of dinuclear rhodium(I) derivatives of the formula [Rh(DIOL)X]₂ with the axially chiral phosphinyl phosphane 2-(diphenylphosphinyl)-2′-(diphenylphosphanyl)-1,1′-binaphthalene ((S)-BINAPO, 1) leads to the formation of cationic compl

Full text of DOI:10.1007/s007060070015

Monatshefte fuÈr Chemie 131, 1351±1361 (2000)
Synthesis, Characterization, and Catalytic
Activity of Rh(I) Complexes with (S)-BINAPO,
an Axially Chiral Inducer Capable of Hemilabile
P,O-Heterobidentate Coordination
1
2
3
Sera®no Gladiali1;Ã, Serenella Medici , Tamas Kegl , and Laszlo Kollar
Á
Â
Á
Á
Á
1
Á
Dipartimento di Chimica, Universita di Sassari, I-07100 Sassari, Italia
Â
Research Group for Petrochemistry of the Hungarian Academy of Sciences, H-8201 Veszprem,
2
3
Hungary
Â
Department of Inorganic Chemistry, University of Pecs, and Research Group for Chemical Sensors
Â
of the Hungarian Academy of Sciences H-7601 Pecs, Hungary
Summary. The reaction of dinuclear rhodium(I) derivatives of the formula [Rh(DIOL)X]2 with the
axially chiral phosphinyl phosphane 2-(diphenylphosphinyl)-20-(diphenylphosphanyl)-1,10-binaphtha-
‡
lene ((S)-BINAPO, 1) leads to the formation of cationic complexes [(BINAPO)Rh(DIOL)] where the
ligand (S)-BINAPO consistently displays a P,O-chelate coordination which is mantained even in
solvents of fair polarity. The mononuclear rhodium(I) complexes (S)-2-diphenylphosphanyl-20-
diphenylphosphinyl-1,10-binaphthalene-(1,5-cyclooctadiene) rhodium tetra¯uoroborate (3b) and (S)-
2-diphenylphosphanyl-20-diphenylphosphinyl-1,10-binaphthalene-(1,4-norbornadiene) rhodium tetra-
¯uoroborate (3c) with 1,5-cyclooctadiene (COD) and 2,5-norbornadiene (NBD) as the diole®n were
isolated and characterized. Both show a ¯uxional behaviour in solution which is due to the mobility of
the diole®n rather than to a displacement-recombination of the oxygenated arm of the ligand. The
mobility of the 1,4-norbornadiene ligand in 3c is extremely pronounced and the coordinated diole®n
¯exibility could be frozen only at about 200 K. These complexes are active but poorly stereoselective
catalysts for the hydrogenation, hydroboration, and hydroformylation of alkenes.
Keywords. Axially chiral auxiliaries; Coordination chemistry; Hemilabile ligands; Homogeneous
catalysis; Rhodium complexes.
Introduction
Mixed ligands with a phosphorus and an oxygen atom as substitutionally inert and
labile donors are the most common type of hemilabile ligands reported in the
literature [1]. A large variety of transition metal complexes with these ligands have
been prepared and extensively studied for their peculiar reactivity, dynamic
properties, and catalytic behaviour [2]. A signi®cant part of these investigations has
dealt with rhodium as the metal centre because the relevant complexes are of
interest for their catalytic activity in several processes.
Ã
Corresponding author

1352
S. Gladiali et al.
Hemilabile P,O-heterobidentate ligands apparently display a positive effect on
the reaction rate and/or on the selectivity of some Rh-catalyzed hydrocarbonylation
reactions. The activating properties of simple phosphane oxides in the hydro-
formylation of ole®ns by Rh-catalysts have been claimed in several patents [3]. The
same holds for aminoalkylphosphane oxides, which perform better than the
corresponding amino phosphanes in the hydroformylation of styrene [4]. The
carbonylation of methanol proceeds at higher rate when Rh-complexes with
diphosphane hemioxides [5] and hemisul®des [6] are used in the place of the
conventional Monsanto catalyst.
More recently, it has been shown that P,O-heterobidentate Rh(I)-complexes
containing phosphane-phosphonates as hemilabile chelating ligands are quite active
catalysts in the carbonylation of methanol to acetic acid [7]. Similar complexes are
capable to hydroformylate styrene at a reasonable rate even at room temperature
and moderate pressures, affording the branched aldehyde with excellent selectivity
[8].
The catalytic applications of Rh-complexes of this type are not limited to the
activation of carbon monoxide, but include also the activation of other small
molecular entities such as hydrogen. For instance, in their extensive investigation on
mixed ether-phosphane ligands, Lindner et al. have demonstrated the bene®cial
in¯uence of the presence of a potentially chelating ether moiety on the performance
of Rh complexes in the hydrogenation of simple alkenes [9].
In spite of the interesting results obtained in these catalytic reactions, to the best
of our knowledge there are no reports on the use of Rh-complexes with chiral P,O-
hemilabile ligands in enantioselective catalysis. It has been suggested that the
unusual ef®ciency displayed in the Rh-catalyzed asymmetric hydrogenation of ꢀ,ꢁ-
unsaturated acids by some ortho-alkoxyphenyl substituted monophosphane such as
cyclohexyl o-anisyl methyl phosphine (CAMP [10]) should be due to the temporary
coordination of the oxygen donor, but no experimental evidence of chelate
coordination to the metal of this type of ligand is yet available.
In recent papers we have shown that the axially chiral phosphinyl phosphane
BINAPO (1, Fig. 1) easily available from 2,20-dihydroxy-1,10-binaphthalene
(BINOL) in four steps, is an effective chiral inducer in the asymmetric hydro-
silylation of styrene with in situ Pd catalysts (ee > 70%) [11] and in the asymmetric
hydroformylation of styrene with in situ Pt±Sn catalysts (ee ꢀ 30%) [12].
In the second case, we have provided evidences that BINAPO behaves as a
hemilabile ligand towards the Pt(II) centre. Bidentate coordination affording an
eight-membered P,O-chelate ring takes place in the absence of competitive donors,
whereas cleavage of the chelate ring through displacement of the oxygenated arm is
easily accomplished by donor solvents such as dimethylsulfoxide or by ꢂ-acidic

Rh(I) Complexes with (S)-BINAPO
1353
ligands such as CO [12]. We have also demonstrated that the insertion of tin(II)-
chloride in the chelate complex, a crucial step for the production of the active
hydroformylation catalyst, proceeds with complete positional selectivity into the
Pt±Cl bond trans to the oxygen donor, thus affording a single Pt±SnCl3 chelate
complex [13].
These results encouraged us to expand the scope of the axially chiral hemilabile
ligand (S)-BINAPO in asymmetric catalysis. Thus, we investigated in some detail
the coordination chemistry of BINAPO towards Rh(I) centres with the aim to exploit
the potential of the relevant complexes in some Rh-catalyzed enantioselective
reactions. Here we report the results obtained in the course of this investigation.
Results and Discussion
The ®rst experiments were performed using [Rh(COD)(ꢃ-OMe)]2 (2; COD ˆ 1,5-
cyclooctadiene) and racemic BINAPO (1) as the starting materials and monitoring
the progress of the reaction by 31P NMR spectroscopy. When two molar equivalents
of 1 were added to a solution of 2 in CDCl3 at room temperature, no reaction took
place, and only the peaks of the free ligand at 14.7 and 27.7 ppm could be
observed. The addition of two molar equivalents of trimethylsilyl tri¯ate to this
solution caused the immediate displacement of the bridging methoxo groups of 2 as
evidenced by a pronounced down®eld shift of both phosphorus resonances, resulting
in a sharp singlet at 42.9 and a sharp doublet at 19.1 ppm (J ˆ 142:5 Hz). The shift
to lower ®eld of both resonances and the presence of a 1JP;Rh coupling demonstrated
that chelate coordination of ligand 1 to the rhodium centre occurred through the
oxygen donor of the phosphinyl group and the phosphanyl phosphorus affording 3a.
Scheme 1

1354
S. Gladiali et al.
Complex 3a did not show any coupling between the metal and the phosphinyl
phosphorus. The same occurred in the case of the related derivatives 3b and 3c (vide
infra). The absence of coupling between these nuclides contrasted with the results
reported by other authors for phosphane-phosphinyl 5- and 6-membered chelate Rh
complexes [5, 7], whereas it is in line with our previous observations on BINAPO-Pt
complexes [12]. We may wonder whether magnetization exchange between
phosphorus and rhodium is suppressed because of the larger size of the chelate
ring in 3a.
1
The H NMR spectrum of this sample showed four fairly broad signals in the
range of the ole®nic protons and two multiplets in the region of allylic protons
which were assigned to the chemically nonequivalent protons of COD coordinated
to the metal. No free COD was noticed in solution.
Trimethylsilyl tri¯ate reacts as well with free BINAPO, giving a product
featuring two singlets at 0.9 and 52.1 ppm in 31P NMR. The formation of this
tri¯ate adduct, tentatively identi®ed as the trimethylsilyl phosphanyl oxonium
tri¯ate 4, is fast and quantitative at room temperature. Addition of the dinuclear
complex 2 to a CDCl3 solution of 4 resulted in the formation of a new complex
showing two peaks at 35.1 (d, J ˆ 169:2 Hz) and 50.5 ppm (s) in the 31P NMR.
These were attributed to the coordinated phosphanyl and to the uncoordinated
phosphanyl oxonium group of complex 5 containing two monodentate ligands 4
with a cis-geometry. Apparently, monodentate coordination to the Rh(I) centre of
two units of BINAPO takes place only when the ligand cannot chelate the metal.
The preparation of the enantiopure complexes 3b and 3c was accomplished by
‡
reacting the rhodium-THF adducts [Rh(NBD)(THF)2] BF4 (6a, NBD ˆ 2,5-
‡
norbornadiene) and Rh(COD)(THF)2] BF4 (6b), respectively, with a stoichio-
metric amount of (S)-BINAPO. The required intermediate complexes 6 were
prepared in situ from the corresponding chloro-bridged dimers and AgBF4
according to Ref. [14]. The isolated yields of the pale yellow products 3b and 3c
were in the range of 80±85%.
The 31P NMR spectrum of the isolated complex 3b gave two sharp phosphorus
resonances (42.9 (s) and 19.1 (d, J ˆ 142:5 Hz) ppm) in accordance with the in situ
1
experiments. Both the aromatic and the ole®nic region in the H NMR (see
experimental) clearly showed the presence of a single complex only. The
coordinated COD gave four broad ole®nic signals at 5.3, 4.7, 3.5, and 3.3 ppm
and two broad multiplets at 1.8 and 2.2 ppm. Line broadening was indicative of a
modest but not negligible ¯uxional behaviour of the complex at room temperature.
The 31P NMR spectrum of the norbornadiene complex 3c showed the same
pattern as that of 3b (39.8 (s) and 16.6 (d, J ˆ 163:5 Hz) ppm, sharp peaks). In the
Scheme 2

Rh(I) Complexes with (S)-BINAPO
1355
1H NMR in CDCl3, the protons of the norbornadiene moiety bound to rhodium
appeared at 1.25, 3.67, 4.0, and 4.12 ppm, each integrating for 2H (see
Experimental). The equivalent nuclei were assigned to the methylene bridge, the
allylic, and the two pairs of vinylic protons, respectively. In the absence of any
dynamic process, the NBD ligand bound to the metal should give eight separated
resonances because all protons are diastereotopic. The observed proton averaging
indicated that complex 3c is highly ¯uxional at room temperature.
The chelate coordination of (S)-BINAPO in 3c is con®rmed in the IR spectrum
by the shift of the P=O stretching band to lower wave numbers (from 1201 to
1
1162 cm ) as a consequence of the slight weakening of the P=O bond.
In acetone-d6, the NBD ole®nic protons of 3c gave a single signal at 4.2 ppm,
whereas the allylic and the methylene protons experienced only an almost negligible
down®eld shift. Whether this is due to an accidental isochronicity or to an additional
dynamic process averaging all vinylic protons of the coordinated NBD could not be
de®nitely clari®ed. Variable temperature measurements showed that the singlet of
the ole®nic protons became very broad at 248 K (ꢄ1=2  60 Hz) and practically
disappeared in the noise at 233 K. Further lowering of the temperature down to
193 K resulted in the separation of four broad ole®nic protons at 2.7, 3.7, 3.95, and
4.7 ppm.
Apparently, the slowing down of the internal motion of the norbornadiene ligand
in 3c at 193 K produced a situation similar to that observed for 1,5-cyclooctadiene
in 3b at 298 K. Taking into consideration the different solvents, from these data the
difference in the activation energy for the dynamic processes of the two complexes
could be estimated to lie in the range of 25±35 kJ/mol. Such a difference in the
¯uxionality of the two complexes 3b and 3c is exceptionally high and, to the best of
our knowledge, it has no precedent in the case of other Rh-diole®n complexes with
bidentate/heterobidentate chelating ligands.
It is as well surprising that the NBD derivative appears de®nitely more ¯uxional
than the COD species; in our experience, the opposite situation is more frequently
encountered. It must be stressed that the ¯uxional behaviour noticed for 3b and 3c is
apparently due to the mobility of the diole®n ligand rather than to a dissociation-
recombination process of the oxygenated arm of the BINAPO ligand. This con-
clusion followed from the observation that even at room temperature the protons of
the binaphtyl backbone showed no line broadening at all but mantained their very
detailed pattern, whereas the diastereotopic protons of NBD were experiencing a
fast dynamic process which made them equivalent pair by pair. The same occurred
to the phosphorus resonances of the NBD complex which remained quite sharp over
the entire range of temperature investigated, showing no evidence of the presence of
traces of an uncoordinated P=O arm.
The preservation of the chelate coordination of 1 even in the presence of a
coordinating solvent such as acetone was fairly surprising and contrasted with the
behaviour observed for the same ligand in the coordination towards a Pt(II) centre
as in the case of (BINAPO)PtCl2 [12]. This fact is even more surprising because in
Rh derivatives such as 3 the metal centre is located at the junction of both chelate
rings, and we would expect that the interplay of the two different rings should
result in an increase of the torsional strain to be accommodated in the bis-chelate
structure.

1356
S. Gladiali et al.
We may wonder whether the different hemilabile character displayed by 1
towards rhodium and platinum may be ascribed to the different softness of the two
metal centres, Pt(II) being probably softer than Rh(I) in spite of the presence of two
hard chlorine ligands. As a matter of fact, literature contains several examples of
chelating P,O-heterobidentate Rh(I) complexes which have been isolated and charac-
terized, whereas the analogous Pt(II) derivatives are found much less frequently [1, 2].
With these Rh complexes at hand, we initiated a screening of their catalytic
activity in some reactions of asymmetric addition to multiple bonds. These catalytic
runs were carried out either using the preformed complexes 3b and 3c or by adding
the required amount of ligand 1 to a suitable dinuclear Rh(I)-diole®n complex with
the aim to generate the catalyst in situ.
The NBD complex 3c displayed no catalytic activity in the H-transfer reduction
of acetophenone from 2-propanol, but it was found to be a good catalyst for
the hydrogenation of the carbon-carbon bond of methyl acetamidoacrylate (7,
Scheme 3). The saturated product 8 (Scheme 3) was isolated in quantitative yield
after 5 h of reaction in benzene-MeOH solution under 2 bar of hydrogen at room
temperature at a substrate/metal ratio of 100:1. The reaction was basically devoid of
stereoselectivity, and the alanine derivative 8 was almost racemic (4% ee; (R)-
con®guration).
Scheme 3. Hydrogenation of methyl ꢀ-acetamidoacrylate
In order to gain some insight into this reaction, a CDCl3 solution of complex 3c
was pressurized with 40 bar of hydrogen in a NMR high-pressure quartz tube, and
the progress of the reaction was monitored by 1H and 31P NMR. Fast and complete
hydrogenation of the coordinated NBD took place in a few seconds, and a complex
mixture of Rh(III)-hydrides was generated. Even if we could not draw any sound
conclusion about the structure of the main hydridic species present in solution from
the NMR spectra, two of them were characterized by one doublet (J ˆ 193:4 and
177 Hz, respectively) in the shift range of bound phosphanes and by one singlet in
the region of a bound phosphane oxide group. This fact lent support to the view that
chelate coordination of BINAPO can survive even under high hydrogen pressure, at
least in part.
The asymmetric hydroboration of styrene (9) was run for 8 h at room tem-
perature in THF solution using catecholborane as the reagent and 3c as the catalyst
at a substrate-to-metal ratio of 500:1 (Scheme 4). The reaction was analyzed by GC
on a chiral phase after an oxidative work-up which converted the borane derivatives
into the corresponding carbinols 10 and 11. The conversion was as low as 22%, and
the branched isomer 10 was formed in 73% yield with 40% ee ((S)-con®guration).
The preformed complex 3c was inspected for its catalytic acitivity in the
hydroformylation of styrene, and its behaviour was compared with the catalyst

Rh(I) Complexes with (S)-BINAPO
1357
Scheme 4. Hydroboration of styrene
Scheme 5. Hydroformylation of styrene
prepared in situ by addition of the appropriate amount of ligand 1 to [Rh(NBD)Cl]2
(Scheme 5).
The catalytic runs were carried out in toluene at a substrate-to-metal ratio of
4000:1 (or 2000:1) at 40±100ꢁC under 80 bar of an equimolar mixture of CO and
H2. The reaction product consisted of a mixture of branched and linear aldehydes
12 and 13. The amount of the hydrogenated product 14 was always below 1%. The
most signi®cant results are collected in Table 1.
The catalytic activity of the preformed complex 3c was quite high, and
quantitative conversions of styrene were obtained at 40ꢁC in 15 h at a substrate-to-
metal ratio of 2000 (Table 1, entry 6). The reaction rate was lower but still high
when acetone was used as the solvent (Table 1, entry 7). These rates compare
favourably with those observed in the hydroformylation of styrene with bidentate
diphosphanes and diphosphites [15] or other chelating phosphorus donor ligands
such as the mixed phosphane-phosphite BINAPHOS [16].
Table 1. Hydroformylation of Styrene by Rhodium/BINAPO Catalysts (2 cm3 of styrene in 6 cm3 of
toluene; substrate:metal ˆ 4000:1; initial pressure: 80 bar; CO:H2 ˆ 1:1)
Entry
1a
T/Cꢁ
L/Rh
t/h
Conv./%
Branched/% (ee)
100
100
80
1
5
2
95
67
76
2a
2
65
3b
4b
5b
6c
1.5
2
6
95
76
80
6
69
77
60
1.5
1
6
36
93
40
15
15
> 99
68
92 (6%, R)
96 (4%, R)
7c,d
40
1
a
b
Racemic BINAPO in the presence of [Rh(NBD)Cl]2; (S)-BINAPO (75% ee) in the presence of
c
[Rh(NBD)Cl]2; the chiral aldehyde was almost racemic; preformed [Rh(NBD)(S)-BINAPO] BF4
‡
d
(3c) used as catalyst; substrate:metal ˆ 2000:1; solvent: acetone

1358
S. Gladiali et al.
Taking into account the temperatures and the higher substrate-to-metal loading
(4000:1), the catalytic activity displayed by the in situ catalysts was de®nitely lower
than that observed for 3c (Table 1, entries 1±5). This was particularly evident in runs
1 and 2, where racemic rather than enantiopure BINAPO was used as the ligand.
Apparently, ligands of different enantiomeric composition can produce catalysts of
different activity. This may be the case if Rh complexes containing two BINAPO
molecules as monodentate ligands play some role in the catalysis of the hydro-
formylation.
Increasing the ligand-to-metal ratio resulted in a further decrease of the
reaction rate (compare entries 1/2 and 3/4 in Table 1) and in a lower selectivity for
the branched aldehyde (compare entries 1/2). This result may as well point towards
an active involvement in the catalysis of Rh-BINAPO species with two units of
ligand for each metal atom.
As usual for Rh-catalyzed hydroformylation, the branched selectivity improved
substantially upon lowering the reaction temperature. At 60ꢁC, the ratio of the chiral
aldehyde 12 was as high as 93% (Table 1, entry 5) and improved up to 96% at 40ꢁC
when the reaction was run in acetone (Table 1, entry 7).
In the enantioselective hydroformylation, the in situ catalysts prepared with a
sample of (S)-BINAPO of 75% enantiomeric purity produced a practically racemic
branched aldehyde. As these experiments were run at fairly high temperatures
(Table 1, entries 3±5), with the aim to minimize the extent of racemization of
the chiral aldehyde 12 in the course of the reaction, the following tests using
the preformed complex 3c containing the enantiopure (S)-ligand were run for 15 h
at 40ꢁC. Even under these mild conditions, however, the stereoselectivity of
the reaction was low (ee < 10%). Since in our experience under these conditions
no racemization of 12 took place in the hydroformylation of styrene with other
Rh-based catalysts, we can con®dently assume that the same has occurred in
the present case. This means that these low ees re¯ect the real enantiodiscrimi-
nation ability of the catalyst and that they are not the consequence of
stereochemical transformations which follow after the formation of the chiral
aldehyde 12.
In conclusion, in the asymmetric hydroformylation of styrene the catalytic
system (S)-BINAPO/Rh is signi®cantly more active but less stereoselective than the
BINAP-based rhodium catalyst under similar conditions [17]. This fact may
indicate that the catalytic species originating from (S)-BINAPO and (S)-BINAP
may involve a different type of binding of the chiral ligand at the metal centre.
Together with the previous observations, this result provides further support that
Rh complexes containing two BINAPO molecules as monodentate ligands may play
an active role in the catalysis of the hydroformylation of styrene. Unfortunately, our
attempts aiming at trapping such a species under CO pressure failed as the 31P NMR
spectra recorded under CO pressure led to inconclusive results.
Conclusions
It has been already shown in our previous work that the binaphthyl phosphinyl
phosphane 1 shows a sharp preference for P,O-chelate coordination towards Pd(II)
and Pt(II) centres [11, 12]. This tendency is even more pronounced in the case of

Rh(I) Complexes with (S)-BINAPO
1359
Rh(I) derivatives, which by reaction with 1 produce the chelate species 3 as the sole
isolable products.
Unlike in the platinum case, no evidence for monodentate coordination of two
equivalents of 1 to the Rh-centre through the phosphanyl donor was noted at any
intermediate stage of this reaction. On the contrary, it seems that for this process to
occur the reaction path leading to chelate coordination must be suppressed, as was
shown in the case of the oxonium tri¯ate 4.
The hemilabile character of 1 towards rhodium is hard to be evidenced: the
oxygenated arm of the ligand is not displaced by acetone, and the chelate
coordination is apparently preserved even after the oxidative addition of hydrogen.
The results obtained in the asymmetric catalytic reactions, however, are better
explained if one assumes that the chelate structure is not preserved and that Rh
complexes containing two monodentate BINAPO molecules are actively involved in
catalysis.
Although we have no direct evidence that displacement of the oxygenated arm
of BINAPO occurs during the hydroformylation, this possibility ®nds some support
from the results observed which are more comfortably assessed if one assumes that
two catalytic species, each one containing two monohapto BINAPO units around
the Rh centre, are present during the reaction. The modest stereoselections, the
signi®cant differences of the reaction rates recorded in changing from the racemic to
the enantiopure BINAPO-based catalyst, and the reversal of handedness observed
with (S)-BINAPO as compared to (S)-BINAP [17] are all factors lending support to
this view.
Experimental
The 1H and 31P NMR spectra were recorded in CDCl3 on a Varian Inova 400 spectrometer at 400 and
161.9 MHz. The 31P chemical shifts are reported relative to external 85% H3PO4. The samples from
the catalytic runs were analyzed with a Hewlet Packard 5890A gas chromatograph ®tted with a 25 m
capillary column coated with diethyl tert-butylsilyl-ꢁ-cyclodextrin PS 086 (i.d. 0.25 mm) from
MEGA (Legnano, Italia) using He as the carrier (head pressure 60 kPa, split ratio 100). The elemental
analyses were performed on a 1108 Carlo Erba apparatus. They agreed favourably with the calculated
values (C,H). The optical rotation of 2-phenylpropanal was measured in benzene solution on a Perkin
Elmer 241 polarimeter. (S)-BINAPO was prepared as described in Ref. [11].
(S)-2-Diphenylphosphanyl-20-diphenylphosphinyl-1,10-binaphthalene(1,5-cyclooctadiene)
rhodium tetra¯uoroborate (3b; C52H44OBF4P2Rh)
To a solution of 200mg [Rh(COD)Cl]2 (0.4 mmol) in 10 cm3 of anhydrous THF under argon, 156mg
AgBF4 (0.8 mmol) were added. After stirring for 1 h, the precipitate was ®ltered through celite. To the
yellow solution of the rhodium-THF adduct, 510 mg (S)-BINAPO (0.8 mmol) were added. The colour
of the solution changed to orange immediately. After stirring for 2 h the solvent was evaporated, and
the resulting complex 3b was obtained as orange solid.
Yield: 600 mg (80%); 1H NMR (CDCl3, ꢅ, 400 MHz): 1.8 (m, 4H), 2.2 (m, 4H), 3.3 (m, 1H), 3.5
(m, 1H), 4.7 (m, 1H), 5.3 (m, 1H), 6.5 (d, 1H), 6.7 (m, 5H), 6.9 (m, 7H), 7.1 (m, 1H), 7.35 (t, 1H), 7.6
(m, 11H), 7.85 (m, 2H), 8.05 (m, 2H), 8.3 (d, 1H) ppm; 31P NMR (CDCl3, ꢅ, 161.9 MHz): 42.9 (s),
19.1 (d, J(103Rh, 31P) ˆ 142.5 Hz) ppm.
1

1360
S. Gladiali et al.
(S)-2-Diphenylphosphanyl-20-diphenylphosphinyl-1,10-binaphthalene-(2,5-norbornadiene)
rhodium tetra¯uoroborate (3c; C51H40OBF4P2Rh)
Following the same procedure as reported above, [Rh(NBD)Cl]2 gave 3c as an orange solid.
Yield: 590 mg (80%); 1H NMR (CDCl3, ꢅ, 400 MHz): 1.25 (s, 2H), 3.67 (s, 2H), 4.0 (s, 2H), 4.12
(s, 2H), 6.2 (d, 1H), 6.4 (m, 2H), 6.6 (m, 4H), 6.8 (t, 1H), 6.9 (m, 4H), 7.1 (m, 2H), 7.3 (m, 2H), 7.45
(m, 2H), 7.6 (m, 8H), 7.8 (m, 2H), 8.1 (d, 1H), 8.3 (dd, 2H), 8.5 (d, 1H) ppm; 31P NMR (CDCl3, ꢅ,
161.9 MHz): 39.8 (s), 16.6 (d, 1J(103Rh, 31P) ˆ 163.5 Hz) ppm; IR (KBr): ꢄ ˆ 1162 (ꢄ(P=O)) cm
.
1
Hydroformylation of styrene
In a typical experiment, 0.005 mmol of [Rh(NBD)Cl]2 and 0.01 mmol of 1 were dissolved in 6 cm3
toluene in a Schlenk tube under Ar, and 2 cm3 styrene were added. The solution was transferred into a
100 cm3 autoclave and pressurized by a CO:H2 ˆ 1:1 mixture (80 bar). The reaction mixture was
thermostatted in an oil bath at the required temperature and agitated by a magnetic stirrer. At the end
of the reaction the mixture was cooled to room temperature, vented, and immediately analyzed.
Conversions as well as chemo-, regio-, and enantio selectivities were determined by means of GC
analysis (initial temperature: 60ꢁC, heating rate: 2ꢁC/min, ®nal temperature: 150ꢁC; retention times:
branched aldehyde 21 min (S) 21.5 min (R); linear aldehyde 26.7 min).
Catalytic hydroboration of styrene
Styrene (0.172 cm3, 1.5 mmol) was added to a solution of 2.9 mg 3c (0.003 mmol, 0.2 mol%) in 2 cm3
THF under Ar. The solution was stirred for 5 min; then, 0.16 cm3 catecholborane (1.5 mmol) was
added. The mixture was stirred at room temperature for 12 h and then quenched with 0.4 cm3 EtOH.
2 cm3 NaOH (2 M in H2O) and 0.2 cm3 H2O2 were added; the mixture was stirred for 10 h, extracted
with Et2O, washed (2 M NaOH, H2O, brine), and dried over Na2SO4. Conversions, regioselectivity
and ee were determined by GC analysis at 105ꢁC (retention times: 3.13 min substrate), 13.8 min ((R)-
1-phenylethanol), 14.8 min ((S)-1-phenylethanol), 14.1 min (2-phenylethanol)).
Asymmetric hydrogenation of methyl acetamidoacrylate
Methyl acetamidoacrylate (143 mg, 1 mmol) and 9.5 mg 3c (0.01 mmol) were placed in a pressure
bottle. The ¯ask was purged with N2, 10 cm3 solvent (benzene/methanol 1:1) were added by means of
a syringe through a septum, and the bottle was connected to the H2 reservoir of a Parr mid-pressure
apparatus. N2 was evacuated, and the vessel was purged twice with H2. Then H2 (2 bar) was applied,
and the reaction mixture was shaken for 16 h at room temperature. Conversion and ee of methyl
alaninate have been determined by GC analysis at 95ꢁC using He (60 kpa) as the carrier (retention
times: substrate, 13.7 min; (S)-methyl alaninate, 14.8 min; (R)-methyl alaninate, 19.1 min).
Acknowledgements
This work was carried out in the frame of the COST Chemistry Action D12 Organic Transformations:
Selective Processes and Asymmetric Catalysis, project No. 0009/98. Financial support from
University of Sassari and from MURST (Rome) is gratefully acknowledged by S. Gladiali.
References
[1] For a comprehensive review on the coordination chemistry of hemilabile ligands see: Slone CS,
Weinberger DA, Mirkin CA (1999) The Transition Metal Coordination Chemistry of Hemilabile
Ligands. In: Karlin KD (ed) Progress in Inorganic Chemistry, vol 48. Wiley, New York, p 233

Rh(I) Complexes with (S)-BINAPO
1361
[2] Bader A, Lindner E (1991) Coord Chem Rev 108: 27
[3] a) Onoda T (1993) Chemtech 34; b) Gipson RM (1976) US Patent 3954877 (CA 85, 77653c); c)
Abatjoglou AG (1983) EP Patent 73961 (CA 99, 53124u); d) Arimitsu S, Shikakura K (1990) JP
Patent 02174740 (CA 113, 39942f); e) Myazawa C, Mikami H, Orita S, Omori Y (1990) JP
Patent 0242040 (CA 113, 190750s)
[4] Abu-Gnim C, Amer I (1996) J Organomet Chem 516: 235
[5] Wegman R, Abatjoglu AG, Harrison AM (1987) J Chem Soc Chem Commun 1891
[6] Baker MJ, Giles MF, Orpen AG, Taylor MJ, Watt RJ (1995) J Chem Soc Chem Commun 197
[7] Bischoff S, Weigt A, Miessner H, LuÈcke B (1996) J Mol Catal A: Chem 107: 339
[8] Le Gall I, Laurent P, Soulier E, SalauÈn J-Y, des Abbayes H (1998) J Organomet Chem 567: 13
[9] Lindner E, Wang Q, Mayer HA, Bader A, KuÈhbauch H, Wegner P (1993) Organometallics 12:
3291
[10] Brown JM, Chaloner PA, Nicholson PN (1978) J Chem Soc Chem Commun 646
[11] Gladiali S, Pulacchini S, Fabbri D, Manassero M, Sansoni M (1998) Tetrahedron Asymm 9: 391
Á
[12] Gladiali S, Alberico E, Pulacchini S, Kollar L (1999) J Mol Catal 143: 155
Á
[13] Kollar L, Gladiali S, Tenorio MJ, Weissensteiner W (1998) J Clust Sci 9: 321
[14] Pertici P, D'Arata F, Rosini C (1996) J Organomet Chem 515: 163
Á
[15] Reviews on asymmetric hydroformylation: a) Gladiali S, Bayon JC, Claver C (1995) Tetrahedron
Asymm 6: 1453; b) Agbossou F, Carpentier JF, Mortreux A (1995) Chem Rev 95: 2485
[16] Nozaki K, Sakai N, Nanno T, Higashjima T, Mano S, Horiuchi T, Takaya H (1997) J Am Chem
Soc 119: 4413
[17] Sakai N, Nozaki K, Mashima K, Takaya H (1992) Tetrahedron Asymm 3: 583
Received June 16, 2000. Accepted (revised) July 24, 2000