Rhodium catalysts build into the structure of a silicate support in the hydroformylation of alkenes

Article abstract of DOI:10.2478/s11532-012-0186-z

Rhodium is build into a nano-structured calcium silicate during the synthesis of the silicate. Thereby, it was desired to create a robust heterogeneous catalyst, which does not suffer from catalyst leaching like rhodium impregnated on a pre-formed silicate. While this was achieved, the silicate structure was adversely affected by the incorporation of rhodium - the surface area and pore volume of the material were found to be comparatively low. Alcohol and acid washing were tested to address this issue. The alcohol treatment proved detrimental as catalytic material was leached from the silicate. The acid washed rhodium containing calcium silicate was quite active in the hydroformylation of alkenes and did not suffer loss of catalyst into the product phase. Acid treated rhodium containing silicates were more active than their untreated counterparts but less selective due to access to the rhodium centers being opened.

Full text of DOI:10.2478/s11532-012-0186-z

Cent. Eur. J. Chem. • 11(4) • 2013 • 561-568
DOI: 10.2478/s11532-012-0186-z
Central European Journal of Chemistry
Rhodium catalysts build into
the structure of a silicate support
in the hydroformylation of alkenes
Research Article
1
*
1
Thomas Borrmann , Andrew J. McFarlane ,
2
1
Uwe Ritter , James H. Johnston
1School of Chemical and Physical Sciences,
Victoria University of Wellington, P. O. Box 600,
6
140 Wellington, New Zealand
2Inorganic Chemistry Department,
Georg-August-University Göttingen,
3
7077 Göttingen, Germany
Received 10 October 2012; Accepted 30 November 2012
Abstract: Rhodium is build into a nano-structured calcium silicate during the synthesis of the silicate. Thereby, it was desired to create a robust
heterogeneous catalyst, which does not suffer from catalyst leaching like rhodium impregnated on a pre-formed silicate. While this
was achieved, the silicate structure was adversely affected by the incorporation of rhodium – the surface area and pore volume of the
material were found to be comparatively low. Alcohol and acid washing were tested to address this issue. The alcohol treatment proved
detrimental as catalytic material was leached from the silicate. The acid washed rhodium containing calcium silicate was quite active in
the hydroformylation of alkenes and did not suffer loss of catalyst into the product phase. Acid treated rhodium containing silicates were
more active than their untreated counterparts but less selective due to access to the rhodium centers being opened.
Keywords: Hydroformylation • Hex-1-ene • Rhodium • Heterogeneous Catalysis • Nano-Structured Calcium Silicate
©
Versita Sp. z o.o.
of the products have driven technology towards the
development of heterogeneous catalytic processes.
Aqueous biphasic hydroformylation systems are
1
. Introduction
Industrial aldehydes are synthesized by the conversion
of alkenes and alcohols using a catalytic process known to work well for short-chained linear alkenes [7-9].
called hydroformylation or oxo-process [1-6]. Aldehyde However, longer chained and branched alkenes present
products range from fine chemicals to substrates for food a problem, as they are not soluble in water. Therefore,
additives, perfumes, cosmetics and medicines. While aqueous processes aimed at synthesizing these show
the majority of aldehydes produced are short-chained, only very low activities. Biphasic approaches employing
the conversion of longer chained and branched alkenes phase transfer catalysts may provide solutions but
is gaining importance due to improvements in the encounter issues like entrainment and loss of catalyst
technology relating to their synthesis and applications.
Most hydroformylation processes in use are
into the product phase [7-9].
Consequently, several research groups have
homogeneous, as they allow more freedom for catalyst attempted the formation of solid-state catalysts, mainly
design and often have higher selectivity and activities, based on rhodium. Substrates for the immobilization of
measured as turn-over-frequencies (TOF) and turn- catalytically active species have been chosen out of the
over-numbers (TON) [1-6]. Concerns regarding the field of zeolites [10-17], silica [18-20], carbon [21], titania
recovery of the metals used as catalysts and the purity [22]andsiloxane[23]substrates.Organicbasedpolymers
*
E-mail: aoc@gmx.li
5
61

Rhodium catalysts build into the structure
of a silicate support in the hydroformylation of alkenes
[
24], dendrimers [25] and liquid ionic crystals [26,27] in
The catalytic conversion of alkenes to aldehydes was
combination with other supports or on their own have determined by gas chromatography using n-nonane as
been used as well.Some of the approaches introduced internal standard in all samples. The relative amounts
are very elegant and promising. For example, Chaudari of the standard, reactants and hydroformylation
et al have synthesized catalysts, where rhodium has products was measured on a HP 6890 Series GC gas
been incorporated into porous zeolites, alumina and chromatograph equipped with a HP-5 column 30 m ×
silica, which open the possibility of use in continuous 0.32 mm × 0.25 μm and FID detector. The carrier gas
processes [23]. One of the issues remaining with some was helium.
of the supported catalysts is that the catalyst can still be
lost into the product phase.
A Micrometrics Flowsorb II 2300 was employed
to record specific surface areas by measurement of
Previously, we have reported synthesis and desorption of 30% nitrogen and 70% helium at 21°C
modification of a nano-structured calcium silicate [28], and 1 atm.
NCaSil, and use of said material in the uptake of metal
X-ray powder diffraction patterns were collected on
ions from solution [29]. It is proposed that rhodium can a Philips PW 3710 MPD diffractometer; anode material
be readily absorbed by NCaSil and that the resulting Cu, λ = 1.54060 Å.
material can be used as a supported, heterogeneous
Oil absorptions were measured using the standard
catalyst in a hydroformylation reaction. However, Cairns test for the oil absorption of pigments by spatula rub-out
et al, and Barassi have found that metals can be released (ASTM D281).
from the surface of NCaSil again [29,30]. Therefore,
a different approach to the synthesis of a supported
catalyst was attempted. Rather than impregnating
preformed NCaSil with rhodium, the metal was built into Rhodium polyethylene glycolate was synthesized
the NCaSil structure during synthesis of the silicate. according to literature [7,8].
2
.2. Synthesis and
catalysts
characterization of
In this study we show, whether rhodium built into the
A modified synthesis of nano-structured calcium
structure of a silicate remains catalytically active, and silicate, NCaSil, was employed to improve binding of
if the metal is bound strongly enough into the silicate rhodium to the silicate. Instead of absorbing rhodium
support matrix to not contaminate the product.
onto a already prepared NCaSil, rhodium trichloride was
added during the silicate synthesis.
Reagent A: A solution of sodium silicate was
prepared in a plastic beaker from 5.65 g (0.0269 mol,
2
. Experimental procedure
2
8.5%wt SiO ) sodium silicate and 8.5 mL of water.
2
2
.1. General
Reagent B: In a separate beaker 2.03 g (0.0270 mol)
Reactions and measurements were carried out in calcium hydroxide and 0.142 g (0.538 mmol) rhodium
air and unless otherwise stated at room temperature. trichloride trihydrate were suspended in 2.5 mL of water.
-1
Synthesis gas, (99.90%) H :CO, 1:1, was obtained 1.71 mL (2.01 g, 33%wt, 1.18 g mL , 0.0364 mol)
2
from Gerling. Solvents used were distilled prior to use. HCl were added to the calcium hydroxide suspension
NMR spectra were measured using a Varian Inova carefully under stirring. Then further 6.2 mL of water
300 MHz spectrometer at room temperature. were added. The suspension formed had a pH-value of
Tetramethylsilane was used as an external standard 11.3 to 11.5.
1
for H NMR. Key to NMR data: s, singlet; d, doublet; t,
triplet; q, quartet; m, multiplet.
Reagent A and Reagent B were mixed vigorously
in a plastic beaker. The resultant slurry was aged for
Infrared spectra were recorded on a Bio-Rad 30 minutes, filtered (vacuum) and the filter cake washed
FTS-7 or a Perkin Elmer System 2000 NIR FT-Raman with 40 mL of water. The filtrate was collected and its
spectrometer. Samples were prepared as KBr pellets. rhodium content analysed according to the literature
Key to IR data: st, strong; m, medium; w, weak; br, [31].
broad; sh, sharp; sd, shoulder.
A Philips FEG-SEM scanning electron microscope slightly yellow filter cake was dried at 110°C for 2 days.
was used to investigate samples, which were coated
2-Ethoxyethanol washed catalysts, Cat-EE: After
with carbon, gold or platinum to avoid charge build-up washing the filter cake with water as described above,
during measurements.
the very pale red filter cake was washed with 20 mL of
The amount of rhodium not incorporated into the 2-ethoxyethanol and then dried at 110°C for 2 days.
Water washed catalysts, Cat-W: The resulting
silicate structure was analysed gravimetrically as the
trichloride [31].
Acid washed catalysts, pH adjusted to a value of
9, Cat-A9: The filter cake was washed with water as
562

T. Borrmann et al.
Table 1. Comparison of rhodium content, oil absorption and specific surface area.
Catalyst
Cat-W
Cat-EE
Cat-A9
Cat-A7
Rhodium content (mol g-1
Rhodium retention (%)
pH adjustment
)
3.29 × 10-5
100
1.64 × 10-5
50
3.19 × 10-5
95
2.31 × 10-5
67
None
127
none
159
to pH 9.0
142
to pH 7.0
123
Oil absorption (g oil per 100 g)
Specific surface area (m2 g-1
)
73
111
109
89
described before and then re-suspended in distilled standard were added. The reaction mixture was stirred
water. The pH-value of the resulting slurry was adjusted (1000 rpm) for 5 h at 80ºC and the consumption of
to 9.0 using hydrochloric acid. Then the slurry was reaction gas was recorded. Catalytic reactions were
filtered (vacuum) and the filter cake washed extensively terminated by depressurization after cooling in a water
with water again. The very pale yellow filter cake was bath. The catalyst was recovered by filtration and the
dried at 110°C for 2 days.
product phase analysed.
Acid washed catalysts, pH adjusted to a value of 7,
Using catalyst, an experiment was carried out with
Cat-A7: Synthesis followed the instructions Cat-A9 with 250 mg of catalyst Cat-W in the absence of toluene.
the difference that the pH-value of the suspension was In a further experiment, 210 mg of recovered catalyst
adjusted to 7.0. The filter cake was very pale grey prior Cat-A7 were employed.
to drying.
All catalysts: IR (KBr):  = 3619 (w, sd), 3561 (w, sh),
3
1
8
438 (st, br), 2931 (w, sh), 1630 (m, sh), 1447 (m, br), 3. Results and discussion
389 (w, sd), 1154 (st, sh), 1094 (st, sh), 1017 (st, sd),
-1
82 (m, sd), 781 (w, sh) 669 (m, sh), 606 (m, sh) cm .
Nano-structured calcium silicates, NCaSil, have been
2
-1
All catalysts: XRD, Cu K radiation, d-spacing [Å] (rel. reported with surface areas up to 500 m g and pore
α
intensity [%]): 6.02819 (55.86), 4.35711 (7.46), 3.47207 volumes able to accommodate 7 times the weight of
(
(
(
(
58.17), 3.00823 (100.0), 2.80728 (80.90), 2.72261 the silicate in linseed oil (determined by spatula rub-
19.47), 2.34454 (13.00), 2.14243 (13.80), 1.90977 out method) [28]. The silicate can be used to absorb
10.99) 1.84801 (33.35), 1.73964 (5.05), 1.69496 metal ions from solution, but doing so has a detrimental
11.20), 1.35544 (0.87), 1.30201 (5.89), 1.28662 (7.43). impact on the silicate structure [29,30]. Calcium and
1
Cat-W, Cat-A9, Cat-A7: H NMR: 1.603 (s, H O), silica are known to be dissolved out of the structure [30].
2
1
.303 (SiO…H O), 0.892 (SiOH, m) ppm.
Additionally, it is possible that the metal ions absorbed
Cat-EE: H NMR: 3.750 (-CH OH, m), 3.530 ( are released again into solution [30]. Therefore, it was
2
1
2
OCH CH O H, t), 3.408 ( CH2OCH CH OCH -, m), decided to include rhodium into the synthesis of the
2
2
2
2
2
1
.890 (-CH OH, t), 1.543 (s, H O), 1.255 (SiO…H O), silicate and create a rhodium-containing calcium silicate.
2
2
2
0
.879 (SiOH, m) ppm.
It was thought that doing so would improve the binding
of rhodium to the silicate backbone and prevent loss of
rhodium during catalytic tests. To test this assumption a
2
.3. Catalysis testing
Hydroformylation reactions were carried out in a Parr sample of rhodium-containing calcium silicate, Cat-W,
Mini-reactor 4560 (300 mL) equipped with temperature, was compared to a sample where rhodium was absorbed
stirring and pressure control device. The consumption onto premade NCaSil. In a simple leaching test 0.1 g
of carbon monoxide and hydrogen gas was monitored of each catalyst were suspended in 100 mL of ethanol
at constant pressure by the pressure drop from a gas for 2 days. Comparative experiments were carried out
reservoir. The pressure in the reservoir was recorded using 100 mL of water and 100 mL of toluene. In toluene
using a pressure transducer and recorder system.
no leaching within the margin of error was observed
For a typical experiment, 270 mg of a rhodium- (Table 1). In ethanol and water Cat-W experienced no
containing calcium silicate was dispersed in 20 mL loss of rhodium within the margin of error but some loss
toluene was placed in the reaction vessel, which then of calcium in water (Table 1). The rhodium absorbed
was charged with the reaction gas (H /CO, 7 MPa). onto NCaSil was leached out to a significant degree
2
Under stirring the autoclave was heated to 80°C both in ethanol and water (Table 1). Consequently, only
for 1 hour. Then the alkene (20.0 ml) and n-nonane rhodium-containing calcium silicates were investigated
-3
(
1.00 mL, 0.718 g cm , 5.60 mmol) as internal in catalysis tests and toluene was used to aid dispersal
5
63

Rhodium catalysts build into the structure
of a silicate support in the hydroformylation of alkenes
Table 2. Leaching test: Exposing rhodium containing calcium silicate Cat-W and rhodium absorbed on NCaSil to different solvents. Numbers
express the percentage of rhodium lost.
Material
Water
Ethanol
Toluene
Rhodium absorbed on NCaSil
58.9
0.1
77.3
0.2
0.1
0.0
Rhodium Containing Silicate Cat-W
of the catalyst during reaction and recovery of the silicate during the silicate formation. The measurements carried
by filtration. out did not allow any distinction between these two
While the rhodium-containing calcium silicates had possibilities. The presence of these groups though is
been proven to be more robust, incorporating rhodium concurrent with the catalysts having a lower surface
into the silicate structure appeared to have a detrimental area compared to NCaSil and might be an indicator of
effect on the surface properties of the silicate (Table 2).
structural differences in the silicate backbone between
Surface areas did only reach about 110 m g for the rhodium modified and unmodified materials.
rhodium-containing silicates. The pore volume appeared
The reduced surface area and porosity of rhodium-
2
-1
to be significantly reduced compared to NCaSil as well. It containing calcium silicates compared to NCaSil[28,32]
is known that during drying of NcaSil, surface tension can was also noticeable in electron microscope images
collapse the silicate structure and reduce its surface area of the material (Fig. 1). The micro-plate morphology
[
32]. Several methods have been developed to prevent of NCaSil was recognizable in Cat-EE, but the plates
the collapse, like washing the silicate with a solvent appeared thicker and the structure resembled more a
with a lower surface tension than water, like ethanol or collapsed NCaSil than one washed with an alcohol. All
2-ethoxyethanol, or washing the silicate with acids [32]. It catalyst samples looked quite comparable independent
was attempted to use these methods to improve the pore of treatment. Energy dispersive X-ray Analysis showed
volume and surface area of rhodium-containing calcium rhodium to be present close to the limit of detection
silicate as well. The rhodium-containing calcium silicate (Fig. 1). The peak associated with rhodium was of
washed only with water will be referred to as Cat-W in similar intensity to the one from the platinum coating
the following. The three variations were termed Cat-EE deposited during sample preparation. As the samples
for the silicate washed with 2-ethoxyethanol, Cat-A9 for were prepared on carbon tape, the carbon signal was
the material washed with HCl to pH 9.0, and Cat-A7 for neglected. However traces of a hydrocarbon impurity
the one adjusted to pH 7.0. Table 2 shows a comparison can be detected by infrared spectroscopy.
of the rhodium content and surface properties of these
The Cat-EE sample was further investigated by
catalysts. It was noted that acid washing resulted in X-ray mapping. The data from the energy dispersive
the loss of rhodium from the structure, with 5% of the X-ray spectroscopy is related to spatial information in
rhodium being leached out at pH 9.0 and 33% at pH regards to the location of elements on a sample. Clear
7.0. The XRD patterns of all catalysts were identical and red, blue or green dots represent presence of these
comparable to NCaSil, the proton NMR spectra revealed elements. Mixed colours originate, where two or more
the residues of 2-ethoxyethanol remaining in Cat-EE. elements are present at the same location. Large areas
Washing with 2-ethoxyethanol also led to a significant in only one colour indicate pure crystals or compounds.
amount of rhodium loss from the silicate. Only about The rhodium content is very low only just above the
half the silicate remained bound in Cat-EE compared background level. This means that conclusions in
to Cat-W.
regards to the presence and distribution of rhodium
The infrared spectra of the rhodium-containing have to be taken as indications only. However, the even
calcium silicates are similar to those of NCaSil [28,32]. spread of signals (green dots) in the overlay chart on
This was expected due to the small amount of rhodium the left and the good correlation between the elements
incorporated into the structure. However, two additional in the ternary chart (Fig. 1) imply an even distribution
-1
weak signals at 3619 and 3561 cm are present. These of all elements investigated. In the ternary correlation
-1
signals shift to 2588 and 2543 cm if the sample is treated chart no dots near the rhodium tip of the triangle are
with D O and can therefore be attributed to hydroxyl observed. Such dots would indicate the presence
2
groups.Thesesignalsarefoundinallrhodium-containing of discrete rhodium species. An even distribution of
calcium silicates studied and indicating the presence of rhodium on the silicate support was considered to be
either additional silanol groups or of rhodium hydroxide beneficial for catalytic conversions as it influences the
formed due to the basic environment, pH > 11.5, present number of available catalytic sites.
564

T. Borrmann et al.
Figure 1. Scanning electron microscope image, energy dispersive X-ray spectrum (excitation energy 15 keV, spot size 3.0) and derived pseudo-
colour X-ray mapping and elemental correlation chart for Cat-EE.
The number of available sites should be in direct reaction time was chosen to make the tests comparable
relation to the activity of a catalyst. Considering the to prior studies [7,8]. In one attempt it was established
amount of rhodium per gram of catalyst and the porosity that close to 100% conversion was reached after about
of the materials, expressed in their surface area and 9 hours. Occurrence of non-volatile products from side
oil absorption, it was assumed that the catalysts active reactions and loss of catalyst was noted as well. The
in the following order: Most active Cat-A9 > Cat-A7 > activity of catalytic conversions was recorded as turn
Cat-EE > Cat-W least active. To investigate catalytic over frequency (TOF) in mol product per mol catalyst, in
activity the conversion of alkenes into the corresponding this case rhodium, per hour.
aldehydes in hydroformylation reactions was attempted.
All rhodium-containing calcium silicates showed
Mainly hex-1-ene was used as alkene as significant data a high activity in the hydroformylation of hex-1-ene
on this hydrocarbon had been collected in prior studies (Table 3) quite comparable in magnitude to the catalyst,
[
7,8]. The conversion of hex-1-ene into heptanals (linear PEG, investigated in a previous study [8]. In comparison
and branched) was recorded after 5 hours along with to the phase transfer catalyst used previously [8], which
the ratio of linear to branched aldehydes. A 5 hour needed to be recovered by phase separation, catalysts
5
65

Rhodium catalysts build into the structure
of a silicate support in the hydroformylation of alkenes
Table 3. Hydroformylation of hex-1-ene using rhodium-containing calcium silicates.
Catalyst
Rh] (mol × 10-6 (ppm))
Cat-W
Cat-W
Cat-W
Cat-EE
Cat-A9
Cat-A7
Cat-A7
PEG
[
8.88 (27.9)
133 (73.8)
0.58
8.20 (26.0)
129 (69.8)
0.55
8.88 (27.9)
104 (76.1)
0.63
4.43 (13.9)
135 (75.1)
0.59
8.61 (27.1)
148 (82.6)
0.54
6.24 (19.6)
150 (83.1)
0.67
4.85 (32.6)
112 (65.5)
0.68
6.9 (18.7)
102 (60.2)
1.08
Conversion (mmol (%))
iso/n-products
TOF [mol mol-1 h-1
]
2978
3050
2529
6096
3447
4810
4618
4818
Special conditions
(1)
(2)
(3)
(4)
(
(
(
(
1) no toluene
2) oct-1-ene not hex-1-ene converted
3) recycled catalyst
4) catalyst from[8]
investigated here were solids. As such they could be the surface area of this catalyst was comparable to
recovered after reaction from the product phase by Cat-W. In contrast the activity of Cat-A9 in relation to
simple filtration. However, due to the small amounts Cat-W could be explained quite well over an increase
of catalyst used, recovery of the pure catalyst was in pore volume and, hence, accessible catalytic sites;
difficult. Therefore toluene was added to the process the amount of improvement in the oil absorption (12%)
to improve catalyst dispersion and product separation. matches the better activity (15%) of Cat-A9 quite well.
Toluene has been shown to not leach rhodium from The turn over frequency of the conversion using Cat-A7
the catalyst earlier. Furthermore, it was proven to not was about 60% higher than that of Cat-W, while their
influence the activity and selectivity of the process surface areas and oil absorptions were very similar. A
significantly. Two catalytic tests were carried out – possible explanation can be found in a closer look at
one in the presence, one in the absence of toluene. the acid washing. During treatment of nano-structured
The selectivity and turn over frequency recorded for calcium silicate with acid to pH 7.0 about 95% of calcium
these tests was comparable within the margin of error ions and silanol groups are removed [32]. While also one
(
Table 3). In all cases some catalyst was lost, trapped in third of the rhodium was lost, it is likely that the residual
a sticky, potentially polymeric, non-volatile residue. The rhodium became far more accessible. Consequently,
total amount of non-volatile solids produced was about Cat-A7 had a considerably higher number of available
3
% for the conversion of hex-1-ene. If oct-1-ene was sites than Cat-W and Cat-A9. This assumption is
converted, this amount increased to 8.9%.
supported by the fact that the conversion of hex-1-ene
ThePEGcatalystinvestigatedintheliteratureshowed using Cat-A7 was less selective than the ones carried
little selectivity towards branched or linear aldehydes out with the other catalysts. It is likely that the calcium
[
8]. The catalysts tested in this study produced nearly and silanol limiting access to the rhodium also influence
twice the amount of linear than of branched aldehydes. the selectivity.
The selectivity was less for the conversion of oct-1-ene
but still comparable.
Finally, the reuse of the spent catalysts was
attempted. Due to having been shown to be robust –
Comparing the different catalysts produced in this no loss of rhodium into the product phase was noticed
study proved the earlier estimation in regards to their – and having the largest activity Cat-A7 was chosen
relative activities wrong. The 2-ethoxyethanol washed for this experiment. First a catalytic conversion of hex-
material, Cat-EE, was the most active by a factor of two 1-ene using Cat-A7 in the presence of toluene was
compared to the water washed catalyst, Cat-W. The carried out. Then the catalyst was recovered from the
reason for this became apparent as it was noticed that autoclave by filtration and dried for several days at
rhodium was lost into the product phase. It is possible 110°C in an oven. Some catalyst was lost due to the
that the presence of the 2-ethoxyethanol residue in presence of a non-volatile residue as mentioned earlier.
Cat-EE facilitated this leaching or that the material was The autoclave was cleaned. The product phase was
the least structurally sound of the ones investigated. placed in the autoclave again with fresh hex-1-ene to
The loss of rhodium means that Cat-EE has to be test for catalytic material being lost into this phase or
considered unsuitable as catalyst. Cat-W as well as the being left in the autoclave. No catalytic activity was
acid washed materials displayed no noticeable leaching noticed indicating that no catalytic material was leached
of rhodium into the product phase. Both acid washed into the product or stuck in the autoclave. The autoclave
materials, Cat-A9 and Cat-A7, performed better than was cleaned again. The recovered catalyst was placed
Cat-W. In the case of Cat-A7 this was unexpected as into the autoclave with toluene and fresh hex-1-ene. A
566

T. Borrmann et al.
catalytic conversion was carried out, and the products of The freeing of access to the rhodium had the effect of
this conversion analysed. It was found that the recycled reducing selectivity. It would be interesting to attempt
catalyst performed very comparable to the fresh catalyst optimization of the catalyst by controlled acid treatment
(
Table 3).
and to find a regime by which activity is increased while
the selectivity is not changed. Furthermore, it would be
interesting to test the rhodium containing silicates in
other catalytic applications, such as hydrogenation, and
also employ different metals.
4
. Conclusions
Nano-structured calcium silicate can be used as catalytic
support for rhodium. If rhodium is impregnated onto
the silicate, the metal is not well bound to the silicate Acknowledgements
support and can be leached into the product phase or
a solvent present. By adding rhodium trichloride to the We would like to thank Katherine Hobbes and Bryony
synthesis of the silicate a more robust catalyst has been J. James, Department of Chemical and Materials
created, which does not loose rhodium into solvents Engineering, School of Engineering, The University of
in contact with the catalyst. However, if alcohols are Auckland, Private Bag 92019, Auckland, New Zealand,
used in the making of the rhodium containing silicate, for the assistance in recording the scanning electron
the material suffers from catalyst leaching. The silicate microscope images. We also thank Professor Dr. Dr.
can be acid washed, which results in the loss of some mult h.c. Herbert W. Roesky for the free use of the
rhodium but appeared to remove blockage around the equipment necessary to carry out the hydroformylation
remaining catalytic centres to increase catalytic activity. tests.
References
[
[
1] B. Cornils, W.A. Herrmann, R.W. Eckl, J. Mol. Catal.
A: Chem. 116, 27 (1997)
2] (a) W.A. Herrmann, C.W. Kohlpainter, Angew.
23, 52 (1984)
[14] E. Rode, M.E. Davis, B.E. Hanson, Chem. Commun.
716 (1985)
Chem. 105, 1588 (1993); (b) W.A. Herrmann, [15] K. Mukhopadhyay, R.V. Chaudari, J. Catal. 213, 73
C.W. Kohlpainter, Angew. Chem. Int. Ed. Engl. 32,
524 (1993)
3] B. Cornils, W.A. Herrmann (Eds.), Applied
(2003)
1
[16] Q. Peng, Y. Yang, Y. Yuan, J. Mol. Catal. A: Chem.
219, 175 (2004)
[
Homogeneous Catalysis with Organometallic [17] K. Mukhopadhyay, A.B. Mandale, R.V. Chaudari,
Compounds (VCH, Weinheim, New York, Basel,
Tokyo, 1996)
4] B. Cornils, W.A. Herrmann (Eds.), Aqueous-Phase
Organometallic Catalysis (VCH, Weinheim, 1998)
5] F. Ungváry, Coord. Chem. Rev. 241, 295 (2003)
6] F. Ungváry, Coord. Chem. Rev. 248, 867 (2004)
Chem. Mater. 15, 1766 (2003)
[18] M.E. Davis, E. Rode, D. Taylor, B.E. Hanson,
J. Catal. 86, 67 (1984)
[
[19] J.A.
Diaz-Aunon,
M.C.
Roman-Martinez,
[
[
[
C. Salina-Martinez de Lecea, H. Alper, Stud. Surf.
Sci. Catal. 143, 295 (2002)
7] U. Ritter, T Borrmann, S. Bogdanovic, H.W. Roesky, [20] H. Zhu, Y. Ding, L. Yan, Y. Lu, C. Li, X. Bao, L. Lin,
Hoechst AG, Patent US 6,225,508 (1997)
Chem. Lett. 33, 630 (2004)
8] T. Borrmann, H.W. Roesky, U. Ritter, J. Mol. Catal. [21] T.A. Kainulainen, M.K. Niemelä, A.O.I. Krause,
A: Chem. 153, 31 (2000)
J. Mol. Catal. A: Chem. 122, 39 (1997)
9] C.W. Kohlpainter, R.W. Fischer, B. Cornils, Appl. [22] T. Garcia, B. Solsona, D.M. Murphy, K.L. Antcliff,
Catal. A: General 221, 221(2001)
S.H. Taylor, J. Catal. 229, 1 (2005)
10] N. Takahashi, M. Kobayashi, J. Catal. 85, 89 [23] R.P.J. Bronger, J.P. Bermon, J.N.H. Reek,
[
[
[
[
(
1984)
11] P.A. Jacobs, N.I. Jaeger, P. Jíru, G. Schulz-Eklov
Eds.), Studies in Surface Science and Catalysis:
P.C.J. Kamer, P.W.N.M. van Leeuwen, D.N. Carter,
P. Licence, M. Poliakoff, J. Mol. Catal. A: Chem.
224, 145 (2004)
(
Metal Microstructures in Zeolites (Elsevier, [24] A.D. Pomogailo, Kinetics and Catalysis 45, 61
Amsterdam, 1982)
(2004)
12] E. Rode, M.E. Davis, B.E. Hanson, Chem. Commun. [25] A. Dahan, M. Portnoy, J. Polym. Sci.: Part A: Polym.
477 (1985)
Chem. 43, 235 (2005)
13] B.E. Hanson, M.E. Davis, D. Taylor, Inorg. Chem. [26] Y. Yang, H. Lin, C. Deng, J. She, Y. Yuan, Chem.
[
[
1
5
67

Rhodium catalysts build into the structure
of a silicate support in the hydroformylation of alkenes
Lett. 34, 220 (2005)
27] A. Riisager, R. Fehrmann, S. Flicker, R. van Hal,
[30] G. Barassi, PhD thesis (Victoria University,
Wellington, NZ, 2012)
M. Haumann, P. Wasserscheid, Angew. Chem. Int. [31] L. Gmelin, Handbuch der Anorganischen Chemie,
Ed. Engl. 44, 815 (2005)
8 edition (Verlag Chemie, Berlin, 1938) Vol. Rh
28] T. Borrmann, A. Nukui, A.J. McFarlane, [32] A.J. McFarlane, PhD thesis (Victoria University,
[
[
[
K.J.D. MacKenzie, J.H. Johnston, Powder
Diffraction 23, 204 (2008)
29] M.J. Cairns, T. Borrmann, J.H. Johnston, W. Hoell,
Micropor. Mater. 95, 126 (2006)
Wellington, NZ, 2007)
568