Aqueous olefin hydroformylation using water-soluble mono- and trinuclear N,O-chelate rhodium(I)-aryl ether precatalysts

Article abstract of DOI:10.1016/j.apcata.2020.117736

Water-soluble, sulfonated, salicylaldiminato-aryl ether mono- (7) and trimeric (8) ligands were prepared and reacted with the [Rh(μ-Cl)(η²:η²?COD)]₂ dimer to yield the corresponding water-soluble mononuclear (9) and trinuc

Full text of DOI:10.1016/j.apcata.2020.117736

Journal Pre-proof
Aqueous olefin hydroformylation using water-soluble mono- and
trinuclear N,O-chelate rhodium(I)-aryl ether precatalysts
Shepherd Siangwata (Investigation) (Data curation) (Writing -
original draft) (Validation) (Formal analysis), Neill J. Goosen
(Supervision) (Resources) (Writing - review and editing), Gregory S.
Smith (Conceptualization) (Resources) (Writing - review and
editing) (Supervision) (Project administration) (Funding acquisition)
PII:
S0926-860X(20)30329-X
DOI:
https://doi.org/10.1016/j.apcata.2020.117736
APCATA 117736
Reference:
To appear in:
Applied Catalysis A, General
Received Date:
Revised Date:
Accepted Date:
29 May 2020
26 June 2020
9 July 2020
Please cite this article as: Siangwata S, Goosen NJ, Smith GS, Aqueous olefin
hydroformylation using water-soluble mono- and trinuclear N,O-chelate rhodium(I)-aryl ether
precatalysts, Applied Catalysis A, General (2020),
doi: https://doi.org/10.1016/j.apcata.2020.117736

This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.
© 2020 Published by Elsevier.

Aqueous olefin hydroformylation using water-soluble mono- and
trinuclear N,O-chelate rhodium(I)-aryl ether precatalysts
Shepherd Siangwataa, Neill J. Goosenb, Gregory S. Smitha,*
aDepartment of Chemistry, University of Cape Town, Rondebosch, 7701, South Africa
bDepartment of Process Engineering, Stellenbosch University, Private Bag X1, Matieland, 7602,
South Africa
* Corresponding author. Tel.: +27 (0)21 650 5279
E-mail address: gregory.smith@uct.ac.za
Graphical abstract
Highlights
 Water-soluble Rhodium(I) precatalysts were evaluated for olefin
hydroformylation.
 Catalysts can be recycled using an aqueous biphasic strategy.
 The trinuclear precatalyst shows greater stability.
1

 Precatalysts show good activity and aldehyde chemoselectivity.
ABSTRACT
Water-soluble, sulfonated, salicylaldiminato-aryl ether mono- (7) and trimeric (8)
ligands were prepared and reacted with the [Rh(μ-Cl)(η2:η2-COD)]2 dimer to yield the
corresponding water-soluble mononuclear (9) and trinuclear (10) Rh(I) complexes.
These rhodium complexes were evaluated and are active as catalyst precursors in the
aqueous biphasic hydroformylation of 1-octene and styrene. Optimal conditions were
realised, using 1-octene as a model reaction, at 85 °C and a syngas pressure of 50
bar, where the best activity and chemoselectivity for aldehydes was obtained. Catalyst
recycling was successfully conducted over 5 cycles in a biphasic medium, with a
gradual loss in catalytic performance for both complexes. However, the dendrimer-
stabilised trinuclear precatalyst (10) showed improved recyclability during “neat”,
aqueous monophasic hydroformylation experiments, while the mononuclear
precatalyst (9) showed a reduced overall performance. Both precatalysts showed
sustained catalytic activity (> 450 h-1) and a total bias toward aldehyde
chemoselectivity during the aqueous biphasic hydroformylation of styrene.
Keywords: hydroformylation; olefins; rhodium; aqueous biphasic
1.
Introduction
The recovery of homogeneous catalysts using a biphasic approach is a very attractive
strategy towards bridging the divide between homogeneous catalysis and
heterogeneous catalysis and recycling expensive metal catalysts [1–5]. Various
strategies have been explored for the phase separation of homogeneous catalysts in
biphasic systems, including fluorous-organic [6–13], supercritical carbon dioxide-ionic
liquids [14–21], and the predominantly applied, aqueous-organic biphasic approach
[22–25]. Using the aqueous-organic biphasic strategy, the catalyst is immobilised in
an aqueous solution and the substrate is contained in an organic layer, allowing for
2

efficient separation and recycling of the catalyst with respect to the reactants and
products (Fig. 1) [26–32]. Aqueous biphasic catalysis is a very attractive strategy for
catalytic transformations aligned with the Green Chemistry Principles, that is aimed at
decreasing the environmental impact of metal-catalysed reactions. This is largely due
to the ease of catalyst recovery through the use of water that is immiscible with most
organic solvents, as well as being an eco-friendly solvent [33–35]. Moreover, water is
relatively cheap, readily accessible, odourless, non-flammable and non-toxic, highly
polar and a good solvent for many gases.
Figure 1. General principle of aqueous biphasic catalysis
The discovery of the hydroformylation reaction by Otto Roelen in 1938 marked a
turning point in the history of homogeneous catalysis [36,37]. Furthermore, the
pioneering industrial scale use of water as a reaction medium for biphasic catalysis,
was exemplified by Ruhrchemie/Rhône-Poulenc in the hydroformylation of propylene
to butanal using the sulfonated, water-soluble Rh(I)-organometallic complex
[Rh(H)(CO)(TPPTS)3] [38–40]. The olefin-to-aldehydes homogeneous catalytic
transformation has become the largest homogeneous transition metal complex-
catalysed reaction in industry, initially producing only 10 000 tonnes of aldehydes per
annum in the 1950s, with a current approximate of 10 million tonnes of aldehydes
produced annually [41–45]. The significance of the hydroformylation reaction cannot
be over-emphasised, being essential to the supply of a wide range of products such
as plasticisers, fragrances, agrochemicals, pharmaceuticals and detergents [46,47].
In view of the wide product range that emanates from the downstream transformation
of the aldehydes, it is prudent to conduct hydroformylation reactions following a
3

sustainable and a circular economy-based approach. However, since the biphasic
reaction takes place at the interface of the two phases, this approach has been limited
to short chain olefins, owing to the poor solubility of longer chain olefins (> C5) in water
[48–52]. As a result, mass transfer limitations pose an economic threat due to low
reaction rates. Moreover, catalyst leaching from the aqueous layer to the organic layer
has been a major challenge in designing catalysts that can fully remain immobilised
throughout repetitive recycling of the aqueous phase [53–56]. The prevention of metal
leaching to the product is vital for the downstream products, especially agrochemicals
and pharmaceutical end-products, which undergo strict product composition quality
analysis.
Therefore, the strategic design of highly active, versatile, robust and reusable catalysts
is of paramount importance in homogeneous catalysis. Such strategies emanate from
the need to beneficiate the principal metal of choice for hydroformylation (rhodium),
which is part of the rare, fast depleting and relatively expensive Platinum Group Metals
(PGMs) [57]. Investigating value-added strategies may help circumvent the major
drawback of homogeneous catalysis, the main complications being catalyst separation
from the substrate for reuse. We have successfully applied sulfonated salicyladimine
ligands in the rhodium-catalysed aqueous biphasic hydroformylation of 1-octene
[53,56,58]. The application of highly water-soluble structures that are anchored to rigid
supports and possessing multiple active sites such as metallodendritic structures, may
open avenues aimed at curbing the unwanted major catalyst loss via leaching, while
beneficiating the metal through good catalytic performance over multiple cycles. The
metals anchored onto a dendrimer surface can act as multiple catalytically active sites,
allowing for potentially positive dendritic effects through enhanced catalytic activity
over the monometallic catalysts. The metal centres on these synthetic catalysts can
act independently (taking advantage of the increase in nuclearity per molecule), or
they can possess synergistic cooperative effects between multiple active sites
contained therein [59]. Such design strategies emanate from naturally occurring
metalloenzymes, which often contain two or more active metal centres leading to
enhanced catalytic activity [60,61]. In this context, we report the evaluation of a water-
soluble mononuclear and a low generation metallodendritic complex of Rh(I) as
catalyst precursors in the hydroformylation of a linear α-olefin (1-octene), and an aryl
substituted olefin (styrene).
4

2.
Experimental
2.1
General
All chemicals and solvents were reagent grade and used as received from Sigma-
t
Aldrich, unless otherwise stated. 5-Sulfonato salicylaldehyde (1) [53], butyl(3-
t
bromopropyl)carbamate (2) [62], butyl(3-phenoxypropyl)carbamate (3) [63], 3-
phenoxypropylamine (5) [64] were prepared according to modified previously reported
literature procedures. RhCl3.3H2O was purchased from Heraeus South Africa (Pty)
Ltd. Carbon monoxide and hydrogen gas were supplied by Air Liquide.
Hydroformylation studies were conducted in a 90 mL stainless steel pipe reactor.
Nuclear Magnetic Resonance (NMR) spectra were recorded on either a Bruker Biospin
GmbH (1H: 400.22 MHz; 13C: 100.65 MHz) or a Varian XR300 MHz (1H: 300.08 MHz;
13C: 75.46 MHz) spectrometer. NMR chemical shifts were reported relative to the
internal standard tetramethylsilane (δ 0.00). FT-IR spectra were recorded using
Attenuated Total Reflectance Infrared spectroscopy. Melting points were determined
using a BÜCHI melting point apparatus B-540. Mass spectrometry was carried out
using a Waters API Quattro Micro Triple Quadrupole electrospray ionisation mass
spectrometer in the positive- and negative-ion modes.
2.2
Preparation of 1,1,1-tris(tbutyl-4-phenoxymethyl)ethane-propylcarbamate
amine (4).
The new protected amine trimeric core 4 was prepared via a Boc-protection synthetic
approach. THPE (0.462 g, 1.51 mmol), K2CO3 (1.88 g, 13.6 mmol), and 18-crown-
ether (9.97 x 10-2 g, 0.377 mmol) were stirred in acetone (20 mL) under reflux for 30
t
minutes. To this was added butyl(3-bromopropyl)carbamate 2 (1.29 g, 5.43 mmol)
and the mixture was refluxed under N2. After 24 h, the reaction was allowed to cool to
room temperature and the solvent was removed under reduced pressure. The
resultant residue was re-dissolved in distilled water (100 mL) and extracted with
dichloromethane (4 x 50 mL). The organic extracts were combined and dried over
anhydrous sodium sulfate and filtered by gravity. The solvent was removed from the
filtrate, yielding the crude product as a colourless viscous oil which was purified via
silica gel column chromatography, eluting the product as a clear oil using toluene:ethyl
acetate (85:15) as eluent. The colourless oil product (4) turns to a white solid upon
1
drying in vacuo. Yield: (0.985 g, 84%). M.P.: 61 – 63 °C. H NMR (CDCl3, δ): 6.99–
5

6.96 (m, 6H, H9), 6.79–6.76 (m, 6H, H10), 4.77 (br s, 3H, H4), 4.02–3.97 (t, 3J = 6.0 Hz,
6H, H7), 3.41–3.24 (q, 3J = 4.5 Hz, 6H, H5), 2.09 (s, 3H, H13), 2.00–1.92 (quin, 3J = 6.0
Hz, 6H, H6), 1.44 (s, 27H, H1). 13C{1H} NMR (CDCl3, δ): 156.9 (C8), 156.2 (C3), 142.1
(C11), 129.8 (C10), 113.8 (C9), 79.34 (C2), 65.94 (C7), 50.77 (C12), 38.32 (C5), 30.94
(C13), 29.71 (C6), 28.56 (C1). FT-IR (νmax/cm-1): 3353 (N–H), 1687 (C=O).
2.3
Preparation of 1,1,1-tris(4-phenoxymethyl)ethane-propylamine (6).
The new trimeric amine 6 was prepared via a TFA-mediated deprotection reaction of
4 in dichloromethane. TFA (1.58 g, 13.8 mmol) was added to a stirring solution of
1,1,1-tris(tbutyl-4-phenoxymethyl)ethane-propylcarbamate 4 (0.359 g, 0.461 mmol) in
dichloromethane at 0 °C. The solution was allowed to stir to room temperature
overnight and then heated at 30 °C for an additional 24 h, with periodic replenishment
of dichloromethane as necessary. The reaction was allowed to cool to room
temperature and then added to a vigorously stirring 1:1 (v/v) mixture of ethyl acetate :
NaOH (1 M). The aqueous layer was washed with ethyl acetate (4 x 50 mL) and the
organic extracts were combined, dried over anhydrous magnesium sulfate and filtered
by gravity. The solvent in the filtrate was removed under reduced pressure, revealing
the product (6) as a clear oil which was collected and dried in vacuo. Yield: (0.164 g,
1
74%). H NMR (CDCl3, δ): 7.01–6.98 (m, 6H, H6), 6.81–6.79 (m, 6H, H7), 4.07–4.03
(t, 3J = 6.0 Hz, 6H, H4), 2.95–2.90 (t, 3J = 6.0 Hz, 6H, H2), 2.12 (s, 3H, H10), 1.98–1.89
3
13
(quin, J = 6.3 Hz, 6H, H3), 1.60 (br s, 6H, H1). C{1H} NMR (CDCl3, δ): 157.3 (C5),
142.9 (C8), 129.4 (C7), 113.0 (C6), 66.81 (C4), 50.76 (C9), 37.38 (C2), 30.91 (C10), 29.83
(C3). FT-IR (νmax/cm-1): 3361 (N–H), 3282 (N–H).
2.4
Preparation of 5-sulfonato phenoxypropylsalicylaldimine (7).
A solution of 3-phenoxypropylamine 5 (0.795 g, 5.26 mmol) in dry methanol (15 mL)
was added to a stirring solution of 5-sulfonato salicylaldehyde 1 (1.18 g, 5.26 mmol)
in dry methanol (60 mL), and the reaction was left to reflux overnight under nitrogen.
The reaction was then allowed to cool to room temperature and filtered by gravity. The
solvent in the filtrate was reduced to a minimum (ca. 3 mL), yielding a yellow precipitate
which was collected by vacuum filtration, and washed with minimum cold methanol
and diethyl ether to remove unreacted 3-phenoxypropylamine. The yellow solid was
then recrystallised by re-dissolving in minimum methanol and chilling at 0 ºC in the
6

refrigerator. The resultant yellow solid product (7) was collected via filtration by
vacuum and washed with minimum cold methanol, and diethyl ether. Yield: (1.26 g,
1
67%). M.P.: decomposes without melting, onset occurs at 298 °C. H NMR (DMSO-
d6, δ): 13.70 (br s, 1H, H1), 8.64 (s, 1H, H8), 7.70–7.68 (m, 1H, H6), 7.56–7.52 (m, 1H,
H4), 7.30–7.23 (m, 2H, H13), 7.17–7.13 (m, 1H, H3), 6.90–6.86 (m, 2H, H14), 6.80–6.77
3
3
(m, 1H, H15), 4.08–4.04 (t, J = 6.0 Hz, 2H, H9), 3.84–3.79 (t, J = 6.6 Hz, 2H, H11),
2.20–2.11 (quin, 3J = 6.6 Hz, 2H, H10). 13C{1H} NMR (DMSO-d6, δ): 166.5 (C8), 161.5
(C2), 155.5 (C12), 140.0 (C5), 130.4 (C6), 129.3 (C13), 126.5 (C4), 123.4 (C15), 117.6
(C7), 116.1 (C3), 112.6 (C14), 65.77 (C11), 55.42 (C9), 31.87 (C10). FT-IR (νmax/cm-1):
3294 (O–H), 1649 (C=N). ESI-MS (negative ion-mode), (m/z) = 334.0746 [M]– (calcd
334.0749). H2Osolubility/25℃ = 21.4 mg/mL.
2.5
Preparation
of
1,1,1-tris(4-phenoxymethyl)ethane-propyl-5-sulfonato
salicylaldimine ligand (8).
A solution of 1,1,1-tris(4-phenoxymethyl)ethane-propylamine 6 (0.399 g, 0.836 mmol)
in dry methanol (10 mL) was added to a stirring solution of 5-sulfonato salicylaldehyde
1 (0.562 g, 2.51 mmol) in dry methanol (50 mL), and the reaction was left to reflux
overnight under nitrogen. The reaction was then allowed to cool to room temperature
and filtered by gravity. The solvent was removed under reduced pressure, to give an
orange solid product which was re-dissolved in distilled water and filtered through a
0.22-micron hydrophilic PTFE syringe filter. The filtrate was lyophilised to yield the
product as an orange solid which was then re-dissolved in minimum methanol and
chilled at 0 ºC in the refrigerator. The resultant orange solid product (8) was collected
by vacuum filtration and washed with minimum cold methanol, and diethyl ether and
dried in vacuo. Yield: (0.661 g, 72%). M.P.: decomposes without melting, onset occurs
at 307 °C. 1H NMR (DMSO-d6, δ): 13.64 (br s, 3H, H1), 8.64 (s, 3H, H8), 7.73–7.67 (m,
3H, H6), 7.55–7.52 (m, 3H, H4), 6.95–6.85 (m, 12H, H13,14), 6.80–6.77 (m, 3H, H3),
4.07–4.03 (t, 3J = 6.0 Hz, 6H, H9), 3.78–3.74 (t, 3J = 6.6 Hz, 6H, H11), 2.15–2.06 (quin,
3J = 6.3 Hz, 6H, H10), 2.03 (s, 3H, H17). 13C{1H} NMR (DMSO-d6, δ): 166.4 (C8), 161.8
(C2), 157.0 (C12), 142.0 (C5), 139.8 (C15), 130.4 (C6), 129.7 (C13), 129.3 (C4), 117.6
(C7), 116.2 (C3), 114.3 (C14), 65.91 (C9), 55.21 (C11), 50.75 (C16), 30.93 (C17), 30.67
(C10). FT-IR (νmax/cm-1): 1637 (C=N), 1609 (C=N). ESI-MS (negative ion-mode), (m/z)
= 1084.2738 [M + Cl]– (calcd 1084.6675). H2Osolubility/25℃ = 19.6 mg/mL.
7

2.6
Preparation of 5-sulfonato phenoxypropylsalicylaldimine Rh(I) complex (9).
NaH (1.31 x 10-2 g, 0.547 mmol) was added to a stirring suspension of 5-sulfonato
phenoxypropylsalicylaldimine 7 (0.107 g, 0.299 mmol) in dichloromethane (15 mL),
and the reaction mixture was stirred for 30 minutes. A half molar equivalent solution
of the dimer [Rh(COD)Cl]2 (7.36 x 10-2 g, 0.149 mmol) in dichloromethane (5 mL) was
added to the stirring suspension, and the reaction was left stirring at room temperature.
After 12 h, methanol (5 mL) was added to quench the NaH, and the mixture was stirred
for a further 30 minutes. The solvent was then reduced to a minimum (ca. 3 mL),
yielding the product (9) as a yellow solid which was collected by vacuum filtration,
washed with minimum cold methanol and diethyl ether, and dried in vacuo. Yield:
1
(0.160 g, 94%). M.P.: decomposes without melting, onset occurs at 397 °C. H NMR
(DMSO-d6, δ): 8.28 (s, 1H, H7), 7.63–7.57 (m, 1H, H5), 7.47–7.44 (m, 1H, H3), 7.32–
7.27 (m, 2H, H12), 6.95–6.93 (m, 3H, H13,14), 6.56–6.53 (m, 1H, H2), 4.36 (br s, 2H, H8),
4.05–4.00 (m, 4H, H15,15’), 3.77 (br s, 2H, H10), 2.40–2.27 (m, 4H, H16), 2.12–2.02 (m,
2H, H9), 1.91–1.76 (m, 4H, H16’). 13C{1H} NMR (DMSO-d6, δ): 166.2 (C7), 164.9 (C1),
158.2 (C11), 134.5 (C4), 132.4 (C5), 132.0 (C3), 129.3 (C12), 120.5 (C14), 119.4 (C6),
117.1 (C2), 114.3 (C13), 83.70 (C15), 70.94 (C15’), 64.30 (C10), 54.69 (C8), 32.90 (C9),
31.18 (C16), 28.21 (C16’). FT-IR (νmax/cm-1): 1606 (C=N). ESI-MS (negative ion-mode),
(m/z) = 578.9556 [M + Cl]– (calcd 579.5195). H2Osolubilty/25℃ = 15.7 mg/mL.
2.7
Preparation
of
1,1,1-tris(4-phenoxymethyl)ethane-propyl-5-sulfonato
salicylaldimine Rh(I) complex (10).
NaH (9.41 x 10-3 g, 0.392 mmol) was added to a stirring suspension of 1,1,1-tris(4-
phenoxymethyl)ethane-propyl-5-sulfonato salicylaldimine ligand 8 (7.82 x 10-2 g, 7.13
x 10-2 mmol) in dichloromethane (15 mL), and the reaction mixture was stirred for 30
minutes. A one and a half molar equivalent solution of the dimer [Rh(COD)Cl]2 (5.28
x 10-2 g, 0.107 mmol) in dichloromethane (5 mL) was added to the stirring suspension,
and the reaction was left stirring at room temperature. After 12 h, methanol (5 mL) was
added to quench the NaH, and the mixture was stirred for a further 30 minutes. The
solvent was reduced to a minimum (ca. 3 mL), yielding the product (10) as a yellow
solid which was collected by vacuum filtration, washed with minimum
dichloromethane, methanol and diethyl ether, and then dried in vacuo. Yield: (1.20 g,
1
98%). M.P.: decomposes without melting, onset occurs at 296 °C. H NMR (DMSO-
8

d6, δ): 8.29 (s, 3H, H7), 7.68–7.57 (m, 3H, H5), 7.49–7.45 (m, 3H, H3), 6.99–6.73 (m,
20H, H12,13), 6.61–6.49 (m, 3H, H2), 4.52–4.17 (br s, 6H, H8), 4.14–3.89 (br s, 12H,
H17,17’), 3.87–3.62 (br s, 6H, H10), 2.39–2.23 (br s, 12H, H18), 2.09–1.99 (br s, 9H,
H9,16), 1.87–1.75 (br s, 12H, H18’). 13C{1H} NMR (DMSO-d6, δ): 165.3 (C7), 164.6 (C1),
157.0 (CAr), 143.7 (CAr), 135.2 (CAr), 134.2 (CAr), 133.9 (CAr), 124.0 (CAr), 121.5 (CAr),
119.3 (CAr), 113.8 (CAr), 88.20 (C17), 74.28 (C17’). 65.34 (C8), 40.01 (C10), 35.49 (C15),
30.19 (C9), 29.64 (C18), 29.36 (C16), 28.04 (C18’). FT-IR (νmax/cm-1): 1605 (C=N), 1586
(C=N). ESI-MS (positive ion-mode), (m/z) = 863.5983 [M + 2H]2+ (calcd 863.5079).
H2Osolubility/25℃ = 8.6 mg/mL.
2.8
General procedure for the hydroformylation experiments.
Hydroformylation studies were conducted in a 90 mL stainless steel pipe reactor. The
conditions of the study (temperature and syngas pressure) were based on our
previously reported work on the hydroformylation of 1-octene using analogous Rh(I)
catalyst precursors bearing N,O-bidentate ligands [53,56,58,65,66]. In a typical
experiment, the reactor was charged with the substrate (1-octene, 7.175 mmol),
internal standard (n-decane, 1.435 mmol) and toluene (5 mL). The catalyst precursor
(2.87 × 10−3 mmol of Rh (9); or 9.57 × 10−4 mmol of Rh (10) substrate : Rh ratio 2500
: 1) was dissolved in distilled water (5 mL) and introduced into the catalytic reactor.
Degassing of the reactor was carried out twice with nitrogen, and twice with syngas
(CO:H2 = 1:1), after which the reactor was heated and pressurised to the desired
temperature and syngas pressure. Samples were analysed after 4 h using a Perkin
Elmer Clarus 580 GC instrument equipped with a flame ionisation detector and SGE
Analytical capillary column (30 m x 0.25 mm). Authentic iso-octenes and aldehydes,
alcohols and n-octane were used to confirm the products. Inductively coupled plasma
optical emission spectroscopy experiments were conducted on a Varian 730 ES ICP-
Optical Emission Spectrophotometer. For the aqueous-biphasic recycling
experiments, the organic layer was decanted off, and a fresh organic layer containing
the substrate and internal standard was introduced to repeat the experiment. In a
typical “neat” hydroformylation experiment, the catalyst was immobilised in the
aqueous phase by dissolution in water, followed with the addition of the substrate and
internal standard only, (toluene excluded). The reaction was then conducted for 4 h
following a similar manner and optimum conditions (85 °C / 50 bar) as described vide
9

supra, after which the reactor was cooled to room temperature, and the contents
transferred to a vial. The organic products were then extracted with toluene and
immediately decanted from the catalyst-containing aqueous-phase and analysed by
gas chromatography. The catalyst-containing aqueous layer was taken through
repetitive cycles with assessment of the catalytic performance at each cycle.
3.
Results and discussion
3.1
Synthesis and characterisation of aryl ether-based mono- and trimeric amines
(5 and 6)
The precursor, Boc-protected amines 3 [63] and 4 were obtained via the Williamson
t
ether method by reacting butyl(3-bromopropyl)carbamate (2) [62] with either phenol
or 1,1,1-tris(4-hydroxyphenyl)ethane respectively, obtaining the products as white
solids in good yields (82 % (3), 84% (4)) (Scheme 1). Subsequently, the monomeric
amine 5 [64] and the new trimeric amine 6 were prepared in moderate yields (71% (5),
74% (6)) via a TFA-promoted Boc-deprotection of 3 and 4 respectively (Scheme 1).
The 1H NMR spectra for 5 and 6 (Fig. S2) both show the successful deprotection of
the amine. In addition, the 1H NMR spectra of 5 and 6 show diagnostic amine proton
signals (H1) at δ = 1.55 (5) and δ = 1.60 (6), and the diagnostic signal for the methyl
protons of the trimeric core of 6 at δ = 2.12 (H10). The infrared (IR) spectra of 5 and 6
reveal absorption bands at 3319 cm-1 and 3038 cm-1 for the ν(N–H) stretching
frequencies of 5, and 3361 cm-1 and 3282 cm-1 for the ν(N–H) stretching frequencies
of 6.
3.2
Synthesis of water-soluble aryl ether salicylaldimine mono- and trimeric ligands
(7 and 8)
The new water-soluble monomeric and trimeric ligands 7 and 8 were synthesised via
Schiff base condensation reactions of the respective amines 5 and 6 with the water-
soluble precursor, 5-sulfonato salicylaldehyde (1) [53] (Scheme 2). The ligands were
isolated as yellow solids in moderate yields (67% (7), 72% (8)) and possess good
solubility in water at room temperature (21.4 mg/mL (7), 19.6 mg/mL (8)). The 1H NMR
spectra of 7 and 8 show diagnostic signals for the imine protons (H8) appearing at δ =
8.64 in both 7 and 8, and the signals for the hydroxyl protons (H1) at δ = 13.70 (7) and
10

δ = 13.64 (8) (Fig. S3). The characteristic signal for the methyl protons of the trimeric
core of 8 is observed at δ = 2.03 (H17). The 13C{1H} NMR spectra show the signals for
the imine carbons at the characteristic region for both ligands, appearing at δ = 166.5
(7) and δ = 166.4 (8) as further evidence to formation of the imine-based ligands. The
infrared (IR) spectra of 7 and 8 show absorption bands characteristic of the imine
stretching frequencies ν(C=N) at 1649 cm-1 (7) and 1637 cm-1 with a shoulder at 1609
cm-1 (8), further corroborating formation of the iminium-based ligands. The ESI–MS
spectra (recorded in the negative ion-mode) show peaks associated with the anionic
complex in the absence of the Na ion(s).
Scheme 1. Preparation of the monomeric and trimeric amines 5 and 6: (i) phenol,
K2CO3, 18-C-6, acetone, reflux (20 h), (ii) TFA, CH2Cl2, 0-30 °C, (iii) 1,1,1-tris(4-
hydroxyphenyl)ethane, K2CO3, 18-C-6, acetone, reflux (20 h), (iv) TFA, CH2Cl2, 0-30
°C.
11

3.3
Synthesis of water-soluble aryl ether Rh(I)-salicylaldimine mono- and trinuclear
complexes (9 and 10)
The new water-soluble mononuclear (9) and trinuclear (10) Rh(I)-salicylaldimine
complexes were prepared via deprotonation of the ligands 7 and 8 respectively, using
NaH as the base, and subsequent complexation with the appropriate molar
equivalents of the Rh(I) dimer [RhCl(COD)]2, (COD = 1,5-cyclooctadiene), (Scheme
2). Both complexes were isolated as yellow solids in good yields (94% (9), 98% (10))
and show appreciably good solubility in water at room temperature (15.7 mg/mL (9),
8.6 mg/mL (10)). The reduced aqueous solubility upon complexation can be ascribed
to the increased hydrophobicity as a result of the cyclooctadiene entity, a trend also
reported previously in the literature for related compounds [53,56]. Evidence of
successful coordination of the Rh-metal to the ligands (7 and 8) is given by the upfield
shift of the signal for the imine protons, from δ = 8.64 (7 and 8) to δ = 8.28 (9) and δ =
8.29 (10) in the 1H NMR spectra of both complexes (Fig. S4 and S5), characteristic of
the synergistic effect involving M-to-L pi-donation and L-to-M sigma donation. Both
spectra do not show the signal for the hydroxyl protons previously observed in 7 and
8, indicative of successful deprotonation of the ligands and subsequent ligand
coordination to the metal centre in a bidentate manner. The signals for the protons of
the cyclooctadiene entity appear as broadened signals and integrate for the
appropriate number of protons in both the mononuclear (9) and trinuclear (10)
complexes. The olefinic carbon atoms of the cyclooctadiene moiety are observed as
doublets in the 13C{1H} NMR spectra, attributed to coupling to the NMR-active rhodium
metal centre (trans to N: 1JRh–C = 12 Hz, and trans to O: 1JRh–C = 14 Hz) and correlate
well with those in literature for related (η2:η2–COD)-Rh–Schiff base complexes [53,67].
The infrared spectra of both complexes show shifts of the imine absorption bands to
lower wavenumbers upon coordination of the ligands to the metal, substantiating the
coordination of the imine nitrogen, as has been reported for similar compounds in the
literature [7,53,56,58]. The ESI–MS spectral data correlate with the proposed
structures.
12

Scheme 2. Preparation of the water-soluble mono- and trinuclear complexes 9 and
10: (i) 5 (1 eq.), MeOH, reflux (4 h), (ii) [Rh(COD)Cl]2 (0.5 eq.), NaH, CH2Cl2, r.t. (24
h), (iii) 6 (0.33 eq.), MeOH, reflux (20 h), (iv) [Rh(COD)Cl]2 (1.5 eq.), NaH, CH2Cl2, r.t.
(24 h).
13

3.4
Aqueous biphasic hydroformylation of 1-octene: preliminary screening using
precatalyst 9
Preliminary hydroformylation experiments were performed using the water-soluble,
mononuclear catalyst precursor 9 and 1-octene as the model substrate. The conditions
of this study (75–105 °C, and 20–50 bar) were based on our previously reported work
of the hydroformylation of 1-octene using analogous Rh(I) catalyst precursors
containing N,O-bidentate ligands [56,58,65,66]. Under these conditions, near-
quantitative conversion of the substrate was attained, indicative of the presence of
good interfacial interaction between the catalytic active species and the substrate (Fig.
S6). A considerable increase in the total aldehydes is observed when the pressure is
varied at a constant temperature of 85 °C, from 63% (30 bar) to 90% (50 bar) and this
could be attributed to the increased CO concentration in the system, limiting the
isomerisation of the olefin while accelerating the CO migration step of the
hydroformylation cycle (Fig. 2). A further increase in temperature and pressure (105
°C / 50 bar), has no significant effect in the chemoselectivity for aldehydes. At this high
temperature and pressure (105 °C / 50 bar), there exists a strong competition between
thermally promoted isomerisation and excess-CO promoted hydroformylation.
However, with excess CO present in the system, each isomerisation cycle can be
interrupted by the formation of the corresponding metal-acyl complex through insertion
of CO into the metal-alkyl bond, and irreversibly converting the intermediates to
aldehydes [66,68]. The catalyst precursor 9 shows good regioselectivity for the linear
aldehyde, nonanal (70%), at the mildest conditions of this study (70 °C and 30 bar),
(Fig. 3). An increase in temperature at a constant pressure (75–105 °C / 30 bar) results
in a gradual decrease in the linear aldehyde, and subsequent increase in the branched
aldehydes. In addition to the generally accepted primary route leading to the formation
of branched aldehydes via the Markovnikov addition of the olefin to the metal-alkyl
bond, the increase in branched aldehydes can also be ascribed to the subsequent
hydroformylation of the iso-octenes that form as a result of temperature-promoted
preferential double bond isomerisation of the olefin. The increase in pressure at
constant temperature results in a gradual increase in activity, for example, from
426 h-1 (75 °C / 30 bar) to 458 h-1 (75 °C / 50 bar) (Table S1, entries 1 and 3). Overall,
the optimum conditions were established at 85 °C / 50 bar, based on the good activity
(557 h-1), chemoselectivity for aldehydes (90%) and conversion of the substrate (99%).
14

Figure 2. Chemoselectivity for aldehydes with temperature and pressure, at constant
time (4 h) for complex 9.
Figure 3. Regioselectivity with temperature and pressure, at constant time (4 h) for
complex 9.
15

3.5
Catalytic evaluation using precatalyst 10 and mercury poisoning experiments
The water-soluble trinuclear complex 10 was successfully evaluated in the
hydroformylation of 1-octene using the previously established optimal conditions of
temperature and pressure (85 °C / 50 bar) (Table 1). The trinuclear precatalyst shows
good conversion (99%, entry 2) of 1-octene after 4 h, as well as comparable aldehyde
selectivity (88%, entry 2) to that of the mononuclear complex (90%, entry 1).
Comparable regioselectivities and activities are also observed after 4 h with both
catalyst precursors (n/iso = 55:45 (9 and 10); TOF = 557 h-1 (9) and 555 h-1 (10)
respectively).
Mercury drop experiments are useful for differentiating between homogeneous and
heterogeneous catalysis mediated by metal nanoparticles. When a drop of mercury
was added to the reaction mixture, a decline in the catalytic performance of both
precatalysts 9 and 10 was observed, indicative of a nanoparticle-mediated
transformation (Table 1, entries 3 and 4). This would further substantiate the trend
observed in Fig. 3 at even higher temperatures (95 °C and 105 °C), wherein
complimentary to the Markovnikov addition, the observed regioselectivity towards
branched aldehydes with increase in temperature can be a result of hydroformylation
of iso-octenes formed from a combination of a temperature-promoted isomerisation of
the substrate and a nanoparticle-influenced formation of iso-octenes. The observed
behaviour of precatalysts 9 and 10 upon addition of mercury is similar to our previously
reported studies on analogous water-soluble salicylaldimine Rh(I)-complexes
[56,58,65].
Table 1 Hydroformylation of 1-octene using catalyst precursors 9 and 10 for 4 h.a
entry Complex Temp. Pressure Conv.
Iso-
n/isoc TOF
(°C)
(bar)
(%) Aldehydes octenes
(h-1)d
(%)b
90
(%)
10
12
33
34
1
2
3
4
9
85
85
85
85
50
50
50
50
99
99
73
78
55:45 557
55:45 555
71:29 305
70:30 323
10
9e
88
67
10e
66
aReactions carried out with (CO–H ) (1:1) in H2O:Toluene (5:5 mL) with 7.175 mmol of 1‐octene and 2.87 × 10−3 mmol of
2
Rh catalyst for 9, and 9.57 × 10−4 mmol of Rh catalyst for 10. The reactor was purged with syngas. GC conversions obtained
using n-decane as an internal standard in relation to authentic standard iso‐ octenes and aldehydes. bTotal aldehydes formed
(from octene and iso-octenes converted), which includes the primary aldehyde product, nonanal, and branched aldehydes.
16
d
cThe molar ratio of primary linear aldehyde product nonanal (n) and branched aldehydes (iso) formed. TOF = (mmol of
aldehydes per mmol of Rh)/time. eMercury poisoning experiments. Average error estimate = ± 0.58.

The substrate and product-distribution profiles of the two catalyst precursors 9 (Fig.
4a) and 10 (Fig. 4b), as a function of time using the optimised conditions (85 °C / 50
bar) indicate that the trinuclear complex has superior performance over the
mononuclear complex within the 1st hour of the reaction. The trinuclear complex shows
2.19 mol% (10, Fig. 4b) of 1-octene present in the system compared to 27.22 mol%
for the mononuclear complex (9, Fig. 4a) over the same time period. The mononuclear
complex reaches closely similar quantities only after 2 h (1.66 mol%). Such catalytic
performance alludes to a slight positive dendritic effect of the trinuclear complex 10
over the mononuclear complex 9, owing to the inherent multiple active sites of 10. This
observation is further complemented by the higher ratio of the linear aldehyde nonanal
at the 1 h interval, when using the trinuclear complex 10 (47.16 mol%, Fig. 4b)
compared to the mononuclear complex 9 (34.68 mol%, Fig. 4a). At this stage, the
bulky nature of the trinuclear complex can be viewed as responsible for the observed
ratios of linear to branched aldehydes of 10 over the ratio depicted by the less bulky
precatalyst 9. However, branched aldehydes are observed to increase in the system
with time and this is ascribed to the ease of hydroformylation of the iso-octenes to
branched aldehydes, wherein at this stage, there exists no competition for the catalytic
active sites between the depleted substrate and the iso-octenes present in the system.
Figure 4. Substrate and product-distribution-time profiles with the mono- 9 (a) and
trinuclear precatalysts 10 (b) at optimum conditions (85 °C / 50 bar).
17

3.6
Recyclability of the catalysts and Rh leaching studies
The catalyst precursors 9 and 10 were successfully recycled for at least 5 times in the
hydroformylation of 1-octene, with periodic decanting of the organic layer and reuse
of the catalyst-containing aqueous layer. The catalyst precursors gave good to
moderate conversion with a gradual decline observed with each cycle under optimum
conditions (85 °C / 50 bar / 4 h.), (Fig. S7). This is attributed to a combination of low
concentration of the active metal species in the aqueous layer, consistent with
leaching of the metal catalyst to the organic phase, as well as catalyst degradation,
often associated with repetitive use of the same catalyst containing phase. This has
been observed in the literature for analogous water-soluble salicylaldimine Rh(I)
complexes [56,58]. Inductively coupled plasma optical emission spectroscopy (ICP-
OES) analysis referenced to the initial metal loading shows an overall rhodium metal
loss of 41% (9) and 57% (10) from the aqueous layer, supporting the observed steady
decrease in 1-octene conversions for both catalyst precursors 9 and 10. Both catalyst
precursors show good to moderate chemoselectivity for aldehydes ( <90% and >60%)
(Fig. 5). It should be noted that no hydrogenation products (alkanes or alcohols) were
observed at any instance during the recycling of the catalysts, indicative of good
selectivity. The catalyst precursors 9 and 10 display a decrease in the activity from the
1st cycle to the 2nd cycle (557 h-1 to 384 h-1 (9); and 555 h-1 to 367 h-1 (10)) (Fig. 6).
This is consistent with the observed chemoselectivity and conversion of both catalysts.
A gradual decrease in activity from the 3rd cycle (ca. 300 h-1) is observed for both
catalysts 9 and 10, possibly due to a combination of active metal degradation with
constant reuse, as well as active metal leaching to the product-containing organic
phase, as attested to by ICP-OES experiments.
18

Figure 5. Chemoselectivity for aldehydes during aqueous-toluene biphasic
recyclability studies in the hydroformylation of 1-octene.
Figure 6. Activity of the complexes during aqueous-toluene biphasic recyclability
studies in the hydroformylation of 1-octene.
We have previously reported good recyclability of water-soluble Rh(I) complexes in
the absence of the organic solvent toluene, providing the potential to circumvent the
leaching of the catalyst during the hydroformylation experiments [69]. In light of that
study, further hydroformylation studies were conducted in a neat aqueous
19

environment. It is interesting to note that good conversion of 1-octene (over 70%) is
maintained by the trinuclear complex 10 throughout the 5 cycles (Fig. S8). The gradual
decline in conversion after the 1st cycle can be ascribed to slight degradation of the
catalyst with constant heating, as the absence of the organic solvent rules out any
excessive leaching of the catalyst to the organic products. However, the mononuclear
complex 9 shows a drastic decline in conversion of the substrate, from 83% (1st cycle)
to 32% (5th cycle). This can be attributed to increased susceptibility to metal
degradation of complex 9, as evidenced by the formation of black particulate matter of
the initial complex with each cycle. This black mass is hardly observed with complex
10, indicative of good stability afforded by the larger macromolecular structure. The
two catalysts maintain good to moderate chemoselectivity for aldehydes (> 65%)
throughout the cycles, retaining consistency in aldehyde selectivity (Fig. 7). A
considerable decline in the activity of 9 is observed with each cycle, from 437 h-1 in
the 1st cycle to 139 h-1 in the 5th cycle (Fig. 8). On the other hand, the trinuclear complex
10 displays appreciably good activity throughout repetitive cycles when compared to
the experiments conducted in the presence of toluene (Fig. 8 vs Fig. 6). This
corroborates that leaching may have been the primary contributor to the observed
decline in activity with the water-toluene biphasic approach depicted in Fig. 6, whereas
active metal degradation is a secondary contributor. In the “neat” studies, inductively
coupled plasma optical emission spectroscopy (ICP-OES) analysis referenced to the
initial metal loading shows an overall rhodium metal loss of 69% (9) and 38% (10).
The affinity of precatalyst 10 for the organic phase (leaching) that is observed in
studies involving toluene could be attributed to the slightly reduced water-solubility of
10 (H2Osolubility/25℃ = 8.6 mg/mL) compared to 9 (H2Osolubility/25℃ = 15.7 mg/mL), resulting
from a combination of the hydrophobic tris(hydroxyphenyl)ethane (THPE) core with
the cyclooctadiene moieties on each metal centre of 10.
20

Figure 7. Chemoselectivity of the complexes during aqueous recyclability studies in
the hydroformylation of 1-octene.
Figure 8. Activity of the complexes during aqueous recyclability studies in the
hydroformylation of 1-octene.
21

3.7
Substrate scope
Both the mononuclear (9) and trinuclear (10) complexes were evaluated in the
hydroformylation reaction against styrene. The aqueous-toluene biphasic experiments
were conducted under the optimised conditions obtained in the hydroformylation of 1-
octene, using 9 (85 °C / 50 bar / 4 h.) at a substrate-to-catalyst molar ratio of 2500:1.
Both complexes evaluated in the hydroformylation of styrene, displayed good (10) to
excellent (9) transformation of the substrate (77% and 98% respectively) (Table 2). It
should be noted that no hydrogenation products (ethylbenzene or alcohols) were
observed using catalyst precursors 9 and 10, indicative of the good absolute aldehyde
chemoselectivity of these complexes. The slightly reduced conversion (77%) and
activity (479 h-1) on styrene by precatalyst 10 can be attributed to the combined steric
effects of the substrate and the bulky THPE anchored precatalyst, leading to poor
substrate accessibility of the active metal sites. The relatively less bulky mononuclear
complex 9 provides ease of access to the active metal centre, leading to improved
conversions (98%) and activity (615 h-1) of the substrate. The hydroformylation of
styrene is a very attractive model in the transformation of vinylarenes, owing to the
usually absolute chemoselectivity for aldehydes, as well as the wide potential
applications of the resultant branched and linear aldehydes [45,70].
Table 2 Aqueous biphasic hydroformylation of styrene with 9 and 10 for 4 ha.
Complex Temp.
(°C)
Pressure
(bar)
Conv.
(%)
Aldehydes
(%)b
n/isoc
TOF
(h-1)d
9
85
85
50
50
98
77
100
100
26:74
26:74
615
479
10
aReactions carried out with (CO–H2) (1:1) in H2O:Toluene (5:5 mL) with 7.175 mmol of styrene and 2.87 × 10−3 mmol of Rh
catalyst for 9, and 9.57 × 10−4 mmol of Rh catalyst for 10. The reactor was purged with syngas. GC conversions obtained
b
using n-decane as an internal standard in relation to authentic standard aldehydes. Total aldehydes formed, which includes
c
the primary aldehyde product, nonanal, and the branched aldehyde. The molar ratio of primary linear aldehyde product (n)
and branched aldehyde (iso) formed. dTOF = (mmol of aldehydes per mmol of Rh)/time. Average error estimate = ± 0.62.
4.
Conclusions
New water-soluble, salicylaldimine aryl ether Rh(I) mononuclear and trinuclear
complexes have been synthesised. The complexes were successfully evaluated in
22

aqueous-toluene biphasic hydroformylation experiments and both catalyst precursors
show near quantitative catalytic conversion of 1-octene, good activities and high
chemoselectivity for aldehydes under the mild optimum conditions (85 °C / 50 bar / 4
h.). This overall catalytic performance is suggestive of good interaction of the substrate
with the catalysts at the interface of the two immiscible solvents (water and toluene)
under hydroformylation conditions. Mercury poisoning experiments showed that the
observed performance was to a dual extent, enhanced by the presence of Rh-
nanoparticles in the system, a phenomenon that we have previously observed and
reported on when using related catalyst precursors [56,58,65]. The reusability of the
active metal species was successfully conducted over 5 cycles, with a gradual decline
in catalytic performance of both catalysts. The decline was primarily attributed to
catalyst leaching to the organic layer, and partly due to catalyst degradation with
constant subjecting to heat and pressure over the cycles. This has been observed and
reported on previously in the literature [56,58]. Inductively coupled plasma optical
emission spectroscopy (ICP-OES) experiments showed a 41% metal loss from the
aqueous layer for 9, and a 57% metal loss for 10. The catalyst precursors were also
successfully evaluated in the “neat” aqueous hydroformylation reactions of 1-octene
and showed good recyclability over 5 cycles. “Neat” monophasic catalytic experiments
were successfully performed in water using the catalyst precursors 9 and 10, as part
of the attempts to alleviate the leaching of the catalyst precursors to the organic layer.
The dendrimer-stabilised trinuclear catalyst 10 showed improved recyclability in the
aqueous hydroformylation experiments, while the mononuclear precatalyst 9 showed
a reduced overall performance. This was attributed to degradation of the mononuclear
precatalyst, as evidenced by the black particulate matter after the reaction, which is
indicative of catalyst degradation. The versatility of the catalyst precursors (9 and 10)
was evaluated in the hydroformylation of styrene under aqueous biphasic
environment. Both catalyst precursors showed good catalytic performance by
displaying sole chemoselectivity for aldehydes (no hydrogenation products present).
The mononuclear complex 9 showed better activity (615 h-1) than the trinuclear
complex 10 (479 h-1). This was attributed to the limited access of the active metal
centres of the sterically hindered dendritic complex 10 by the relatively bulky styrene
substrate.
23

CRediT author statement
Shepherd Siangwata: Investigation, Data Curation, Writing-Original Draft, Validation,
Formal Analysis, Neill J. Goosen: Supervision, Resources, Writing-Review and Editing.
Gregory S. Smith: Conceptualisation, Resources, Writing-Review and Editing, Supervision,
Project Administration, Funding Acquisition
Declaration of interests
The authors declare that they have no known competing financial interests or personal relationships
that could have appeared to influence the work reported in this paper.
Acknowledgements
We would like to thank the University of Cape Town (UCT), Stellenbosch University,
The Department of Science and Technology of South Africa and NRF-DST Centre of
Excellence in Catalysis – c*change for financial support.
References
[1]
B. Cornils, W.A. Herrmann, I.T. Horvath, W. Leitner, S. Mecking, H. Olivier-
Bourbigou, D. Vogt, in: Multiph. Homog. Catal., 2nd ed., Wiley-VCH Verlag
GmbH & Co. KGaA, Weinheim, 2005, pp. 3–21.
[2]
[3]
[4]
A.E.C. Collis, I.T. Horváth, Catal. Sci. Technol. 1 (2011) 912–919.
W. Keim, Green Chem. 5 (2003) 105–111.
T. Rösler, T.A. Faßbach, M. Schrimpf, A.J. Vorholt, W. Leitner, Ind. Eng.
Chem. Res. 58 (2019) 2421–2436.
[5]
[6]
X. Jin, J. Feng, H. Song, J. Yao, Q. Ma, M. Zhang, C. Yu, S. Li, S. Yu, Green
Chem. 21 (2019) 3583–3596.
Y. Huang, E. Perperi, G. Manos, D.J. Cole-Hamilton, J. Mol. Catal. A Chem.
24

210 (2004) 17–21.
[7]
[8]
[9]
L. Maqeda, B.C.E. Makhubela, G.S. Smith, Polyhedron 91 (2015) 128–135.
A.P. Dobbs, M.R. Kimberley, J. Fluor. Chem. 118 (2002) 3–17.
L.P. Barthel-Rosa, J.A. Gladysz, Coord. Chem. Rev. 190–192 (1999) 587–
605.
[10] R.H. Fish, Chem. Eur. J. 5 (1999) 1677–1680.
[11] D.F. Foster, D.J. Adams, D. Gudmunsen, A.M. Stuart, E.G. Hope, D.J. Cole-
Hamilton, Chem. Commun. (2002) 722–723.
[12] W. Chen, L. Xu, J. Xiao, Chem. Commun. (2000) 839–840.
[13] D.J. Adams, D.J. Cole-Hamilton, E.G. Hope, P.J. Pogorzelec, A.M. Stuart, J.
Organomet. Chem. 689 (2004) 1413–1417.
[14] S.L. Desset, U. Hintermair, D.J. Cole-Hamilton, Top. Catal. 53 (2010) 963–
968.
[15] J. Walkowiak, G. Franciò, W. Leitner, in: Appl. Homog. Catal. with Organomet.
Compd., 3rd ed., Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, 2018, pp.
1221–1258.
[16] M. Haumann, P. Wasserscheid, in: Appl. Homog. Catal. with Organomet.
Compd., 3rd ed., Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, 2018, pp.
999–1067.
[17] D.R. Palo, C. Erkey, Ind. Eng. Chem. Res. 38 (1999) 3786–3792.
[18] A.C.J. Koeken, L.J.P. Van Den Broeke, B.J. Deelman, J.T.F. Keurentjes, J.
Mol. Catal. A Chem. 346 (2011) 1–11.
[19] X. Jin, J. Feng, Q. Ma, H. Song, Q. Liu, B. Xu, M. Zhang, S. Li, S. Yu, Green
Chem. 21 (2019) 3267–3275.
[20] Q. Lin, W. Jiang, H. Fu, H. Chen, X. Li, Appl. Catal. A Gen. 328 (2007) 83–87.
[21] Y. Xu, Y. Wang, Y. Zeng, J. Jiang, Z. Jin, Catal. Letters 142 (2012) 914–919.
25

[22] F. Joó, Á. Kathó, J. Mol. Catal. A Chem. 116 (1997) 3–26.
[23] E.A. Karakhanov, A.L. Maksimov, Russ. J. Gen. Chem. 79 (2009) 1370–1383.
[24] B. Cornils, Org. Process Res. Dev. 2 (1998) 121–127.
[25] L.C. Matsinha, S. Siangwata, G.S. Smith, B.C.E. Makhubela, Catal. Rev. - Sci.
Eng. 61 (2019) 111–133.
[26] J. Hagen, in: Ind. Catal., 2nd ed., Wiley-VCH Verlag GmbH & Co. KGaA,
Weinheim, 2006, pp. 9–14.
[27] C. Li, W. Wang, L. Yan, Y. Ding, Front. Chem. Sci. Eng. 12 (2018) 113–123.
[28] M. Lombardo, C. Trombini, in: RSC Green Chem. Ser. Eco-Friendly Synth.
Fine Chem., 1st ed., Royal Society of Chemistry, 2009, pp. 1–79.
[29] I. Vural Gürsel, T. Noel, Q. Wang, V. Hessel, Green Chem. 17 (2015) 2012–
2026.
[30] D.J. Cole-Hamilton, R.P. Tooze, in: Catayst Sep. Recover. Recycl., 1st ed.,
Springer, Dordrecht, 2006, pp. 4–35.
[31] V.S. Shende, V.B. Saptal, B.M. Bhanage, Chem. Rec. 19 (2019) 2022–2043.
[32] X. Jin, J. Feng, S. Li, H. Song, C. Yu, K. Zhao, F. Kong, Mol. Catal. 475 (2019)
110503.
[33] P.T. Anastas, J.B. Zimmerman, Environ. Sci. Technol. 37 (2003) 95–101.
[34] R.A. Sheldon, Green Chem. 7 (2005) 267–278.
[35] R.A. Sheldon, Curr. Opin. Green Sustain. Chem. 18 (2019) 13–19.
[36] R.L. Pruett, J. Chem. Educ. 63 (1986) 196–198.
[37] G.D. Frey, J. Organomet. Chem. 754 (2014) 5–7.
[38] B. Cornils, E.G. Kuntz, J. Organomet. Chem. 502 (1995) 177–186.
[39] C.W. Kohlpaintner, R.W. Fischer, B. Cornils, Appl. Catal. A Gen. 221 (2001)
219–225.
26

[40] A. Bényei, F. Joó, J. Mol. Catal. 58 (1990) 151–163.
[41] J. Halpern, Pure Appl. Chem. 73 (2001) 209–220.
[42] T. Vanbesien, F. Hapiot, E. Monflier, Lipid Technol. 25 (2013) 175–178.
[43] B. Cornils, W.A. Herrmann, M. Rasch, Angew. Chem. Int. Ed. 33 (1994) 2144–
2163.
[44] B. Cornils, in: Top. Curr. Chem., Springer-Verlag Berlin Heidelberg, Berlin,
Germany, 1999, pp. 133–149.
[45] R. Franke, D. Selent, A. Bꢀrner, Chem. Rev. 112 (2012) 5675–5732.
[46] G.T. Whiteker, C.J. Colbey, Top Organomet Chem 42 (2012) 35–46.
[47] B. Cornils, J. Mol. Catal. A Chem. 143 (1999) 1–10.
[48] C. De Rumpa, S. Sumanta, Res. Chem. Intermed. 39 (2013) 3463–3474.
[49] H. Fu, M. Li, H. Mao, Q. Lin, M. Yuan, X. Li, H. Chen, Catal. Commun. 9
(2008) 1539–1544.
[50] L. Obrecht, P.C.J. Kamer, W. Laan, Catal. Sci. Technol. 3 (2013) 541–551.
[51] J. Leblond, J. Potier, S. Menuel, H. Bricout, D. Landy, S. Tilloy, E. Monflier, F.
Hapiot, Catal. Sci. Technol. 7 (2017) 3823–3830.
[52] C. De Rumpa, S. Sumanta, A. Ghosh, K. Mukherjee, S. Bhattacharyya,
Sekhar, B. Saha, Res. Chem. Intermed. 39 (2013) 3463–3474.
[53] E.B. Hager, B.C.E. Makhubela, G.S. Smith, Dalton Trans. 41 (2012) 13927–
35.
[54] Y. Brunsch, A. Behr, Angew. Chem. Int. Ed. 52 (2013) 1586–9.
[55] S. Paganelli, M. Marchetti, M. Bianchin, C. Bertucci, J. Mol. Catal. A Chem.
269 (2007) 234–239.
[56] S. Siangwata, N. Baartzes, B.C.E. Makhubela, G.S. Smith, J. Organomet.
Chem. 796 (2015) 26–32.
27

[57] B.R. James, P.W.N.M. Van Leeuwen, C. Claver, in: Rhodium Catalyzed
Hydroformylation, 1st ed., Kluwer Academic Publishers, New York, 2002, pp.
6–277.
[58] L.C. Matsinha, S.F. Mapolie, G.S. Smith, Dalton Trans. 44 (2015) 1240–1248.
[59] J. Park, S. Hong, Chem. Soc. Rev. 41 (2012) 6931–6943.
[60] P.J. Deuss, R. Denheeten, W. Laan, P.C.J. Kamer, Chem. Eur. J. 17 (2011)
4680–4698.
[61] M.H. Pørez-Temprano, J.A. Casares, P. Espinet, Chem. Eur. J. 18 (2012)
1864–1884.
[62] H. Li, M.A. Hao, L. Wang, W. Liang, K. Chen, Org. Prep. Proced. Int. 41 (2009)
301–307.
[63] P. Hidalgo Ramos, P. Saisaha, J.A.A.W. Elemans, A.E. Rowan, R.J.M. Nolte,
Eur. J. Org. Chem. (2016) 4487–4495.
[64] E. Robinson, E. Leung, A.M. Matuszek, N. Krogsgaard-Larsen, D.P. Furkert,
M.A. Brimble, A. Richardson, J. Reynisson, Med. Chem. Commun. 6 (2015)
239–246.
[65] S. Siangwata, S. Chulu, C.L. Oliver, G.S. Smith, Appl. Organomet. Chem. 31
(2017) e3593.
[66] S. Siangwata, N.C.C. Breckwoldt, N.J. Goosen, G.S. Smith, Appl. Catal. A
Gen. 585 (2019) 117179.
[67] C. Janiak, A.-C. Chamayou, A.K.M.R. Uddin, M. Uddin, S.K. Hagen, M.
Enamullah, Dalton Trans. 19 (2009) 3698–3709.
[68] M. Vilches-Herrera, L. Domke, A. Börner, ACS Catal. 4 (2014) 1706–1724.
[69] N.N. Omosun, G.S. Smith, Eur. J. Inorg. Chem. (2019) 2558–2564.
[70] O.L. Eliseev, T.N. Bondarenko, S.N. Britvin, P.P. Khodorchenko, A.L. Lapidus,
Mendeleev Commun. 28 (2018) 264–266.
28