Water-Soluble, Disulfonated alpha-Diimine Rhodium(I) Complexes: Synthesis, Characterisation and Application as Catalyst Precursors in the Hydroformylation of 1-Octene

Article abstract of DOI:10.1002/ejic.201900317

The synthesis and characterisation of two new water-soluble Rh^I mononuclear 1,4-diazabutadiene (DAD^s) complexes of general formula: [Rh(DAD^s-R)(COD)], where (DAD^s = sulfonated-tagged–1,4-diazabutadiene, COD = cy

Full text of DOI:10.1002/ejic.201900317

DOI: 10.1002/ejic.201900317
Full Paper
Rhodium Hydroformylation Catalysts | Very Important Paper |
Water-Soluble, Disulfonated alpha-Diimine Rhodium(I)
Complexes: Synthesis, Characterisation and Application as
Catalyst Precursors in the Hydroformylation of 1-Octene
Nikechukwu N. Omosun[a] and Gregory S. Smith*[a]
Abstract: The synthesis and characterisation of two new water- ised using 1H NMR, 13C NMR, FT-IR spectroscopy and mass spec-
soluble RhI mononuclear 1,4-diazabutadiene (DADs) complexes trometry. Their performance as catalyst precursors in the hydro-
of general formula: [Rh(DADs-R)(COD)], where (DADs = sulfon- formylation of 1-octene was assessed and compared to those of
ated-tagged–1,4-diazabutadiene, COD = cyclooctadiene; and the non-sulfonated RhI mononuclear 1,4-diazabutadiene (DAD)
R = H or C10H8), are described. The rhodium(I) complexes were complexes. Utilisation of the water soluble [Rh(DADs-R)(COD)]
obtained via a complexation reaction of [Rh(COD)(MeCN)2]BF4 lead to high conversions and regioselectivity (linear/branched
with the sulfonated α-diimine (DAD) ligands, which were previ- aldehyde ratio) in the aqueous biphasic hydroformylation of 1-
ouslyobtainedfromaSchiffbasecondensationreactionof4-amino- octene. Additionally, the catalysts were recovered by phase sep-
phenol with either 1,2-ethanedione or acenapthenequinone. All aration and are reusable over four consecutive catalytic runs
rhodium complexes and their precursor ligands were character- without significant loss in catalytic activity.
Introduction
oxygen and this process is limited to only short chain olefins
(< C4).[13,17] The development of new water-soluble catalysts is
thus an important and ongoing research activity at both aca-
demic/industrial levels and a significant advancement is been
made to improve aqueous-phase hydroformylation reac-
tions.[18,12] The underlying idea is to introduce hydrophilic sub-
stituents into sterically demanding and electron-rich ligands that
display high catalytic activity in conventional organic solvents. In
our previous work, we have reported on RhI complexes bearing
Schiff base ligands containing N,O- and N,P-chelates.[19–22] These
catalyst precursors are highly active and chemo- and regioselec-
tive in the hydroformylation of 1-octene.
The contribution of catalysis to economic development and en-
vironmental sustainability cannot be over-emphasised and this
is due to the high activity and selectivity of catalysts to the
desired product.[1–3] In the industrial chemical sector, 85 % of
most organic transformations use various transition metal
based catalysts.[4] Rhodium-catalysed reactions such as hydro-
formylation convert olefins to aldehydes that are useful inter-
mediates to synthesise pharmaceuticals,[5] fine chemicals,[6] and
agrochemicals.[7] This transformation proceeds favourably using
homogeneous catalysts and over the years significant progress
has been achieved for these catalysed reactions.[8] Despite
these improvements, the challenges of the homogeneous catal-
ysis is the difficulty of recovery and recyclability or reuse of
these expensive metal catalysts.[9,10] One strategy that has been
established to address these challenges is the use of aqueous
phase catalytic systems which immobilise metal complexes in
water, used as the reaction medium.[11–16] The Rührchemie/
Rhône Poulenc “Oxo” process is a well-established industrial ap-
plication which exemplifies aqueous biphasic catalysis utilizing
a water soluble rhodium complex based on the trisodium salt
of tris(m-sulfonatophenyl)phosphine (TPPTS) ligand. However,
TPPTS based ligands are highly unstable in the presence of air/
In aqueous reactions, the catalyst precursor is dissolved in
water and the substrate is contained in the organic phase, form-
ing two layers (biphasic). Upon heating, there is an interfacial
interaction between the catalyst and the substrate and on cool-
ing the phases separate. The water-soluble catalyst is easily sep-
arated from the product mixture by simple decantation as
shown in Figure 1. This biphasic method allows for catalyst re-
covery, possible multiple recycling and high product turnover.
This strategy correlates well with Green Chemistry principles in
the use of water as a solvent and minimising leaching which in
turn promotes environmental conservation and sustainabil-
ity.[23,24]
Following our previous studies, we were interested in the α-
diimine (1,4-diazabutadiene skeleton) ligand scaffold, which has
been used in catalytic reactions such as olefin polymerisa-
tion[25–27] and C–C cross-coupling reactions because of their
versatile coordination behaviour and the interesting properties
of their metal complexes.[28,29] The two key features of α-di-
imine ligands are: (i) the presence of two exo imines to the two
[a] Department of Chemistry, University of Cape Town,
Rondebosch, 7701, South Africa
E-mail: Gregory.Smith@uct.ac.za
http://www.gregsmith-research.uct.ac.za
Supporting information and ORCID(s) from the author(s) for this article are
available on the WWW under https://doi.org/10.1002/ejic.201900317.
Eur. J. Inorg. Chem. 0000, 0–0
1
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
1
with the primary amine of 4-aminophenol is absent in the H
NMR spectrum of ligand L1 and L2 confirming a successful
Schiff base condensation reaction had occurred. All the ex-
pected peaks of aromatic protons were observed in ligand L1
and L2 between 6.92 ppm and 9.40 ppm respectively. Notably,
the peak for phenolic protons of ligand L1 and L2 appears as
a broad singlet at δ = 9.78 ppm and δ = 9.40 ppm respectively.
The signal for the phenyl protons in ligand L2 appears as a
singlet due to the symmetrical nature of the molecule as a re-
sult of the rigid acenaphthene backbone which prevents rota-
tion around the imine carbon–carbon bond and forces the
imine nitrogen atoms to remain in a fixed orientation.[35,36] It
has been reported that bis(N-arylimino)acenaphthene deriva-
tives of alpha diamine ligands are more rigid and sterically
bulky compared to diimine ligands derived from glyoxal or
acyclic diketones and this rigidity may impose stable chelation
to metal complexes exhibiting a more limited range of bonding
modes and impacting high chemical stability with respect to
hydrolysis and cleaving of the central C–C bond.[37–39] The IR
spectra of L1 and L2 show absorption bands for the imine
C=N bond at 1603 cm–1 and 1659 cm–1 and the O–H bond at
–1
3075 cm–1 and 3289 cm respectively.
The alkylation of the hydroxy group in ligand L1 and L2 was
performed in dry tetrahydrofuran using 1,3-propanesultone for
the introduction of the alkyl-sulfonate group (Scheme 1). Ini-
tially, the phenoxide was generated using the deprotonating
agent NaH, followed by ring-opening of the cyclic sulfonate
(1,3-propanesultone) via nucleophilic attack from the
phenoxide species which occurs with cleaving of the C=O.[40,41]
The modification occurred in high yield (82 %) when these rea-
gents were added in slight excess. The absence of hydroxyl pro-
Figure 1. Strategy to recycle and reuse a water-soluble metal precatalyst.
alpha carbons system which leads to better σ-donating and
better π-accepting properties,[30–32] and (ii) the stereoelectron-
ics of the substituents bonded to the imine.[33] Extensive chem-
istry has been carried out with similar imine systems such the
2,2′-bipyridine and phenanthroline. These are under-explored
andthesimplestrepresentativeofthisclassofcompounds(1,4-di-
substituted, 1,4-diaza1,3-butadienes) have been reported to
have a flexible N=C–C=N skeleton and unusual electron donor
and π-accepting properties.[34] In this study, we chose α-diimine
ligands as model compounds with para hydroxyl groups at the
focal point which allows for the attachment of water-soluble
entities and varied the backbone which impacts on the steric
and electronic properties.
1
ton signal in the H NMR spectra of ligands L3 and L4 respec-
tively confirms the successful deprotonation or formation of a
phenoxide species. The phenyl protons attached to the imine
appear as quartet signals alluding to the fluxionality resulting
from a change from a symmetrical to asymmetrical molecule
upon sulfonation. Both ligands display excellent solubility in
water and DMSO at room temperature.
Complexation was achieved using [Rh(COD)(MeCN)2]BF4,
prepared in situ from [(Rh(COD)Cl)2] and AgBF4 in acetonitrile
at room temperature. The water-soluble rhodium(I) complexes
3 and 4 were obtained in quantitative yield upon reaction of
[Rh(COD)(MeCN)2]BF4 with the appropriate ligand at room
temperature. DMSO proved useful as the solvent of choice
as it possesses sufficient polarity to dissolve the salt-like ligands
and complexes. The displacement of acetonitrile from
[Rh(COD)(MeCN)2]BF4 with the α-diimine ligands was facile, af-
fording new rhodium cationic complexes. The hydrophilic com-
plexes were simply purified by washing the product mixture
with acetone to remove any unreacted [Rh(COD)(MeCN)2]BF4.
The 1H NMR spectrum of the metal complexes show signals
consistent with the proposed structure. Comparing the 1H NMR
spectrum of ligand L4 and complex 4, a significant down field
shift was observed for the acenaphthene ring proton upon co-
ordination caused by increase in electron density. However, the
To the best of our knowledge, hydroformylation involving
disulfonated α-diimine rhodium(I) complexes has not been
explored. Herein, we report the synthesis of water-soluble RhI
precatalysts bearing sulfonated N,N-bis(arylimino)ethane and
sulfonated α-diimine ligands of the type N,N- bis(arylimino)-
acenaphthene (aryl-BIAN) for the hydroformylation reaction of
1-octene in an aqueous medium.
Results and Discussion
Synthesis and Characterisation of Ligands and Rhodium(I)
Catalyst Precursors
The alpha-diimine ligands L1 and L2 were synthesised via a
Schiff base condensation reaction of the carbonyl compounds
(1,2-ethanedione and acenapthenequinone) with two molar
equivalents of 4-amino phenol in methanol and using a cata- protons of the phenyl bonded to the imine are observed as two
lytic amount of acetic acid (Scheme 1). The peak associated doublets which may be attributed to the coordination of the
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
2
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Scheme 1. Synthesis of RhI catalyst precursors 1–4.
imine nitrogen to the Rh metal centre. The signals for the cyclo- Preliminary Screening Using Catalyst Precursor 1 in the
octadienyl entities are observed as broad signals each peak in- Hydroformylation of 1-Octene
1
tegrating for 4 protons. The H NMR spectrum for complex 3
Catalyst precursor complex 1 (where R = H) being the simplest
and most representative structure of all the precatalysts synthe-
sised was used for optimisation experiments. The hydroformyl-
ation reaction was attempted by varying the temperature
(55 °C, 75 °C and 95 °C) and syngas pressure (20 bar, 30 bar
and 40 bar). All the reactions were performed for 4 hours in
toluene (5 mL) and the products were analysed using gas chro-
matography with n-decane as the internal standard.
shows trends similar to those observed in the NMR spectrum
of complex 4. The protons of the cyclooctadienyl entities are
observed in complex 3 as a singlet and multiplets at 3.92 ppm,
2.35 ppm and 1.72 ppm. Coordination-induced down field shift
was observed for the imine proton H-1 from 8.44 ppm to
8.68 ppm. The absorption bands of the imine functionality in
the infrared spectra of the RhI-COD complexes (3 and 4) de-
crease on coordination of the metal compared to that observed
in the metal-free ligands. This is due to the degree of back
Under conditions of 55 °C and 20 bar, minimal conversion of
bonding from the metal into the π* (antibonding) orbitals of 1-octene was observed and increasing the temperature to 75 °C
the imine (synergic effect). The non-sulfonated α-diimine at the same pressure resulted in an increase conversion of 1-
rhodium complexes (1 and 2) used in this study were readily octene with the formation of more isooctenes. This is because
synthesised by reacting the ligands either L1 or L2 with an increase in temperature leads to more alkene activation.[42]
[Rh(COD)(MeCN)2]BF4 in acetonitrile (Scheme 1). The 1H NMR However, at much higher temperature of 95 °C at the same
spectra of complex 1 and 2 display all the expected aromatic pressure (20 bar) the isomerisation of 1-octene was greatly re-
and COD signals.
duced, and an overall increase in conversion to aldehydes was
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
3
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
observed. This might be attributed to the increased solubility Recyclability of Catalyst Precursors 3 and 4
of the catalyst at higher temperatures thereby leading to higher
isomerisation and subsequent hydroformylation of 1-octene.
Generally, with respect to substrate conversion, an increase in
temperature leads to increased catalyst activity. At constant
temperature and increasing pressure (20, 30 and 40 bar), a gen-
eral increase in the turnover frequency conversion of 1-octene
to aldehydes and lower formation of isooctene was observed.
It has been reported that increasing the CO pressure in hydro-
formylation results in the increase of CO concentration in the
reaction medium and scavenges the vacant site of the 16-elec-
tron alkyl rhodium species reducing isomerisation and acceler-
ating the CO migration step.[43] This ideally, results in a higher
conversion rate of the olefin to aldehydes. However, under
these conditions, the gradual reduction of linear aldehydes was
observed and a considerable amount of branched aldehydes is
formed. This is indicated by the decrease in n:iso ratio. Optimum
temperature and pressure conditions were established at 75 °C
and 40 bar and further used for catalyst comparison with re-
spect to conversion, n:iso ratio and turnover frequency.
The recyclability studies over four catalytic runs was performed
as follows. After a hydroformylation reaction, the reaction mix-
ture was cooled to room temperature and the product (con-
tained in the organic phase) was separated by decantation and
the catalyst-containing solution (aqueous phase) was reused
over three consecutive cycle by introducing the fresh substrate.
Overall conversions greater than 65 % was observed over four
cycles, with the first two cycles showing conversions greater
than 90 % with high regioselectivity towards linear aldehydes
(Figure 3).
Hydroformylation Reaction Using Catalyst Precursors 3–4
in the Aqueous Phase
Figure 3. Recyclability of catalyst precursors 3 and 4 in the hydroformylation
of 1-octene.
The water-soluble 1,4-diazabutadiene (DADs) rhodium(I) com-
plexes (3 and 4) were used as catalyst precursors under the
established optimum condition (75 °C / 40 bar) in water (5 mL)
and compared to the non-sulfonated 1,4-diazabutadiene rho-
dium(I) complexes (1 and 2).
The regioselectivity of each catalyst varied upon recycling as
shown in Figure 4. This may be attributed to changes in the
structure of the active catalyst under hydroformylation condi-
tions as it is recycled. Despite this, a significant drop in conver-
sion and turn over frequencies was observed after the third
cycle for both catalyst precursors 3 and 4 with an increase in
the formation of isooctenes. In all experiments, noticeable cata-
lyst decomposition was observed by the darkened colouration
of the aqueous phase. Also, the dark colour intensity increased
after each cycle and the efficiency of the catalyst (3 and 4)
upon recycling dramatically decreased (< 65 % conversion).
With respect to regioselectivity, increased formation of linear
aldehydes was observed using catalyst precursor 4. It has been
reported that the n/iso ratio of the product depends mainly on
steric effects which favours anti-Markovnikov addition making
the catalyst more selective towards the formation of the linear
aldehyde.[44]
All catalyst precursors gave high conversions (Table 1)
greater than 95 % and showed good regioselectivity for linear
aldehydes (> 65 %). Complexes 3 and 4 showed high conver-
sions and turnover frequencies in comparison with complexes
1 and 2, in favour of aldehydes (> 85 %) of which > 65 % is
nonanal. The excellent solubility in water can, in part, account
for the better activity as there is more substrate to catalyst in-
teraction upon increase in temperature. The percentage conver-
sion of the water-soluble RhI complexes (3 and 4) were also
evaluated at various times (Figure 2); this was conducted using
the optimised conditions 75 °C and 40 bar. The precursors were
evaluated at intervals of 1 hour, 2 hours, 4 hours, 6 hours and
8 hours. In general, near-quantitative conversions of 1-octene
(> 99 %) was observed after 6 hours and slight differences in
catalytic rates are seen in the first 2 hours.
Figure 2. Percentage conversion of 1-octene over 8 hours using catalyst pre-
Figure 4. Regioselectivity of catalyst precursors 3 and 4 in the hydroformyl-
cursors (3 and 4). Reactions performed at 75 °C and 40 bars.
ation of 1-octene.
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
4
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Table 1. Hydroformylation of 1-octene using catalyst precursors 1–4.[a]
Entry
Precatalyst
Conversion
[%]
Aldehydes
[%]
Linear
aldehydes
[%]
Branched
aldehydes
[%]
Iso-
octenes
[%]
n:iso
TOF
[h–1
]
1
2
3
4
5
6
7
8
1
2
3
4
95
97
98
99
54
56
98
98
72
70
86
88
41
38
86
88
69
70
66
73
63
61
64
74
31
30
34
27
37
29
34
26
28
30
15
12
59
60
14
13
2.21
2.36
1.91
2.73
1.72.
2.14
1.90
2.70
329
317
522
539
142
206
566
556
1+Hg
2+Hg
3+Hg
4+Hg
[a] Reaction conditions: The reactor (90 mL) was loaded with toluene (5 mL) for catalyst precursors 1 and 2, or H2O (5 mL) for catalyst precursors 3 and 4,
1-octene (0.805 g, 6.37 mmol), internal standard n-decane (0.204 g, 1.435 mmol) and Rh–metal loading (2.87 × 10–3 mmol). The reactor was purged with
nitrogen three times, followed by purging three times with syngas. Catalyst to substrate ratio used was (1:2500). The reactions were carried out for 4 hours
at 75 °C/40 bar and the data was analysed using GC-FID. TOF = (mmol of aldehydes/mmol of Rh)/time. Average error estimate: (1) = 0.14, (2) = 0.11, (3) =
0.18, (4) = 0.10.
Inductively Coupled Plasma Spectrometry Studies
thesised and fully characterised using various spectroscopic
and analytical techniques. The complexes are highly soluble in
water at room temperature, and the efficiency of the complexes
was evaluated as catalyst precursors in the aqueous hydroform-
ylation of 1-octene and compared to those of their non water-
soluble analogues. Excellent conversion of 1-octene to alde-
hydes with higher turnover frequencies and chemo- and regio-
selectivities were observed using catalyst precursors 3 and 4.
The recyclability tests revealed that the water-soluble precata-
lysts can be reused up to four times with a loss in catalytic
activity observed after the third cycle. The mercury drop test
confirmed the molecular nature of the water-soluble precata-
lysts and the absence of any nanoparticles. Further in situ tem-
perature-programmed decomposition (TPD) studies under con-
trolled conditions are required to fully establish the stability of
these complexes.
To investigate the origin of this decrease in catalytic perform-
ance, the recycled organic layer was analysed using ICP-MS
analysis. The data showed lower amounts of rhodium concen-
tration in the organic layer in first cycle and negligible leaching
was observed in the second cycle (less than < 0.003 %) and no
rhodium leaching in the third and fourth cycle suggesting no
significant decrease of rhodium concentration in the aqueous
phase. However, the decrease efficiency of the catalyst after the
second cycle may be due to the degradation of the catalyst
upon further recycling thereby suggesting a decrease in the
formation of the active rhodium species in the reaction me-
dium. This phenomenon is consistent with catalyst decomposi-
tion.
Mercury Poisoning Experiments
A
mercury poisoning test was used to determine the
Experimental Section
heterogeneity of a catalytic process and it is based on the abil-
ity of mercury to combine with metals or to get adsorbed unto
their surface. This therefore establishes if a catalyst acts as a
molecular species or is heterogeneous in nature (nanoparticles)
as mercury is believed to only interact with heterogeneous cat-
alysts forming amalgam and has no influence on molecular cat-
alysts. The mercury poisoning test was performed by adding a
drop of mercury into each reactor containing catalyst precursor
using the temperature (75 °C), time (4 hours) and pressure
(40 bar) (Table 1, entries 5–8). A substantial drop in conversion
is observed for the non-water-soluble rhodium precatalysts 1
and 2 in the presence of mercury suggesting a combination of
homogeneous and heterogeneous catalysis while no significant
changes in conversion was observed with the water-soluble an-
alogues (3 and 4) in the presence of mercury. This suggests that
the water-soluble precatalysts act as homogeneous molecular
precatalysts in solution.
General: All solvents were purchased and purified according to
standard methods. All reagents were purchased from Merck and
used without further purification. 1,2-bis(4-hydroxylphenylimino)
glyoxal,[45] L1[46] and L2[47] were synthesised according to modified,
previously reported methods. Nuclear Magnetic Resonance (NMR)
spectra were recorded on either a Bruker Ultrashield 400 Plus (1H:
400.22 MHz; 13C: 100.65 MHz) or a Bruker 300 MHz (1H: 300.08 MHz;
1
13C: 75.46 MHz) spectrometer at ambient temperature. H and 13C
NMR chemical shifts are referenced to the deuterated solvent. Infra-
red (IR) spectra were determined using a Perkin–Elmer Spectrum
100 FT-IR spectrometer, and were recorded using Attenuated Total
Reflectance Infrared spectroscopy (ATR-IR). Melting points were de-
termined using a Büchi melting point apparatus B-540. Mass spec-
trometry was carried out using a Waters API Quattro Micro Triple
Quadrupole electrospray ionisation mass spectrometer. Hydroform-
ylation samples were analysed on a Perkin Elmer Clarus 580 gas
chromatography. Inductively coupled plasma emission spectro-
scopy experiments were carried out using Thermo-Fisher X-Series II
quadrupole ICP-MS.
Synthesis of Disulfonated α-Diimine Ligands: Synthesis of L3:
Sodium hydride (0.033 g, 1.37 mmol) dispersed in mineral oil was
suspended in 10 mL of dry THF, in a two-neck flask. A solution of
Conclusion
In summary, two new water-soluble disulfonated α-diimine li-
gands and their corresponding rhodium complexes were syn- L1 (0.100 g, 0.420 mmol) dissolved in dry THF (2 mL) was added.
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
5
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
The mixture was allowed to stir under argon for one hour at room 2.43 (d, 4H J = 1.5 Hz CH2COD), 1.87 (d, 4H J = 4.1 Hz CH2COD). FT-IR
temperature and then 1,3-propanesultone (0.15 g, 1.23 mmol) in (ATR, cm–1) ν = 3325 (C- O), 2880 (C–HCOD), 1623(C=N), 1593 (C=C).
dry THF (2 mL) was added and allowed to stir for 8 h under reflux
Rhodium(I) Complex 3: A mixture of silver tetrafluoroborate
at 50 °C. The reaction mixture turned orange. After cooling to room
(0.0250 g, 0.128 mmol) and [RhCl(COD)]2 (0.0300 g, 0.0607 mmol)
temperature, a precipitate was collected by suction filtration, and
was stirred in acetonitrile (10 mL) for one hour at room tempera-
washed with THF and acetone to remove the unreacted excess
ture. The precipitated silver chloride was removed using a syringe
sultone. The product was purified by precipitation from DMSO/acet-
filter (PTFE, 0.25 μm) and the resulting solution was reduced under
one mixture and dried under vacuum for 8 h. L3 was isolated as an
vacuum and ligand L3 (0.0600 g, 0.113 mmol), dissolved in DMSO,
orange powder. Yield: 0.180 g, (82 %). M.p: 229–230 °C. 1H NMR
was added and stirred for 0.5 h at room temperature. The reaction
(300 MHz, D2O-d2) (ppm) = 7.96 (s, 2H, N=C-H), 7.25 (d, 4H, 3J =
mixture was poured in to acetone. The precipitate that formed was
8.80 Hz Ar-H), 6.93 (t, 4H, 3J = 7.20 Hz Ar-H) 3.99 (t, 4 H, 3J = 8.30 Hz
washed with acetone and dried under vacuum for 8 h. Yield:
O-CH2), 2.98 (t, 4H, 3J = 7.40 Hz CH2-SO3), 2.12 (m, 4 H, CH2-CH2-
1
0.0530 g, (56 %). H NMR (300 MHz, [D6]DMSO): δ (ppm) = 8.68 (s,
CH2). 13C NMR (151 MHz, [D6]DMSO) δ = 159.25, 157.93, 143.09,
2H, Himine), 7.97 (d, J = 8.4 Hz, 4H, Ar), 7.17 (d, J = 9.0 Hz, 4H, Ar),
4.20 (t, J = 6.5 Hz, 4H), 3.92 (s, 4H, CHCOD), 2.63–2.56 (m, 4H), 2.36
(s, 4H, CH2COD), 2.09–2.01 (m, 4H), 1.77 (d, J = 5.9 Hz, 4H, CH2COD).
123.56, 115.73, 67.69, 48.39, 25.80. FT-IR (ATR, cm–1) ν = 1185 and
1054 (O=S=O). HR-ESI-MS (m/z) = 526.9700 [M + H]+.
FT-IR (ATR, cm–1) ν = 1185 and 1054 (O=S=O). 1599 (C=N), 1571
(C=C). Rhodium(I) Complex 4: A mixture of silver tetrafluoroborate
(0.0240 g, 0.123 mmol, 1.0 equiv.) and [RhCl(COD)]2 (0.0300 g,
0.0607 mmol, 0.5 equiv.) was stirred in acetonitrile (10 mL) for one
hour at room temperature. The precipitated silver chloride was re-
moved using a syringe filter (PTFE, 0.25 μm) and the resulting solu-
tion was reduced under vacuum and ligand L4 (0.0793 g,
0.122 mmol) dissolved in DMSO was added and stirred for 0.5 h at
room temperature. The reaction mixture was poured in to acetone.
The resultng precipitate was washed with acetone and dried under
vacuum for 8 h. Yield: 0.0520 g, (62 %). 1H NMR (300 MHz,
[D6]DMSO): δ (ppm) = 8.26 (d, J = 8.3 Hz, 2H, Himine), 7.67 (t, J =
7.8 Hz, 2H, Ar), 7.35 (d, J = 8.6 Hz, 4H, Ar), 7.18 (d, J = 8.7 Hz, 4H,
Ar), 6.60 (d, J = 7.2 Hz, 4H, Ar), 4.19 (t, J = 6.3 Hz, 4H), 3.97 (s, 4H,
CHCOD), 2.62 (d, J = 7.1 Hz, 4H), 2.43 (d, J = 8.4 Hz, 4H, CH2COD ),
Synthesis of L4: Sodium hydride (0.015 g, 0.700 mmol) dispersed
in mineral oil was suspended in 10 mL of dry THF, in a two-neck
flask. A solution of L2 (0.100 g, 0.270 mmol) dissolved in dry THF
(2 mL) was added. The mixture was allowed to stir under argon for
one hour at room temperature and then 1,3-propanesultone
(0.08 g, 0.65 mmol) in dry THF (2 mL) is added and allowed to stir
for 8 h under reflux at 40 °C. The reaction mixture turned dark red.
After cooling to room temperature, the precipitate was collected by
suction filtration and washed with THF and acetone to remove the
unreacted excess sultone. The product was purified by recrystallisa-
tion using a water/ethanol mixture and dried under vacuum for 8 h.
L4 was isolated as a red powder. Yield: 0.122 g, 67 %. M.p: 154
1
3
–156 °C. H NMR (300 MHz, [D6]DMSO): δ (ppm) = 8.08 (d, 2H, J =
7.9 Hz, ArH), 7.57 (t, 2H, ArH), 6.96 (d, 2H, 3J = 7.2 Hz ArH), 7.05 (dd,
3
8H, ArH), 4.15 (t, 4 H, J = 6.5 Hz OCH2), 2.63 (t, 4H, CH2-SO3), 2.09
(m, 4 H, CH2CH2CH2). 13C NMR (151 MHz, [D6]DMSO) δ = 179.32,
161.60, 157.26, 124.29, 120.60, 116.88, 68.35, 49.46. FT-IR (ATR,
cm–1) ν = 1181 and 1036 (O=S=O). HR-ESI-MS (m/z) = 653.0217
([M + H]+, 65 %).
2.10–2.03 (m, 4H, ), 1.86 (d, J = 8.6 Hz, 4H, CH2COD). FT-IR (ATR, cm–1
ν = 1181 and 1036 (O=S=O). 1601 (C=N), 1571 (C=C).
)
Catalytic Experiment: Hydroformylation reactions were conducted
using a 90 mL stainless-steel pipe reactor equipped with a Teflon-
coated magnetic stirrer bar. The reactor was charged with the
RhI catalyst precursor (2.87 × 10–3 mmol), substrate [1-octene
(6.37 mmol)], internal standard n-decane (1.435 mmol) and either
toluene (5 mL) for catalyst precursors 1 and 2 or water (5 mL) for
catalyst precursors 3 and 4. The reactor was and purged with nitro-
gen three times, followed by purging with syngas (CO/H2, 1:1) twice.
The reactor was pressurised with syngas to the desired pressure
and heated to the required temperature. Samples collected at the
beginning and at the end of the reaction were analysed by gas
chromatography, and the products were confirmed against isooct-
ene and aldehyde standards. All catalytic reactions were performed
in triplicate in order to ensure reproducibility and are recorded as
an average of three identical experiments.
Synthesis of Rhodium(I) Complexes
Rhodium(I) Complex 1: A mixture of silver tetrafluoroborate
(0.243 g, 1.25 mmol) and [Rh(COD)Cl]2 (0.308 g, 0.624 mmol) was
stirred in acetonitrile (10 mL) for one hour at room temperature.
The precipitated silver chloride was removed using a syringe filter
(PTFE, 0.45 mm) and the solution was added to a suspension of
ligand L1 (0.300 g, 1.25 mmol) in acetonitrile (10 mL). The reaction
mixture was then stirred for one hour at room temperature. The
desired complex (1) was isolated by filtration, washed with aceto-
nitrile and dried under vacuum. Yield: 0.134 g, (59 %) 1H NMR
(300 MHz, [D6]DMSO) δ = 10.20 (br s, 2H Ar-OH), 8.62 (s, 2H, N=C-
H), 7.90 (d, 4H, 3J = 6.3 Hz Ar-H), 6.99 (d, 4H, 3J = 8.8 Hz Ar-H ), 3.90
(s, 4H, CHCOD), 2.35 (m, 4H, J = 9.8 Hz CH2COD), 1.86 (d, 4H, J = 8.0 Hz
CH2COD). 13C NMR (151 MHz, [D6]DMSO) δ = 158.34, 157.00 (Cimine),
141.65, 123.89, 116.36 (s),86.82 (CCOD) 30.68 (CCOD). FT-IR (ATR, cm–1
)
Acknowledgments
ν = 3424 (C- O), 2841 (C–HCOD), 1607(C=N), 1581 (C=C).
We gratefully acknowledge the University of Cape Town and
the NRF-DST Centre of Excellence in Catalysis – c*change for
financial support.
Rhodium(I) Complex 2: A mixture of silver tetrafluoroborate
(0.160 g, 0.823 mmol) and [Rh(COD)Cl]2 (0.203 g, 0.411 mmol) was
stirred in acetonitrile (10 mL) for one hour at room temperature.
The precipitated silver chloride was removed using a syringe filter
(PTFE, 0.45 mm) and the solution was added to a suspension of
ligand L2 (0.300 g, 0.823 mmol) in acetonitrile (10 mL). The reaction
mixture was then stirred for one hour at room temperature. The
desired complex (2) was isolated by filtration, washed with aceto-
nitrile and dried under vacuum. Yield: 0.214 g, (61 %) 1H NMR
(300 MHz, [D6]DMSO) δ = 9.99 (br s, 2H Ar-OH), 8.28 (d, 2H, J =
Keywords: Hydroformylation · Rhodium · Diimine ligands ·
Homogeneous catalysis
[1] J. Fang, H. Jin, T. Ruddy, K. Pennybaker, D. Fahey, B. Subramaniam, Ind.
Eng. Chem. Res. 2007, 46, 8687–8692.
[2] J. R. Ludwig, C. S. Schindler, Chem 2017, 2, 313–316.
[3] P. L. Holland, Chem 2017, 2, 443–444.
2
3
8.2 Hz, ArH), 7.65 (t, J = 6.9 Hz, 2H, ArH), 7.24 (d, 4H, J = 6.7 Hz),
7.01 (d, 4H, 3J = 7.0 Hz), 6.64 (d, 2H, J = 1.5 Hz), 3.98 (s, 4H, CHCOD),
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
6
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[4]
[5]
[6]
[27] F. Z. Wang, S. S. Tian, R. P. Li, W. M. Li, C. L. Chen, Chinese J. Polym. Sci.
2017, 2, 157–162.
[28] P. Huo, J. Li, W. Liu, G. Mei, Chinese J. Chem. 2017, 35, 247.
[29] G. A. Grasa, A. C. Hillier, S. P. Nolan, Org. Lett. 2001, 3, 1077–1080.
[30] S. Filipuzzi, P. S. Pregosin, A. Albinati, S. Rizzato, Organometallics 2008,
27, 437–444.
[31] S. A. Svejda, M. S. Brookhart, Organometallics 1999, 18, 65–74.
[32] S. Ernst, W. Kaim, J. Am. Chem. Soc. 1986, 108, 3578–3586.
[33] C. O. Brien, M. Y. Wong, D. B. Cordes, A. M. Z. Slawin, E. Zysman-colman,
Organometallics 2015, 34, 13–22.
J. Heveling, J. Chem. Educ. 2012, 89, 1530–1536.
J. C. Doucet, H. Hierso, Curr. Opin. Drug Discov. Devel. 2007, 10, 672–690.
H. U. Blaser, A. Indolese, A. Schnyder, H. Steiner, M. Studer, J. Mol. Catal.
2001, 173, 3–18.
[7]
[8]
C. Torborg, M. Be, Adv. Synth. Catal. 2009, 351, 3027–3043.
B. Breit, in Recent Adv. Alkene Hydroformylation (Ed.: M. J. Krische),
Springer Berlin Heidelberg, 2007, pp. 139–172.
[9]
[10]
[11]
M. Arai, F. Zhao, Catalysts 2015, 868–870.
D. J. Cole–Hamilton, Science 2003, 299, 1702–1707.
[34] G. van Koten, K. Vrieze, Advances in Organometallic Chemistry, Academic
L. C. Matsinha, S. F. Mapolie, G. S. Smith, Dalton Trans. 2015, 44, 1240–
1248.
Press. Inc., Amsterdam, 1982.
[35] S. Anga, T. Pal, R. K. Kottalanka, M. Paul, T. K. Panda, Can. Chem. Trans.
2013, 1, 105–115.
[12]
J. Zhou, X. Guo, C. Tu, X. Li, H. Sun, J. Organomet. Chem. 2009, 694, 697–
[36] M. Gasperini, F. Ragaini, S. Cenini, Organometallics 2002, 21, 2950–2957.
[37] J. Wang, R. Ganguly, L. Yongxin, H. Sen Soo, Inorg. Chem. 2017, 56, 7811–
7820.
[38] X. Yu, F. Zhu, D. Bu, H. Lei, RSC Adv. 2017, 7, 15321–15329.
[39] V. Rosa, T. Avile, G. Aullon, B. Covelo, C. Lodeiro, Inorg. Chem. 2008, 47,
7734–7744.
[40] G. M. T. Smith, P. M. Burton, C. D. Bray, Angew. Chem. Int. Ed. 2015, 54,
15236–15240; Angew. Chem. 2015, 127, 15451.
[41] V. Sikervar, e-EROS Encycl. Reagents Org. Synth. 2014, 1–4.
[42] A. M. Kluwer, M. J. Krafft, I. Hartenbach, B. de Bruin, W. Kaim, Top. Catal.
2016, 59, 1787–1792.
702.
[13]
[14]
S. K. Sharma, R. V. Jasra, Catal. Today 2015, 247, 70–81.
M. Benatmane, K. Cousin, N. Laggoune, S. Menuel, E. Monflier, P. Woisel,
F. Hapiot, J. Potier, ChemCatChem 2018, 10, 5306–5313.
[15] E. Monflier, S. Tilloy, G. Fremy, Y. Barbaux, A. Mortreux, Tetrahedron Lett.
1995, 36, 387–388.
[16] G. Fremy, R. Grzybek, E. Monflier, A. Mortreux, A. M. Trzeciak, J. Ziolkow-
ski, J. Organomet. Chem. 1995, 505, 11–16.
[17] K. V. Katti, H. Gali, C. J. Smith, D. E. Berning, Acc. Chem. Res. 1999, 32, 9–
17.
[18] A. L. Casalnuovo, J. C. Calabrese, J. Am. Chem. Soc. 1990, 112, 4324–4330.
[19] S. Siangwata, N. Baartzes, B. C. E. Makhubela, G. S. Smith, J. Organomet.
Chem. 2015, 796, 26–32.
[20] C. Williams, M. Ferreira, E. Monflier, S. F. Mapolie, G. S. Smith, Dalton
Trans. 2018, 47, 9418–9429.
[21] L. Maqeda, B. C. E. Makhubela, G. S. Smith, Polyhedron 2015, 91, 128–
135.
[43] A. Stefani, G. Consiglio, C. Botteghi, P. Pino, J. Am. Chem. Soc. 1973, 95,
6504–6505.
[44] J. B. Claridge, R. E. Douthwaite, M. L. H. Green, R. M. Lago, S. C. Tsang,
A. P. E. York, J. Mol. Catal. 1994, 89, 113–120.
[45] S. H. Chan, L. S. M. Lam, C. W. Tse, K. Y. K. Man, W. T. Wong, A. B. Djurišić,
W. K. Chan, Macromolecules 2003, 36, 5482–5490.
[46] L. Biancalana, L. K. Batchelor, T. Funaiolo, S. Racchini, G. Pampaloni, P. J.
Dyson, F. Marchetti, Inorg. Chem. 2018, 57, 6669–6685.
[47] I. Matei, T. Lixandru, E. Comanita, Buletinul Institutului Politehnic din Iasi
1960, 6, 171–176.
[22] E. B. Hager, B. C. E. Makhubela, G. S. Smith, Dalton Trans. 2012, 41, 13927.
[23] S. R. SHARMA, Int. J. Chem. Phys. Sci. 2015, 4, 000–000.
[24] T. Welton, Proc. R. Soc. London Ser. A 2015, 471, 20150502.
[25] S. Dai, S. Zhou, W. Zhang, C. Chen, Macromolecules 2016, 49, 8855–8862.
[26] X. Sui, C. Hong, W. Pang, C. Chen, Mater. Chem. Front. 2017, 1, 967–972.
Received: March 20, 2019
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
7
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
Rhodium Hydroformylation
Catalysts
N. N. Omosun, G. S. Smith* ............ 1–8
Water-Soluble, Disulfonated alpha-
Diimine Rhodium(I) Complexes: Syn-
thesis, Characterisation and Appli-
cation as Catalyst Precursors in the
Hydroformylation of 1-Octene
Two water-soluble rhodium(I) com- hydroformylation of 1-octene, leading
plexes based on sulfonated alpha-di- to high, chemoselective conversions of
imine scaffolds have been synthesised the substrate to aldehydes and the
and characterised by NMR and IR spec- regioselective formation of largely lin-
troscopy, and mass spectrometry. They ear products.
are catalyst precursors for the aqueous
DOI: 10.1002/ejic.201900317
Eur. J. Inorg. Chem. 0000, 0–0
www.eurjic.org
8
© 0000 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim