Rhodium(I) macrocyclic and cage-like structures containing diphosphine bridging ligands

Article abstract of DOI:10.1007/s11243-016-0107-7

Abstract: Three series of rhodium organometallic complexes, mono-(1c, 2c, 5c, 6c), di-(1a-6a) and tetranuclear (1b, 5b, 6b), containing six different diphosphines 1,1′-bis(diphenylphosphino)methane or dppm (1), 1,2-bis(diphenylphosphino)ethane or dppe (2), 1,4-bis(diphenylphosphino)butane or dppb (3), bis(diphenylphosphino)acetylene or dppa (4), 1,2-bis(diphenylphosphino)benzene or dppbz (5) and 4,5-bis(diphenylphosphino)-9,9′-dimethylxanthene or xantphos (6) were successfully synthesised. These Rh(I) complexes were characterised by conventional techniques. The influence of the flexibility/rigidity of these P-donor ligands was carefully analysed, including their effect on both synthesis and catalysis. The luminescent properties of the dinuclear and tetranuclear complexes were investigated, and only those containing dppa, dppbz and xantphos displayed luminescence. Structures of the dinuclear complexes were modelled by using DFT methods in order to elucidate the most reasonable conformation. The different types of complexes were applied in the catalytic hydrogenation of (E)-4-phenylbut-3-en-2-one, showing high activity and similar catalytic behaviour. No cooperative effect could be inferred. Graphical Abstract: Three series of 1D, 2D and 3D rhodium complexes were successfully synthesised, and their emission properties were analysed. The large number of isomeric forms of the 2D were analysed by DFT methods. Their use in catalytic hydrogenation of (E)-4-phenylbut-3-en-2-one was studied showing high selectivity towards the formation of 4-phenylbutan-2-one.[Figure not available: see fulltext.]

Full text of DOI:10.1007/s11243-016-0107-7

Transit Met Chem
DOI 10.1007/s11243-016-0107-7
Rhodium(I) macrocyclic and cage-like structures containing
diphosphine bridging ligands
1
1
•
1
Julia Arcau Montserrat Ferrer1 Elisabet Aguilo Laura Rodrıguez
•
•
`
´
´
Received: 26 October 2016 / Accepted: 24 November 2016
Ó Springer International Publishing Switzerland 2016
Abstract Three series of rhodium organometallic com-
plexes, mono-(1c, 2c, 5c, 6c), di-(1a-6a) and tetranuclear
(1b, 5b, 6b), containing six different diphosphines 1,10-bis
(diphenylphosphino)methane or dppm (1), 1,2-bis(diphenylphos-
phino)ethane or dppe (2), 1,4-bis(diphenylphosphino)butane
or dppb (3), bis(diphenylphosphino)acetylene or dppa (4),
1,2-bis(diphenylphosphino)benzene or dppbz (5) and 4,5-
bis(diphenylphosphino)-9,90-dimethylxanthene or xantphos
(6) were successfully synthesised. These Rh(I) complexes
were characterised by conventional techniques. The influ-
ence of the flexibility/rigidity of these P-donor ligands was
carefully analysed, including their effect on both synthesis
and catalysis. The luminescent properties of the dinuclear
and tetranuclear complexes were investigated, and only
those containing dppa, dppbz and xantphos displayed lumi-
nescence. Structures of the dinuclear complexes were mod-
elled by using DFT methods in order to elucidate the most
reasonable conformation. The different types of complexes
were applied in the catalytic hydrogenation of (E)-4-
phenylbut-3-en-2-one, showing high activity and similar
catalytic behaviour. No cooperative effect could be inferred.
Introduction
The use of transition metal-based self-assembly processes
has emerged as one of the most efficient ways to build
artificial supramolecular structures [1–3]. Typically, a
metal complex with two or more available coordination
sites is reacted with an appropriately designed multifunc-
tional ligand to give the product in one step in high yield.
Square-planar Pd(II) and Pt(II) complexes have been
commonly used to prepare metallosupramolecules which
contain square or near orthogonal angles in their geometries
[1–8]. However, other metals have also been used in these
reactions. In particular, Rh(I) and Ir(I) units have been
explored in the synthesis of bi- [9–14] and three-dimensional
compounds [15–18], they are of great interest due to their
luminescent behaviour [19] and their potential catalytic
activity [20–25] and they could also be very useful in
molecular recognition studies [26, 27], among others.
In recent years, bimetallic complexes have been used in
homogeneous catalysis as an effective tool for promoting
organic transformations [28, 29]. Bimetallic catalysts can
display cooperative effects, because of the proximity
between the metal centres, leading to reactivities and
selectivities that cannot be attained using their corre-
sponding monometallic counterparts [30, 31].
Over the last few decades, bidentate ligands have been
widely employed to improve selectivities, activities and
stabilities of metal-complex-based homogeneous catalysts
[32–35]. The key to optimise the performance of a certain
species in a catalytic reaction is to adjust the electronic and
steric properties together with the topology exhibited by
the ligands.
Electronic supplementary material The online version of this
article (doi:10.1007/s11243-016-0107-7) contains supplementary
material, which is available to authorized users.
´
& Laura Rodrıguez
laura.rodriguez@qi.ub.es
The use of binucleating ligands with ligand isolated
donor sets is one of the more recent developments where
two metal centres are held in close proximity to
1
´
`
` ´
Departament de Quımica Inorganica i Organica, Seccio de
Quımica Inorganica, Universitat de Barcelona, Martı i
´
`
´
`
Franques 1-11, 08028 Barcelona, Spain
123

Transit Met Chem
cooperatively catalyse a given organic transformation [31].
Cooperative effects that have been observed in bimetallic
systems usually occur through electronic and/or spatial
interactions. In particular, concerning the reactivity of
group 9 bimetallic catalysts, it is important to underline the
work reported by Stanley and co-workers with a highly
selective and active dirhodium hydroformylation catalyst,
[Rh2(norbornadiene)(P4)], where P4 is the binucleating
tetraphosphine-based ligand, (Et2PCH2CH2)(Ph)PCH2-
P(Ph)(CH2CH2PEt2) [36]. Another example was shown by
Messerle and co-workers with a dual metal catalyst system
(Rh and Ir) using cationic Rh(I) and Ir(I) complexes of the
type [M(bpm)(CO)2]BArF4 (where M = Rh or Ir,
of different isomers and their relative stabilities as depicted
in Fig. 1. 4a displayed two set of signals with different
intensities, showing close chemical shifts (24.0 and
24.6 ppm) and the same coupling constant value (J(P-
Rh) = 124 Hz). This behaviour has been already reported
in the literature for closely related dinuclear rhodium
derivatives [44] and attributed to the existence of different
conformational isomers arising from the relative disposi-
tion of the phosphine backbones (chair/boat or parallel/
antiparallel in Figs. 2, 3) isomers and also from the relative
orientation of the carbonyl and chloride ligands in both
rhodium centres (syn and anti isomers in Figs. 2, 3).
Some broadening of the signals was only observed for
1a and could be due to a dynamic process that involves the
coordination–decoordination of the diphosphines, as pre-
viously observed for its analogue three-dimensional com-
plex 1b [40].
bpm = bis(1-pyrazolyl)methane,
and
BArF4 = te-
trakis[3,5-bis(trifluoromethyl)-phenyl]borate) which is
observed to be a highly efficient catalyst for the two step
intramolecular dihydroalkoxylation reaction of alkyne
diols in the formation of spiroketals [37]. The two metal
centres worked cooperatively to provide greater efficien-
cies than the individual monometallic complexes for the
concurrent cyclisation of alkyne diols.
The sharp and well defined doublet exhibited by 2a, 5a–
6a could be due to the fact that these diphosphines form
exclusively one complex in solution, or that several iso-
mers exist in a fast equilibrium at the temperature of study.
Van Leeuwen and co-workers claimed the involvement
of bimetallic catalytic species using wide bite angle
diphosphine ligands, in particular in carbonylation reac-
tions [38, 39].
Structure optimisation of the dinuclear complexes
As indicated above, a wide range of different conforma-
tions could be plausible for our face-to-face (trans-P)
complexes depending both on the syn/anti or cis/trans
(only for 4a) disposition of the chloride and carbonyl
ligands coordinated to the metal atoms and the relative
arrangement of the backbone chain of the diphosphine with
respect to an imaginary Rh–Rh edge (chair/boat for com-
plexes 1a and 4–6a or parallel/antiparallel for 2a and 3a).
In order to find out the different possible isomers and the
most stable in each case, we have performed theoretical
calculations at the DFT level, using the RB3LYP hybrid
functional (see ‘‘Experimental section’’). The optimised
structures are displayed in Figures S1–S6, and the calcu-
lated minimum energy in all cases is shown in Table 1.
Inspection of Table 1 shows a clear dependence of the
nature of the diphosphine upon the stability of the corre-
sponding isomers. However, the obtained values do not
allow establishing a direct correlation between length,
bulkiness and/or flexibility of the phosphine backbone
chain and the energetically favoured isomer. In spite of
this, a general preference for the anti disposition of the
chloride and carbonyl ligands is observed although in some
cases the difference between both conformations is too
small to be considered reliable (see for example 4a boat
and 6a chair).
For all the reasons above, we decided to go on with our
previous work based on the synthesis of bi- and three-
dimensional rhodium structures containing diphosphine
ligands as linkers between metal atoms [40]. Their catalytic
behaviour in hydrogenation reactions was analysed, and the
possible cooperative effects on these processes was studied
and compared with the parent mononuclear rhodium
complexes.
Results and discussion
Synthesis of rhodium dinuclear complexes
[RhCl(CO)2]2 was reacted with six different diphosphines
1–6 (Rh:P = 1:2) in dichloromethane at room temperature
for 1 h (Scheme 1). The reaction was monitored by 31P-
NMR until the disappearance of the free diphosphine sig-
nal, which was ca. 40–60 ppm downfield shifted upon
coordination to the metal atom. The corresponding dinu-
clear complexes [RhCl(CO)]2(l-diphos)2 (1a–6a) were
obtained in high yields after concentration of the solution
and precipitation with hexane.
Complexes 1a and 3a were previously described fol-
lowing different synthetic procedures [41–43]. Evidence
for the proposed structures was obtained from IR, 1H-NMR
and 31P-NMR spectroscopy and MS spectrometry. 31P-
NMR spectra present different profiles due to the existence
The rhodium atom presents, in all cases, a square-planar
geometry and the inner cavities are observed to be in the
˚
range ca. 3.5 9 6.3 A for 1a, 5.1 9 5.6 A for 2a,
˚
˚
7.2 9 5.4 A for 3a, 3.0 9 5.1 A for 5a and 4.1 9 7.1 A
˚
˚
123

Transit Met Chem
Scheme 1 Synthesis of
dinuclear complexes 1a–6a
Ph
P
Ph
P
CO
Rh
Cl
Ph
Ph
Ph
OC
Cl
Cl
CO
CO
CH2Cl2
rt, 1h
Rh
+ 2 diphos
Rh
X
OC
X
CO
Rh
Cl
P
P
Ph
Ph
Ph
diphos = dppm
dppe
1a; X = CH2
2a; X = (CH2)2
3a; X = (CH2)4
dppb
dppa
dppbz
4a; X = C
5a; X = C6H4
C
xantphos 6a; X = xanthene
Fig. 1 31P-NMR spectra of 1a,
4a and 6a in CDCl3
6a
4a
1a
for 6a for all possible conformations. A particular case is
4a whose cavity presents a completely different size in the
˚
case of chair conformation (3.4 9 7.5 A) and boat
1 h. Actually compounds 1b [40], 5b and 6b were effi-
ciently isolated, while the use of the other diphosphines
(dppe, dppb, dppa) did not give rise to the formation of the
desired products. The high rigidity of the dppa probably
prevents the cage-type structure and the high flexibility of
dppe and dppb favours the formation of oligomers.
31P-NMR spectra of 5b and 6b complexes exhibit one
broad doublet centred at 20–60 ppm with a Rh–P coupling
constant of ca. 135 Hz. The value of these coupling con-
stants is in the same range as that previously reported for
1b [40].
˚
(5.3 9 5.4 A). Moreover, the only square-like cavity is the
one corresponding to the acetylide derivative 4a, in boat
conformation.
Synthesis of rhodium tetranuclear complexes
Following the same experimental procedure previously
reported by some of us [40] (Scheme 2), we planned the
synthesis of different cage-like rhodium structures. Our
main goal was the formation of new three-dimensional
complexes with different inner size cavities in order to
explore their influence on different applications, such as
catalysis.
IR spectra show more complex m(C:O) stretching
vibration patterns with respect to the two IR bands
observed for the precursor, [RhCl(CO)2]2(l-4,40-bipy)
(Figures S7 and S8) [40]. Each band splits into two with an
additional lower energy vibration. This shift is in agree-
ment with a higher retro-donation effect from the rhodium
centre to the antibonding molecular orbital of the carbonyl
[RhCl(CO)2]2(l-4,40-bipy) was reacted with the corre-
sponding diphosphine (Rh:P = 1:2) in dichloromethane for
123

Transit Met Chem
Fig. 2 Different possible conformers expected for 1a–3a complexes
ligand because of the loss of one CO ligand upon diphos-
phine coordination. Maldi/TOF mass analyses evidenced
peaks consistent with the tetranuclear structure.
charge-transfer character stems from interactions between
filled metal d orbitals and antibonding ligand p* orbitals.
The MLCT transition is observed to be less favoured for
the tetrametallic derivatives, being only clearly defined for
the dppbz derivative, 5b, probably due to the higher aro-
maticity and electron deficiency of the system due to the
presence of the benzene unit.
Photophysical characterisation
Absorption and emission spectra of complexes 1a, 2a, 4a–
6a, 1b, 5b and 6b were recorded at 1 9 10-5 M in
dichloromethane. The results are summarised in Table 2;
Figs. 4 and 5.
The only emissive dinuclear species are 4a and 6a (dppa
and xantphos derivatives, Fig. 6). Excitation spectra col-
lected at the emission maxima match the lowest energy
absorption band being indicative of an intraligand emission
origin. It should be noticed that, as displayed in Fig. 6, the
emission spectrum of 4a presents two different bands. The
vibronically well-resolved highest energy emission band is
in agreement with the involvement of the acetylene group on
this transition. The broad structureless lowest energy band
can be tentatively attributed to metal centred transitions,
since dr* ? pr excited states are common for dinuclear
Rh(I) complexes coordinated by bidentate ligands [47].
The spectra of both bimetallic and tetrametallic com-
plexes are dominated by strong absorption bands attributed
to intraligand (IL) p ? p* transitions, with diphosphine-
based p ? p* transitions in the near-UV region between
280 and 300 nm along with a relatively weak, broad band
in the visible range (between 350 and 450 nm). The broad
visible bands are similar to the metal-to-ligand charge-
transfer (MLCT) transitions observed in rhodium com-
plexes containing p-acceptor ligands [45, 46], in which the
123

Transit Met Chem
Fig. 3 Different possible conformers expected for 4a–6a complexes
Compounds 1b and 6b (dppm and xantphos derivatives)
displayed a broad luminescence at ca. 350 nm upon exci-
tation the samples at the intraligand absorption band
(Fig. 7). The small Stokes’ shift (ca. 5000 cm-1) suggests
a singlet emissive IL origin state. An additional lower
energy emission is observed for 6b when the sample is
excited at the MLCT transition being indicative of the
different origin transitions. Excitation spectra collected at
the maxima are in agreement with these assignments
(Figure S9–S10). No emission was displayed by complex
5b.
The corresponding reaction with dppb gave rise to the
formation of different species in equilibrium whose 31P-
NMR signals showed similar chemical shifts, preventing
the isolation of pure 3c. Moreover, the rigidity of the
alkynyl group of the diphosphine dppa prevents the
diphosphine from acting as a chelating ligand and conse-
quently precludes the formation of the mononuclear com-
plex 4c [45, 49].
IR spectra of the isolated complexes, in particular the
region corresponding to v(C:O) absorptions, revealed the
expected cis-CO arrangement in agreement with the
reported data [38, 48].
Synthesis of rhodium mononuclear complexes
Catalytic hydrogenation
Mononuclear derivatives were also synthesised in order to
compare the catalytic behaviour of the different complexes.
For this, the reaction of [RhCl(CO)2]2 with the appropriate
diphosphine in stoichiometric conditions (after extraction
of the chloride ligand by addition of silver triflate) was
applied to give the monomeric species 1c, 2c, 5c and 6c
(Scheme 3).
The dinuclear complexes 1a, 2a and 4a–6a were used as
catalytic precursors in the hydrogenation reaction of the
polyfunctional substrate (E)-4-phenylbut-3-en-2-one, in
order to analyse their ability towards the reduction of the
different groups present on the same substrate. 3a was not
used for this purpose since it was observed to decompose
123

Transit Met Chem
dihydrogen for 3 h of reaction (entry 6). Using the most
active catalytic system, 6a, under hasher conditions (at
80 °C or 40 bar H2), no differences in the reactivity were
observed (entry 5 vs. 7 and 8). With a higher load of cat-
alyst and longer time, only A and B were obtained in a ratio
of 77:23 (entry 9). The highest amount of B was obtained
under 40 bar H2 for 24 h (entry 10).
Table 1 Calculated minimum energies for the different isomers of
1a–6a compounds
Compound
Energy (kcal/mol)
1a chair
syn
16.7
syn
4.5
syn
6.4
syn
10.7
syn
0.0
syn
6.2
cis
anti
0.0
1a boat
anti
13.9
anti
14.3
anti
0.0
We also tested the tetranuclear complexes 1b and 6b.
Xanthphos-based complex (6b) was slightly more active
than 1b, giving 92% of A and 8% of C, while 1b led
exclusively to the formation of A. Both catalytic trends are
similar to those obtained with the corresponding dinuclear
complexes, 1a (under smooth conditions) and 6a, respec-
tively, (entries 5–6 vs. 11–12), pointing to a lower activity
of the tetranuclear systems in comparison with the dinu-
clear ones.
2a antiparallel
2a parallel
3a antiparallel
3a parallel
4a chair
anti
12.1
anti
7.3
trans
11.6
anti
0.0
The monometallic complex 2c used as catalytic pre-
cursor, gave the same reactivity trend compared to the
related bimetallic system 2a (entry 13 vs. 2), without
observing any influence concerning the presence of the
chloride anion, present in 2a [51]. The similar reactivity
exhibited by the dinuclear complexes 1a, 2a, 5a and 6a and
the monometallic complex 2c, together with the lower
activity of the tetranuclear complexes 1b and 6b, seems to
suggest that the active catalytic species are mononuclear
complexes generated in situ under catalytic conditions
from the corresponding dinuclear ones.
8.2
syn
0.9
syn
14.3
syn
3.4
syn
1.8
syn
11.6
4a boat
5a chair
anti
1.7
5a boat
anti
0.0
6a chair
anti
0.0
6a boat
anti
10.7
The most stable isomer (considered to have an energy of 0.0 kcal/-
mol) is highlighted in bold
Conclusions
The reaction of six different diphosphines with
Rh(I) organometallic precursors led to the formation of five
new dinuclear (1a, 2a, 4a–6a) and two tetranuclear (5b,
6b) complexes. The use of the highly flexible dppb ligand
precluded the formation of the complexes in pure form and
rapid decomposition was observed in solution. The use of
more rigid diphosphines (dppbz and xantphos) favoured the
formation of three-dimensional cage-like derivatives.
The compounds displayed very weak emissions in
dichloromethane with some intraligand luminescence
rapidly in solution. Under the conditions indicated in
Scheme 4, we could not observe important differences
among the different Rh(I) complexes leading to full con-
version (entries 1–5 in Table 3), except for 4a which was
less active (42% conversion, entry 3), probably due to an
induction period corresponding to the hydrogenation of the
C:C triple bond present in the ligand skeleton, leading to
the same catalytic system generated by 2a. As expected
[50], the hydrogenation mainly gave 4-phenylbutan-2-one
(A in Scheme 4) in all the cases, being 6a the most active
system (able to hydrogenate the aromatic ring), which gave
up to 14% of 4-cyclohexylbutan-2-one (entry 5), certainly
formed from the hydrogenation of the non-conjugated
compound A.
3
observed for 4a, 6a, 1b and 6b and an additional MLCT
band for 6b.
Theoretical calculations at density functional level
(DFT) carried out to know the most stable conformation of
the dinuclear complexes have shown a clear influence of
the backbone of the diphosphine on the stability of the
different conformers.
The system 1a was the only one giving butylcyclohex-
ane (E in Scheme 4) as by-product coming from the
hydrogenative dehydration process of alcohol D.
Mono-, di- and tetrametallic complexes were assayed as
catalytic precursors for the hydrogenation of (E)-4-
phenylbut-3-en-2-one. The similar behaviour observed
between polynuclear and monometallic complexes sug-
gests that the active catalytic species are mononuclear
In order to selectively attain one hydrogenated product,
milder conditions were applied. Actually, 4-phenylbutan-2-
one, A, was exclusively obtained under 5 bar of
123

Transit Met Chem
Scheme 2 Synthesis of
tetranuclear complexes 1b, 5b
and 6b
CO
Cl
OC Rh
CO
CH2Cl2
rt, 1h
2
+ 4 diphos
N
N
Rh CO
Cl
Ph
Ph
Ph
Cl
Cl
Rh
Ph
X
P
P
N
N
Rh
Ph
Ph
Ph
CO
P
P
CO
P
P
Ph
+
4
CO
X
X
X
Ph
Ph
Ph
Ph
CO
CO
Rh
Cl
N
N
Rh
Cl
Ph
P
P
Ph
Ph
Ph
diphos = dppm
dppbz
1b; X = CH2
5b; X = C6H4
xantphos 6b; X = xanthene
Table 2 Absorption and
emission wavelengths (nm) of
complexes 1a–6a
Compounds
Absorption (e 9 10-3, M-1 cm-1
)
Emission (CH2Cl2, 298 K)
1a
2a
3a
4a
5a
6a
1b
5b
6b
a
443 (5.8), 324 (6.1), 263 (27.2)
390 (5.6), 343 (7.1), 275 (36.1)
375 (3.21), 328 (5.4), 254 (31.2)
270 (31.3)
–
(kexc = 280 nm) and 1b–6b
(kexc = 274 nm) at
–
1 9 10-5 M concentration
–
340, 410
428 (4.1), 325 (12.5), 272 (37.6)
331 (5.0), 283 (24.0)
–
335
335sh (7.1), 285 (45.9)
335
428 (4.6), 325 (14.3), 265 (58.6)
335sh (12.9), 280 (37.8)
–
335, 415a
kexc = 333 nm
moieties generated under the catalytic conditions used in methane (dppm) [52], 1,2-bis(diphenylphosphino)ethane
these studies.
(dppe) [52] 1,4-bis(diphenylphosphino)butane (dppb) [52],
[RhCl(CO)]2(l-dppm)2 [53] and ([RhCl(CO)]2(l-4,40-bipy))2
(l-dppm)4 (1b) [40] were synthesised as described previously.
The compounds 4,40-bipyridine (Avocado, 98%), trans-4-phenyl-
3-butene-2-one (Janssen Chimica, 99%), 1,2-bis(diphenylphos-
phino)-acetylene (dppa), 4,5-bis(diphenylphosphino)-9,90-dime-
thylxanthene (xantphos), 1,2-bis(diphenylphosphino)benzene
Experimental section
General
All manipulations were performed under prepurified N2 using (dppbz) were used as received.
standard Schlenk techniques. All solvents were distilled from
Infrared spectra were recorded on an FTIR 520 Nicolet
appropriate dryingagents. The complexes [RhCl(CO)2]2 [49], spectrophotometer. 31P{1H}-NMR [d(85% H3PO4) = 0.0
[RhCl(CO)2]2(l-bipy) [40], 1,10-bis(diphenylphosphino) ppm] and 1H-NMR [d(TMS) = 0.0 ppm] spectra were
123

Transit Met Chem
9.0x104
6.0x104
3.0x104
0.0
1b
6b(λexc = 333nm)
6b(λexc = 274 nm)
6
4.0x10
1a
2a
3a
4a
5a
6a
6
2.0x10
0.0
300
400
500
600
300
400
500
Wavelength (nm)
Wavelength (nm)
Fig. 4 Absorption spectra of the dinuclear complexes (1a, 2a, 4a–6a)
at 1 9 10-5 M in dichloromethane
Fig. 7 Emission spectra of 1b (solid line) and 6b in dichloromethane
solution (dashed and dotted line)
1.0x105
obtained on a Bruker DXR 250, Varian Inova 300 and
Varian Mercury 400 spectrometers. ES(?) mass spectra
were recorded on a Fisons VG Quatro spectrometer.
Absorption spectra were recorded on a Varian Cary 100
Bio UV-spectrophotometer and emission spectra on a
Horiba–Jobin–Yvon SPEX Nanolog spectrofluorimeter.
GC analyses were carried out on an Agilent GC6890 with a
flame ionisation detector, using an SGE BPX5 column
composed by 5% of phenylmethylsiloxane. ES(?) mass
spectra were recorded with a Fisons VG Quatro spec-
trometer. Maldi/TOF spectra were recorded on a Maldi Tof
Voyager DE STR (Applied Biosystems) instrument.
Molecular modelling was carried out using Spartan’10
V1.1.0 for Mac as software. DFT calculations were carried
out with the Spartan software using the B3LYP hybrid
functional [54, 55]. The basis set was chosen as follows:
for Rh LANL2DZ was used [56, 57] while for the
remaining atoms the 6–31G basis [58] was used. The
structures were previously optimised at MM? molecular
mechanics level.
1b
5b
6b
5.0x104
0.0
300
400
500
600
Wavelength (nm)
Fig. 5 Absorption spectra of 1b, 5b and 6b complexes at
1 9 10-5 M in dichloromethane
4a
6a
4.0x106
Synthesis and characterisation
Synthesis of [RhCl(CO)]2(l-dppm)2 (1a) Similar proce-
dure previously described in the literature [53] was fol-
lowed, but dichloromethane instead of cyclohexane was
used as solvent. Yield: 74%. 31P-NMR (121.42 MHz,
298 K, CDCl3): 19.6 ppm (d, 1J(Rh-P) = 115 Hz). 1H-
NMR (300 MHz, 298 K, CDCl3): 7.67–7.27 (m, 40H, Ph),
3.96 (br, 4H, P–CH2–P). ESI–MS(?) m/z: 1102.9
(M ? H?, calc.: 1102.5), 1045.0 (M - 2CO ? H?, calc.:
1046.1), 1010.0 (M - 2CO - Cl-, calc.: 1010.0). IR
(KBr, cm-1) 1970s, m(C:O).
2.0x106
0.0
300
400
500
Wavelength (nm)
Fig. 6 Emission spectra of 4a (dashed line) and 6a (solid line) in
dichloromethane solution (dashed line) upon excitation the samples at
280 nm
Synthesis of [RhCl(CO)]2(l-dppe)2 (2a) A dichlor-
omethane solution (5 mL) of dppe (103 mg, 0.26 mmol)
123

Transit Met Chem
Scheme 3 Synthesis of
monometallic complexes 1c, 2c,
5c and 6c
Scheme 4 Hydrogenation of
(E)-4-phenylbut-3-en-2-one
catalysed by Rh(I) complexes
Table 3 Hydrogenation of (E)-4-phenylbut-3-en-2-one catalysed by
polynuclear Rh complexes
concentrated to ca. 4 mL and hexane (20 mL) was added.
The obtained pale orange solid was filtered off, washed
with hexane and vacuum dried. Yield: 75% (109 mg). 31P-
Entry
Complex
Selectivity (%)a
A/B/C/D/E
1
NMR (121.42 MHz, 298 K, CDCl3): 57.5 ppm (d, J(Rh-
P) = 101 Hz). 1H-NMR (300 MHz, 298 K, CDCl3):
8.78–7.38 (m, 40H, Ph), 2.38 (br, 4H, P–(CHHa)2)-P), 2.27
(br, 4H, P–(CHbH)2)-P). ESI–MS(?) m/z: 501.0
(M - 2CO - 2Cl-, calc.: 501.3). IR (KBr, cm-1): 1986
versus, m(C:O).
1
1a
2a
4a
5a
6a
1a
6a
6a
6a
6a
1b
6b
2c
87/4/7/–/2
85/8/7/–/–
95/5/–/–/–
88/5/7/–/–
82/14/4/–/–
100/–/–/–/–
85/13/2/–/–
86/12/2/–/–
77/23/–/–/–
48/49/3/–/–
100/–/–/–/–
92/8/–/–/–
85/9/6/–/–
2
3b
4
5
6c
7d
8e
9f
10g
11
12
13
Synthesis of [RhCl(CO)]2(l-dppa)2 (4a) A similar pro-
cedure that described for 2a was used in the synthesis of
4a, but dppa instead of dppe was used to obtain a pale
yellow solid. Yield: 70%. 31P-NMR (121.42 MHz, 298 K,
CDCl3): 3.6 ppm (d, 1J(Rh-P) = 133 Hz). 1H-NMR
(300 MHz, 298 K, CDCl3): 7.87–7.25 (m, Ph). ESI–
MS(?) m/z: 497.0 (M - 2CO - 2Cl-, calc.: 497.2). IR
(KBr, cm-1): 2105 m, m(C:C); 1986 versus, m(C:O).
Conditions: 1 mmol of (E)-4-phenylbut-3-en-2-one, 0.01 mmol of
Rh, 20 bar H2, 20 mL of ethanol, 50 °C, 2 h. Full conversion for all
the systems except for 4a
Synthesis of [RhCl(CO)]2(l-dppbz)2 (5a) Similar proce-
dure described for 2a was used in the synthesis of 5a using
dppbz instead of dppe to obtain a pale yellow solid. Yield:
70%. 31P-NMR (121.42 MHz, 298 K, CDCl3): 62.3 ppm
(d, 1J(Rh-P) = 134 Hz). 1H-NMR (300 MHz, 298 K,
CDCl3): 7.59–6.83 (m, Ph ? P - C6H4 - P). ESI–MS(?)
m/z: 995.0 ([M–3 Ph]?, calc.: 994.9). IR (KBr, cm-1):
1973 versus, m(C:O).
a
b
c
d
e
f
1
Determined by GC and H NMR using decane as internal standard
42% conversion
Under 5 bar H2 for 3 h
At 80 °C
Under 40 bar H2
0.02 mmol of Rh for 24 h
Under 40 bar H2 for 24 h
g
Synthesis of [RhCl(CO)]2(l-xantphos)2 (6a) Similar pro-
cedure described for 2a was used in the synthesis of 6a by
using xantphos instead of dppe to obtain an orange solid.
Yield: 80%. 31P-NMR (121.42 MHz, 298 K, CDCl3):
21.1 ppm (d, 1J(Rh-P) = 131 Hz). 1H-NMR (300 MHz,
298 K, CDCl3): 7.59–7.07 (m, 56H, Ph ? HArom,Xanthene),
was added to a solution (5 mL) of [RhCl(CO)2]2 (50 mg,
0.13 mmol) in the same solvent. After 1 h of stirring the
bright orange solution became dark red. The solution was
123

Transit Met Chem
1.85–1.55 (m, 12H, CH3). ESI–MS(?) m/z: 1491.0
(M ? H?, calc.: 1490.9), 1426.1 (M - CO - Cl-, calc.:
1426.6), 710.0 (M - 2Cl-, calc.: 709.6). IR (KBr, cm-1):
1986 versus, m(C:O).
Synthesis of [Rh(CO)2(dppe)](OTf) (2c) Similar proce-
dure described for 1c was used in the synthesis of 2c by
using dppe instead of dppm to obtain a pale yellow solid.
Yield: 92%. 31P-NMR (121.42 MHz, 298 K, CDCl3):
57.6 ppm (d, 1J(Rh-P) = 132 Hz). 1H-NMR (300 MHz,
298 K, CDCl3): 7.33–7.11 (m, 20H, Ph), 2.11 (br, 4H, P–
CH2CH2–P). ESI–MS(?): 453.2 (M?–CO–Ph, calc.:
453.2). IR (KBr, cm-1) 1972, m(C:O); 1101, 1031
(OTf-).
Synthesis of ([RhCl(CO)]2(l-4,40-bipy))2(l-dppbz)4 (5b)
dichloromethane solution (5 mL) of dppbz (37 mg,
0.08 mmol) was added to solution (5 mL) of
A
a
[RhCl(CO)2]2(l-bipy) (22 mg, 0.04 mmol) in the same
solvent. After 1 h of stirring, the orange solution turned
bright yellow. The solution was concentrated to ca. 4 mL and
hexane (15 mL) was added. The obtained pale orange solid
was filtered off, washed with hexane and vacuum dried.
Yield: 80% (44 mg). 31P-NMR (121.42 MHz, 298 K, ace-
tone-d6): 61.7 ppm (d, 1J(Rh-P) = 133 Hz). 1H-NMR
(300 MHz, 298 K, acetone-d6): 8.90 (br, 8H, Ha-pyr),
8.05–7.02 (m, 104H, Ph ? P - C6H4–P ? Hb-pyr). Maldi/
TOF m/z: 2131.1 (M - dppz - 4CO - 2Cl ? H?, calc.:
Synthesis of [Rh(CO)2(dppbz)](OTf) (5c) Similar proce-
dure described for 1c was used in the synthesis of 5c by
using dppbz instead of dppm to obtain a pale yellow solid.
Yield: 94%. 31P-NMR (121.42 MHz, 298 K, CDCl3):
62.4 ppm (d, 1J(Rh-P) = 134 Hz). 1H-NMR (300 MHz,
298 K, CDCl3): 7.38–6.75 (m, 24H, Ph). ESI–MS(?):
501.3 (M?–CO–Ph, calc.: 501.3). IR (KBr, cm-1) 1970,
m(C:O); 1103, 1032 (OTf-).
2131.0),
1047.0
([RhCl(CO)]2(l-dppbz)2 - 2-
Synthesis of [Rh(CO)2(xantphos)](OTf) (6c) Similar pro-
cedure described for 1c was used in the synthesis of 6c by
using xantphos instead of dppm to obtain a pale yellow solid.
Yield: 90%. 31P-NMR (121.42 MHz, 298 K, CDCl3):
36.7 ppm (d, 1J(Rh-P) = 123 Hz). 1H-NMR (300 MHz,
298 K, CDCl3): 7.83–7.42 (m, 26H, Ph), 1.74 (s, 6H, CH3).
Ph - Cl ? CH3CN, calc.: 1047.1). IR (KBr, cm-1): 2097,
2059 s, 2012 versus, 1978 versus, m(C:O); 1612 s, m(C=N).
0
Synthesis of ([RhCl(CO)]2(l-4,4 -bipy))2(l-xantphos)4 (6b) Simi-
lar procedure described for 5b was used in the synthesis of 6b
by using xantphos instead of dppbz. Yield: 85%. 31P-NMR
(121.42 MHz, 298 K, acetone-d6): 20.9 ppm (d, 1J(Rh-
P) = 137 Hz). 1H-NMR (300 MHz, 298 K, acetone-d6):
8.94 (br, 8H, Ha-pyr), 8.08 (m, 8H, Hb-pyr), 7.92–6.83 (m,
112H, Ph ? HArom,Xanthene), 2.30 (m, 24H, (CH3)2,Xanthene).
Maldi/TOF m/z: 1599.2 (M - 2Cl- - CO, calc.: 1599.1),
1586.2 (M - 2Cl- - 2CO, calc.: 1586.10), 1572.2
(M - 2Cl- - 3CO, calc.: 1572.1), 1557.2 (M - 2Cl-
- 4CO, calc.: 1557.1), 1543.3 (M - 2Cl- - 4CO - 2
CH3, calc.: 1543.2), 1525.2 (M - 2Cl- - 4CO - 4 CH3,
calc.: 1525.2), 1510.2 (M - 2 Cl- - 4 CO - 6 CH3, calc.:
1510.2), 1373.9 (M - xantphos - 2 Cl- ? CH2Cl2 ?
ESI–MS(?): 710.1 (M?–CO, calc.: 710.1). IR (KBr, cm-1
1974, m(C:O); 1101, 1029 (OTf-).
)
Catalytic hydrogenation procedure
(E)-4-phenyl-but-3-en-2-one (146 mg, 1.0 mmol) was
added to a stirred solution of Rh catalyst (0.01 mmol of
rhodium: 6 mg for 1a, 2a, 4a and 5a; 7.5 mg for 6a; 6 mg
for 2c; 6.3 mg for 1b) in ethanol (20 mL) in the presence
of decane (142 mg, 1.0 mmol) as internal standard. The
solution was then pressurised with H2 (between 5 and
40 bar) at 50 °C during 6 h. The crude solution was filtered
off through Celite and analysed by GC/MS.
H2O, calc.: 1373.8), 1353.1 (M - xantphos - 2 Cl-
?
MeOH ? 2 H2O, calc.: 1353.1), 1322.0 (M - xant-
phos - 2 Cl-, calc.: 1322.0). IR (KBr, cm-1): 2099, 2012 s,
1986 m, 1966 w, m(C:O); 1612 m, m(C=N).
Acknowledgements Authors would like to thank Dr. Isabelle Favier
´
and Prof. Montserrat Gomez from the Centre National de la
Recherche Scientifique (CNRS) and the Universite de Toulouse 3—
´
Synthesis of [Rh(CO)2(dppm)](OTf) (1c) Solid AgOTf
(37 mg, 0.14 mmol) was added to a stirred solution of
[Rh(l-Cl)(CO)2]2 (25 mg, 0.06 mmol) and dppm (49 mg,
0.13 mmol) in CH2Cl2 (20 mL). After 15 min stirring,
AgCl was eliminated by filtration through Celite, the
solution was concentrated and hexane (50 mL) was added
and a yellow–orange solid was obtained. Yield: 90%
(75 mg). 31P-NMR (121.42 MHz, 298 K, CDCl3):
19.4 ppm (d, 1J(Rh-P) = 114 Hz). 1H-NMR (300 MHz,
298 K, CDCl3): 7.70–7.11 (m, 20H, Ph), 4.08 (br, 1H, P–
CHHa–P), 3.57 (br, 1H, P–CHbH–P). ESI–MS(?) m/z:
515.3 (M? - CO, calc.: 515.0). IR (KBr, cm-1) 1970,
m(C:O); 1101, 1030 (OTf-).
Paul Sabatier in Toulouse (France) for the catalytic experiments.
´
Financial support from the Ministerio de Economıa y Competitividad
of Spain (Projects CTQ2012-31335 and CTQ2015-65040-P) are also
acknowledged.
References
1. Cook TR, Zheng YR, Stang PJ (2013) Chem Rev 113:734
2. Cook TR, Zheng YR, Stang PJ (2015) Chem Rev 115:7001
3. Smulders MMJ, Riddell IA, Browne C, Nitschke JR (2013) Chem
Soc Rev 42:1728
´
4. Ferrer M, Pedrosa A, Rodrıguez L, Rossell O, Vilaseca M (2010)
Inorg Chem 49:9438
123

Transit Met Chem
´
5. Angurell I, Ferrer M, Gutierrez A, Martınez M, Rocamora M,
Rodrıguez L, Rossell O, Lorenz Y, Engeser M (2014) Chem Eur J
´
29. Wauchwalter P, Rose J, Braunstein P (2015) Chem Rev 115:28
30. Ho JHH, Wagler J, Willis AC, Messerle BA (2011) Dalton Trans
40:11031
´
20:14473
6. Ferrer M, Gutierrez A, Rodrıguez L, Rossell O, Ruiz E, Engeser
´
M, Lorenz Y, Schilling R, Gomez-Sal P, Martın A (2012)
´
´
31. Bratko I, Gomez M (2013) Dalton Trans 42:10664
´
´
32. Terrade FG, Lutz M, Reek JNH (2013) Chem Eur J 19:10458
33. Kokan Z, Kirin SI (2013) Eur J Org Chem 36:8154–8161
Organometallics 31:1533
´
C, Quintela JM (2013) Chem Eur J 19:15329
´
´
7. Alvarin˜o C, Pıa E, Garcıa MD, Blanco V, Fernandez A, Peinador
´
´
34. Sanhes D, Gual A, Castillon S, Claver C, Gomez M, Teuma E
(2009) Tetrahedron Asymmetry 20:1009
35. Bravo MJ, Ceder RM, Muller G, Rocamora M (2013) Organo-
metallics 32:2632
´
8. Ferrer M, Gomez-Bautista D, Gutierrez A, Ordun˜a-Marco G, Oro
LA, Perez-Torrente JJ, Rossell O, Ruiz E (2016) Organometallics
´
´
35:336
9. Ferrer M, Gomez-Bautista D, Gutierrez A, Miranda JR, Ordun˜a-
36. Matthews RC, Howell DK, Peng WJ, Train SG, Treleaven WD,
Stanley GG (1996) Angew Chem Int Ed Engl 35:2253
37. Ho JHH, Hodgson R, Wagler J, Messerle BA (2010) Dalton
Trans 39:4062
´
´
Marco G, Oro LA, Perez-Torrente JJ, Rossell O, Garcıa-Ordun˜a
´
P, Lahoz FJ (2014) Inorg Chem 53:1699
´
´
´
10. Fornies J, Gomez J, Lalinde E, Moreno MT (2004) Chem Eur J
38. Freixa Z, Kamer PCJ, Lutz M, Spek AL, van Leeuwen PWNM
(2005) Angew Chem Int Ed 44:4385
39. Feliz M, Freixa Z, van Leeuwen PWNM, Bo C (2005) Organo-
metallics 24:2608
10:888
11. Wang H, Guo XQ, Zhong R, Lin YJ, Zhang PC, Hou XF (2009) J
Organomet Chem 694:3362
´
12. Han YF, Lin YJ, Jia WG, Jin GX (2009) Dalton Trans
2009:2077–2080
40. Rodrıguez L, Ferrer M, Rossell O, Coco S (2009) J Organomet
Chem 694:3951
´
13. Alvarez-Vergara MC, Casado MA, Martın ML, Lahoz FJ, Oro
´
LA, Perez-Torrente JJ (2005) Organometallics 24:5929
41. Lamb G, Clarke M, Slawin AMZ, Williams B, Key L (2007)
Dalton Trans 47:5582–5589
´
14. Wang GL, Lin YJ, Jin GX (2011) Chem Eur J 17:5578
15. Han YF, Jia WG, Lin YJ, Jin GX (2008) J Organomet Chem
693:546
42. Lamb GW, Clarke ML, Slawin AMZ, Williams B (2008) Dalton
Trans 36:4946–4950
43. Cowie M, Dwight SK (1980) Inorg Chem 19:209
´
16. Grote Z, Bonazzi S, Scopelliti R, Severin K (2006) J Am Chem
Soc 128:10328
17. Han YF, Lin YJ, Weng LH, Berke H, Jin GX (2008) Chem
Commun 3:350–352
18. Govindaswamy P, Lindner D, Lacour J, Su¨ss-Fink G, Therrien B
(2007) Dalton Trans (39):4457–4463
19. Tran NT, Stork JR, Pham D, Chancellor CJ, Olmstead MM,
Fettinger JC, Balch AL (2007) Inorg Chem 48:7998
44. Lopez-Vallbuena JM, Escudero-Adan EC, Benet-Buchholz J,
Freixa Z, van Leeuwen PWNM (2010) Dalton Trans 39:8560
45. de Pater BC, Fru¨hous HW, Vrieze K, de Gelder R, Baerends EJ,
McCormack D, Lutz M, Spek AL, Hartl F (2004) Eur J Inorg
Chem 8:1675–1686
46. Kunkely H, Vogler A (1999) J Organomet Chem 577:358
47. Indoli MT, Chiorboli C, Sandola F (2007) Top Curr Chem
280:215
´
´
´
20. Escarcega-Bobadilla MV, Rodrıguez-Perez L, Teuma E, Serp P,
48. Youssef TE (2010) Polyhedron 29:1776
´ ´
49. Rojas S, Murcia-Mascaros S, Terrero P, Garcıa Fierro JL (2001)
New J Chem 25:1430
50. Chahdoura F, Pradel C, Gomez M (2013) Adv Synth Catal
355:3648
´
´
Masdeu-Bulto AM, Gomez M (2011) Catal Lett 141:808
21. Giustra ZX, Ishibashi JSA, Li SY (2016) Coord Chem Rev
314:134
´
22. Delgado-Rebollo M, Prieto A, Perez PJ (2014) Chem Cat Chem
6:2047
23. Sarmentero MA, Fernandez-Perez H, Zuidema E, Bo C, Vidal-
Ferran A, Ballester P (2010) Angew Chem Int Ed 49:7489
24. Besset T, Norman DW, Reek JNK (2013) Adv Synth Catal
355:348
´
51. Guyonnet Bile E, Sassine R, Denicourt-Nowicki A, Launay F,
Roucoux A (2011) Dalton Trans 40:6524
´
´
52. Hewertso W, Watson HR (1962) J Chem Soc 1490–1494
53. Mague J, Mitchener JP (1969) Inorg Chem 8:119
54. Becke AD (1993) J Chem Phys 98:5648
25. Li Y, Wang Z, Ding K (2015) Chem Eur J 21:16387
26. Vargiu AV, Magistrato A (2012) Inorg Chem 51:2046
27. Mirtschin S, Krasniqi E, Scopelliti R, Severin K (2008) Inorg
Chem 47:6375
55. Lee C, Yang W, Parr RG (1988) Rev Phys B 37:785–789
56. Wadt WR, Hay PJ (1985) J Chem Phys 82:284
57. Hay PJ, Wadt WR (1985) J Chem Phys 82:299
58. Hehre WJ, Ditchfield R, Pople JA (1972) J Chem Phys 56:2257
28. Park J, Hong S (2012) Chem Soc Rev 41:6931
123