Selective production of glycols from xylitol over Ru on covalent triazine frameworks-suppressing decarbonylation reactions

Article abstract of DOI:10.1039/c8gc00208h

Ru on covaltent triazine frameworks (CTF) are highly active and selective catalysts for the conversion of xylitol to glycols (80% C-yield) in basic media. With increasing N-content decarbonylation reactions are suppressed leading to high glycol selectivit

Full text of DOI:10.1039/c8gc00208h

Green Chemistry
View Article Online
View Journal
PAPER
Selective production of glycols from xylitol over
Ru on covalent triazine frameworks – suppressing
decarbonylation reactions†
Cite this: DOI: 10.1039/c8gc00208h
a
a
a
b
Anna Katharina Beine, Andreas J. D. Krüger, Jens Artz, Claudia Weidenthaler,
c
a
a
Christoph Glotzbach, Peter J. C. Hausoul
and Regina Palkovits
*
Ru on covaltent triazine frameworks (CTF) are highly active and selective catalysts for the conversion of
xylitol to glycols (80% C-yield) in basic media. With increasing N-content decarbonylation reactions are
suppressed leading to high glycol selectivity. The suppression can be attributed to the presence of N in
the support and to metal-support interactions. The catalysts exhibit high stability and could be recycled
5 times with minor loss of activity.
Received 19th January 2018,
Accepted 20th February 2018
DOI: 10.1039/c8gc00208h
rsc.li/greenchem
fraction can be used for the production of ethanol and plat-
form chemicals such as 5-hydroxymethylfurfural and levulinic
Introduction
5
–7
The use of lignocellulosic biomass for the production of acid.
renewable commodity chemicals can decrease the dependence mainly used to produce furfural. However hemicellulose also
The hemicellulose fraction, on the other hand, is
8
1
–3
on fossil resources and reduce CO2 emissions.
Especially possesses great potential for the production of sustainable
9
,10
the conversion of non-edible (hemi)cellulose to glycols is a alkanes and alcohols.
In particular the selective hydrogeno-
promising route for valorisation (Scheme 1). Ethylene and pro- lysis of xylitol (XYL) to glycols represents an interesting reac-
pylene glycol (EG and PG) are important commodity chemicals tion in green chemistry.
that find wide application as monomers and antifreezes and
are currently produced from petroleum–based ethylene and alysed dehydrogenation, base-promoted retro-aldol splitting
The conversion of XYL to EG and PG proceeds via metal-cat-
1
1–16
propylene. Recent studies have demonstrated that ligno- and rehydrogenation.
Under basic conditions the most
1
7
cellulosic biomass can be fractionated to insoluble cellulose, important side reaction is the formation of lactic acid (LA).
soluble lignin and hemicellulose-based sugars. The cellulose Often used metals in the reaction are Ni, Pt and Ru.
4
18,19
For
Ni-based catalysts different supports and additional metals
were used. Ni supported on activated carbon (AC)-LaOH3
enabled the conversion of cellulose to glycols and resulted in a
combined yield of 64% while Ni supported on Ce–TiO con-
2
2
0,21
verted xylitol to glycols with a combined yield of 67%.
Using a catalyst containing Ni and CaO on carbon Sun et al.
were able to produce glycols from xylitol with a selectivity of
1
9
7
0%. Ni- and W-containing catalysts on carbon can even
facilitate higher selectivity and activity of up to 80% for sugars
Scheme 1 Production of glycols from biomass via metal-catalysed as well as sugar alcohols.2
2–25
Nevertheless, major challenges
hydrogenolysis.
of these catalysts are leaching, oxidation and agglomeration
which lead to an unsatisfactory stability of these catalysts.
For Pt/C a higher activity was found for the hydrogenolysis of
xylitol compared to Pd, Rh and Ru supported on carbon. The
obtained selectivity to glycols of 48%, though, was lower than
for Ru/C. For the conversion of cellulose also a low glycol
2
4,25
a
RWTH Aachen University, Worringerweg 2, 52074 Aachen, Germany.
1
8
E-mail: palkovits@itmc.rwth-aachen.de; Fax: +49 241-80-22177;
Tel: +49 241 80 26497
b
Max-Planck-Institut für Kohlenforschung, Kaiser-Wilhelm-Platz 1, 45470 Mülheim,
Germany
26
selectivity of 24% was reached. The selectivity could be
c
Thyssenkrupp Industrial Solutions AG, ThyssenKrupp Allee 1, 45143 Essen,
increased to 62% by using a bifunctional Pt and W containing
Germany
2
6
catalyst on carbon support. One drawback of Pt-based cata-
lysts is the high raw material price. For Ru-based catalysts, the
†
Electronic supplementary information (ESI) available. See DOI: 10.1039/
c8gc00208h
This journal is © The Royal Society of Chemistry 2018
Green Chem.

View Article Online
Paper
Green Chemistry
Table 1 Surface area, total pore volume and elemental analysus for
prepared CTF-materials
BET [m g−1]
2
V
P(total) [cm g−1]
3
N-Content [%]
Catalyst support
S
CTF-b
CTF-c
CTF-d
1399
1595
1949
0.92
0.97
1.61
17.2
10.4
3.7
2
−1
range between 1400–1950 m g (see Table 1 and Fig. S1 in
Scheme 2 Monomers for the production of CTF-materials, polymeris-
ation and impregnation.
3
7,38
the ESI†) and are comparable to those reported before.
The N-content was determined by elemental analysis and
varied from 3.7% for CTF-d to 17.2% for CTF-b. The CTF-
occurrence of side reactions leads to a wide product spectrum materials were loaded with 5 wt% Ru by coordination with
1
5,16,27,28
and rather low selectivity.
2 2
[RuCl (p-cymene)] in refluxing methanol.
A common strategy for improving the selectivity of catalysts
is variation of the support material. For Ru-based catalysts
Catalyst screening
Reactions were carried out at 473 K under 8 MPa H
was used as base as it was found to be most selective in pre-
vious studies. A screening of the most prominent metals (Ni,
Pt and Ru) on carbon support was carried out for XYL hydro-
genolysis (Table 2). These results are compared to the results
of the prepared Ru/CTF catalysts (Fig. 1). All tested carbon sup-
materials such as AC, carbon nano-fibers (CNF), Al
and TiO were tested. Up to 20% EG and 38% PG was obtained
using CNFs, although the formation of LA was not
2 3 2
O , ZrO
2
. Ca(OH)
2
2
1
8
7
,29,30
reported.
In addition, doping carbon materials with
heteroatoms such as phosphorus and sulfur has been studied
3
1–33
for sugar alcohol hydrogenolysis.
The use of a nickel phos-
phate phase led to an increase of catalytic activity, while
selectivity was only changed minimally. The use of sulfur
increased the selectivity for PG to 41% and EG to 26% using
sorbitol as substrate. However, the activity decreased signifi-
cantly and after 4 h only 14% conversion was observed, indi-
Table 2 Screening of metal on carbon catalysts in the hydrogenolysis
a
of XLY
3
3
cating that S-doping leads to catalyst inhibition. Also in this
study the formation of LA was not investigated. Recent studies
have shown that N-doping can lead to improved performance
in electrochemistry, gas adsorption and separation as well as
Metal
loading
wt%]
X (PentOH)
[%]
S (EG)
[%]
S (PG)
[%]
S (GLY)
[%]
S (LA)
[%]
[
3
4–38
catalysis.
Especially, the basic surface properties, high Ni/C
Pt/C
10
5
5
100
97
100
36
37
17
34
30
29
2
13
1
15
20
17
metal dispersion and strong metal–support interactions lead
Ru/C
3
3
to improved performance and catalyst stability. Recently,
covalent triazine frameworks (CTFs) were introduced as tailor-
made porous materials with tuneable N-content based on the
choice of monomer (Scheme 2). They were used as supports
in noble metal-catalysed oxidation reactions and showed a
a
Conditions: 473 K, 3 h, 8 MPa H
2
, 2.0 g xylitol, 0.2 g catalyst, 0.3 g
Ca(OH)
2
, 20 mL H O. PentOH describes XYL and all other pentitols.
2
3
7
3
5,38
molecular metal dispersion and good stability.
CTFs enable a systematic variation of factors assumed
crucial for selectivity such as N-content Therefore they were
studied in XYL hydrogenolysis and we are able present CTFs as
selective support for Ru-based catalysts.
Results and discussion
Catalyst characterisation
CTF-materials based on the monomers, pyridine-2,6-dicarbo-
nitrile (CTF-b), terephthalonitrile (CTF-c) and biphenyl-4,4′-
dicarbonitrile (CTF-d) were prepared by polymerization in
ZnCl at 873 K for 10 h (nomenclature of the CTF-materials
2
Fig. 1 Influence of the nitrogen content of the support on the Ru-cata-
lysed hydrogenolysis of xylitol (conditions: 473 K, 8 MPa H , (1) 4 h, (2)
h, 2.0 g xylitol, 0.2 g catalyst, 0.3 g Ca(OH) , 20 mL H O, PentOH
of the prepared CTF-materials calculated by the BET method describes XYL and all other pentitols).
3
8
was adopted from previous publications). The general syn-
thesis route is shown in Scheme 2. The specific surface areas
2
3
8
3
2
2
Green Chem.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Green Chemistry
Paper
ported catalysts show close to full conversion of XYL and all
other pentitols (PentOH) like ribitol and arabitol after a reac-
tion time of 3 h. The highest selectivity is observed for Pt/C
and Ni/C showing a combined glycol selectivity of 67% and
7
0%, respectively. For Ru/C only low glycol selectivity of 46% is
observed. Besides the main products EG and PG, GLY and LA
are observed in varying amounts. For Ru/C the HPLC-based
mass balance was incomplete indicating the formation of
further side products. Extending the reaction time to 4 h leads
to a further decrease in glycol selectivity for Ru/C to only 12%
for EG and 23% for PG (Fig. 1). In contrast for Ru/CTF-based
catalysts the selectivity to glycols is higher and increases with
increasing N-content whereas the selectivity to LA decreases.
For Ru/CTF-b 36% selectivity to EG and 44% to PG was
obtained. This combined glycol selectivity of 80% exceeds the
highest selectivity of 67% reached for Ru-based systems by
Tronci et al. as well as the metal on carbon catalysts investi-
Fig. 3 Liquid, gaseous and solid product yields for the Ru/C and Ru/
CTF-b catalysed hydrogenolysis of xylitol (conditions: 473 K, 4 h, 8 MPa
2 2 2
H , 2.0 g xylitol, 0.2 g catalyst, 0.3 g Ca(OH) , 20 mL H O).
analysis of the gas phase after 4 h of reaction revealed only
CH4 with a C-yield of 24% for Ru/C whereas Ru/CTF-b only
gave 8% (Fig. 3). It is proposed that CH is formed from the
4
3
3
gated above. The selectivity is comparable to Ni and W con-
1
8–21
taining systems that often suffer from leaching.
It is also
decarbonylation product CO under high H pressure. CO can
2
noteworthy that the selectivity to LA and GLY is low at 11.1%
and 5.8%, respectively.
also react to form CH OH or CO however, CH OH was not
3
2;
3
detected. It is likely that any CO
2
formed will react with the
present Ca(OH) to CaCO . Metal carbonates can further react
2
3
4
0
Product spectrum
with CO to metal oxalates. The formation of CaC
O was
2 4
already observed in the hydrogenolysis of xylitol over Cu/
SiO2. To quantify CO3 formation, the reaction solution was
Fig. 2 shows the time dependence of the reactions of Ru/C and
Ru/CTF-b. In case of Ru/C, PentOH are converted within 2 h of
reaction (Fig. 2, left). Initially, EG, PG, GLY and LA are formed
with high selectivity but are subsequently consumed and the
mass balance decreases rapidly over time. In addition, glycolic
acid (GA) was detected as side-product in <1%. The GC data of
peracetylated reaction mixtures (Fig. S2 in the ESI†) also reveal
the formation of threitol (THR) which, together with GLY and
EG, is rapidly consumed. In line with previous observations
4
1
2−
acidified and the evolved CO volume measured. Oxalate for-
2
mation was analysed by HPLC of the acidified solution. For
Ru/C, 2% CO in form of CaCO was formed, 1% for Ru/CTF-b,
2
3
whereas CaC O could not be observed in either case. For Ru/C
2
4
1
a combined yield of C -products of 26% can be found whereas
for Ru/CTF-b only 9% were formed leading to an improved
glycol yield.
3
9
Considering the reaction mechanism (Scheme 3), in the
first reaction step dehydrogenation takes place yielding
adsorbed 1-/5-aldoses or 2-/3-/4-ketoses via micro-reversibil-
this indicates the occurrence of decarbonylation reactions.
In contrast, for Ru/CTF-b, PentOH are fully converted after 3 h
and much higher yields and selectivities are obtained (Fig. 2,
right). Only after 4 h a slight decrease in C-balance is observed.
In addition, the GC-data of the peracetylated reaction solution
reveals only small amounts of THR (Fig. S3 in the ESI†). GC
1
1
ity. For the 1-/5-aldose (i.e. xylose) decarbonylation yields
THR, which is observed for Ru/C but not for the Ru/CTF-cata-
Fig. 2 Time-dependence of the Ru/C (5 wt%)(left) and Ru/CTF-b (right)
catalysed hydrogenolysis of xylitol (conditions: 473 K, 8 MPa H
xylitol, 0.2 g catalyst, 0.3 g Ca(OH) , 20 mL H O).
2
, 2.0 g
Scheme 3 Proposed reaction mechanism for the hydrogenolysis of
pentitols.
2
2
This journal is © The Royal Society of Chemistry 2018
Green Chem.

View Article Online
Paper
Green Chemistry
Fig. 4 Lactic acid hydrogenation over Ru/C and R/CTF-b (conditions:
4
H
73 K, 4 h, 8 MPa H
2 2
, 0.65 g LA, 0.2 g catalyst, 0.3 g Ca(OH) , 20 mL
Fig. 5 Conversion and selectivity for Ru/CTF-b using different metal
2
O).
precursors (conditions: 473 K, (1) 2 h, (2) 3 h, 8 MPa H
.2 g catalyst, 0.3 g Ca(OH) , 20 mL H O).
2
, 2.0 g xylitol,
0
2
2
lysts. Retro-aldol condensation (RA) of the 1-/5-aldose and 2-/
-ketose leads to C and C intermediates, which after hydro-
genation yield GLY and EG. RA of the 3-ketose leads to a C /C
4
3
2
Nevertheless, in all cases the selectivity to EG was compar-
split forming THR and erythritol in a 1 : 1 ratio; however, the able and the mass balance approached 100%, indicating sup-
4
1
GC data emphasise that this does not occur.
pression of decarbonylation.
The increased PG selectivity and decreased LA formation
for Ru/CTF may be the result of increased LA hydrogenation. Ru nanoparticles
To test this we have carried out reactions on LA using Ru/C
Based on the data above it is proposed that Ru is reduced to
nanoparticles during the reaction. To gain more insight into
and Ru/CTF-b (Fig. 4). After 4 h, Ru/C gave full conversion with
5% selectivity to PG and gas phase analysis revealed 94%
CH yield (C-balance over 100%). In contrast, Ru/CTF-b only
gave 34% conversion with 28% selectivity to PG and only 1%
CH . These data clearly show that CTF supports can effectively
1
nanoparticle formation on Ru/CTF-catalysts, H -chemisorption
2
4
was performed on the reduced materials (Table 3). The results
confirm a clear trend between the active metal surface and the
N-content of the support. Going from CTF-d to CTF-b the
N-content increases from 3.7% to 17.2% and the accessible Ru
4
suppress decarbonylation and have a pronounced effect on the
selectivity of Ru. However, the higher PG selectivity cannot be
attributed to LA hydrogenation. Hence the cause for this effect
appears to be due to the electronic structure of the active
metal or the presence of nitrogen in the support.
2
−1
surface increases from 1.8 to 8.7 m
Conversely the calculated crystallite size decreases from 11.1
cubic)/13.3 (hemispherical) nm to 2.3/2.8 nm. This indicates
g , respectively.
(
that a high N-content improves the stabilization of small Ru
nanoparticles (e.g. by inhibiting sintering). Interestingly, the
particles on Ru/C exhibit comparable properties as Ru/CTF-b.
Hence it can be concluded that the dramatic difference in
selectivity between these catalysts can be attributed to the pres-
ence of nitrogen in the support.
Influence of the Ru-oxidation state
The employed impregnation method leads to a molecular dis-
persion of Ru on the support. However, hydrogenolysis is
carried out at high H pressure and temperature, which are
2
conditions that facilitate nanoparticle formation. This raises
the question if the observed selectivity change is the result of
metal complexes or a modulation of the reactivity of Ru nano-
particles by the CTF-support. To test this, we have performed a
reaction with [RuCl (p-cymene)] in the presence of 2,2′-bipyri-
Influence of nitrogen in the support
As discussed above, higher yields are obtained for Ru/CTF cat-
alysts due to a suppression of the decarbonylation activity typi-
2
2
dine (3 eq. w.r.t. Ru) as a homogeneous model for CTF-b
Fig. S3 in the ESI†). The data reveal only 20% conversion after
h reaction while LA is the main product, indicating that
molecular species are much less reactive and selective. In
addition, we have studied the influence of the oxidation state
(
2
Table 3
2
H -Chemisorption and elemental analysis results for Ru/CTF-
catalysts and Ru/C (samples reduced under H
prior to measurement)
2
-flow at 623 K for 16 h,
of the metal precursor (Fig. 5). CTF-b loaded with RuCl gave
similar activity and selectivity as [RuCl (p-cymene)] , although
2 2
the formation of LA was higher at the expense of PG. To obtain
Crystal size
3
(
cubic/
N-Content
[%]
Ru surface
Dispersion
[%]
hemispherical)
[nm]
2
−1
Catalyst
[m g
]
a Ru(0) catalyst, the RuCl loaded material was reduced under
3
Ru/CTF-b
Ru/CTF-c
Ru/CTF-d
Ru/C
17.2
10.4
3.7
0
8.7
6.0
1.8
7.2
26.3
18.0
5.5
2.3/2.8
3.4/4.0
11.1/13.3
2.8/3.4
2 3
H at 623 K for 3 h (RuCl -red.). For this catalyst the selectivity
to PG was significantly less due to increased formation of LA
and GLY.
21.6
Green Chem.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Green Chemistry
Paper
cally observed for Ru/C. In addition the selectivity to LA already described in literature, CTF materials exhibit a high
4
3
decreases with increasing N-content and active metal surface. surface basicity. Especially pyridinic, pyrollic and graphitic
This suggests that nitrogen in the support influences not only nitrogen species present in the CTF-b material possess lewis
sintering but also metal particle size. XPS studies were carried basic functions. This way nitrogen affects the local catalyst
out to gain more insight into the influence of the support on environment and may possess a high affinity for reaction inter-
the surface chemistry of Ru. Due to the overlapping regions of mediates. However when considering that the reaction pro-
C 1s and Ru 3d, it was chosen to use N 1s of pyridinic nitrogen ceeds under strongly basic conditions, we suggest that the
at 398.0 eV as reference. This binding energy was confirmed electronic effect on Ru is the most likely cause for the suppres-
for CTF-b (398.04 eV) and unreduced Ru/CTF-b (398.06 eV) sion of decarbonylation. Moreover it is proposed that the elec-
when referenced to C 1s of adventitious carbon (284.5 eV). The tron donation of N to Ru decreases adsorption energies and
data show that Ru-loading does not influence the peak posi- facilitates desorption and as a result decreases the occurrence
tion of the pyridinic N 1s signal (Fig. S5 in the ESI†). Besides of side reactions.
this the CTF materials also exhibit pyrollic, graphitic, oxidic,
and chemisorbed nitrogen species (Fig. 6a). For the [RuCl (p- Catalyst stability
2
2
cymene)] -loaded CTF-b also four different C species can be
found possibly corresponding to the varying C–N bindings
Finally, the stability of Ru/CTF-b was studied over five runs of
h (Fig. 7). Activity only slightly decreases over five cycles.
1
while the Ru 3d5/2 signal is observed at 281.5 eV (Fig. 6b) con-
Conversion decreases gradually from 80% to 67% and the
selectivities to the desired products are comparable in all
cases. ICP-OES of the reaction mixtures indicates only
minimal leaching which decreases to below the detection limit
after 3 runs (Table S1 in the ESI†). TEM images of the spent
catalyst reveal well dispersed nanoparticles of 2–3 nm size
4
2
firming the Ru(II) state. After reduction (Fig. 6c) the Ru 3d5/2
signal is observed at 297.7 eV indicating a complete reduction
4
2
to Ru(0). Compared to the signal of freshly reduced Ru/C
Fig. 6d), which exhibits a binding energy of 280.3 eV for the
(
Ru 3d5/2 signal, the signal of Ru/CTF-b is shifted 0.6 eV to
lower binding energy. We propose that this shift is the result
of the donating property of N in the CTF-support. Therefore,
these data confirm that the electronic structure of the Ru
surface is affected by the CTF support. Besides this the change
in selectivity might also be the result of the support itself. As
(
Fig. 8). This is similar to the particle size observed by chemi-
sorption and supports the proposition that the CTF-support is
able to stabilise metal nanoparticles. These data clearly
demonstrate that Ru/CTF-b is a highly active, selective as well
as stable catalyst for the hydrogenolysis of xylitol.
Fig. 7 Recycling of Ru/CTF-b in the hydrogenolysis of xylitol (con-
ditions: 473 K, 8 MPa H
Ca(OH) , 20 mL H O).
2
, 1 h reaction, 2.0 g xylitol, 0.2 g catalyst, 0.3 g
2
2
Fig. 6 XPS spectra for (a) N 1s of CTF-b, (b) C 1s and Ru 3d for Ru/CTF-
b, (c) C 1s and Ru 3d of reduced Ru/CTF-b, (d) C 1s and Ru 3d for freshly
reduced Ru/C (a–c: referenced to N 1s at 398.0 eV, d: referenced to C 1s
at 284.5 eV).
Fig. 8 Dark field TEM-image of the spent Ru/CTF-b catalyst after 5 re-
cycling runs.
This journal is © The Royal Society of Chemistry 2018
Green Chem.

View Article Online
Paper
Green Chemistry
The monochromatised Al K
operated at 15 kV and 15 mA. Binding energies were referenced
α
X-ray source (E = 1486.6 eV) was
Conclusions
In conclusion, Ru/CTF-b presents a highly selective catalyst for to N 1s at 398.0 eV and C 1s to 284.5 eV.
the production of glycols from XYL. Up to 80% yield of glycols
CTF-synthesis
were obtained in the presence of Ca(OH) while successfully
2
suppressing the formation of LA. Variation of the support Dicarbonitrile (2,6-pyridinedicarbonitrile (CTF-b), 1,4-di-
revealed that the selectivity to EG and PG increases with the cyano-benzene (CTF-c) or 4,4′-Biphenyldicarbonitrile (CTF-d),
nitrogen content of the CTF-support, whereas the selectivity to 4.85 mmol, 1 eq.) was ground with ZnCl (24.25 mmol, 5 eq.)
2
LA decreases. In contrast to Ru/C, the products are not decom- and transferred into a quartz ampoule. After drying for 3 h at
posed (via decarbonylation) leading to much higher yields of RT in high vacuum, the ampoule was flame-sealed and placed
the desired products. It is proposed that this superior perform- in a furnace. The ampoule was heated first to 400 K for 10 h
ance of Ru/CTF catalysts results from the Lewis basic support and then to 600 K for 10 h. The ampoules were cooled to RT
material, which facilitates product desorption from the active and opened. The black solid was ground and washed with HCl
metal centre. Also N has a profound impact on the Ru surface (1 M, 3 L) and dried. The obtained powder was ground using a
chemistry; in XPS the Ru 3d5/2 signal is shifted to lower ball mill (Fritsch Pulverisette 23, 5 min, 30 Hz). The material
binding energy for the reduced sample compared to Ru/C. was washed with dilute HCl (2 M, 0.25 L) and neutralized with
Furthermore, recycling studies using Ru/CTF-b show a con- water, dilute NaOH (1 M, 0.40 L) and neutralized with water
stant activity and selectivity over multiple runs emphasising and THF (0.40 L). Drying took place under vacuum for 12 h.
the stability of the catalyst.
Catalyst preparation
Metal salt [RuCl
RuCl ·xH O (86.42 mg, 0.346 mmol) and CTF (700 mg) were
dispersed in ethanol (200 mL). The suspension was refluxed
All synthesis steps were carried out under inert atmosphere until the filtrate was clear and colourless. The catalyst was fil-
and the catalyst was stored under N . Ru/C (5 wt%) was tered and dried in vacuum at 333 K. The filtrate was concen-
obtained from Sigma-Aldrich and used as delivered. trated in vacuum and the residue dissolved in H O (25 mL)
2 2
(p-cymene)] (106 mg, 0.173 mmol) or
3
2
Experimental
2
2
3 2
Chemicals were purchased from Abcr (RuCl ·xH O), and analysed by ICP-OES. If no clear solution could be found
Chemsolute (HCl, pyridine, THF), Merck (acetic anhydride, after impregnation, the solvent was evaporated using a rotation
Ca(OH) , NaOH), NORIT (CABOT A SUPRA (activated carbon)), evaporator and the catalyst was dried. For the synthesis of
2
Roth (xylitol) Sigma-Aldrich (4,4′-Biphenyldicarbonitrile, 1,4- RuCl
dicyano-benzene, Ni(NO ·6H O, 5 wt% Pt/C, 2,6-pyridine- for 3 h at 623 K (6 L h ) using a heating rate of 8 K min . To
dicarbonitrile, [RuCl (p-cymene)] , ZnCl ), and westfalen (H ). obtain the 10 wt% Ni/C catalyst activated carbon CABOT A
3
-red./CTF catalyst reduction under hydrogen took place
−
1
−1
3
)
2
2
2
2
2
2
The catalyst supports were analysed by elemental analysis SUPRA was impregnated with Ni(NO
2400 CHNS/O Series II System by PerkinElmer), N -physisorp- ent wetness impregnation method. After drying the catalyst
tion (Quadrasorb SI by Quantachrome Instruments), materials was reduced under hydrogen for 2 h at 673 K (6 L h ) using a
3 2 2
) ·6H O using the incipi-
(
2
−1
1
were degassed in
Instruments in vacuum at 393 K for at least 4 h and TEM
a .
FloVac Degasser by Quantachrome heating rate of 5 K min−
Catalysis
(HD-2700 by Hitachi). ICP-OES was performed on an ICP
Spectroflame D by Spectro. H
2
-Chemisorption was measured Hydrogenolysis and hydrogenation reactions were carried out
on a 3flex by Micromeritics (prior to measurement the cata- in a 50 mL stainless steel autoclave equipped with a Teflon
lysts were reduced under H at 623 K for 16 h). Metal surface, inlet, a sampling tube. All reactions were performed at 473 K
2
dispersion and crystal size were calculated using the while stirring with a magnetic stirrer at 750 rpm. In a typical
Freundlich and Langmuir models. GC-analysis was performed experiment, the autoclave was charged with XYL (2.00 g,
on a Thermo Scientific Trace GC system equipped with a DB 13.1 mmol, 1 eq.) or LA (0.65 g, 7.2 mmol, 0.55 eq.), catalyst
2
3 column by Agilent after peracetylation using a mixture of (0.20 g, 0.1 eq.), Ca(OH)
acetic anhydride and pyridine (3 : 1 (v : v), 1 mL per 0.1 mL of (deionized, 20 mL). The autoclave was flushed 3 times using
reaction solution). HPLC analysis was performed on H and pressurized with 8 MPa H . Samples were taken period-
Shimadzu system (Rezex ROA-Organic Acid H+ (8%) column by ically and filtered over a syringe filter (PA 45/25). The samples
Phenomenex, (eluent: 0.05 M H SO )). Gas phase GC analysis were analysed by HPLC and GC. For the recycling experiments,
2
(0.30 g, 4.0 mmol, 0.3 eq.) and water
a
2
2
2
4
was performed on an Agilent HP6890 system with a 2 m Shin the catalyst was filtered and washed with aqueous acetic acid
carbon ST 100/120 mesh micro packed column using a TCD (60 mL, 1 M), water (until the washing solution was at pH 7)
detector. Helium was used as transport gas with a constant and ethanol (60 mL). The catalyst was dried in vacuum at
−
1
flow rate of 24.6 mL min . X-ray photoelectron spectroscopy 333 K. All steps for catalyst recycling were carried out under N2
XPS) measurements were carried out using a Kratos HSi atmosphere.
spectrometer with a hemispherical analyser. Reduced samples In case of the homogeneously catalysed reaction, [RuCl
were prepared and transferred under inert gas atmosphere. cymene)] (0.049 mmol) and bipyridine (0.085 mmol) were co-
(
2
(p-
2
Green Chem.
This journal is © The Royal Society of Chemistry 2018

View Article Online
Green Chemistry
Paper
ordinated in EtOH (2.0 mL) and added to a solution of XYL 19 J. Sun and H. Liu, Catal. Today, 2014, 234, 75–82.
2.0 g), water (20 mL) and CaOH2 (0.3 g). The reaction was 20 R. Sun, T. Wang, M. Zheng, W. Deng, J. Pang, A. Wang,
(
carried out and analysed as described above.
X. Wang and T. Zhang, ACS Catal., 2015, 5,
74–883.
8
2
2
2
1 Z. Zhou, S. Dai, J. Qin, P. Yu and W. Wu, RSC Adv., 2015, 5,
70410–70416.
2 Y. Zhang, A. Wang and T. Zhang, Chem. Commun., 2010, 46,
Conflicts of interest
There are no conflicts to declare.
862–864.
3 R. Ooms, M. Dusselier, J. A. Geboers, B. Op de Beeck,
R. Verhaeven, E. Gobechiya, J. A. Martens, A. Redl and
B. F. Sels, Green Chem., 2014, 16, 695–707.
Acknowledgements
2
2
2
2
4 N. Ji, T. Zhang, M. Zheng, A. Wang, H. Wang, X. Wang and
J. G. Chen, Angew. Chem., 2008, 120, 8638–8641.
5 N. Ji, M. Zheng, A. Wang, T. Zhang and J. G. Chen,
ChemSusChem, 2012, 5, 939–944.
6 M.-Y. Zheng, A.-Q. Wang, N. Ji, J.-F. Pang, X.-D. Wang and
T. Zhang, ChemSusChem, 2010, 3, 63–66.
7 F. Auneau, M. Berchu, G. Aubert, C. Pinel, M. Besson,
D. Todaro, M. Bernardi, T. Ponsetti and R. Di Felice, Catal.
Today, 2014, 234, 100–106.
We would like to thank Carina Kirstein from the ICVT in
Clausthal for chemisorption experiments, Norbert Pfänder
from the MPI Mülheim for TEM-images. We thank Noah
Avraham, Simon Petring, Heike Bergstein and Elke Biener for
HPLC, GC and ICP-OES analyses.
Notes and references
1
2
R. A. Sheldon, Catal. Today, 2011, 167, 3–13.
A. M. Ruppert, K. Weinberg and R. Palkovits, Angew. Chem.,
Int. Ed., 2012, 51, 2564–2601.
28 M. Rivière, N. Perret, A. Cabiac, D. Delcroix, C. Pinel and
M. Besson, ChemCatChem, 2017, 9, 2145–2159.
29 L. Zhao, J. H. Zhou, Z. J. Sui and X. G. Zhou, Chem. Eng.
Sci., 2010, 65, 30–35.
3
4
R. Rinaldi, R. Jastrzebski, M. T. Clough, J. Ralph,
M. Kennema, P. C. A. Bruijnincx and B. M. Weckhuysen, 30 J. Lee, Y. Xu and G. W. Huber, Appl. Catal., B, 2013,
Angew. Chem., Int. Ed., 2016, 55, 8164–8215. 140–141, 98–107.
T. vom Stein, P. M. Grande, H. Kayser, F. Sibilla, W. Leitner 31 T. Soták, T. Schmidt and M. Hronec, Appl. Catal., A, 2013,
and P. Dominguez de Maria, Green Chem., 2011, 13, 1772–
777.
X. Liu, X. Wang, S. Yao, Y. Jiang, J. Guan and X. Mu, RSC
Adv., 2014, 4, 49501–49520.
459, 26–33.
1
32 C. Montassier, J. C. Ménézo, L. C. Hoang, C. Renaud and
J. Barbier, J. Mol. Catal., 1991, 70, 99–110.
33 S. Tronci and B. Pittau, RSC Adv., 2015, 5, 23086–23093.
5
6
A. Corma, S. Iborra and A. Velty, Chem. Rev., 2007, 107, 34 W. Shen and W. Fan, J. Mater. Chem. A, 2013, 1, 999–1013.
2
411–2502.
35 M. Soorholtz, R. J. White, T. Zimmermann, M.-M. Titirici,
M. Antonietti, R. Palkovits and F. Schuth, Chem. Commun.,
2013, 49, 240–242.
36 C. Galeano, J. C. Meier, M. Soorholtz, H. Bongard,
C. Baldizzone, K. J. J. Mayrhofer and F. Schüth, ACS Catal.,
2014, 4, 3856–3868.
7
8
P. Gallezot, Chem. Soc. Rev., 2012, 41, 1538–1558.
T. vom Stein, P. M. Grande, W. Leitner and P. Domínguez
de María, ChemSusChem, 2011, 4, 1592–1594.
K. Chen, M. Tamura, Z. Yuan, Y. Nakagawa and
K. Tomishige, ChemSusChem, 2013, 6, 613–621.
9
1
1
1
1
1
1
1
0 S. Liu, Y. Okuyama, M. Tamura, Y. Nakagawa, A. Imai and 37 P. Kuhn, M. Antonietti and A. Thomas, Angew. Chem., 2008,
K. Tomishige, Green Chem., 2016, 18, 165–175. 120, 3499–3502.
1 C. Montassier, D. Giraud and J. Barbier, Stud. Surf. Sci. 38 J. Artz, S. Mallmann and R. Palkovits, ChemSusChem, 2015,
Catal, Elsevier, 1988,pp. 165–170. 8, 672–679.
2 D. K. Sohounloue, C. Montassier and J. Barbier, React. 39 P. J. C. Hausoul, A. K. Beine, L. Neghadar and R. Palkovits,
Kinet. Catal. Lett., 1983, 22, 391–397. Catal. Sci. Technol., 2017, 7, 56–63.
3 K. Wang, M. C. Hawley and T. D. Furney, Ind. Eng. Chem. 40 K. Kudo, F. Ikoma, S. Mori, K. Komatsu and N. Sugita,
Res., 1995, 34, 3766–3770. J. Chem. Soc., Perkin Trans. 2, 1997, 679–682.
4 M. A. Andrews and S. A. Klaeren, J. Am. Chem. Soc., 1989, 41 H. Liu, Z. Huang, C. Xia, Y. Jia, J. Chen and H. Liu,
11, 4131–4133. ChemCatChem, 2014, 6, 2918–2928.
5 P. J. C. Hausoul, L. Negahdar, K. Schute and R. Palkovits, 42 J. F. Moulder, W. F. Stickle, P. E. Sobol and K. D. Bomben,
1
ChemSusChem, 2015, 8, 3323–3330.
Handbook of X-ray Photoelectron Spectroscopy, PerkinElmer
Corporation, Eden Prairie, 1992, pp. 114–115.
6 K. L. Deutsch, D. G. Lahr and B. H. Shanks, Green Chem.,
2
012, 14, 1635–1642.
43 G. Tuci, M. Pilaski, H. Ba, A. Rossin, L. Luconi, S. Caporali,
C. Pham-Huu, R. Palkovits and G. Giambastiani, Adv.
Funct. Mater., 2017, 27, 1605672.
1
1
7 E. P. Maris and R. J. Davis, J. Catal., 2007, 249, 328–337.
8 J. Sun and H. Liu, Green Chem., 2011, 13, 135–142.
This journal is © The Royal Society of Chemistry 2018
Green Chem.