Catalytic isomerization-hydroformylation of olefins by rhodium salicylaldimine pre-catalysts

Article abstract of DOI:10.1039/d0nj01970d

A series of new Schiff-base rhodium(i) water-soluble complexes (C1-C3), were prepared and characterized. These complexes served as catalyst precursors for the hydroformylation of 1-octene and resulted in excellent substrate conversions (>98%) with 100% chemoselectivities to aldehydes, under mild conditions. Notably, good regioselectivities towards branched aldehydes were observed clearly demonstrating the catalysts’ ability in thermodynamically favoured isomerization followed by hydroformylation (n/iso ratio ranging between 0.7-1.2). Interestingly, catalystC1uniquely promoted contra-thermodynamic isomerization of 2-octene to 1-octene with up to 50% conversion. The efficacy of catalystC1was further evaluated in the hydroformylation of longer chain olefins (C₁₀-C₁₂), methyl acrylate, ethyl acrylate and styrene. The catalyst displayed conversions >99% with the long chain substrates and much lower conversions with the acrylates. These water-soluble (pre)catalysts were recycled up to three times with no significant loss in catalytic activity and selectivity. Mercury poisoning tests were conducted and the experiments revealed that the conversion of the substrates into aldehydes was due to molecular active catalysts and not as a result of colloidal particles that could have formedin situthrough the decomposition of the catalyst precursor. Finally, the molecular catalyst responsible for activity was established using preliminary computational calculations.

Full text of DOI:10.1039/d0nj01970d

View Article Online
View Journal
NJC
New Journal of Chemistry
Accepted Manuscript
A journal for new directions in chemistry
This article can be cited before page numbers have been issued, to do this please use: P. N. Sekoto, T. M.
Magengenene, L. C. Matsinha, R. Tia, J. Darkwa and B. C.E. Makhubela, New J. Chem., 2020, DOI:
10.1039/D0NJ01970D.
Volume 42
This is an Accepted Manuscript, which has been through the
Number
6
21 March 2018
Pages 3963-4776
Royal Society of Chemistry peer review process and has been
accepted for publication.
NJC
New Journal of Chemistry
rsc.li/njc
A journal for new directions in chemistry
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
shall the Royal Society of Chemistry be held responsible for any errors
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
ISSN 1144-2546
PAPER
Chechia Hu, Tzu-Jen Lin et al.
Yellowish and blue luminescent graphene oxide quantum dots
prepared via a microwave-assisted hydrothermal route using
H2O2 and KMnO4 as oxidizing agents
rsc.li/njc

Page 1 of 30
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
7
8
9
Catalytic isomerization-hydroformylation of olefins
by rhodium salicylaldimine pre-catalysts
Pamela N. Sekoto,a Tseliso M. Magengenene,a Leah C. Matsinha,a Richard Tia,b James Darkwa
a and Banothile C. E. Makhubela*a
aResearch Centre for Synthesis and Catalysis, Department of Chemical Sciences, University of Johannesburg,
Auckland Park, 2006, South Africa.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
bDepartment of Chemistry, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana
Abstract
A series of new Schiff-base rhodium(I) water-soluble complexes (C1-C3), were prepared and
characterized. These complexes served as catalyst precursors for the hydroformylation of 1-
octene and resulted in excellent substrate conversions (>98%) with 100% chemoselectivities to
aldehydes, under mild conditions. Notably, good regioselectivities towards branched aldehydes
were observed clearly demonstrating the catalysts‘ ability in thermodynamically favoured
isomerization followed by hydroformylation (n/iso ratio ranging between 0.7-1.2). Interestingly,
catalyst C1 uniquely promoted contra-thermodynamic isomerization of 2-octene to 1-octene with
up to 50% conversion. The efficacy of catalyst C1 was further evaluated in the hydroformylation
of longer chain olefins (C10-C12), methyl acrylate, ethyl acrylate and styrene. The catalyst
displayed conversions >99% with the long chain substrates and much lower conversions with the
acrylates. These water-soluble (pre)catalysts were recycled up to three times with no significant
loss in catalytic activity and selectivity. Mercury poisoning tests were conducted and the
experiments revealed that the conversion of the substrates into aldehydes was due to molecular
active catalysts and not as a result of colloidal particles that could have formed in situ through
the decomposition of the catalyst precursor. Finally, the molecular catalyst responsible for
activity was established using preliminary computational calculations.
______________________________________________________________________
Email: bmakhubela@uj.ac.za
Keywords: isomerization-hydroformylation, rhodium-catalysed, aqueous-biphasic, green
chemistry; catalyst recycling; 1-octene.
1

New Journal of Chemistry
Page 2 of 30
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
Introduction
5
6
7
8
The hydroformylation (oxo) reaction is an important transition metal-catalysed reaction that
converts olefins to aldehydes in the presense of syngas. The first hydroformylation catalysts were
cobalt based, and operated at 200 to 450 bars pressure and at temperatures between 140 ˚C and
180 ˚C. It was later discovered that rhodium catalysts were better, since they had the unique
ability to catalyse the reaction at low temperature and pressure.[1–5]Aldehydes produced in this
process are used in the production of detergents, fragrances and solvents. These aldehydes are
also important starting materials for the preparation of a wide range of phamaceuticals, fine
chemicals and agrochemicals. Over 90% of the world‘s demand for oxo chemicals comes from
Europe, Asia and North America as reported in 2012. [2,6-10]
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Rhodium is the most active metal for catalysing this reaction but the challenge is that it is a very
expensive metal. As a result, researchers in academia and industry have come up with various
approaches to recover this precious metal during catalysis. Biphasic approaches have been
explored. In this approach, the catalyst is immobilised in one of the phases whilst the substrate
and products are contained in the other phase. This allows easy recovery of the expensive
catalyst from the products and hence the recovered catalyst can be reused in another catalytic
cycle. Such systems include combinations of water with either organic solvent, fluorous solvent
or ionic liquids with organic solvent. Apart from that approach, catalysts may be immobilised on
a solid support which results in a heterogeneous catalytic system. [10-20]
The aqueous biphasic hydroformylation reaction has been successfully conducted on an
industrial scale in the RuhrСhemie/Rhone-Poulenc process (OXEA Process).[2,21-22] In this
system, a Rh/TPPTS catalyst is employed. The advantage of this system is that it employs water,
which is considered a green solvent because it is non-toxic, non-flammable, easily accessible and
a cheap alternative that is readily available.[23] The aqueous biphasic hydroformylation of olefins
has been widely explored with short chain olefins and poor solubility of the longer chain
substrates results in suppressed catalytic activities.
In light of this, we developed new water-soluble organometallic complexes and evaluated their
efficacy as catalyst precursors for aqueous biphasic hydroformyaltion of long chain olefins (C8-
2

Page 3 of 30
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
C12). We also extended the study to acrylates and styrene as substrates, and further investigated
catalyst recyclability and stability.
7
8
Results and discussion
9
Synthesis and characterization of ligands L1 and L2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Ligands (L1-L3) were synthesised by reacting substituted salicylaldehydes with 4-aminobenzoic
acid (Scheme 1) with 4-aminobenzoic acid. In order to prepare L2, the sulfonated
salicylaldehyde precursor was first prepared according to a protocol reported in literature.[24]
Ligands L1 - L3 were obtained as yellow to orange solids in relatively good yields of between
64-89%, and are highly thermally stable (melting points in the range of 195 ˚C to 304 C˚.
Furthermore, the ligands L1 and L2 are insoluble in common solvents (toluene, MeOH, EtOH,
DCM, EtOAc) at room temperature but dissolved readily in water at ≥ 65 ˚C. This may be due to
better hydrogen bonding interactions between the carboxylic acid functional group of the ligands
and water at temperatures above 65 ˚C.
Scheme 1: Outline for the synthesis of L1-L3.
The 1H NMR spectra of L1–L3 display expected imine proton signals upfield in the region of ≈
9.0 ppm. The aromatic proton signals in these ligands are seen in the region of 6.73 to 8.02 ppm.
Notably, the aromatic protons of the para-substituted phenyl ring appear as two doublets
integrating for 2 protons each in all the ligands. The carboxylic acid and phenolic proton signals
1
were not always seen in the H NMR spectra, possibly due to the deuterated solvent used,
however the phenolic and carboxylic acid protons in L1 and L3 gave rise to singlets at 12.55
ppm (L1, OH) and 12.71 ppm (L3, OH) and 12.95 (L3, COOH) ppm.
13
The C{1H} NMR spectra of ligands L1 – L3 all are similar and display all expected signals.
The carbonyl carbon of the carboxylic acid groups (COOH) appears downfield in the region of ≈
3

New Journal of Chemistry
Page 4 of 30
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
7
8
9
167 ppm for all the ligands. Slightly upfield to this are the imine (C=N) carbon signals, which
resonate at ≈ 160 ppm. The aromatic carbon signals are seen in the region of 113.01 ppm to
134.30 ppm. The carbon signal directly bonded to the oxygen atom of the phenol group is more
deshielded than the other aromatic carbon signals and resonates at 153.60 ppm (L1), 153.10 ppm
(L2) and 153.61 ppm (L3).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The high resolution electrospray ionisation mass spectra results for ligands L1 - L3 validated the
integrity of the ligands by displaying fragmentation peaks at m/z = 285.0633 (L1, [M+H]+), m/z
= 317.1280 (L2, [M-Na-2H]+), m/z = 242.0809 (L3, [M+H]+).
Synthesis and characterization of Rh(I) complexes C1 – C3
The salicylaldimine ligands (L1-L3) were dissolved in a H2O/EtOH (1:1) mixture and
deprotonated using KOH base for 30 minutes at room temperature. A solution in a
ligand/rhodium dimer molar ratio of 2:1 was left to stir at room temperature for 3 hours (Scheme
2). The solvent was removed to afford bright yellow to orange water-soluble (4 mg/mL – 5
mg/mL) complexes which were obtained in good yields (of 90% for C1, 78% for C2 and 77%
for C3).
Scheme 2: Outline for the synthesis of C1 - C3.
1
In the H NMR spectra of the complexes, the imine protons appear as singlets at around 8.2-8.3
ppm. The evidence of the complex formation was confirmed by the upfield shifts of the imine
protons upon coordination to the rhodium metal from the ligands L1, L2 and L3 (9.03 ppm, 9.00
ppm and 8.98 ppm) respectively to the complexes C1, C2 and C3 (8.35 ppm, 8.20 and 8.22 ppm)
respectively. The aromatic protons of all three complexes were observed at around 7.95 ppm and
4

Page 5 of 30
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
7
8
9
7.32 ppm. The upfield shifts of the imine peaks was caused by the reduction of electron density
around the imine functionality and the de-shielding caused by the electron back- donation from
the rhodium centre. A split of the olefinic 1,5-cyclooctadiene protons was not observed as
expected upon the complexation due to the asymmetric environment induced by the N^O
chelating ligand, however, this has also been observed in literature. [24]
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The 13C{1H} NMR spectra of the complexes C1 – C3 are similar and show the expected signals.
The carbon peaks for the carbons directly bonded to the oxygen of the phenol shifted slightly
downfield from around 153 ppm in the ligands to 156 ppm in the complexes. This is partly due
to the de-shielding caused by the decreased electron density upon rhodium coordination. There
are new signals observed on the spectrum at around ≈ 80 ppm (CH COD) and ≈ 30 ppm (CH2
COD) due to the cyclooctadiene’s olefinic and alkyl carbons respectively.
The high resolution electrospray mass spectra (ESI-MS) validated the synthesis of the complexes
at m/z = 497.9933 [M+3H]+ for C1, m/z = 555.0132 [M+2H]+ for C2 and m/z = 453.2105
[M+2H]+ for C3 in the positive ion mode for all complexes.
Single Crystal X-ray structure of C3
Complex C3 was also characterised by single crystal X-ray diffraction. The crystals were
obtained by slow diffusion of THF into a concentrated solution of C3 dissolved in dimethyl
sulfoxide. Complex C3 is shown in Figure 1 in a ball and stick representation.
The molecular crystal structure of complex C3 shows a distorted square planar geometry around
the Rh metal centre that is coordinated to the N^O-chelating salicylaldimine ligand and the
cyclooctadiene moiety. The bond angles around the Rh metal centre range between 82° and 98°
which slightly deviate from the 90° angle around the metal for a square planer complex. Similar
results have been reported in literature.[16,17] The bond angles around the central carbon atom of
the carboxylic acid moiety (C22) are in agreement with the angles for a trigonal planar molecule
with all three angles approximately 120°. A THF solvent molecule also co-crystallised with the
complex and this has not been included in Figure 1 for clarity. Table 1 and Table 2 summarise
selected crystallographic data, bond angles and bond angles
5

New Journal of Chemistry
Page 6 of 30
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 1. Molecular structure of C3 determined by single crystal X-ray diffraction. Hydrogen
atoms have been omitted in the structure for clarity.
Table 1 Selected bond angles and bond distances for the molecular structure of C3.
Bond Lengths (Å)
Rh1-N1
Bond Angles (°)
N1-Rh1-O3
N1-Rh1-C4
N1-Rh1-C5
O3-Rh1-C1
O3-Rh1-C8
C1-Rh1-C4
C4-Rh1-C8
C19-C22-O1
C19-C22-O2
O1-C22-O2
2.082(5)
2.015(4)
2.129(6)
2.106(6)
2.116(7)
2.136(6)
1.258(8)
1.282(7)
1.290(7)
89.57(17)
93.5(2)
97.8(2)
87.4(2)
85.5(3)
82.4(3)
96.0(3)
119.4(5)
116.9(5)
123.6(5)
Rh1-O3
Rh1-C1
Rh1-C4
Rh1-C5
Rh1-C8
C22-O1
C22-O2
N1-C15
6

Page 7 of 30
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
Table 2 Crystallographic data selected for the molecular structure of C3.
5
6
7
C3
Empirical formula
C23.5H22NO3.5Rh
8
9
Formula weight
477.33
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Temperature/K
100 (2)
Crystal system
monoclinic
Space group
C2/c
a/Å
37.918(5)
b/Å
6.9343(9)
c/Å
16.008(2)
α/°
90
β/°
102.150(3)
γ/°
90
Volume/Å3
4114.9(9)
Z
8
ρcalcg/cm3
1.541
μ/mm-1
0.857
Crystal description,colour
Crystal size/mm3
Radiation
Block, Red
0.857 × 0.205 × 0.072
Mo Kα (λ = 0.71073)
5.2 to 56.82
2Θ range for data collection/°
Index ranges
-50 ≤ h ≤ 49, -9 ≤ k ≤ 9, -21 ≤ l ≤ 21
5139
Nref
Independent reflections
Data/restraints/parameters
Goodness-of-fit on F2
Final R indexes [I>=2σ (I)]
Final R indexes [all data]
Largest diff. peak/hole / e Å-3
5161 [Rint = 0.0480, Rsigma = 0.0311]
5161/0/243
1.069
R1 = 0.0714, wR2 = 0.2160
R1 = 0.0811, wR2 = 0.2286
5.46/-0.75
Hydroformylation of 1-octene using complexes C1-C3 as catalyst precursors
7

New Journal of Chemistry
Page 8 of 30
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
7
8
9
The salicyldamine (C1-C3), rhodium(I) complexes were evaluated as catalyst precursors in
aqueous biphasic hydroformylation of 1-octene. The reaction of 1-octene with syngas (1:1,
CO/H2) in the presence of a water-soluble Rh(I) catalyst precursor to form a mixture of linear
and branched aldehydes and isomerization products (internal olefins) is shown below in Scheme
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3.
Rh(I) catalyst
30-50 bar syngas
O
Me
O
Me
5
5
Me
5
1:1 H2O/Toluene
Scheme 3 Aqueous biphasic hydroformylation of 1-octene.
The catalyst precursors C1-C3 were separately dissolved in 5 mL water, then 1-octene substrate
and n-decane (internal standard) were dissolved in 5 mL toluene. The two solutions were poured
into a stainless steel reactor which was then pressurised and heated to the desired temperature.
After the reaction period, the reaction solution was cooled and the organic layer containing the
products and unreacted substrate were separated from the catalyst containing aqueous layer by
decantation. The organic layer was analysed by gas chromatography. Toluene was chosen since
it has been previously used in similar hydroformylation reactions.[25,26] C3 was used to optimise
the reactions.
Conversion as a function of time
At 30 bar and 75 °C, the reactions were set up for 2, 4, 6 and 8 hours. Conversion of 1-octene
increases steadily over the time intervals from 30 % conversion at 2 hours, to 78 % at 6 hours.
Complete substrate conversion (99 %) was realised at 8 h and no further conversion was
observed at 10 h (Figure 2), thus 8 h was chosen as the optimum time for the reactions. This
kinetic profile is expected, since no induction period is observed and the conversion increases
with respect to time up to 100%. This suggests kinetics controlled by mass transfer of the
substrate, which is expected for the biphasic hydroformylation of 1-octene.
8

Page 9 of 30
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2 Conversion as a function of time during the hydroformylation of 1-octene by C3.
Conditions: Reaction carried out using C3 as a catalyst precursor (Catalyst loading 1.29 x 10-3 g, 2.87 x 10-3 mmol),
at 30 barss syngas pressure (1:1 ratio of CO:H₂), and 75°C, in H2O (5 mL) and toluene (5 mL), with 1-octene
(0.721 g, 6.37 mmol) and n-decane internal standard (0.180 g, 1.26 mmol) over 8 hours. The reactor was flushed
with nitrogen thrice, followed by flushing twice with syngas prior to beginning the reaction. Average error estimate
± 0.31 (2 h), ± 0.77 (4 h), ± 0.65 (6 h), ± 0.72 (8 h), ± 0.85 (10 h).
Effect of temperature
At 30 bars and 75 °C (Table 3, Entry 1) the conversion of 1-octene was excellent. Of the 98 %
converted substrate, 78 % aldehydes formed and 22 % products were due to isomerisation of the
1-octene to form iso-octenes. The activity of the catalyst was found to be 211 h-1. The catalyst
however demonstrates an n/iso ratio of 1.2, which means that the catalyst was not very selective
to the formation of linear or branched aldehydes. Increasing the temperature to 85 °C (Table 3,
Entry 2) resulted in similar conversion of 1-octene, however, an increase in aldehydes formed
was seen. This increased from 78% at 75 °C to 97% at 85 °C, while the catalyst’s activity also
increased from 211 h-1 to 267 h-1.
This may suggest that the syngas mixture diffuses more readily into the solvents at higher
temperatures, thus promoting hydroformylation. The selectivity for linear aldehydes is not
favoured at higher temperatures (n;iso ratio = 0.6) and this is expected since at higher
temperatures (at 85 °C and 95 °C) isomerization is known to dominate and the resultant internal
olefins are ultimately hydroformylated to yield mostly branched aldehydes. Therefore, 85 °C was
chosen as the optimum temperature and was used for further pressure optimization.
9

New Journal of Chemistry
Page 10 of 30
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
Table 3 Effect of temperature in hydroformylation of 1-octene using C3.
6
7
8
Entry aTemp. bConv.
Iso-octenes n:iso
(%)
TON
Aldehydes (%)
(˚C)
(%)
9
Total linear branched
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
2
3
75
85
95
98
99
99
78
97
92
56
39
39
44
61
61
22
3
1.2
0.6
0.6
1688
2136
2056
8
Conditions: Reactions carried out in toluene (5 mL) and distilled water (5 mL) with C3, (Catalyst loading
(1.29 x 10-3 g, 2.87 x 10-3 mmol), 1-octene (0.721 g, 6.37 mmol) and n-decane as an internal (0.180 g, 1.26
mmol) over 8 hours. The reactor was flushed with nitrogen thrice, followed by flushing twice with syngas
(CO:H2 = 1:1). Average error estimate ± 0.78 (75˚C), ± 0.69 (85˚C), ± 0.56 (95 ˚C). TON = (mmol of
aldehydes/mmol of Rh) and is based on total aldehydes. aTemp. = temperature, bConv. = conversion.
Effect of pressure
Pressure optimisations were then carried out at 85 °C. The pressure was varied from 30 bars to
40 bars (Table 4, Entry 2). Much like at 30 bars and 85 °C, substrate conversion was 99 %, and
this time 1-octene was converted entirely to aldehydes. The reaction was 100 % chemoselective
towards aldehyde, while the n/iso ratio remained almost constant. Upon increasing the pressure
to 50 bars, the substrate conversion dropped slightly to 95 % (Table 4, Entry 3), together with a
slight drop in chemoselectivity. 10 % iso-octenes were observed in this case.
Table 4. Effect of pressure in the hydroformylation of 1-octene using C3.
Entry Pressure aConv. (%)
(bars)
Iso-octenes
n:iso
TON
Aldehydes (%)
(%)
Total
97
linear
39
branched
1
2
3
30
40
50
99
99
95
61
57
60
3
0.6
0.7
0.6
2136
2176
1896
100
90
43
0
40
10
Conditions: Reactions carried out in toluene (5 mL) and distilled water (5 mL) with C3, (Catalyst loading (1.29 x
10-3 g, 2.87 x 10-3 mmol), 1-octene (0.721 g, 6.37 mmol) and n-decane as an internal (0.180 g, 1.26 mmol) over 8
hours. The reactor was flushed with nitrogen thrice, followed by flushing twice with syngas (CO:H2 = 1:1).
Average error estimate ± 0.70 (30 barss), ± 0.25 (40 barss), ± 0.67 (barss). TON = (mmol of aldehydes/mmol of
Rh) and is based on total aldehydes. a Conv. = conversion.
10

Page 11 of 30
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
7
8
9
The TON, determined based on total desired aldehydes, increased on increasing the pressure
from 30 bars to 40 bars, but dropped slightly at 50 bars syngas pressure. The increase in CO
partial pressure is known to result in increased rates of CO insertion which results in suppressed
isomerization rates. However, in this system the CO seems to displace the cyclooctadiene
moiety. This results in the formation of a Rh-hydrido carbonyl species which is not a very active
hydroformylation catalysts under mild conditions but rather active for the isomerization reaction.
Therefore, 40 bar and 85 °C was chosen as the optimum reaction conditions since under these
conditions, the 100 % substrate was conversion was achieved with 100 % selectivity to aldehyde.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Hydroformylation of 1-octene using C1–C3 as (pre)catalysts
Hydroformylation reactions with catalyst C1–C3 were performed at the optimum conditions of
40 bars, 85 °C and 8 h. The catalysts precursors C1–C3 all displayed excellent catalytic
conversions as high as 99 % (Table 5). The 1-octene was converted to aldehydes exclusively,
and no n-octane and iso-octenes were formed within the detectable limits of the GC.
The catalyst precursors showed turnover numbers in the range of 2000-2400, with the most
water-soluble catalyst precursor (C2) exhibiting the highest TON of 2314 followed by C1 (with
a TON of 2208). In general, the less water-soluble the catalyst precursor the lower the TON,
suggesting that the catalyst precursor’s solubility in water impacts on performance (Table 5).
The catalyst precursors gave almost equal amounts of branched and linear aldehydes and
therefore are not very regioselective.
Table 5 Hydroformylation of 1-octene using C1-C3 as catalyst precursors.
Entry aCat.
bConv.
(%)
Iso-octenes
TON
n:iso
Aldehydes (%)
(%)
Total
100
linear branched
1
2
3
C1
C2
C3
99
99
99
43
53
43
57
47
57
0
0
0
0.8
1.2
0.7
2208
2318
2176
100
100
Conditions: Reactions carried out in toluene (5 mL) and distilled water (5 mL) with C1-C3, (Catalyst loading
(1.29 x 10-3 g, 2.87 x 10-3 mmol) separately. Substrate used was 1-octene (0.721 g, 6.37 mmol) and n-decane
as an internal (0.180 g, 1.26 mmol) over 8 hours. The reactor was flushed with nitrogen thrice, followed by
flushing twice with syngas (CO:H2 = 1:1). Average error estimates: ± 0.21 (C1), ± 0.86 (C2), ± 0.69 (C3).
TON = (mmol of aldehydes/mmol of Rh) and is based on total aldehydes. aCat. = catalyst precursor. b Conv. =
conversion.
11

New Journal of Chemistry
Page 12 of 30
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
7
8
9
These results are comparable to what has been reported in the literature for similar catalyst
precursors.[24,26] C2 slightly favours the formation of linear aldehydes, whereas, C1 and C3 gave
slightly more branched than linear aldehydes (Table 5). This indicates that the steric bulk around
the rhodium is not sufficient to cause a discrimination of the linear over the branched aldehydes
and vice versa. Introducing a bulky ligand such a water-soluble phosphine such as 1,3,5-triaza-7-
phosphaadamantane (PTA), to add to the steric bulk around the rhodium with an aim to fine-tune
the regioselectivity and water-solubility of the pre-catalysts. N^N and N^P chelating catalyst
precursors have been reported for hydroformylation and they proved to have superior
regioselectivity.[26a]
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In order to compare the perfomance of the catalysts, the reactions were perfomed under the
optimum conditions but instead of a reaction time of 8 hours, the reactions were stopped at 4
hours. The results show that there is significant isomerization activity in the first few hours of the
reaction and catalysts favouring the linear aldehyde nonanal in the first four hours. This indicates
that the branched aldehydes observed at 8h are as a result of the hydroformylation of 1-octene
isomerization products. The results obtained for this study are summarised in Table 6.
Table 6. Hydroformylation of 1-octene using C1-C3 as catalyst precursors.
Entry aCat.
bConv.
(%)
Iso-octenes
TON
n:iso
Aldehydes (%)
(%)
Total
58
linear branched
1
2
3
C1
C2
C3
53
49
50
44
42
36
18
13
21
42
45
43
2.4
3.2
1.7
752
748
750
55
57
Conditions: Reactions carried out in toluene (5 mL) and distilled water (5 mL) with C1-C3, (Catalyst loading
(1.29 x 10-3 g, 2.87 x 10-3 mmol) separately. Substrate used was 1-octene (0.721 g, 6.37 mmol) and n-decane
as an internal (0.180 g, 1.26 mmol) over 4 hours. The reactor was flushed with nitrogen thrice, followed by
flushing twice with syngas (CO:H2 = 1:1). Average error estimates: ± 0.32 (C1), ± 0.98 (C2), ± 0.87 (C3).
TON = (mmol of aldehydes/mmol of Rh) and is based on total aldehydes. aCat. = catalyst precursor. b Conv. =
conversion.
The results show that the catalysts have very comparable activity and selectivity. At 4 hours
approximately 50% conversion is achieved with isomerization competing effectively with the
12

Page 13 of 30
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
hydroformylation reaction. It is very clear that as the reaction proceeds to 8 h as shown in Table
5, the hydroformylation of internal olefins results in the production of branched aldehydes.
7
8
9
An investigation onto the extent to which these catalytic systems act as isomerization catalysts
was inverstigated by subjecting 2-octene under isomerization conditions for 6 hours at 30 bar
hydrogen pressure and 120 °C in the presence of hydrogen only using C1. Analysis of the
reaction mixture showed that 2-octene isomerized to 1-octene (Figure 3).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(a)
(b)
Figure 3. (a) Chromatogram after isomerization of 2-octene and (b) chromatogram after mixure
in (a) was hydroformylated.
In the isomerization reaction about 40 % of 2-octene was converted to 1-octene exclusively,
showing that the reaction is selective towards formation of the terminal olefin. When this
mixture was subjected to the optimized hydroformylation conditions, the same distribution of
linear (39%) and branched aldehydes (61%) was observed. This is not a new phenomenon as has
been reported by van Leeuwen and Kamer who earlier observed 2-octene to 1-octene
isomerization followed by hydroformylation to give mostly nonanal. Furthermore, they attributed
contra-thermodynamic isomerization of 2-octene to 1-octene (that evetually leads to n-aldehyde
selectivity) to be driven by the large bite angles (ranging from 105.7° to 131.2°) in the catalysts‘
[26b,c]
phenoxaphosphino modified xanthene-type ligands.
A measurement of the salicylaldimine
ligand‘s N-Rh-O bite angle, in C1, gave a significantly smaller value of 89.79°, therefore
explaining the observed lower nonanal selectivity in this study.
13

New Journal of Chemistry
Page 14 of 30
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
Catalyst recyclability
Upon completion of each reaction, the biphasic solutions were separated by decantation then a
fresh substrate was introduced to the aqueous layer and the reaction was repeated for recycling
studies. Catalysts C1-C3 can be reused up to three time without significant loss of activity and
selectivity (Figure 4). On the fourth cycle the catalysts displayed conversions below 10%.
Decrease in conversion is an indication that with time the active catalyst leaches into the organic
layer.
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The ICP-MS experiments were conducted to analyse the organic layer after each cycle. The
results show significant leaching of the Rh metal into the toluene layer with 64% (C1) and 66%
(C2) loss of Rh after the first cycle. Catalyst C3 displayed the most leaching because this
catalyst shows some partial solubility in toluene, with 83 % loss observed. In the second and
third cycles, further losses were observed with catalyst C3 showing the least conversion in the 3rd
cycle. After this no activity was observed for all catalysts. This may suggest that the water-phase
is acting as a reservoir for the active Rh species responsible for the hydroformylation activity.
Figure 4 Results obtained after catalyst (C1–C3) recyclability studies.
Conditions: Reactions carried out in toluene (5 mL) and distilled water (5 mL) with C1-C3, (Catalyst loading (1.29 x 10-3 g, 2.87
x 10-3 mmol) separately. Substrate used was 1-octene (0.721 g, 6.37 mmol) and n-decane as an internal (0.180 g, 1.26 mmol)
over 8 hours. The reactor was flushed with nitrogen three times, followed by flushing twice with syngas (CO:H2 = 1:1).
Mercury Poisoning Tests
A homogenous catalyst, can form heterogeneous metal film strips, metal powders and metal
nano-particles which end up catalysing the reactions, instead the molecular catalysts.[27-28] The
14

Page 15 of 30
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
presence of Hg(0) around metal nano-particles forms amalgams which then retard the metal’s
ability to perform catalytic conversions. This tool is used to distinguish true homogenously
catalysed reactions from those catalysed by metal nano-particles.[29]
7
8
9
In separate experiments, the hydroformylation reactions were conducted with a drop of mercury
(in addition to all necessary inputs such as the substrate, catalyst precursor, internal standard,
syngas and a solvent). The reaction was carried out, and after 8 h the aqueous and organic layers
were separated and the organic layers were analysed by GC. The catalyst’s activity,
chemoselectivity and regioselectivity were similar to the reactions done in the absence of the
mercury (Figure 5), this proved that molecular homogenous catalysts are responsible for the
hydroformylation of 1-octene.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 5 Mercury poisoning experiments.
Conditions: Reactions carried out in toluene (5 mL) and distilled water (5 mL) with C1-C3, (Catalyst loading (1.29 x 10-3 g, 2.87
x 10-3 mmol) separately. Substrate used was 1-octene (0.721 g, 6.37 mmol) and n-decane as an internal (0.180 g, 1.26 mmol)
over 8 hours. The reactions done with Hg(0) had a few drops of metallic mercury added, and reactor was flushed with nitrogen
three times, followed by flushing twice with syngas (CO:H2 = 1:1).
Preliminary modelling of the active catalyst
The pathway was modelled (computed) and investigated using Spartan ’10 V1.1.0 Molecular
Modelling Program at Semi-Empirical PM3 (D) level at 1 atm and 298.15 K in order to simulate
the experimental conditions. The PM3 (D) method has been specifically parameterized for
transition metal systems for quick exploratory calculations on trends.[30,31]
15

New Journal of Chemistry
Page 16 of 30
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
7
8
9
It has been reported that a Rh-hydride species formed in situ is the active catalyst.[32-33] Since
1,5-cyclooctadiene (COD) is a good leaving group, it was assumed that in the presence of excess
syngas, it is initially displaced by carbon monoxide and hydrogen to form a dicarbonyl,
dihydride species, as observed by Ramarou et al. in the synthesis of Rh carbonyl complexes.[34a]
The reaction pathway is considered to go through 3-centred concerted addition mechanism
across a square planar complex to generate a dihydride active species. The first step towards the
generation of the active species involves the coordination of carbon monoxide to the pre-catalyst to
form an intermediate (3.1) (Scheme 4).[34b]
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 4. Possible reaction pathway for generation of catalytically active species.
After the formation of 5-coordinate intermediate, the first transition state (TS1) is achieved. In this
transition state one Rh-olefin bond dissociates as a new Rh-CO is formed in intermediate (3.2). This
leaves a vacant site for coordination of a second CO molecule to give intermediate (3.3). As the
second CO ligand begins to coordinate to the rhodium centre via a σ-bond, the second Rh-olefin
bond (of the COD ligand) dissociates in transition state (TS2). This results in the expulsion of COD
16

Page 17 of 30
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
7
8
9
and the formation of a 4-coordinate square planar intermediate (3.4). The process from 3.3 to 3.4
overcomes an energy barrier of 24.9 kcal/mol. Intermediate 3.4 then associates with a molecule of
H2 in the next step to afford an intermediate with a Rh-H2 bond (3.5). Intermediate 3.5 may exist in
an equilibrium with the species (3.6) which is formed via a transition state TS3. Reaction energy
profile for the possible pathway towards the formation of the active species 3.6 is shown in
Scheme 5.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 5. Reaction energy profile for the third possible pathway for the formation of the active
species.
17

New Journal of Chemistry
Page 18 of 30
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
The hydroformylation of olefins mechanism has been extensively studied over the years and it is
5
6
7
generally accepted that a Rh(I) is the catalytically active species. Though rare, Rh(III) has been
[34c-e]
found to promote hydroformyaltion in a some ocassions.
It is possible that the imine
8
9
1
nitrogen dissociates from the rhodium centre, as has been seen before through in situ HNMR
studies,[34f].This would give way for coordination of the olefinic substrate, in a Rh(III)-catalysed
hydroformylation pathway (Scheme 6).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 6. Rearrangement that results in formation of a Rh(III) species potentially the active
catalyst.
Scope of catalyst precursor C1 in hydroformylation of various olefins
Since the optimum hydroformylation reaction conditions have been established, the scope of C1
in the hydroformylation of longer chain olefins (C10-C12), methyl/ethyl acrylates and styrene was
investigated. Reaction mixtures were analysed using 1H NMR spectroscopy.
Long chain olefins are known to have poor solubility in the aqueous layer which would
ultimately affect catalyst contact with organic substrate. However, it has been reported in the
literature that conversions observed may be attributed to enriched concentration of the Rh
catalyst in the interfacial layer and possible high mobility of substrate between the organic and
interfacial layer.[35] From the results summarised in Table 7, the catalyst precursor C1 gives
conversions of 1-decene, 1-undecene and 1-dodecene of >99%.
18

Page 19 of 30
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
Table 7. Hydroformylation of various substrates using C1 as catalyst precursor.
5
6
7
8
9
Entry Substrate
bConv.
(%)
Iso-octenes n:iso TON
Aldehydes (%)
(%)
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Total linear branched
1
2
3
4
1-Decene
1-Undecene
1-Dodecene
Methyl
>99
>99
>99
23
100
71
30
32
36
-
70
-
0.42 2220
0.47 2181
0.57 2200
68
29
19
-
81
64
100
100
-
510
acrylate
5
6
Ethyl
35
9
100
100
-
100
60
-
-
-
777
acrylate
Styrene
40
0.67 200
Conditions: Reactions carried out in toluene (5 mL) and distilled water (5 mL) with C1-C3, (Catalyst loading (1.29
x 10-3 g 2.87 x 10-3 mmol) separately. Substrate used was 1-octene (0.721 g, 6.37 mmol) and n-decane as an
internal (0.180 g, 1.26 mmol) over 8 hours. The reactor was flushed with nitrogen thrice, followed by flushing twice
with syngas (CO:H2 = 1:1). Average error estimates: ± 0.22 (decene), ± 0.30 (undecene), ± 0.19 (docecene), ± 0.34
(methyl acrylate), ± 0.11 (ethyl acrylate), ± 0.36 (styrene). TON = (mmol of aldehydes/mmol of Rh) and is based on
total aldehydes. bConv. = conversion.
The chemoselectivity for aldehydes was >70% with a mixture of linear and branched aldehydes
observed. This is similar to what was observed when 1-octene was used as the substrate using
catalysts C1-C3. In the case of C10-C12 more branched aldehydes are observed which can be
attributed to greater isomerisation in the longer chain substrates before hydroformylation occurs.
Lower conversion of 23% and 35% were observed when methyl acrylate and ethyl acrylate were
employed as substrates respectively. However, the systems displayed excellent aldehyde
chemoselectivity and this could largely be due to the fact isomerization is not possible with the
two substrates. Lower TONs of 510 and 777 h-1 for methyl acrylate and ethyl acrylate were
observed respectively. It is expected that the conversion of 1-dodecene (C12) be lower than that
of 1-undecene (C11), however we notice the opposite in this catalytic system with conversion of
1-dodecene being 10% higher than that observed for 1-undecene. However, when the TONs are
19

New Journal of Chemistry
Page 20 of 30
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
considered the catalyst seems to maintain more or less similar activity. The 10% difference is not
a significant difference.
7
8
9
The hydroformylation of styrene is an important process that give access to both linear and
branched aldehydes that find use in the production of anti-inflammatory drugs, detergents
andplasticizer industries.[36] However, the hydroformylation of styrene usually favours the
formation of the branched aldehyde product. This is largely due to the formation of a very stable
benzylic rhodium species due to the ɳ2 electron donor emanating from the aromatic system thats
induces this regioselectivity.[36] The same phenomenon was observed when C1 was employed
with 60 % branched aldehydes being formed from the aqueous biphasic hydroformylation of
styrene. A very poor conversion of only 9% was recorded with 100% aldehyde formation.
Styrene also showed very suppressed conversions to the desired products. This could firstly due
to the biphasic nature of the catalytic system that hinders maximum contact between the catalyst
and substrate. Secondly, our catalysts are clearly not very active as observed by the suppressed
results.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Conclusions
Water-soluble N^O complexes C1-C3 were prepared and characterised and further evaluated as
isomerization-hydroformylation catalyst precursors.
Optimum conditions were realised at 85 °C and 40 syngas pressure. The three catalysts
displayed excellent substrate conversion of >98% and 100% aldehyde chemoselectivity,
interestingly, good selectivity for branched aldehydes was observed and this indicates that
isomerization is favoured prior to hydroformylation. Further studies have shown that the
catalysts are only not effective in thermodynamically favoured isomerization (isomerization of 1-
octene to 2- and 3-octenes) but also promotes contra-thermodynamic isomerization (2-octene to
1-octene). The later is a rare phenomenon which means that these catalytic systems can be
further developed for upgrading internal olefins (such as in fatty acids) into high value terminal
olefins.
The catalyst precursors with additional water-solubilising groups (C1 and C2) were more active
for hydroformylation displaying turnover numbers (TON) of 2208 and 2318, and this may be as
a result of better solubility enhancing the activity as compared to C3. The aqueous biphasic
20

Page 21 of 30
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
7
8
9
approach to separation (of the products-catalyst solution) was demonstrated for the
salicyldamine-based catalysts, since these were successfully separated and recycled for up to
three times without any significant loss in catalytic activity and selectivity. The efficacy of C1
was further evaluated in the aqueous biphasic isomerization-hydroformylation of 1-decene, 1-
undecene and 1-dodecene. The catalyst proved to be versatile and excellent substrate conversions
with the long chain olefins while lower conversions were recorded with acrylates and styrene.
TONs in the range 2100-2300 were observed with the long chain olefins which is very similar to
what is observed when 1-octene is used as the substrate. Furthermore, mercury poisoning studies
using 1-octene as a substrate confirmed that the catalysis was due to molecular homogenous
catalysts, more specifically potentially a Rh(III) hydride catalytically active species as obtained
from preliminary computations.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Experimental
Materials and methods
All compounds were prepared using standard Schlenk and vacuum line techniques. Solvents and
reagents were purchased from Sigma-Aldrich and were of full analytical grade and were used as
received. These reagents include 4-aminobenzoic acid, 4-formyl-3-hydroxy benzoic acid, aniline,
salicylaldehyde, 1,5-cyclooctadiene. Sodium 3-(((4-carboxyphenyl) imino) methyl) – 4
hydroxybenzene sulfonate), 4-((2-hydroxybenzylidene) amino) benzoic acid and (4-[(2-
thinylmethylene) amino] benzoic acid) were synthesised according to literature. [24-25] RhCl3
was purchased from Heraeus South Africa Pty Ltd. [RhCl(COD)]2 was synthesised following a
literature protocol.[26]
Nuclear Magnetic Resonance (1H NMR, 13C{1H} NMR) spectra were recorded on a Bruker
Ultrashield 400 (1H NMR 400.17 MHz and 13C{1H} 100.62 MHz) in DMSO-d6 at ambient
temperatures while the coupling constants were calculated in Hertz (Hz). The chemical shifts
were reported in parts per million (ppm) and relative to tetramethylisane ( 0.00) as an internal
standard. FT-IR spectra were recorded using a Perkin Elmer FT-IR Spectrum BX II fitted with
an ATR probe. Elemental analyses were performed on a Thermo Scientific FLASH 2000 CHNS-
O analyser. The HR-MS (ESI) spectra were recorded on Walters Synapt G2 Spectrometer. A
Gallenkamp digital melting point apparatus was used to determine the melting point.
Hydroformylation samples were analysed on a Scion 456-GC with 30 m x 0.25 mm
21

New Journal of Chemistry
Page 22 of 30
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
cyanopropylphenylmethylpolysiloxane phase column. ICP-MS experiments were conducted on a
Perkin Elmer PESCIEX ELAN 6100.
Crystal data was collected on a Bruker APEXII diffractometer with Mo Kα (λ = 0.71073 Å)
radiation (Cu Kα (λ = 1.54178 Å) used for 1) and diffractometer to crystal distance of 4.00 cm.
The initial cell matrix was obtained from three series of scans at different starting angles. Each
series consisted of 12 frames collected at intervals of 0.5º in a 6º range about with an exposure
time of 10 s per frame. The data were collected using the full sphere data collection routine to
survey the reciprocal space to the extent of a full sphere to a resolution of 0.75 Å. Data were
harvested by collecting 2982 frames at intervals of 0.5º scans in ω and φ with exposure times of
10 s per frame.[37] A successful solution by the direct methods of SHELXS 2013 provided all
non-hydrogen atoms from the E-map. All nonhydrogen atoms were refined with anisotropic
displacement coefficients. All hydrogen atoms (except the ones participating in hydrogen
bonding interactions) were included in the structure factor calculation at idealized positions and
were allowed to ride on the neighbouring atoms with relative isotropic displacement
coefficients.[38] The systematic absences in the diffraction data were uniquely consistent for the
space group P21/n that yielded chemically reasonable and computationally stable results of
refinement.[39-43] CCDC 1896188 for C3 contains the supplementary crystallographic data for
this paper.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis of ligands L1 – L3
Synthesis of L1
(40 mg, 0.3 mmol) 4-amino benzoic acid was dissolved in ethanol, (50 mg, 0.3 mmol) 4-formyl-
3-hydroxy benzoic acid was added to the reaction vessel and the mixture was heated at 50 oC for
16 hours. The mixture turned orange within minutes (approximately 5 minutes) of stirring. After
16 hours, a peach precipitate had formed and was isolated by suction filtration then dried using a
high vacuum pump for 2 hours.
1
Yield 0.055 g (64 %). Melting point: 300-304 oC. H NMR (400 MHz, DMSO-d6, 22 oC) (ppm)
3
3
= 12.55 ( br s, 1H, OH) 9.04 (s, 1H, Himine), 8.02 (d, JH-H = 8.8 Hz, 2H, HAr), 7.85 (d, JH-H = 8
o
Hz, 1H, HAr), 7.52 (m, 4H, HAr). 13C{1H} NMR (400 MHz, DMSO-d6, 22 C) (ppm) = 167.9,
160.9, 153.6, 131.7, 131.2, 128.9, 122.1, 121.0, 120.0, 89.0, 120.2, 118.5, 117.3, 113.0.
Elemental analysis: Calculated for C15H11NO5: C 63.16, H 3.89, N 4.91 %. Found: C 62.93, H
22

Page 23 of 30
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
3.93, N 4.80 %. HR-MS (ESI+): Calculated for C11H11NO5, m/z = 285.0637 [M]+, Found:, m/z =
286.0256 [M+H]+. FT-IR (νmax/cm-1) (ATR): 1678 (C=N), 1698 (C=O). Solubility: not soluble
in DCM, MeOH, EtOH, and H2O, soluble in H2O at 65 oC.
7
8
9
Synthesis of L2
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Monosodium 5-sulfonatosalicylaldehyde (0.58 g, 0.0026 mol) and 4-aminobenzoic acid (0.36 g,
o
0.0026 mol) were heated in ethanol at 60 C for about 15 hours. After 15 hours, a yellow
precipitate (L2) formed which was then isolated by suction filtration and washed with portions of
cold ethanol (2 x 30 ml). Following the washing of L2, was dried under vacuum for 3 hours.
o
1
Yield 0.63 g (71%). Melting point: decomposes without melting, onset occurs at 195 C. H
o
NMR (400 MHz, DMSO-d6, 22 C): 9.00 (s, 1H, HImine), 7.98 (m, 3H, HAr), 7.65 (m, 1H, HAr),
3
3
7.45 (d, JH-H = 8 Hz, 2H, HAr), 6.94 (d, JH-H = 8 Hz, 1H, HAr). 13C{1H} NMR (100 MHz,
DMSO-d6, 22 oC) (ppm) = 167.4, 164.4, 160.2, 153.1, 152.0, 140.0, 130.6, 121.5, 118.0, 116.8,
115.8, 112.5. Elemental analysis calculated for C14H10NNaO6S: C 48.98, H 2.94, N 4.08 %.
Found: C 48.73, H 2.97, N 4.63 %. HR-MS (ESI-): calculated for C14H10NNaO6S, m/z =
343.0253 [M]+, found: m/z =317.1280 [M-Na-2H]+. FT-IR (νmax/cm-1, ATR): 1618 cm-1(C=N),
1700 (C=O). Solubility: Not soluble in: toluene, MeOH, EtOH, DCM, EtOAc. Soluble in H₂O
(65°C).
Synthesis of L3
4-Aminobenzoic acid (1.37 g, 10.0 mmol) was dissolved in 50.0 mL EtOH in a 100 mL round
bottom flask. Then 2 drops of glacial acetic acid were added followed by salicylaldehyde (1.06
mL, 10.0 mmol), and the reaction solution was heated at 50 oC for 3 hours during which time an
orange precipitate formed. The reaction mixture was filtered by suction to isolate the orange
precipitate from the solvent. The prodcut was washed with EtOH (2 x 30 mL) then dried under
vacuum for 18 hours.
Yield: 2.16 g (89%). Melting point: 265-268 oC. 1H NMR (400 MHz, DMSO-d6, 22 oC) (ppm) =
3
12.95 (br s, 2H, HCOOH)12.71 (br s, 1H, HOH) 8.98 (s, 1H, Himine), 8.01 (d, JH-H = 8.8 Hz, 2H,
HAr), 7.69 (d, 3JH-H = 1.2 Hz, 2H, HAr), 7.48 (m, 3H, HAr) 6.98 (m, 2H, HAr). 13C{1H} NMR (100
o
MHz, DMSO-d6, 22 C) (ppm) = 167.8, 161.2, 160.8, 153.6, 136.8, 134.3, 131.2, 129.3, 121.9,
119.9, 117.7, 113.0. Elemental analysis calculated for C14H11NO3: C 69.70, H 4.60 N 5.81 %.
Found: C 70.39 H 4.71, N 5.79 %. HR-MS (ESI-): Calculated for C14H11NO3: m/z = 241.0673
23

New Journal of Chemistry
Page 24 of 30
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
[M]+, found: m/z = 242.0809 [M+H]+. FT-IR (νmax/cm-1, ATR): 1677 (C=N), 1700 (C=O).
Solubility: Sparingly soluble in toluene, MeOH, EtOH, completely soluble in DCM and EtOAc.
Soluble in H₂O (65°C).
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis of complexes C1-C3
Synthesis of [Rh(L1)(COD)], C1
L1 (25 mg, 0.1 mmol) was dissolved in an EtOH:H2O mixture (50:50), (20 mL). 0.1 M KOH(aq)
(1.00 ml, 0.1 mmol) was added to the solution and it was left to stir at room temperature for 30
minutes. [RhCl(COD)]2 (25 mg, 0.05 mmol) was then added to the reaction solution and this was
stirred at room temperature for a further 3 hours. An orange precipitate formed which was then
isolated by suction filtration, washed with EtOH (2 x 20 mL) and dried under vacuum for 4
hours.
1
o
Yield 0.026 g, (100%). Melting point: 340-344 °C. H NMR (400 MHz, DMSO-d6, 22 C)
(ppm) = 8.35 (s, 1H, Himine), 7.96 (d, 3J = 8Hz, 2H, HAr), 7.54 (d, 3J = 8 Hz, 1H, HAr), 7.25 (d, 3J
3
= 8 Hz, 2H, HAr), 7.21 (s, 1H, HAr), 7.04 (d, J = 8 Hz, 1H, HAr), 4.38 (m, 4H, HCOD), 2.29 (m,
4H, HCOD), 1.71 (m, 4H, HCOD). C{1H} NMR (100 MHz, DMSO-d6, 22 C) (ppm) = 167.9,
160.9, 153.6, 131.7, 131.2, 129.8, 128.9, 122.1, 120.2, 118.5, 117.3, 115.4, 113.0, 78.3, 30.5.
Elemental analysis: Calculated for C23H22NO5Rh: C 55.77, H 4.48, N 2.83 %. Found: C 55.44, H
4.79, N 2.77 %. HR-MS (ESI+): Calculated for C23H22NO5Rh, m/z = 495.0612 [M]+, found m/z =
497.9933 [M+3H]+, m/z = 388.9996 [M-(COD)]+. FT-IR (νmax/cm-1, ATR): 1584 (C=N).
Solubility: Soluble in THF and H₂O (65°C), insoluble in MeOH, EtOH, DCM.
13
o
Synthesis of [Rh(L2)(COD)], C2
L2 (0.18 g, 0.5 mmol) was dissolved in an EtOH-H2O mixture (50:50), (20 mL). 0.1 M KOH(aq)
(5.00 ml, 0.5 mmol) was added to the solution and this was left to stir at room temperature for 30
minutes. [RhCl(COD)]2 (0.12 g, 0.25 mmol) was then added to the solution and this was stirred
at room temperature for a further 3 hours. A yellow precipitate formed which was isolated by
suction, washed with EtOH (2 x 20 mL) and dried under vacuum for 5 hours.
1
Yield 011 g (78%). Melting point: decomp without melting, onset at 280 °C. H NMR (400
o
3
MHz, DMSO-d6, 22 C) (ppm) = 9.03 (s, 1H, HImine), 8.20 (s, 1H, HAr), 7.90 (d, J = 8 Hz, 2H,
HAr), 7.54 (d, 3J = 4 Hz, 1H, Hg), 7.09 (d, 3J = 8 Hz, 1H, HAr), 6.63 (d, 3J = 8 Hz, 1H, HAr), 4.34
24

Page 25 of 30
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
7
8
9
13
(m, 4H, HCOD), 2.29 (m, 4H, HCOD), 1.73 (m, 4H, HCOD). C{1H} NMR (100 MHz, DMSO-d6,
22 oC) (ppm)= 167.4, 166.8, 164.4, 160.2, 153.1, 152.0, 140.0, 131.1, 130.6, 121.5, 118.0, 116.8,
81.6, 29.9. Elemental analysis: Calculated for C22H21O6NRhNa: C 48.98, H 2.94, N 4.08 %.
Found: C 48.77, H 2.23, N 4.63 %. HR-MS (ESI+): Calculated for C22H21O6NRhNa m/z =
553.3786 [M]+, found m/z = 555.0132 [M+2H]+. FT-IR (νmax/cm-1, ATR): 1590s (C=N), 1702
(C=O. Solubility: Soluble in H2O, insoluble in toluene, DCM, EtOH, MeOH, EtOAc.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Synthesis of [Rh(L3)(COD)], C3
L3 (25 mg, 0.1 mmol) was dissolved in an EtOH-H2O mixture (50:50), (20 mL). 0.1 M
KOH(aq) (1.00 ml, 0.1 mmol) to the solution and this was left to stir at room temperature for 30
minutes. [RhCl(COD)]2 (25 mg, 0.05 mmol) was added to the reaction solution and this was
stirred at room temperature for a further 3 hours. A yellow precipitate formed which was isolated
by suction filtration, washed with EtOH (2 x 20 mL) and dried under vacuum pump for 5 hours.
Yield 0.026 g (85%). Melting Point: 298-301 °C. 1H NMR (400 MHz, DMSO-d6, 22 oC) (ppm)
8.22 (s, 1H, HImine), 7.95 (d, 3J = 8 Hz, 2H, HAr), 7.43 (d, 3J = 8 Hz, 1H, HAr), 7.33 (t, 3J = 4 Hz,
1H, HAr) 7.20 (d, 3J = 8 Hz, 2H, HAr), 6.72 (d, 3J = 8 Hz, 1H, HAr), 6.55 (t, 3J = 8 Hz, 1H, HAr),
4.43 (s, 4H, HCOD), 2.28 (s, 4H, HCOD), 1.74 (m, 4H, HCOD). C{1H} NMR (100 MHz, DMSO-
13
o
d6, 22 C) (ppm) = 166.0, 165.7, 155.4, 136.5, 135.7, 130.2, 129,3, 125,5, 123.9, 121.3, 118.5,
114.8, 79.0, 30,9. Elemental analysis: Calculated for C22H22O3NRh: C 58.55, H 4.91, N 3.10 %.
Found: C 57.4, H 4.94, N 3.15 %. HR-MS (ESI+): Calculated for C22H22O3NRh m/z = 451.3209
[M]+, found m/z = 453.2105 [M+2H]+, m/z = 348.9327 [M-(COD)+3H]+. FT-IR (νmax/cm-1,
o
ATR): 1564s (C=N), 1698 (C=O). Solubility: Soluble in H2O (65 C) DCM, EtOAc. Sparingly
soluble in: toluene and MeOH.
General procedure for the catalytic hydroformylation experiments
Hydroformylation reactions were done in a 50.0 mL stainless steel pipe reactor. The reactor was
charged with water (5 mL) containing the rhodium(I) catalyst precursor (2.87 x 10-3 mmol) and
with the organic solvent toluene (5.00 mL) containing the substrate 1-octene (6.37 mmol) and the
internal standard n-decane (1.26 mmol). N2 gas was used to flush the reactor three times
followed 1:1 syngas (30-50 barss). This was followed by heating the reactor to the desired
temperatures, 75-95 °C and stirring. The reactions were performed for 8 h with the samples
25

New Journal of Chemistry
Page 26 of 30
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
being taken at an initial time = 0 h and final time = 8 h and these were analysed by GC. All
products were confirmed with respect to authentic iso-octenes and aldehydes.
In the recycling experiments, upon reaction completion, the organic layer was separated by
decantation and a fresh organic layer containing the substrate and the internal standard was
introduced for the catalyst recycling experiments. All reactions were conducted in duplicates and
the results were reported as averages for the identical experiments.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Acknowledgements
This work was supported by the National Research Foundation of South Africa under Grant
(Numbers: 117989 and 99269); DST-NRF Centre of Excellence in Catalysis (c*change);
University of Johannesburg Centre for Synthesis and Catalysis and SASOL SA. Many thanks
also goes to Dr Gift Mehlana from the Midlands State University, Zimbabwe, for the single
crystal X-ray crystallographic data refinement and to Dr Boitumelo Makume from SASOL South
Africa for the bite angle measurement.
Conflict of interest
There are no conflicts to declare.
References
[1]
Roelen O., Roelen. Ger. Pat. 849548, 1938/1952; US Pat. 2327066, (1943).
Cornils, B.; Herman, W. A.; Rasch, M.; Cornils, B.; Herman, W. A.; Rasch, M.; Roelen,
O.; Pioneer in Industrial Homogenous Catalysis, Angew. Chem. Int. Ed. Engl., (1994), 33,
2144.
[2]
[3]
[4]
Hebrard, F.; Kalck, P. Cobalt-Catalyzed Hydroformylation of Alkenes: Generation and
Recycling of the Carbonyl Species, and Catalytic Cycle. Chem. Rev., (2009), 109, 4272.
Behr A.; Neubert, P., Applied Homogeneous Catalysis, 1st Ed., Wiley-VCH Verlag
GmbH & Co. KGaA, (2012), 27.
[5] van Leeuven, P. W. N. M.; Claver C, Rh-Catalysed Hydroformylation, Eds.; Kluwer
Academic Publishers: Dordrecht, Netherlands, (2000), 1.
[6]
[7]
G.D. Frey, J. Organomet. Chem., (2014), 754, 5.
Franke, R.; Selent, D.; Bo, A., Applied Hydroformylation, Chem. Rev., (2012), 112,
26

Page 27 of 30
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5675.
5
6
7
8
[8]
[9]
Frohning, C.D.; Kohlpaintner, C.W.; Bohnen, H. W in: Applied Homogeneous Catalysis,
2nd Ed., Vol. 1, (Eds.: B. Cornils, W. A. Herrmann), Wiley-VCH, Weinheim, (2002), 31.
Bizzari, S.; Blagoev, M.; Kishi, A., Chemical Economics Handbook Report: Oxo
Chemical, Stanford Research Institute (SRI) International, (2006), Report No. 682.7000,
http://www.sriconsulting.com/CEH/Public/Reports/682.7000/ (accessed 13 June, 2018).
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
[10] S.K. Sharma, R. V. Jasra, Catal. Today., (2014), 247, 70.
[11] Neves, Â.C. B.; Calvete, M. J. F.; Pinho e Melo, T.M.V.D.; Pereira, M. M., Immobilized
Catalysts for Hydroformylation Reactions: A Versatile Tool for Aldehyde Synthesis. Eur.
J. Org. Chem., (2012), 6309.
[12] Likholobov, V.A.; Moroz, B.L.; Handbook of Heterogeneous Catalysis, Wiley-VCH
Verlag GmbH & Co. KGaA, Weinheim, (2008), 3663.
[13] Desset, S. L.; Hintermair, U.; Cole-Hamilton, D. J.; Biphasic and Flow Systems Involving
Water or Supercritical Fluids. Top Catal., (2010), 53, 963.
[14] Haumann, M.; Riisager, A.; Hydroformylation in Room Temperature Ionic Liquids
(RTILs):ꢀ Catalyst and Process Developments. Chem. Rev., (2008), 108, 1474.
[15] Subramaniam, B; Gas-expanded liquids for sustainable catalysis and novel materials:
Recent advances. Coord. Chem. Rev., (2010), 254, 1843.
[19] Zhao, X.; He, D.; Mika, L. T.; Horvȃth, I.; Fluorous Hydroformylation. Top Curr. Chem.,
(2012), 308, 275.
[20] Almeida, A. R.; Peixoto, A. F.; Calvete, M. J. F.; Góis, P. M. P.; Pereira, M.M.,
Rhodium(I) N-Heterocyclic Carbene Complexes as Catalysts for Hydroformylation of
Olefins: An Overview. Curr. Org. Synth., (2011), 8, 5, 764.
[21] Cornils B.; Kuntz, E.G.; Aqueous Phase Organometallic Catalysis, Concepts and
Applications (Eds. Cornils B.; and Herrmann W.A.) Wiley-VCH Verlag GmbH,
Weinheim, (1998), 271.
[22] Kuntz, E.G.; FR Pat. 2.349.562.1976, (1976).
[23] (a) Adams, D. J.; Dyson, P. J.; Tavener, S.; Chemistry In Alternative Reaction Media, John
Wiley and Sons Ltd., West Sussex, (2004), (b) Matsinha, L. C.; Siangwata, S.; Smith, G. S.;
Makhubela, B. C. E.; Catalysis Reviews: Science and Engineering, (2019), 61, 111.
27

New Journal of Chemistry
Page 28 of 30
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
[24] L. C. Matsinha, S. F. Mapolie and G. S. Smith, Dalton Trans., (2015), 44, 1240.
[25] E.B. Hager, B. C. E. Makhubela, G. S. Smith, Dalton Trans, (2012), 14, 13927.
7
8
9
[26] (a) J. Chatt and L. M. Venanzi, J. Chem. Soc., (1957), 4735, (b) C. P. Casey, G. T.
Whiteker, M. G. Melville, L. M. Petrovich, J. A. Gavney, Jr., and D. R. Powell, J. Am.
Chem. Soc., (1992) 114, 5535. (c) R. P. J Bronger, P. C. J. Kamer and P. W. N. M. Van
Leewen, Organometallics, (2003), 22, 5358.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
[27] J. P. Collman, L. S. Hedegus and R. G. Finke, Principles and Applications of
Organotransition Metal Chemistry, University Science Books, Mill-Valey, C. A. (1987),
1145.
[28] J. A. Widegren and R. G. Finke, J. Mol. Catal. A: Chem., (2003), 198, 317.
[29] J. A. Widegren, M. A. Bennet and R. G. Finke, J. Am. Chem. Soc., (2003), 125, 10301.
[30] S. Aguado-Ullate, Catal. Sci. Technol., (2014), 4, 979.
[31] P. Pino, F. Piacenti and M. Bianchi, Reactions of CO and H2 with Olefinic Substrates.
The Hydroformylation Reaction. Wiley-Interscience, (1997).
[32] R. Kummer, H. J. Nienburg, H. Hohenschutz and M. Strohmeyer, Am. Chem. Soc.,
(1974), 1, 19.
[33] C. Huo, Y. Li, M. Beller and H. Jiao, Organometallics, 2003, 22, 4665.
[34] (a) D. S. Ramarou, B. C. E. Makhubela and G. S. Smith, J. Organomet. Chem., (2018),
870, 23, (b) S. Wolf and H Plenio, J. Organomet. Chem., (2009), 694, 1487, (c) P.
Govender,
S. Ngubane, B Therrien and G. S. Smith, J. Organomet. Chem., (2017),
848, 281. (d) S.-J. Chena, Y.-Y. Wanga, W.-M. Yaoa, X.-L. Zhaoa, G. VO-Thanhb, Y.
Liu, J. Mol. Catal. A: Chem., (2013) 378, 293. (e) B. El Ali, J. Tijani, M. Fettouhi, M. El-
Faer, A. Al-Arfaj, Appl. Catal. A: Gen., (2005), 283, 185. (f) N. K. Oklu and B. C. E.
Makhubela, Inorg. Chim. Acta, (2018), 482, 460.
[35] H. Fu, M Li, H. Chen and X. Li, J. Mol. Catal. A: Chem., (2006), 259, 156
[36] S. Yu, Y. Chie, Z. Guan, Y. Zou, W. Li and X. Zhang, Org. Lett., (2009), 11, 241
[37]
Bruker-AXS. 2009 APEX2, SADABS, and SAINT Software Reference Manuals.
Bruker-AXS, Madison, Wisconsin, USA.
[38]
(a) G. M. Sheldrick, Acta Cryst. A short history of SHELX. (2008), A64, 112. (b)
G. M. Sheldrick, XPREP. Version 2013/1. Georg-August-Universitat Gottingen,
Gottingen, Germany. 2013b.
28

Page 29 of 30
New Journal of Chemistry
View Article Online
DOI: 10.1039/D0NJ01970D
1
2
3
4
5
6
[39] G. M. Sheldrick, The SHELX homepage, http://shelx.uni-ac.gwdg.de/SHELX/. 2013a.
[40] G. M. Sheldrick, Acta Cryst. A, 2015a, 71, 3.
[41] G. M. Sheldrick, Acta Cryst. C, 2015b, 71, 3.
7
8
9
[42] O.V. Dolomanov, L.J. Bourhis, R.J. Gildea, J.A.K. Howard and H. Puschmann,
OLEX2: J. Appl. Cryst., (2009), 42, 339.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
[43] I. A. Guzei, Programs Gn. University of Wisconsin-Madison, Madison, Wisconsin,
USA. (2007-2013).
29