From microcrystalline cellulose to hard- and softwood-based feedstocks: Their hydrogenolysis to polyols over a highly efficient ruthenium-tungsten catalyst

Article abstract of DOI:10.1039/c5gc00421g

The utilization of cellulose and its integration in a biorefinery concept is essential even in the near future due to the growing global shortage of crude oil. Here, we report the catalyzed one-pot hydrogenolysis of cellulosic materials to valuable bio-derived molecules, especially polyols (e.g. ethylene glycol). We demonstrate how a very promising bifunctional catalyst, Ru/W/AC, converted not only 100% of microcrystalline cellulose to polyols in repeated experiments with a maximum yield of 84% and an ethylene glycol productivity of 3.7 g (g_catalyst h)^-1, but also pine-, birch-, and eucalyptus-derived materials. Moreover, we systematically investigated the problem of catalyst stability with time by studying the changes in both the catalyst structure and the liquid phase, which have often been overlooked when biomass is converted to fuels and chemicals. Control of the active sites for the conversion of cellulosic feedstocks coupled with reaction engineering and strategies to prevent catalyst deactivation, is a prerequisite to understanding how high yields of platform chemicals can be achieved.

Full text of DOI:10.1039/c5gc00421g

View Article Online
View Journal
Green
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: P. Claus, K.
Fabiovicová and M. Lucas, Green Chem., 2015, DOI: 10.1039/C5GC00421G.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
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.
www.rsc.org/greenchem

Page 1 of 12
Green Chemistry
Dynamic Article Links ►
Journal Name
View Article Online
DOI: 10.1039/C5GC00421G
Cite this: DOI: 10.1039/c0xx00000x
www.rsc.org/xxxxxx
ARTICLE TYPE
From microcrystalline cellulose to hardꢀ and softwoodꢀbased feedstocks:
their hydrogenolysis to polyols over a highly efficient rutheniumꢀ
tungsten catalyst
Katarína Fabičovicováa, Martin Lucasa and Peter Claus*a
5
Received (in XXX, XXX) Xth XXXXXXXXX 20XX, Accepted Xth XXXXXXXXX 20XX
DOI: 10.1039/b000000x
The utilization of cellulose and its integration in a biorefinery concept is essential even in the near future of growing global shortage of
crude oil. Here, we report the catalyzed oneꢀpot hydrogenolysis of cellulosic materials to valuable bioꢀderived molecules, especially
polyols (e.g. ethylene glycol). We demonstrate how a very promising bifunctional catalyst, RuꢀW/AC, converted not only 100 % of
10 microcrystalline cellulose to polyols in repeated experiments with a maximum yield of 84 % and ethylene glycol productivity of 3.7 g
(gcatalyst h)ꢀ1) but was also active for pineꢀ, birchꢀ, and eucalyptusꢀderived materials. Moreover, we systematically investigated the
problem of catalyst stability with time by detailed studies of changes of both the catalyst structure and the liquid phase, which has been
often overseen when biomass is converted to fuels and chemicals. Control of active sites for the conversion of cellulosic feedstocks
coupled with reaction engineering and strategies to prevent catalyst deactivation is a prerequisite to understand how high yields of
15 platform chemicals can be achieved.
challenge but also necessary because of the worldwide shortage
of crude oil as raw material for the production of fuels and
chemicals. For example, ethylene glycol is usually produced via
1 Introduction
The increasing demand for a new effective technology to process
cellulose, which presents perspective resource for the
hydrolysis of ethylene oxide which is obtained from petroleumꢀ
55 based ethylene8. Reversibly formed acid sites in hot water
together with ruthenium on activated carbon were effective for
the formation of hexitols (39.3% yield) at 518 K, 60 bar
hydrogen pressure within 3 h9. Various carbon materials like
carbon nanotubes (CNTs)10 or mesoporous carbon materials
60 (CMKs)11,12 were developed and enabled, in combination with
noble metals, the conversion of cellulose to glucose and sugar
alcohols. Moreover, if a catalyst with nickel and tungsten carbide
on activated carbon was applied, product distribution differed
strongly and the hydrogenolysis reaction to ethylene glycol was
65 facilitated13. Catalysts containing nickel or noble metals and
different forms of tungsten (W14 or WO315) were also active, and
a binary catalyst system composed of tungsten acid (H2WO4) and
Ru/C16 or Raney Ni17 gave ethylene glycol as main product.
a
20 production of valuable fuels and chemicals, incites many
research activities in the last decade. Besides the sustainability
and the possible CO2 neutrality, other advantages like
inexpensiveness and abundance of cellulose feedstock can help
to resolve the problem of decreasing fossil reserves in our
25 society1. Although cellulose degradation offers a big variety of
products2, the great stability caused by interꢀ and intraꢀ molecular
hydrogen bonds yielding a robust crystalline structure makes a
highly selective conversion of this biopolymer very challenging.
The depolymerization and hydrolysis of cellulose to
30 celloꢀoligosaccharides and monomer building blocks (glucose
molecules) is the first key reaction step. Homogenous acidic
catalysis showed good results, but the negative influence of
concentrated or diluted mineral acids (separation/recovery of
acid, corrosion) prohibits largeꢀscale application of these
35 processes3. Therefore, the use of solid acid catalysts in an
Besides tungsten acid, catalyst systems with heteropoly acids17,
18, 19
70
(Ru/C + CsHPA, H4[Si(W3O10)4] or H3[P(W3O10)4]) are
able to convert cellulose to polyols like sorbitol or ethylene
glycol. Not only activated carbon, but also SiO2 or SiO2ꢀAl2O3 as
support in combination with Ni and W were active for the
polyols production20, 21. Beside the drawbacks of using mineral
aqueous solution under hydrogen atmosphere became
a
promising new way for the production of desired chemicals from
cellulose4. Various solid acid materials like zeolites, sulfated
zirconia, silica (SBAꢀ15) or carbon materials were investigated5,
40 6, 7. Additionally, the presence of different active sites on solid
supports allows that in a oneꢀpot reaction system various
complex reaction pathways can occur. For example, in the
presence of supported metal catalysts (Ru or Pt on support), C6
sugar alcohols like sorbitol and mannitol can be produced via the
45 tandem reaction of hydrolysis and hydrogenation5.
75 acids or expensive ionic liquids for the degradation of cellulose
another critical point is that sugar alcohols can cause Ni leaching
because of their chelating properties22, 23
.
Despite the increasing research activities for processing
cellulose and other biomass feedstock to chemicals and fuels, the
80 problem of catalyst stability with timeꢀonꢀstream, which is selfꢀ
evident due to the knowledge of catalyst action in petroleum
industry but must not necessarily follow the same mechanisms,
has been seldom reported22, 23, 24, 25 or overseen.
Other than that mentioned, the direct hydrogenolysis of
cellulose presents an interesting route to obtain various shortꢀ
chain polyols (e.g. ethylene glycol, propylene glycol, and
glycerol). Polyols are an important class of platform molecules,
50 and their sustainable production within a biorefinery is a
In this work, we demonstrate the effective cellulose
85 hydrogenolysis over
a
selfꢀprepared ruthenium containing
This journal is © The Royal Society of Chemistry [year]
[journal], [year], [vol], 00–00 | 1

Green Chemistry
Page 2 of 12
View Article Online
DOI: 10.1039/C5GC00421G
Carbon efficiency coefficient (CEL) is based on the ratio of
carbon of the known soluble products to carbon of all soluble
products (known + unknown soluble products).
catalyst supported on W/AC starting with microcrystalline
cellulose. Recycling tests of this bifunctional Ru/W/AC catalyst
and the detailed studies of changes of both the catalyst and the
liquid phase occurring during the reaction were performed in
order to understand the long term stability and the degradation
behaviour of the catalyst. Additionally, the prepared catalyst was
used in combination with ballꢀmilled cellulose in order to
investigate the influence on cellulose reactivity and product
distribution. In the end, we transfer the optimal hydrogenolysis
5
65 2.3 Catalyst characterisation
Powder Xꢀray diffraction (XRD) analysis of samples was carried
out on a powder diffraction system using CuKα1ꢀradiation at
voltage of 40 kV and current of 40 mA. Scans were carried out
from 2 theta = 10 to 90o with 0.067o steps and 40 s per step.
70 The amount of acid sites was determined by temperatureꢀ
programmed desorption (TPD) of NH3. After calcination at 573
K the sample was exposed to NH3 at 308 K for 1 h. Weakly
adsorbed NH3 was removed by flowing nitrogen at 308 K.
Finally, the temperature was increased with a rate of 5 K minꢀ1 up
75 to 923 K and the amount of NH3 was measured by FTꢀIR
spectrometry (Thermo Fischer Scientific).
10 process to different kinds of cellulosic materials.
2 Experimental
2.1 Catalyst synthesis
The support material (activated carbon, (AC), specific surface
15 area 1200 m2 gꢀ1, particle size 0.063–0.2 ꢁm) SX Ultra was
purchased from Cabot Norit Company/Nederland, and
impregnated in the first step with an aqueous solution of
ammonium metatungstate hydrate followed by drying at 383 K
for 15 h, pretreatment in Ar at 823 K for 300 minutes and
20 reduction in hydrogen in the following way: applying a heating
ramp from room temperature to 808 K in 26 minutes, then to
1128 K in 64 minutes and holding this temperature for 30
minutes. After cooling down, the preꢀcatalyst W/AC was
obtained. The latter was impregnated with an aqueous solution of
25 ruthenium nitrosyl nitrate followed by drying at 383 K for 15 h
and reduction in hydrogen flow (200 ml/min, 623 K for 3 h).
This procedure gives the final catalyst Ru/W/AC which was
introduced in this form into the batch reactor of the cellulose
hydrogenolysis reaction. The loading of tungsten and ruthenium
30 is 40 wt%, related to WO3, and 2 wt%, respectively. A
monometallic ruthenium catalyst was prepared in an analogous
way, however, without the first step.
Temperatureꢀprogrammed reduction (TPR) measurements were
performed on a TPD/R/O 1100 from Thermo Fisher Scientific
using a quartz Uꢀtube reactor. The catalyst was pretreated in Ar
80 flow (30mL/min) at 383 K for 60 minutes and cooled to 303 K.
The experiments were performed using 30mL/min of 5.1 vol.%
H2/Ar by heating the sample from 303 K to 1073 K at a heating
rate of 5 K/min while monitoring the TCD signal. Additionally,
the apparatus was coupled with a mass spectrometer to detect
85 water formed during the TPR.
3 Results and discussion
The bimetallic catalyst containing ruthenium and
tungsten supported on activated carbon (Ru/W/AC) was carefully
90 prepared according to the twoꢀstep incipient wetness (IW)
technique starting with ammonium metatungstate hydrate on SX
Ultra as described above. After heating this preꢀcatalyst, denoted
as W/AC, in flowing hydrogen up to 1128 K, Xꢀray diffraction
(XRD) and temperatureꢀprogrammed reduction (TPR)
95 experiments revealed the formation of tungsten carbides, WC and
W2C (Fig. S1, S2). For the second IW step an aqueous solution
of ruthenium nitrosyl nitrate was used. With this Ru precursor
and after reduction at 623 K, ruthenium is highly dispersed in the
virgin heterogeneous Ru/W/AC catalyst as indicated by the
100 absence of diffraction signals for ruthenium. Moreover, in this
catalyst no crystalline phases of tungsten carbides were observed
indicating their dissolution and redispersion in the presence of
water and hydrogen, respectively. This is supported by the TPR
of Ru/W/AC which shows a highꢀtemperature behavior similar to
105 W/AC due to the reduction of tungsten species until the carbides
were formed. Accordingly, traces of tungsten dioxide and
metallic tungsten were observed.
2.2 Hydrogenolysis of cellulose
35 General procedure: Typically, 0.5 g catalyst, 5 g cellulose
microcrystalline (Merck) and 100 mL water were converted in a
batch reactor (Parr Instrument, 300 mL). At the beginning of
each experiment, the reaction solution consisting of 0.5 g
catalyst, 5 g cellulose, and 100 mL water was closed in the
40 reactor. Then, air atmosphere was flushed up with argon, the
reactor was pressurized with 20 bar of hydrogen and the reaction
solution was heated up to 493 K under stirring (1000 rpm). After
reaching the reaction temperature, the reactor was second time
pressurized to 65 bar. This was confirmed as the start of the
45 reaction. After 3 h of reaction time, the reaction solution was
cooled down to room temperature. During and after the reaction,
liquid samples were taken and analyzed by HPLC, GC and GCꢀ
MS.
Catalyst recycling: In the first run, 0.5 g of Ru/W/AC was
50 introduced for the experiment. After the reaction, the product
solution was filtered, the catalyst was washed with deionized
water and dried at 383 K under air atmosphere over night. After
drying, the catalyst was scraped from the folded filter, reꢀweight
and recovered for the next run (catalyst mass: Table 1).
55 Conversion of cellulose (X) was determined gravimetrically
based on the weight loss during the reaction.
Ru/W/AC showed remarkable performance for the
hydrogenolysis of cellulose (Table 1, entry 1). Not only the
110 cellulose conversion was very high (90 %) at 493 K, 65 bar
hydrogen and within 3 h, but also the carbon efficiency
coefficient (CEL= 76.7 %) which describes the ratio of carbon of
the known soluble products to carbon of all soluble products
(known and unknown). Interestingly, the catalyst produced the
115 desired polyols with an overall yield of 66.6 %, and the dominant
product was ethylene glycol with a yield of 34.2 %. Additionally,
other valuable products such as sorbitol and erythritol were
obtained with a yield of 18.5 % and 5.2 %, respectively. The
monometallic catalyst Ru/AC exhibited much lower reactivity
120 (cellulose conversion = 60 %) and gave only 19.7 % polyols
(Fig. S4). With the addition of tungsten in the first step of the
The yields (Yi) in this work are based on the moles of carbon in
the product i divided by the moles of carbon in cellulose:
,
Yi = (zi / z
)×(ni ⋅ M
/ mCellulose,start )
C6H10O5
C6H10O5
60 zi is carbon number in a molecule.
2
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

Page 3 of 12
Green Chemistry
View Article Online
DOI: 10.1039/C5GC00421G
preparation protocol, acidic sites were formed which were probed
conversion and polyol yield was detected during the six
consecutive runs which demonstrates that the loss of catalyst is
not the main reason for the decrease of activity of Ru/W/AC in
the last seventh cycle. To confirm this point, one experiment with
70 0.372 g fresh catalyst was carried out at the same reaction
conditions as the recycling test (Table 1, entry 9). Note that the
results are comparable to those obtained in the fifth recycling
run, where the catalyst amount was in the same range. Therefore,
leaching of active components and/or chemical and structural
75 changes of catalyst surface must be considered. The ICPꢀOES
analysis showed that ruthenium is very stable under our reaction
conditions, and no traces were found in the liquid phase. This is
in line with our previous study of glucose hydrogenation over
supported ruthenium demonstrating the absence of any leaching
80 during 1000 hours timeꢀonꢀstream in a trickleꢀbed reactor23. On
the other hand, tungsten in ppm range was detected in the liquid
samples after each recycling run (Table 1), the overall loss of
weight corresponds to 21 %. In the first run, the concentration of
tungsten in the liquid phase was the highest, 121 ppm.
85 Interestingly, when the tungsten concentration in the last cycle
decreased to only 5 ppm, both the activity of the catalyst and the
yield of desired short chain polyols decreased. The product
solution obtained after the reaction was blue and during the
filtration the solution became colorless. Typical blue compounds
90 of tungsten are hydrogen tungsten bronzes (H0.1WO3, H0.33WO3),
which can be easily oxidized by contact with air32. Probably, the
leached tungsten species are in form of hydrogen tungsten bronze
during the reaction, and during the filtration they quickly oxidize.
In which form tungsten during the reaction exists, is not known
95 yet. Although the virgin heterogeneous Ru/W/AC catalyst clearly
exhibits high concentration of acidic tungsten sites which are
responsible for cleavage of CꢀO and CꢀC bonds, the confirmed
loss of tungsten due to leaching indicates that these species may
by temperatureꢀprogrammed desorption of ammonia (NH3ꢀTPD):
the acid site concentrations of Ru/W/AC and Ru/AC amount to
1.157 mmol/g and 0.098 mmol/g, respectively. Remarkably, the
high concentration of acid sites in the former corresponds to a
level which is known from zeolites (e.g., HꢀBEA26). However,
with Ru and Ni modified zeolites hexitols were the mainly
formed polyols27, 28. The results imply that tungsten species must
be involved in the formation of shortꢀchain polyols such as
5
10 ethylene glycol.
In general, the product distribution obtained during the
degradation of cellulose under hydrothermal conditions and in
presence of hydrogen strongly depends on reaction conditions
and type of the used catalysts. The predominant reactions, which
15 casually explain the product distribution and occurred during the
cellulose hydrogenolysis over our catalyst Ru/W/AC, are shown
in Scheme 1. They can be classified in cellulose hydrolysis,
retroꢀaldol condensation of glucose (and other saccharides) and
hydrogenation reactions. In the first step, cellulose is hydrolyzed
20 to celloꢀoligosaccharides and glucose due to H+ ions, which are
inꢀsitu produced in water at high temperatures (autoprotolysis of
water)9, 29. Tungsten species, which are responsible for CꢀC and
CꢀO bond scissions, also participate in the hydrolysis of cellulose
to glucose and cause formation of shortꢀchain aldehydes and
25 ketones (e.g. glycolaldehyde, glyceraldehyde, 1ꢀhydroxyꢀ2ꢀ
propanone,
1ꢀhydroxyꢀ2ꢀbutanone, 3ꢀhydroxyꢀ2ꢀbutanone).
Importantly, retroꢀaldol condensation is the main reaction to form
glycolaldehyde and erythrose (from glucose) or other aldehydes
(e.g., glyceraldehyde from fructose). Erythrose is then rapidly
30 hydrogenated to erythritol or converted into two glycolaldehyde
molecules via retroꢀaldol condensation. The latter is also of
synthetic value applying subcritical and supercritical water for
the conversion of glucose30 and cellobiose31. The aforementioned
unsaturated intermediates are finally hydrogenated, by ruthenium
be involved in
a homogeneously catalyzed step in the
35 sites, to the desired polyols like ethylene glycol, propylene 100 degradation of cellulose. With decreasing concentration of
glycol, 1,2ꢀbutanediol, and 2,3ꢀbutanediol (2b, 6b, 7b, 8b,
Scheme 1). Besides the threeꢀstep reaction route consisting of
hydrolysis, retroꢀaldol condensation, and hydrogenation entailing
shortꢀchain polyols, direct hydrogenation of glucose is
leached tungsten species from run to run (Table 1), the retroꢀ
aldol condensation of glucose to glycolaldehyde and erythrose
(with subsequent hydrogenation of both) got suppressed, and the
main reaction route switches from ethylene glycol formation to
40 competitive and yields sorbitol. Degradation products like 105 the direct hydrogenation of glucose. Therefore, the yield of
methanol and ethanol are also obtained during the reaction of
microcrystalline cellulose over Ru/W/AC. Gas phase analysis
showed the presence of methane, ethane, CO, and CO2 with a
summarized yield of ~ 3 %, whereas methane was the dominant
45 product.
sorbitol achieved a high value (28 %) in the last run (Fig. 1). The
ratio of ethylene glycol/sorbitol at this point corresponds to ∼
1:5.8 compared to ∼ 2.4:1 in the previous runs. Remarkably,
although loss of ppm of tungsten is occurred, the heterogeneous
110 catalyst still contains metallic tungsten (Fig. S2) and is still active
for cellulose hydrogenolysis after repeated runs and produces the
desired polyols switching now the selectivity from one valueꢀ
added product (ethylene glycol) to the other (sorbitol).
Recycling tests of the Ru/W/AC were performed in
order to test the stability of this catalyst. Due to the difficult
catalyst separation from the nonꢀconverted solid cellulose rest,
full conversion of cellulose during the recycling test is necessary
50 for the reꢀuse of the catalyst. For this reason, different from the
standard reaction conditions, a reaction temperature of 498 K
was chosen during the recycling test because full cellulose
conversion was achieved at this temperature. Presented in Table
1 (entry 2ꢀ8), the catalyst Ru/W/AC maintained its stability
55 during six cycles (100 % cellulose conversion, and also the
overall polyol yield was always very high, in the range of 64ꢀ70
%). In the last (seventh) run, the yield of polyols decreased to
46.4 % and the product distribution differed strongly. The yield
of ethylene glycol was very small, 4.9 % (note, that the yield of
60 ethylene glycol in the first run was 36.6 %), meanwhile the yield
of sorbitol increased from 15.6 % in the first run to 28.4 % in the
last recycling run. As noticed in Table 1, a small loss of catalyst
in each recycling run was caused because of the complicated reꢀ
use method of a heterogeneous catalyst in this gas/liquid/
65 solid/solid reaction system. However, no decrease in cellulose
To further elucidate the performance of our
115 heterogeneous Ru/W/AC catalyst in light of the interplay of
soluble hydrogen tungsten bronze (HxWO3) and tungsten acid
(H2WO4) which can act as precursor for the former under
hydrothermal conditions and in presence of hydrogen15, we
compared the standard reaction with 0.5 g of Ru/W/AC catalyst
120 containing 36.7 % of tungsten (ICPꢀOES analysis, corresponds to
1.8 g/L tungsten) and 0.5 g of Ru/AC + 2.5 g/L tungsten acid
(corresponds to 1.84 g/L tungsten). Here, slightly higher
cellulose conversions between 3ꢀ8 % were obtained in the
reactions performed with Ru/AC + H2WO4 (Table 2). The
125 formation of dissolved active tungsten species occurs slowly via
continuously leaching from heterogeneous Ru/W/AC. Leaching
kinetics follows
a 0.5 order with respect to tungsten
concentration (Fig. S3). The system with tungsten acid exhibits a
higher concentration of homogenous active species in the liquid
130 phase from the beginning, and because of this, the higher
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 3

Green Chemistry
Page 4 of 12
View Article Online
DOI: 10.1039/C5GC00421G
conversion of cellulose can be obtained. However, the carbon
efficiency (CEL) was lower (69ꢀ80 %) in this case compared to
the Ru/W/AC catalyst (76ꢀ82 %). For all of these experiments,
noticeable differences could be seen in polyol distributions. Note
that the yield of ethylene glycol was still 11ꢀ15 % greater
(independent on reaction time), if the heterogeneous catalyst
Ru/W/AC was used. On the other hand, the combination of
Ru/AC + H2WO4 allowed higher production of sorbitol. In this
case, the yield of sorbitol of 30.5 % was obtained after one hour
10 reaction time, and a further increase was not monitored neither
with increasing reaction time nor with varying concentration
(Fig. S4). Under identical reaction conditions, Ru/W/AC
produced only half of the sorbitol yield of the Ru/AC + H2WO4
system. The differences in the distribution of the main products,
15 ethylene glycol and sorbitol, are due to the kinetics and mass
transport of several steps of the complex reaction system, namely
(i) tungsten leaching, (ii) retroꢀaldol condensation of glucose
catalyzed by both the dissolved tungsten species formed via the
leaching process and the remaining acid tungsten sites of the
20 heterogeneous Ru/W/AC catalyst (in contrast to H2WO4 of the
combined catalyst system), and (iii) glucose and glycolaldehyde
hydrogenation.
Before we demonstrate the catalyst performance for
this challenge, we kept in mind that such types of feedstock are
usually pretreated. To further increase the reactivity of cellulose,
various pretreatment techniques (mechanical, chemical) can be
70 used. Ballꢀmilling is a mechanical process, which effectively
degrades the resistant cellulose crystallinity, particle size and the
high degree of polymerization33. Usually, the time of ballꢀmilling
is between 1 hour and few days, and such pretreated cellulose
was applied for the hydrolysis of cellulose7, 18, 34, 35. Because of
75 the high costs of this pretreatment, it is necessary to optimize
both the time for ballꢀmilling of cellulose and for the
5
hydrogenolysis
reaction.
Therefore,
we
ballꢀmilled
microcrystalline cellulose using ZrO2 balls in a ZrO2 bottle of a
planetary ball mill for various times (4, 15, 30, 120, 720 minutes)
80 and we investigated the influence of the pretreated cellulose on
the hydrogenolysis reaction over the Ru/W/AC catalyst.
Additionally, for
4 and 15 minutes ballꢀmilled cellulose,
experiments with shorter reaction time of 1 hour and 2 hours
were carried out in order to better comparison and understanding
85 of the influence of this cellulose pretreatment method. Higher
cellulose conversions were obtained if ballꢀmilled cellulose was
used (Fig. 2). Surprisingly, very shortꢀtime ballꢀmilling of 4
minutes is enough to increase the cellulose conversion from 65 %
(no pretreatment) to 81 % after one hour reaction time, and to 94
90 % for 15 minutes long ballꢀmilled cellulose. Moreover, the
overall polyol yield achieved a value of 70.0 % (therefrom 39.6
% ethylene glycol yield) if the reaction took three hours for 4
minutes ballꢀmilled cellulose. In contrast to shortꢀtime ballꢀ
milled cellulose, if the longꢀtime ballꢀmilled cellulose was used
95 (30, 120, and 720 minutes), the overall polyols yield slowly
decreased and for 720 minutes ballꢀmilled cellulose achieved a
value of only 15.4 % after three hours of reaction time. The
mechanical pretreatment method also influenced the product
distribution (Fig. 2). Interestingly, the amount of sugar
The real active site of tungsten during the reaction is
not known yet because the lack of in situ methods for gasꢀliquidꢀ
25 solid reactions. Because of high reaction temperature and usage
of water as a solvent, possible reaction of tungsten with water in
this hydrothermal environment can occur22. Therefore, the
formation of real active sites from metallic tungsten is possible.
This leads to tungsten in different valence states (WO3, WO2,
30 hydrogen bronze), all of them exhibit acidic properties and
principally they all can take part in the reaction beside the
redispersed tungsten carbides. Therefore, regarding tailoring the
active sites for hydrolysis and CꢀO/CꢀC bond scissions we
conclude that the following conditions should be fulfilled: a) high
35 tungsten loading in the preꢀcatalyst, b) its highꢀtemperature 100 hydrogenation products, i.e. sorbitol (from glucose) and
reduction and c) redispersion of tungsten carbide species by the
aqueous environment of the second impregnation step followed
by reduction. The latter step also ensures a close position of
hydrogenation and acidic sites on the catalyst surface. This is
erythritol (from erythrose), decreased with increasing time of
cellulose ballꢀmilling. On the other hand, the amount of shortꢀ
chain polyols like propylene glycol or 1,2ꢀbutanediol increased
with increasing time of ballꢀmilling and achieved yields in the
40 also indicated by an experiment with physical mixture of Ru/AC 105 range of 4ꢀ10 %. Based on these results we can conclude that the
and W/AC in order to see the importance of synergy between
ruthenium and tungsten active sites. The conversion of cellulose
obtained during this reaction was 85 % and the overall polyol
yield was similar to the experiment with standard Ru/W/AC
shortꢀtime ballꢀmilling of cellulose can improve its reactivity and
the yield of desired polyols. Moreover, the reaction time of
cellulose hydrogenolysis can be shorted to one hour instead of
three hours: this optimization led to a modified spaceꢀtime yield
45 catalyst. However, the product distribution differs and is similar 110 (productivity) of 3.2 – 3.7 g ethylene glycol (gcatalyst h)ꢀ1.
to them obtained after the reaction with the catalyst system
Ru/AC + H2WO4. In both these cases, more sorbitol in the range
of 30ꢀ45 % was obtained, but the yield of ethylene glycol was ~
11ꢀ15 % smaller. Therefore, we conclude that for the production
In the final step of this pretreatment we investigated the effect of
mixꢀmilling of catalyst and cellulose, i.e. a mechanocatalytic
activation of cellulose. It was reported, that mixꢀmilling of
simple carbon and cellulose (with small amounts of HCl) were
50 of ethylene glycol with a high yield the close position of 115 effective for its hydrolysis in order to achieve a high yield of
ruthenium and tungsten species is necessary. On the other hand,
for the high production of sorbitol catalyst systems with
separated active species of Ru and W are more essential.
It must be noted that the ICPꢀOES analysis of the
sugars, especially glucose36. We mixꢀmilled 5 g of cellulose with
0.5 g of Ru/W/AC catalyst for only four minutes. After milling,
we flushed out the celluloseꢀcatalyst mixture with 100 g of water
and used it for the hydrogenolysis reaction. After three hours of
55 liquid phase obtained after the reaction with Ru/AC + H2WO4 120 reaction time, full conversion of cellulose was achieved. The
showed, in contrast to Ru/W/AC, the presence of small traces of
nickel and iron. Therefore, the possible corrosion of the stainless
steel reactor can prohibit an industrial application of the catalyst
system consisting of tungsten acid and ruthenium catalyst. After
enhanced contact between solid cellulose and solid catalyst
caused by mixꢀmilling could increase the overall polyol yield to
84 % (Fig. 2), which is the highest polyol yield achieved during
cellulose hydrogenolysis with our Ru/W/AC. Besides a high
60 standard reaction time of 3 hours, polyols were produced in the 125 ethylene glycol yield of 37.4 %, the yields of propylene glycol,
same level for both systems (overall yield of 66.1 % and 66.6 %
for Ru/AC + H2WO4 and Ru/W/AC, respectively). Taking into
account all these results, we decided to use the heterogeneous
catalyst for the next step, namely the hydrogenolysis of different
65 cellulosic materials based on pine, birch and eucalyptus.
butanediol, and sorbitol were also high, 14 %, 10.8 %, and 17.3
%, respectively. Remarkeably the carbon balance in the case of
mixꢀmilling is the highest (90 %, illustrated by the CEL data of
Table S1).
4
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

Page 5 of 12
Green Chemistry
View Article Online
DOI: 10.1039/C5GC00421G
Real biomass can be provided from agriculture (e.g.
exhibits a very high concentration of acid sites. Therefore the use
65 of mineral acids can be avoided. We are aware that the next step
of processing biomass to valueꢀadded chemicals by
heterogeneously catalyzed hydrogenolysis should include real
lignocellulosic feedstocks. Also in this case catalysts have to
withstand deactivation which is expected to occur. Our aim was
70 to show how a carefully synthesized bifunctional catalyst
converts not only model cellulose but also real cellulosic
materials to valuable platform chemicals and can be reꢀused
several times without loss of activity. It is necessary to critical
scrutinize the catalyst stability under the conditions of such kind
75 of liquid phase reactions (high temperatures, aqueous
environment, presence of solvents, biogenic impurities, metalꢀ
chelating properties of reactants/products).
dedicated crops and residues), forestry (e.g. wood), industry and
household (solid residues, e.g. wastepaper), and aquaculture37.
Complex chemical composition of biomass consisting of the
three main components cellulose, hemicellulose, lignin, and other
small fractions (proteins, fats, etc.) images a significant challenge
in its utilization. Therefore, the raw material has to be separated
by physical or chemical methods. After the fractionation, the
conversion of each component can improve the economical
5
10 efficiency of a bioꢀrefinery38. We showed above that the
heterogeneous catalyst Ru/W/AC can convert microcrystalline
cellulose into polyols with high yield. This catalyst system was
then applied to different kinds of cellulosic materials which
typically incur, for example, in wastepaper. Concretely,
15 cellulosic material of pine (unbleached and bleached), birch, and
eucalyptus were used for the hydrogenolysis reaction over
Ru/W/AC at 493 K, 65 bar of hydrogen pressure within three
hours of reaction time. Also the low catalyst/cellulosic material
ratio of 1/10 which was used for the hydrogenolysis of
20 microcrystalline cellulose was preserved. Again we carried out
the reaction in pure water and without the use of mineral acids.
All raw materials were firstly cut with scissors, preꢀmilled with a
grinder (particle size ~ 5 mm) and finally milled in the planetary
ball mill (particle size < 0.4 ꢁm). The results obtained in the
25 experiments are shown in Fig. 3. Surprisingly, the activity of
Ru/W/AC was very high as indicated by conversions of the
different cellulosic materials in the range of 89 % and 100 %. As
expected, the product distribution depends strongly on the
individual cellulose type. The reaction of unbleached pine
30 cellulose gave 9.9 % of polyols and the yield of ketones was even
higher, 19.3 %. On the other hand, the hydrogenolysis of
bleached pineꢀ, birchꢀ and eucalyptusꢀbased cellulosic feedstocks
showed a high overall polyol yield, 32.1 %, 41.0 % and 60.2 %,
respectively. Note that among the polyols produced by the
35 eucalyptusꢀderived feedstock, ethylene glycol is the most
abundant component (∼ 28 %), and no formation of sorbitol was
observed. This indicates that among the aforementioned
competitive reaction steps the retroꢀaldol condensation of sugar
molecules and the subsequent hydrogenation of the unsaturated
40 intermediates to shortꢀchain polyols proceeds much faster than
the sugar hydrogenation. Moreover, we analyzed the four
materials with respect to their αꢀcellulose content which amounts
to 91, 87, 85 and 88 % for unbleached pine, bleached pineꢀ,
birchꢀ and eucalyptusꢀbased cellulosic feedstocks, respectively.
45 Obviously, there is no correlation between the catalyst
performance and the content of αꢀcellulose indicating that the
latter is not a key parameter. Therefore other factors must be
considered such as pretreatment (bleaching or not, sulfite vs.
sulfate process) and the cellulose source (hardwood, e.g.
50 eucalyptus vs. softwood, e.g. pine) influencing the fiber
morphology, especially length and shape. In comparison to the
results obtained during the hydrogenolysis of microcrystalline
cellulose over Ru/W/AC, the yields of ethylene glycol and C4/C6
polyols decreased, whereas the yield of propylene glycol and 1,2ꢀ
55 butanediol was doubled for the reaction with bleached pineꢀbased
feedstock, and tripled in the case of birchꢀ and eucalyptusꢀ
derived feedstock. Indeed, the hydrogenolysis over our Ru/W/AC
catalyst could be successfully transferred to other cellulosic
materials.
80
Acknowledgements
We thank the Fritz und Margot Faudi Stiftung for the financial
support. The help of J. Hunold with XRD measurements is
greatly acknowledged.
85
Notes and references
a Technical University Darmstadt, Dept. Chemistry, Chemical
Technology II, Alarich-Weiss-Straße 8,64287 Darmstadt, Germany. Fax:
+49 6151 164788; Tel: +49 6151 165369; E-mail: claus@tc2.tu-
90 darmstadt.de
† Electronic Supplementary Information (ESI) available: [details of any
supplementary information available should be included here]. See
DOI: 10.1039/b000000x/
95
100
105
110
115
120
125
1
2
3
G. W. Huber, J. A. Dumesic, Catal. Today 111, 2006, 119ꢀ
132.
J. J. Bozell, M. K. Patel, Feedstocks for the Future.
(American Chemical Society Washington, DC, 2006).
J. A. Geboers, S. Van de Vyver, R. Ooms, B. Op de Beeck,
P. A. Jacobs, B. F. Sels, Catal. Sci. Technol. 1, 2011, 714ꢀ
726.
4
5
6
7
X. Liu, X. Wang, S. Yao, Y. Jiang, J. Guan, X. Mu, RSC
Adv. 4, 2014, 49501.
A. Fukuoka, P. L. Dhepe, Angew. Chem. Int. Ed. 45, 2006,
5161ꢀ5163.
A. Onda, T. Ochi, K. Yanagisawa, Green Chem. 10, 2008,
1033ꢀ1037.
S. Suganuma, K. Nakajima, M. Kitano, D. Yamaguchi, H.
Kato, S. Hayashi, M. Hara, J. Am. Chem. Soc. 130, 2008,
12787ꢀ12793.
8
9
S. Rebsdat, D. Mayer, Ullmann's Encyclopedia of Industrial
Chemistry. (WileyꢀVCH Verlag GmbH & Co. KGaA, 2000).
C. Luo, S. Wang, H. Liu, Angew. Chem. 119, 2007, 7780ꢀ
7783.
60 Conclusions
10 W. Deng, X. Tan, W. Fang, Q. Zhang, Y. Wang, Catal. Lett.
133, 2009, 167ꢀ174.
Polyols, a class of biomassꢀderived platform molecules, were
obtained in high yield by hydrolytic hydrogenation of different
cellulosic feedstocks using a bifunctional RuꢀW catalyst which
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 5

Green Chemistry
Page 6 of 12
View Article Online
DOI: 10.1039/C5GC00421G
11 H. Kobayashi, T. Komanoya, K. Hara, A. Fukuoka,
ChemSusChem 3, 2010, 440ꢀ443.
29 M. Sasaki, Z. Fang, Y. Fukushima, T. Adschiri, K. Arai, Ind.
Eng. Chem. Res. 39, 2000, 2883ꢀ2890.
55
12 Y. Zhang, A. Wang, T. Zhang, Chem. Commun. 46, 2010,
5
862ꢀ864.
30 M. Sasaki, K. Goto, K. Tajima, T. Adschiri, K. Arai, Green
Chem. 4, 2002, 285ꢀ287.
13 N. Ji, T. Zhang, M. Zheng, A. Wang, H. Wang, X. Wang, J.
G. Chen, Angew. Chem. 120, 2008, 8638ꢀ8641.
60 31 M. Sasaki, M. Furukawa, K. Minami, T. Adschiri, K. Arai,
Ind. Eng. Chem. Res. 41, 2002, 6642ꢀ6649.
10 14 M.ꢀY. Zheng, A.ꢀQ. Wang, N. Ji, J.ꢀF. Pang, X.ꢀD. Wang, T.
Zhang, ChemSusChem 3, 2010, 63ꢀ66.
32 O. Glemser, C. Naumann, Z. Anorg. Allg. Chem. 265, 1951,
288ꢀ302.
15 Y. Liu, C. Luo, H. Liu, Angew. Chem. 124, 2012, 3303ꢀ3307.
65
33 M. Schwanninger, J. C. Rodrigues, H. Pereira, B.
15 16 Z. Tai, J. Zhang, A. Wang, J. Pang, M. Zheng, T. Zhang,
Hinterstoisser, Vib. Spec. 36, 2004, 23ꢀ40.
ChemSusChem 6, 2013, 652ꢀ658.
34 A. Onda, T. Ochi, K. Yanagisawa, Top. Catal. 52, 2009, 801ꢀ
17 J. Geboers, S. Van de Vyver, K. Carpentier, P. Jacobs, B.
Sels, Green Chem. 13, 2011, 2167ꢀ2174.
70
75
807.
20
35 H. Kobayashi, Y. Ito, T. Komanoya, Y. Hosaka, P. L. Dhepe,
K. Kasai, K. Hara, A. Fukuoka, Green Chem. 13, 2011, 326ꢀ
333.
18 J. Geboers, S. Van de Vyver, K. Carpentier, K. de Blochouse,
P. Jacobs, B. Sels, Chem. Comm. 46, 2010, 3577ꢀ3579.
19 R. Palkovits, K. Tajvidi, A. M. Ruppert, J. Procelewska,
36 H. Kobayashi, M. Yabushita, T. Komanoya, K. Hara, I.
25
Chem. Comm. 47, 2011, 576ꢀ578.
Fujita, A. Fukuoka, ACS Catalysis 3, 2013, 581ꢀ587.
20 S. J. You, I. G. Baek, E. D. Park, Appl. Catal., A, 466, 2013,
37 F. Cherubini, Energy Convers. Manag. 51, 2010, 1412ꢀ1421.
161ꢀ168.
80
38 B. Kamm, M. Kamm, Appl. Microbiol. Biotechnol. 64, 2004,
30 21 I. G. Baek, S. J. You, E. D. Park, Bioresour. Technol. 114,
137ꢀ145.
2012, 684ꢀ690.
22 K. Fabicovicova, O. Malter, M. Lucas, P. Claus, Green
85
Chem. 16, 2014, 3580ꢀ3588.
35
23 B. Kusserow, S. Schimpf, P. Claus, Adv. Synth. Catal. 345,
2003, 289ꢀ299.
90
24 A. Bienholz, R. Blume, A. KnopꢀGericke, F. Giergsdies, M.
40
45
Behrens, P. Claus,
J. Phys. Chem. C 115, 2011, 999ꢀ1005.
25 A. Bienholz, F. Schwab, P. Claus, Green Chem. 12, 2010,
95
290ꢀ295.
26 J. Plößer, M. Lucas, P. Claus, J. Catal. 320, 2014, 189ꢀ197.
27 G. Liang, H. Cheng, W. Li, L. He, Y. Yu, F. Zhao, Green
Chem. 14, 2012, 2146ꢀ2149.
100
50
28 J. Geboers, S. Van de Vyver, K. Carpentier, P. Jacobs, B.
Sels, Chem. Comm. 47, 2011, 5590ꢀ5592.
105
110
115
120
6
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

Page 7 of 12
Green Chemistry
View Article Online
DOI: 10.1039/C5GC00421G
Tables
5
Table 1. Cellulose conversions (X), carbon efficiency coefficients (CEL) and product yields of cellulose hydrogenolysis over Ru/W/ACa
catalyst including recycling.
#
Exp
X
[%]
CEL
[%]
Yield
EG
[%]
PG
Leaching
of
tungsten
mCAT
[g]
BD
Ery
Sor
Other
Polyols
MeOH+ Others
EtOH
[ppm]
nm
127
71
STD
90
76.7
70.4
68.8
75.9
72.3
73.2
67.3
55.6
76.1
34.2
36.7
34.4
34.8
39.1
39.7
33.8
4.5
4.2
3.4
3.8
2.8
4.6
3.8
3.9
3.1
5.0
3.5
3.3
3.8
3.1
4.4
4.4
4.2
1.3
4.7
5.2
5.1
5.0
5.3
4.0
3.7
3.5
4.5
5.1
18.5
15.4
18.2
20.4
16.0
14.7
15.7
28.4
12.7
1.0
1.4
1.7
2.0
1.8
2.6
3.1
4.6
2.1
(1.3+0.6)
(1.4+1.0)
(1.0+0.8)
(1.2+0.8)
(1.7+0.9)
(1.7+0.9)
(1.0+0.9)
(0.3+0.2)
(2.0+1.5)
0.4
0.1
0.8
1.1
0.9
1.0
0.8
0.6
0.04
0.501
0.505
0.505
0.450
0.450
0.369
0.360
0.357
0.372
1
2
3
4
5
6
7
8
9
1.run 100
2.run 100
3.run 100
4.run 100
5.run 100
6.run 100
7.run 85
66
45
40
21
5
nm
RED
100
42.3
EG = ethylene glycol, PG = propylene glycol, BD = 1,2ꢀbutanediol, Ery = erythritol, Sor = sorbitol, Other Polyols = (glycerol+1,2ꢀ
hexanediol), MeOH+EtOH = (methanol+ethanol), Others = (cellobiose, glucose, mannose, fructose, mannitol, xylitol, 3ꢀ
hydroxytetrahydrofurane, 1ꢀhydroxyꢀ2ꢀpropanone, 1ꢀhydroxyꢀ2ꢀbutanone, 3ꢀhydroxyꢀ2ꢀbutanone, 2,5ꢀhexanedione, acetic acid, levulinic
10 acid), nm = not measured, STD = experiment at standard reaction conditions, RED = experiment with reduced amount of catalyst.
a Tungsten loading = 40 wt% relating to WO3. Ruthenium loading = 2 wt%.
Table 2. Cellulose conversions (X), carbon efficiency coefficients (CEL) and product yields of cellulose hydrogenolysis over Ru/W/ACa
b
and Ru/AC + H2WO4 .
#
Exp
Time
[h]
X
CEL
Yield
EG
[%]
PG
[%] [%]
BD
Ery
Sor
Other
Polyols
2.8
MeOH+
EtOH
Others
Ru/W/AC
1
2
3
1
2
65
84
90
73
92
76.4
81.5
76.7
69.6
73.7
24.4
34.5
34.2
10.4
19.7
2.0
2.8
4.2
1.3
3.6
2.0
2.9
3.5
0.7
2.1
3.9
5.8
5.2
6.5
7.8
13.8
17.0
18.5
30.5
30.2
(1.0 + 1.1)
(1.4 + 1.5)
(1.3 + 0.6)
(0.5 + 0.3)
(1.2 + 0.8)
0.6
0.9
0.4
0.3
0.7
1
2
3
4
5
Ru/W/AC
1.4
Ru/W/AC
1.0
Ru/AC+ H2WO4
Ru/AC+ H2WO4
0.5
1.2
Ru/AC+ H2WO4
3
95
79.5
23.1
4.2
2.2
7.2
27.0
2.4
(1.2 + 0.6)
0.1
6
15 EG = ethylene glycol, PG = propylene glycol, BD = 1,2ꢀbutanediol, Ery = erythritol, Sor = sorbitol, Other Polyols = (glycerol+1,2ꢀ
hexanediol), MeOH+EtOH = (methanol+ethanol), Others = (cellobiose, glucose, mannose, fructose, mannitol, xylitol, 3ꢀ
hydroxytetrahydrofurane, 1ꢀhydroxyꢀ2ꢀpropanone, 1ꢀhydroxyꢀ2ꢀbutanone, 3ꢀhydroxyꢀ2ꢀbutanone, 2,5ꢀhexanedione, acetic acid, levulinic
acid).
a Tungsten loading = 40 wt% relating to WO3. Ruthenium loading = 2 wt%.
20 b Ru/AC: Ruthenium loading = 2 wt%. Mass of catalyst = 0.5 g. Mass of H2WO4 = 0.25 g.
25
30
35
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 7

Green Chemistry
Page 8 of 12
View Article Online
DOI: 10.1039/C5GC00421G
Figures
Scheme 1. Simplified reaction scheme of cellulose hydrogenolysis over Ru/W/AC catalyst: main reactions to glucose 1,
glycolaldehyde 2a, ethylene glycol 2b, erythrose 3a, erythritol 3b, sorbitol 4, byꢀproduct formation: fructose 5, 1ꢀhydroxyꢀ2ꢀpropanone
6a, propylene glycol 6b, 1ꢀhydroxyꢀ2ꢀbutanone 7a, 1,2ꢀbutanediol 7b, 3ꢀhydroxyꢀ2ꢀbutanone 8a, 2,3ꢀbutanediol 8b. Further degradation
products which were formed during the reaction (often in trace amounts: acetic acid, levulinic acid, cellobiose, and mannose) are not
included in this scheme.
5
10
8
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

Page 9 of 12
Green Chemistry
View Article Online
DOI: 10.1039/C5GC00421G
5
Fig. 1. Yields of ethylene glycol (EG), propylene glycol (PG), 1,2ꢀbutanediol (BD), and sorbitol in dependence on tungsten
leaching in the recycling test.
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 9

Green Chemistry
Page 10 of 12
View Article Online
DOI: 10.1039/C5GC00421G
Fig. 2 Conversions and product yields for Ru/W/AC with nonꢀpretreated cellulose (0 min) and ballꢀmilled cellulose (4, 15, 30, 120,
and 720 min). Reaction conditions: 0.5 g catalyst, 5 g cellulose, 100 mL water, reaction temperature 493 K, 65 bar of hydrogen pressure
(measured at reaction temperature), 1000 rpm, * ꢀ mixꢀmilling of catalyst and cellulose.
5
Fig. 3. Conversions and product yields for hydrogenolysis of different kinds of cellulosic material (pine unbleached, pine
10 bleached, birch, eucalyptus) over Ru/W/AC catalyst. Reaction conditions: 0.5 g catalyst, 5 g cellulosic material, 100 mL water,
reaction temperature 493 K, 65 bar of hydrogen pressure (measured at reaction temperature), 1000 rpm, 3 hours reaction time. 1ꢀOHꢀ2ꢀP
= 1ꢀhydroxyꢀ2ꢀpropanone, 1ꢀOHꢀ2ꢀB = 1ꢀhydroxyꢀ2ꢀbutanone, 3ꢀOHꢀ2ꢀB = 3ꢀhydroxyꢀ2ꢀbutanone.
10 | Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

Page 11 of 12
Green Chemistry
Polyols, a class of biomass-derived platform molecules, were obtained in high yield by hydrogenolysis
of different cellulosic feedstocks using a bifunctional Ru-W catalyst which exhibits a very high concentration of acid sites.

Green Chemistry
Page 12 of 12
View Article Online
DOI: 10.1039/C5GC00421G