Isosorbide synthesis from cellulose with an efficient and recyclable ruthenium catalyst

Article abstract of DOI:10.1039/c7gc01821e

Herein, we describe an efficient two-step pathway for isosorbide synthesis from cellulose with the use of new recyclable Ru-catalysts. We show that the oxidative and sulfonation treatments of the new Ru-catalysts increase the acidity and the hydrophilicity of the activated carbon support material, thus reducing the catalyst fouling caused by the build-up of insoluble products. Accordingly, the new Ru-catalysts are more resilient towards lignin containing cellulose than a commercial Ru/C catalyst, and the best Ru-catalyst maintains its high catalytic activity in four consecutive runs with dissolving pulp, microcrystalline cellulose and even with residual lignin containing bagasse pulp. Overall, our two-step approach provides isosorbide in high yields of 56-57 mol% (49-50 wt% of the substrate) from the cellulosic substrates.

Full text of DOI:10.1039/c7gc01821e

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: J. Keskiväli, S.
Rautiainen, M. Heikkilä, T. T. T. Myllymäki, J. Karjalainen, K. lagerblom, M. L. Kemell, M. Vehkamäki, K.
Meinander and T. Repo, Green Chem., 2017, DOI: 10.1039/C7GC01821E.
Volume 18 Number
7
7
April 2016 Pages 1821–2242
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Green
Chemistry
Cutting-edge research for a greener sustainable future
www.rsc.org/greenchem
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
author guidelines.
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, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1463-9262
CRITICAL REVIEW
G. Chatel et al.
Heterogeneous catalytic oxidation for lignin valorization into valuable
chemicals: what results? What limitations? What trends?
rsc.li/green-chem

Page 1 of 9
Pleas eG dr eo e nn o Ct ha de jmu si st tmr yargins
View Article Online
DOI: 10.1039/C7GC01821E
Journal Name
ARTICLE
Isosorbide Synthesis from Cellulose with Efficient and Recyclable
Ruthenium Catalyst
a
a†
a
b
Juha Keskiväli, Sari Rautiainen, Mikko Heikkilä, Teemu T.T. Myllymäki, Jaakko-Pekka
Received 00th January 20xx,
Accepted 00th January 20xx
a
a
a
a
c
Karjalainen, Kalle Lagerblom, Marianna Kemell, Marko Vehkamäki, Kristoffer Meinander, and
a
Timo Repo *
DOI: 10.1039/x0xx00000x
Herein, we describe an efficient two-step pathway for isosorbide synthesis from cellulose with the use of new recyclable Ru-
catalysts. We show that the oxidative and sulfonation treatments of the new Ru-catalysts increase the acidity and the
hydrophilicity of the activated carbon support material, thus reducing the catalyst fouling caused by build-up of insoluble
products. Accordingly, the new Ru-catalysts are more resilient towards lignin containing cellulose than a commercial Ru/C
catalyst, and the best Ru-catalyst maintains its high catalytic activity in four consecutive runs with dissolving pulp,
microcrystalline cellulose and even with the residual lignin containing bagasse pulp. Overall, our two-step approach provides
isosorbide in high yields of 56-57 mol% (49-50 w% of the substrate) from the cellulosic substrates.
www.rsc.org/
Introduction
The concern of increasing energy and material consumption
and their impacts on the environment have led to growing
demand of sustainable chemicals and fuels from renewable
1
resources. In this respect, non-edible lignocellulosic biomass is
considered as one of the best resource due to its abundancy,
versatility and price. The hydrolysis of lignocellulose offers
access to a wide variety of sugars, which can serve as diverse
substrates in the synthesis of platform- and fine-chemicals,
2–9
biofuels, and solvents. One of the most studied biomass-
derived platform molecules is isosorbide, which can be
synthesised from polysaccharides, starch and cellulose, as well
as from glucose and sorbitol.7
,10–12
The interest towards
isosorbide has risen during recent years due to its high potential
as a monomer in biopolymer production and pharmaceutical
Scheme 1: The reaction pathway for the two-step synthesis of isosorbide from cellulose.
12–14
applications.
Despite the promising results obtained with
starch, glucose and sorbitol, efficient isosorbide synthesis from
inedible resources, such as lignocellulose, is still in high demand.
(
anhydrosorbitols) and isosorbide (Scheme 1). The reported
acid catalysts include homogeneous H
well as heterogeneous catalysts
phosphate, Amberlyst-70 and Dowex-H.
cellulose hydrolysis, Ru- and Ni-based catalysts are commonly
used in hydrogenation of glucose to sorbitol.
3
PO
, H SO
4 2 4
and HCl as
H
4
SiW12
O
40
,7,15–20
,
niobium
After the
Isosorbide is available from cellulose by acid-catalysed
hydrolysis of the glycosidic bonds, followed by hydrogenation of
glucose to sorbitol and further dehydration to sorbitans
6
21–24
The formed
sorbitol undergoes an acid-catalysed two-fold dehydration into
isosorbide through 1,4- and 3,6-sorbitan‡ intermediates.
Typically isosorbide is synthesised under high hydrogen
pressures and high temperatures to ensure efficient
hydrogenation of glucose and dehydration of sorbitol and
25
a
Department of Chemistry, University of Helsinki, A. I. Virtasen aukio 1, P.O. box
5, 00014 University of Helsinki, Finland
Department of Applied Physics, Aalto University, P.O. box 15100, 00076 Aalto,
5
b
Finland
c
Department of Physics, University of Helsinki, Gustaf Hällströmin katu 2a, P.O.
6,7,19
sorbitans.
box 64, 00014 University of Helsinki, Finland
†
Present address:Department of Organic Chemistry, Stockholm University, Svante
Arrhenius väg 16 C, 10691 Stockholm, Sweden
Corresponding author, E-mail: timo.repo@helsinki.fi
Electronic Supplementary Information (ESI) available: General experimental details,
Optimization of one-step reactions, optimization of two-step reactions, catalyst
characterization, catalysts on sulfonated carbon. See DOI: 10.1039/x0xx00000x
Generally, the isosorbide synthesis from cellulose is conducted
with one-step reaction setups, exposing hydrogenation
catalysts to the harsh reaction conditions required to dehydrate
*
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

Pleas eG dr eo e nn o Ct ha de jmu si st tmr yargins
Page 2 of 9
ARTICLE
Journal Name
sorbitol and sorbitans.6,17 Unfortunately, typical one-step Table 1: Highest isosorbide yields with different substrates underVoiepwtimAratlicolenOe-nsltineep
conditions seems to facilitate the deactivation of the reaction conditions with the commercial Ru/C as the hydrogenation catalyst.a
DOI: 10.1039/C7GC01821E
hydrogenation catalysts, thus preventing their efficient
Entry
Substrate
Time (h) Temperature
°C)
Isosorbide
yield (mol%)
50
recycling. Reasons for the deactivation include build-up of
insoluble by-products, metal particle sintering and leaching.26,27
Although the catalyst recycling is vital for the efficient
production of isosorbide, few methods to increase the
d
(
1
2
3
Bagasse Pulp
2
2
2
220
220
220
b
(BP)
Microcrystalline
cellulose (MCC)
Dissolving pulp
51
50
2
8
recyclability have been reported.
(
DP)c
The reports on isosorbide synthesis cover mostly pure cellulose
and sugars as substrates, such as microcrystalline cellulose
4
5
Glucose
Sorbitol
2
2
220
220
60
67
(
MCC) and glucose, while studies involving isosorbide synthesis
15
a
from lignin containing cellulosic substrates are sparse. In
addition, activated charcoal is commonly used as
1.18 mmol of cellulose, 0.5 M H
0.02 mmol of Ru (5 w% Ru/C); Bagasse pulp (95.37 w% cellulose and 4.63 w%
lignin) obtained from Stora Enso, used as received, and the amount of the
substrate was calculated to match 1.18 mmol of cellulose; Dissolving pulp
obtained from Stora Enso, used as received; The yields were obtained with HPLC
2
SO
4
(aq) 30 ml, 220 °C, 40 bar H
2
, 2 hours and
b
a
hydrogenation catalyst support material, but the effect of its
hydrophilicity and acidity on the recyclability of the
hydrogenation catalyst in the isosorbide synthesis has not yet
been established. Considering this, the optimisation of the
c
d
equipped with refractive index detector (HPLC-RID) using calibration curves.
reaction conditions and the development of the hydrogenation Under the optimized one-step reaction conditions, the best
catalysts to tolerate the presence of lignin, enabling the use of isosorbide yields were obtained within 2 hour reaction at 220 °C
genuine cellulose streams from pulp industry as substrates, using 0.5 M H SO aqueous solution and 40 bar hydrogen
2
4
should be further investigated.
pressure. Longer reaction times resulted in degradation of
isosorbide and the colour of the solution turned pale yellow.
Herein, we report a high yield two-step synthesis of isosorbide The highest yield of isosorbide from cellulosic substrates was 51
from cellulose with new Ru-based hydrogenation catalyst mol%, and the use of glucose and sorbitol as substrates
comprising high hydrogenation activity and good recyclability produced 60 and 67 mol% isosorbide yields, respectively (Table
even with lignin containing substrate, bagasse pulp (BP). We 1, entries 2, 4 & 5). However, under the one-step reaction
also focused on the influence of the support’s hydrophilicity on conditions the hydrogenation catalyst, Ru/C suffered severe
the deactivation of the hydrogenation catalyst. We were able to deactivation and the recycling of the catalyst was not possible;
synthesise isosorbide in good yields of 56-57 mol% from as a result, glucose underwent competing acid-catalysed
different cellulosic substrates. Furthermore, the new cascade reaction of isomerisation-dehydration-rehydration
hydrogenation catalysts could be recycled for at least four runs forming levulinic and formic acid during the second run (Scheme
2
9–31
using different cellulosic substrates.
2).
Results and discussion
Optimization of the one-step reaction conditions
We initialized the study by optimizing one-step high
temperature reaction setup and comparing the reaction
outcome with glucose, sorbitol and cellulosic materials as isomerisation, dehydration and rehydration sequence.
substrates. We also studied the effect of various Brønsted acids
Scheme 2: Formation of levulinic and formic acid from glucose through
and hydrogenation catalysts on the outcome of the synthesis of
isosorbide (Table 1, See Table S1 in ESI). Simultaneously to the Optimization of the two-step reaction conditions
optimization, we studied the recyclability of the hydrogenation
catalysts (See Tables S1 & S4 in ESI).
In order to improve the recyclability of Ru/C, the isosorbide
synthesis was divided into two steps. The first step includes
hydrolysis of cellulose to glucose, hydrogenation of glucose into
sorbitol and partial dehydration of sorbitol to sorbitans (Scheme
The catalyst studies revealed that the effectivity of acids to
catalyse hydrolysis of cellulose and dehydration of sorbitol and
1
, Step 1). We investigated reaction temperature, time, H
2
sorbitans order
decreased
in
the
of
pressure and dilute acid concentrations, to find the ideal
reaction conditions using BP as the substrate (See Table S2 in
ESI). To minimize the deactivation of Ru/C, the reaction
temperatures were kept relatively low compared to the one-
step reactions. Overall, the optimized conditions were found to
be 170 °C, 2 h, 0.5 M H SO , 20 bar of H , which generated
2 4 2
combined yield of 59 mol% of sorbitol, 1,4-sorbitan and
isosorbide. After the reaction, the hydrogenation catalyst was
H
2
SO >HOTf>TFA>Amberlyst-38>HCl>HNO . Additionally, Ru/C
4
3
turned out to be the best hydrogenation catalyst, as using Pt/C,
Pd/C and Rh/C catalysts only small yields of isosorbide were
detected (See Table S1 in ESI). Based on these results, H SO and
2 4
Ru/C were selected for further studies. At this point, we
considered only activated carbon as the support material of the
catalysts, due to its relatively low price and versatility.
2
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 9
Pleas eG dr eo e nn o Ct ha de jmu si st tmr yargins
Journal Name
ARTICLE
collected via filtration, washed with milli-q water and stored for respectively. The 6 hour reaction time was used in the following
View Article Online
DOI: 10.1039/C7GC01821E
further use.
investigations.
The effective acid-catalysed dehydration of sorbitol and 1,4- After the optimization we used different substrates under the
sorbitan to isosorbide in the second step requires higher best two-step reaction conditions to compare the reaction
temperatures than those used in the first step. The dehydration outcome with the one-step process (Table 3). The two-step
was investigated at high temperatures (≥200 °C) in order to setup generated isosorbide from cellulosic substrates in 52-54
obtain high isosorbide yields within reasonable reaction time mol% yields, are comparable to the one-step setup 50-51 mol%
(Scheme 1, Step 2, See Table S3 in ESI). The reactions were yields of isosorbide (Table 3). Similar elevations in yields were
conducted with the clear colourless acidic filtrate collected after noticed with glucose and sorbitol as substrates. Throughout the
the first reaction step (Table 2). With short reaction times (1-2 optimization studies the yields of 2,5- and 1,5-sorbitans
h) isosorbide was generated in good yields of 43 mol% and 47 remained at same level around 3% and 8% respectively. In both,
mol% at 240 °C and 260 °C, respectively. However, extended one- and two-step reactions, the screening of the reaction
reaction time at these temperatures resulted in significant loss conditions revealed that the relatively high temperatures and
of overall yield through side reactions, thus lowering the acid concentrations resulted in accelerated dehydration, and as
isosorbide yield (Table 2 entries, 3-4 & 7-8). Curiously, during a consequence, high yields of the dehydration products
the first step studies we noticed similar decrease in the overall (sorbitans and isosorbide) were obtained. Also, in all of the
yields at lower temperatures in the presence of Ru-catalyst experiments full conversion of cellulosic substrates was
(>180 °C, See Figure S1 in ESI). This implies that the Ru-catalyst observed. Although, BP is more fibrillary and porous than MCC,
amplifies the side reactions, thus the separation of the Ru- according to SEM, the isosorbide yields are equal after the
catalyst after the first step benefits the process in two ways; it second step of the reaction (See Figure S18). Apparently, the
decreases the side reactions and increases the recyclability. The different microstructures of the cellulosic substrates has little
outcome of the second step reaction was enhanced by lowering effect on the two-step reaction outcome.
the reaction temperature and increasing the reaction time.
Increase of reaction time at 200 °C resulted in gradual change Catalyst studies
of product distribution from favouring sorbitol and 1,4- sorbitan
The recyclability of commercial Ru/C in the two-step setup was
as the main products to isosorbide with only a small loss of the
studied with different cellulosic substrates. After the optimized
overall yield (Table 2 entries 1, 5, 9-11). Similar observation was
hydrogenation reaction (first step) the Ru/C was separated via
made with reactions conducted at 170 °C, and up to 58 mol%
filtration and used for four consecutive runs with MCC,
yield of isosorbide was obtained after considerably long 24 hour
dissolving pulp (DP) or BP as substrates (Figure 1). The Ru/C
reaction time (See Figure S2 in ESI). Unfortunately, the long
catalyst was reused without any regeneration procedures, but
reaction times are detrimental on the recyclability of the Ru/C,
2
simply washed with milli-q H O, dried and ground between the
thus preventing the use of 24 hour one-pot reaction at 170 °C
See Table S5 in ESI). The highest isosorbide yields of 53 mol%
runs. Unexpectedly, the recycling experiments revealed, that
the yields of 1,4-sorbitan and sorbitol were significantly higher
in the second run than in the first run. After the second run,
sorbitol, 1,4-sorbitan and isosorbide were detected in the
overall yields of 74% with MCC, 70% with DP and 72% with BP
(
and 54 mol% from BP were obtained at 200 °C in 5 and 6 hours
Table 2: The effects of reaction temperature and time on the second step reaction with
BP as the substrate.
(
Figure 1). To examine if the increased yields were caused by the
Yield (%)a
Entry
Time
Temp.
(°C)
200
220
240
260
200
220
240
260
200
200
200
1,4-
Isosorbide
Overallb
Table 3: Highest isosorbide yields with different substrates under optimal two-step
reaction conditions with the commercial Ru/C as the hydrogenation catalyst.a
(
h)
1
1
1
1
2
2
2
2
4
5
6
Sorbitan+Sorbitol
1
2
3
4
5
6
7
8
9
62
43
35
3
37
5
0
0
20
8
4
7
78
66
64
55
74
58
54
40
71
74
69
Entry
Substrate
Temperature
°C)
Time (h)
Isosorbide yield
12
22
43
27
44
47
37
41
53
54
d
(
(%)
1
2
3
4
5
BPb
MCC
DPc
Glucose
Sorbitol
170+200
170+200
170+200
170+200
200
2+6
2+6
2+6
2+6
6
54
54
52
62
73
a
1.18 mmol of cellulose, 0.5 M H
170 °C, 20 bar H , 2 h and 0.02 mmol of second run Ru/C. Between the reaction
steps the Ru/C is collected via filtration and the second step is conducted with the
filtrate. The second step reaction conditions: 200 °C, 40 bar H
2 4
SO (aq) 30 ml; the first step reaction conditions:
1
1
0
1
2
b
2
, 6 h; Bagasse pulp
Prior to all of the second step reactions, the first step was conducted under the
optimized reaction conditions (170 °C, 2h, 20 bar H , 0.5 M H SO , second run Ru/C
0.02 mmol of Ru)) using BP as a substrate, used as received; The yields were
(
95.37 w% cellulose and 4.63 w% lignin) obtained from Stora Enso, used as received
2
2
4
and the amount of the substrate was calculated to match 1.18 mmol of cellulose;
a
(
c
d
Dissolving pulp obtained from Stora Enso, used as received; The yields were
obtained with HPLC-RID using calibration curves.
b
obtained with HPLC -RID using calibration curves; Sum of 1,4-sorbitan, sorbitol,
,5-sorbitan, 2,5-sorbitan and isosorbide yields.
1
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 3
Please do not adjust margins

Pleas eG dr eo e nn o Ct ha de jmu si st tmr yargins
Page 4 of 9
ARTICLE
Journal Name
µmol/g of the once used Ru/C (See Table S9 in ESI). In
View Article Online
DOI: 10.1039/C7GC01821E
comparison to the one-step setup described above, the clear
benefit of the two-step setup is higher yields of isosorbide and
greatly improved recycling of the Ru/C catalyst with the lignin-
free substrates. Although the commercial Ru/C catalyst could
be recycled in the two-step setup with MCC and DP, however,
with the lignin containing BP the overall catalytic activity of Ru/C
decreased drastically after the second run, generating levulinic
and formic acid as the main products (Figure 1 and Scheme 2).
In addition, after the second use of the catalyst, the amount of
the strong acid sites had decreased to 409 µmol/g with BP as
the substrate. We presumed that the catalyst deactivation with
BP could arise from build-up of insoluble lignin based products
on the surface of the catalyst. Due to good catalyst recyclability
with MCC and DP, it seems that the presence of sulphuric acid
did not result into the widely reported sulphur poisoning of the
hydrogenation catalyst under the conditions in hand (Figure
2
6,37,38
1
).
From the point-of-view of sustainability, low-purity
lignocellulosic raw materials available from pulp industry are
more preferable substrates than the pure cellulosic substrates.
To further improve Ru/C catalyst performance and recyclability
using the lignin containing substrate BP, we studied the effect
of different acidifying treatments, sulfonation and oxidation, on
the carbon surface. The generated strong acid sites make the
carbon surface more hydrophilic and should, therefore,
increase the accessibility of hydrophilic molecules on the
39
catalyst surface. Thus, we prepared a series of sulfonated
carbon supports, with a procedure described by Foo et al.,32
using different sulfuric acid treatments (Table 4). After the
sulfonation, we determined the amount of acidic sites in the
support material with Boehm-titration and measured the point
35
of zero charge (PZC). Hereinafter, the acidified support
materials are denoted as Csulfonated with subscript indicating the
2 4
H SO concentration and time of the acid treatment.
Ruthenium (RuCl
support materials as described by Komanoya et al., followed
by the reduction of ruthenium(III) to ruthenium(0) under
3
) was impregnated on the acidified carbon
Figure 1: Results from commercial Ru/C catalyst recycling with different cellulosic
materials under the optimized first step reaction conditions. Reaction conditions:
22
1
2
.18 mmol of cellulose, 0.02 mmol of Ru (5 w% Ru/C), 170°C, 20 bar H , 30 ml of
0
.5 M H
2 4
SO . Yields obtained with HPLC-RID using calibration curves. 1,4-Sorbitan
and sorbitol signals overlap in the chromatogram.
Table 4: Different sulfonated carbon support materials and the amounts of their acidic
sites.
adsorption of the products on the catalyst, the once used Ru/C
catalyst was subjected to the same reaction conditions without
any substrate. Only 3 mol% of the products were detected in
the solution (See Table S4, entry 16 in ESI). Thus, it is likely that
during the first run, the catalyst was altered in a manner that
prevents the side reactions. For example, the carbon support is
readily modifiable in terms of hydrophilicity and acidity by
Sulfonation
conditions
Amount of acid sites (µmol/g) a
Support
AC Norit
2 4
c(H SO )
Time Weak and Strongd
Total
PZC c
(
h)
-
medium
103
(pH)
8.4
-
47
150
b
SX
2 4
adding sulfonic acid groups in the presence of strong H SO and
C
C
C
C
10M30min
10M3h
conc30min
conc3h
10 M
10 M
conc.
conc.
0.5
3
0.5
3
235
39
10
314
438
447
447
549
477
457
573
5.8
5.3
5.3
4.0
in hot liquid water varying the quantity of the oxygen containing
functionalities, of which carboxylic acid, phenol, lactone and
32–36
carboxylic acid anhydride are the most typical.
126
a
The amount of the acid sites were determined by Boehm-titration; b Untreated
To gain insight of the effect of the reaction conditions on the
catalyst, the amount of acid sites was determined using Boehm-
titration. The amount of strong acid sites was found to almost for 2 hours, and then measuring the pH of the solution. sulfonic and carboxylic
support material, purchased from Sigma Aldrich; c The point-of-zero-charge (PZC)
was determined by vigorously stirring 1 g of the support material in 25 ml of water
d
35
acids are not separable by Boehm-titration as both are strong acids.
double from 285 µmol/g of the fresh commercial Ru/C to 530
4
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 5 of 9
Pleas eG dr eo e nn o Ct ha de jmu si st tmr yargins
Journal Name
forming gas atmosphere (5% H
ARTICLE
2
2
/ 95% N ) at 400 °C for 2 hours
View Article Online
DOI: 10.1039/C7GC01821E
in an oven (See detailed description in ESI). The catalysts were
then cooled to room temperature under forming gas, resulting
in reduced Ru/Csulfonated catalysts. The synthesized catalysts
were studied under the optimized first-step reaction conditions
using the lignin containing BP as the substrate (See Table S10 in
ESI). Depending on the acidifying treatment, the combined
yields of sorbitol, 1,4-sorbitan and isosorbide were 20-60 mol%
after the first run. However, upon recycling the catalysts
exhibited low activity and even total deactivation. The Boehm-
titrations of the acidified supports and the fresh Ru/Csulfonated
catalysts showed that the amount of the strong acid sites was
dramatically decreased, from 314-447 µmol/g to 233-333
µmol/g, during the catalyst preparation (See Table S9 in ESI).
The decomposition of the carboxyl acid groups has been
reported to occur at around 200 °C and sulfonic acids at above
3
50°C under inert atmosphere, thus explaining the decrease of
the strong acid sites during the heating in the catalyst
3
5,36,40
preparation.
To preserve the strong acid sites, we
introduced an oxidative treatment for the catalysts. After the
reduction of ruthenium (400 °C, 2h under forming gas
atmosphere), the oven was cooled to 290 °C and the
atmosphere was changed to air followed by cooling to room
temperature. The series of the oxidised catalysts are denoted
with RuO /Csulfonated. Notably, after this oxidative treatment the
x
amount of the strong acid sites was 325-425 µmol/g, similar to
the original levels of the sulfonated support material. As a
x
result, the RuO /Csulfonated catalysts showed high catalytic
activity under the optimized first-step reaction conditions and
gave the desired hydrogenation products in good yields, even in
x
the first run using BP as a substrate (Figure 2). Using RuO /C10M3h
as catalyst the highest first run overall yield (sorbitol, 1,4-
sorbitan and isosorbide) of 75 mol% was obtained, which was
significantly higher than the corresponding 59 mol% with
commercial Ru/C and the 60 mol% with Ru/Csulfonated
.
Remarkably, activity of the RuO /C10M3h catalyst remained
x
constant during four consecutive runs even with the lignin
containing BP as substrate, which to our knowledge is the first
reported recyclable catalyst for synthesis of isosorbide from
Figure 2. Results from RuO
x
/C recycling experiments using BP as a substrate under
the optimized first step reaction conditions. Reaction conditions: 1.18 mmol of
x 2
cellulose, 0.02 mmol of RuO /C catalysts (5 w% Ru), 170°C, 20 bar H , 30 ml of 0.5
2 4
M H SO . Yields obtained with HPLC-RID using calibration curves. 1,4-Sorbitan and
sorbitol signals overlap in the chromatogram.
lignin containing substrates (Figure 2). In contrast, RuO
x
deposited on untreated carbon (AC Norit SX) showed decreased
activity after second run, similar to the commercial catalyst. In
explains the higher yields in the first runs and the superior
recyclability of the more acidic RuO /Csulfonated catalysts
x
x
general, using lignin-free cellulose DP all of the RuO /Csulfonated
compared to the less acidic Ru-based catalysts.
catalysts could be recycled for four runs without loss of activity,
and they showed enhanced recyclability compared to
commercial Ru/C with the use of BP as the substrate (See Table
S10 in ESI). Cellulosic substrates and their oligomers are known
to adsorb on to the hydrophobic sites of the carbon supports
through CH-π hydrogen bonding and hydrophobic
interactions. Similarly the hydrophobic lignin is capable to
form these interactions with the catalyst. The adsorption of
these compounds could cover the active hydrogenation Ru-sites
on the catalyst, and result in the deactivation. Seemingly, the
addition of the hydrophilic oxygen containing functionalities,
such as carboxylic acids, reduces the adsorption of the glucose
based compounds and lignin on the surface of the catalyst. This
Interestingly, the catalysts with supports sulfonated in 10 M
H
2
SO
acid. Though the amount of strong acid sites in both
RuO /Cconc3h were similar, the latter has lower
recyclability and should contain more sulfonic acid groups. The
lower recyclability of the RuO /Cconc3h could be explained with
4
performed better than those sulfonated in concentrated
x
/C10M3h and RuO
x
4
1
x
faster loss of the strong acid sites, as the sulfonic acid groups
are hydrolysed more easily than the carboxylic acid groups
35
under hydrothermal conditions. In this respect, the preferred
strong acid sites would be carboxylic acids over sulfonic acids.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 5
Please do not adjust margins

Pleas eG dr eo e nn o Ct ha de jmu si st tmr yargins
Page 6 of 9
ARTICLE
Journal Name
Further investigation of the catalyst was carried out using X-ray
photoelectron spectroscopy (XPS). Despite the overlap of the
C(1s) and Ru(3d) core levels inhibiting the detailed studies, the
XPS analysis supports the titration results regarding the impact
of the catalyst treatments on surface acid functionalities.
According to both C(1s) and O(1s) spectra, increased relative
amounts of species corresponding to carboxylic acid
View Article Online
DOI: 10.1039/C7GC01821E
x
functionalities were detected on C10M3h and RuO /C10M3h
compared to the reduced Ru/C10M3h (See Figures S15-S17 and
Tables S7-S8 in ESI). The XPS and titration results are in
agreement with the previous reports stating that oxidative
treatment of carbon increases the amount of carboxylic acid
groups on carbon support and consequently increasing the
34,42
hydrophilicity.
Ru(3d) signals in the XPS spectra of both
reduced and oxidized catalyst can be assigned to an oxidized Ru
species on the surface of the particles, presumably due to
handling of the samples in air (See Figures S15 & S16 in ESI).43
Figure 3: TEM-images of catalysts A) fresh RuO /C
x
10M30min
catalyst with mean
particle size of 2.48 nm and B) four times used RuO
x
/C10M30min catalyst with mean
To further elucidate the effect of treatment of the catalysts, the particle size of 4.84 nm C) fresh commercial Ru/C catalyst with mean particle size
of 2.66 nm D) five times used commercial Ru/C with mean particle size of 2.98.
two catalyst series, reduced (Ru/Csulfonated) and oxidized
(
RuO
x
/Csulfonated), were studied by powder X-ray diffraction
deactivation of the hydrogenation catalyst was fouling or
poisoning of the catalysts, which seems to be slowed down by
strong acid sites on the surface of the catalyst.
(XRD). In both cases, reflections corresponding to metallic Ru
were found, and with the oxidized catalysts low intensity signals
of RuO were identified (See Figures S11 & S12 in ESI). This
2
suggests that despite the oxidative treatment most of the
particles were metallic, oxidation occurring mostly on the
surface of the Ru-particle. However, comparison of XRD
x
Finally, the best developed RuO /C10M3h catalyst was taken into
the optimized two-step isosorbide synthesis to compare the
overall efficiency of the one- and two-step setups catalysed by
the commercial Ru/C (Tables 1, 3 and 5). The isosorbide yields
patterns of fresh and four times used RuO
x
/C10M30min revealed
more intensive peaks corresponding to RuO
2
in the recycled
x
using cellulosic substrates with RuO /C10M3h were 56-57 mol%
catalyst and build-up of an unknown substance (See Figure S13
in ESI). The hydrogenation activity of the catalyst decreases only
slightly upon recycling, and indeed, bulk RuO was found active
2
in the hydrogenation reaction (See Table S10, entry 47 in ESI).
Furthermore, the two times used commercial Ru/C composed
mainly metallic Ru, according to XRD (See Figure S14 in ESI).
Since this catalyst was less active than the partially oxidized
along with quantitative conversion, the molar yields are equal
to 50 w% of MCC and DP and 49 w% of BP (lignin was included
in the substrate weight). In comparison, the commercial Ru/C
gave isosorbide in the yields of 52-54% with the two-step setup.
The use of pure glucose as a substrate with RuO /C10M3h
x
furnished isosorbide in good 65 mol% yield, which was slightly
higher than the yield with the commercial Ru/C catalyst (Tables
RuO
was the cause of catalyst deactivation. In fact, mild oxidation of
Ru/C to Ru/RuO /C was previously reported to facilitate higher
activity and selectivity in transfer hydrogenation than either
x
/C10M30min, it was highly unlikely that the oxidation of Ru
1
x
and 3). The elevated yields with the recyclable RuO /C10M3h
can be attributed to more selective conversion of cellulose and
glucose to sorbitol and 1,4-sorbitan in the first step.
x
44,45
phase alone.
Table 5: Highest isosorbide yields with different substrates under optimal two-step
reaction conditions using the best RuO
x
/C10M3h hydrogenation catalyst.a
Transmission electron microscope (TEM) was used to
investigate the growth of the Ru-particle size upon recycling.
Mean particle size of the fresh commercial Ru/C catalyst was
Entry
Substrate
Temperature
°C)
170+200
170+200
170+200
170+200
Time (h)
Isosorbide yield
d
(
(%)
BPb
1
2
3
4
2+6
2+6
2+6
2+6
57
56
56
65
2
.66 nm and after five runs under the optimal first step
MCC
conditions with MCC as a substrate the average particle size had
grown to 2.98 nm (Figure 3, See Table S6 in ESI). With the
DP
Glucose
c
RuO
x
/C10M30min the mean particle size grew markedly from 2.48
a
1
.18 mmol of cellulose, 0.5 M H
70 °C, 20 bar H , 2 h and 0.02 mmol of second run RuO
/C10M3h was collected via filtration and the second step was
2
SO
4
(aq) 30 ml; the first step reaction conditions:
nm to 4.87 nm after four runs in the optimized first step
conditions but the catalytic activity decreased only slightly reaction steps the RuO
1
2
x
/C10M3h. Between the
x
when BP was used as a substrate. With the lignin free substrate, conducted with the filtrate. The second step reaction conditions: 200 °C, 40 bar H
2
,
b
6
h; Bagasse pulp (95.37 w% cellulose and 4.63 w% lignin) obtained from Stora
MCC, similar particle growth was observed, but catalytic activity
remained the same. Evidently, sintering of metal particles
cannot explain differences of the reaction outcome between
different substrates under identical reaction conditions. Our
analysis strongly indicates that the main reason causing the
Enso, used as received and the amount of the substrate was calculated to match
c
1
.18 mmol of cellulose; Dissolving pulp obtained from Stora Enso, used as
d
received; The yields were obtained with HPLC-RID using calibration curves.
6
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 7 of 9
Pleas eG dr eo e nn o Ct ha de jmu si st tmr yargins
Journal Name
ARTICLE
class insert was then placed into an autoclave whicVhiewwAartsiccleloOsnleinde
Conclusions
and pressurised with N
2
. The slurry w
D
asOI:
h
1
e
0
a
.1
t
0
e
3
d
9/C
a
7
t
GC
2
0
0
1
0
821E
°
C
As shown herein, modification of Ru/C hydrogenation catalyst
surface with sulfonation and oxidative treatment has significant
effect on the catalyst properties in the isosorbide synthesis.
With the increase of strong acid sites on the catalyst surface,
the activity and recyclability of the catalyst can be improved
when using lignin containing substrates. Consequently, we have
developed new efficient RuO /Csulfonated catalysts that maintain
x
high catalytic activity at least for four consecutive runs with the
cellulosic substrates, even with lignin containing substrate.
overnight under the nitrogen atmosphere. After cooling the
autoclave, the slurry was once more filtered and washed with
water until the pH of the filtrate was neutral. In the end the
sulfonated carbon was dried in an oven.
Impregnation of ruthenium: Ruthenium was impregnated on
the sulfonated carbon supports in the way described by
Komanoya et al.22 Desired amount of sulfonated carbon was
Using the optimized reaction conditions efficient two-step weighted and mixed into 20 ml of milli-q water. Appropriate
synthesis setup and the best catalyst RuO
/C10M3h, isosorbide amount of RuCl •xH O was diluted into milli-q water, which was
x
3
2
was generated in high yields of 56-57% (49-50 w%) from then added dropwise into the sulfonated carbon slurry. The
different cellulosic substrates. According to our catalyst studies, mixture was stirred vigorously overnight. After stirring water
deactivation of the Ru-catalyst with lignin containing substrates was evaporated using rotary evaporator, and the forming
seems to occur through fouling and poisoning. The sulfonation
and oxidative treatments described above generate strong acid
sites on the surface of the catalyst support thus increasing its
hydrophilicity simultaneously reducing the interactions with
residual lignin. In general, the acidifying surface treatments
provide an intriguing approach for catalyst modification aimed
at transformation of lignin containing biomass.
precatalyst was dried in vacuo. The precatalyst was then placed
into an oven and heated at 400 °C under forming gas
atmosphere (5% H / 95% N ), forming metallic ruthenium
2 2
nanoparticles, which after cooling would be ready to use. When
the ruthenium nanoparticles were treated with air, instead of
cooling to room temperature, the oven was cooled to 290 °C
and the atmosphere was changed to air. Then the air treated
catalyst was cooled to room temperature under air.
Experimental
General experimental setup
Acknowledgements
Desired amount of substrate and catalyst (if recycled the
catalyst was ground to powder before use) were weighed and
placed into a class insert with magnetic stirrer. After this the
aqueous solution was added. The class inserts were then placed
into the autoclaves (Roth 30S), which were purged three times
with nitrogen, after which the autoclave was filled with desired
hydrogen pressure. The reaction mixture was then heated at
chosen temperature for chosen time, followed by cooling of the
autoclave and pressure release. The catalyst was then filtered
from the reaction solution and dried in an oven for the possible
next use. The filtrate could be used in the second step reaction
Stora Enso is gratefully acknowledged for providing the funding
and materials (Bagasse pulp and dissolving pulp) for this study.
Special thanks to Dr. Jari Kavakka and Dr. Piia Keskiväli-Bond for
their ideas and guidance. Also, this work made use of the Aalto
University Nanomicroscopy Center (Aalto-NMC) premises.
Notes and references
‡
1
From now on the 1,4-sorbitan and 3,6-sorbitan are referred as
,4-sorbitan in the text. Only commercial standard of 1,4-sorbitan
(setup similar to the above mentioned) or analysed with HPLC- was available at the time when the investigation was conducted.
RID. For the analysis, the filtrate was diluted to 50 ml with milli-
q water, from this a 1 ml aliquot was taken and diluted to 10 ml
with milli-q water. 2 ml of this solution was then filtered
through ion exchange resin and analysed with HPLC-RID. Citric
acid was used as an internal standard.
1
A. J. Ragauskas, C. K. Williams, B. H. Davison, G. Britovsek,
J. Cairney, C. A. Eckert, W. J. Frederick Jr, J. P. Hallett, D. J.
Leak and C. L. Liotta, Science, 2006, 311, 484–489.
G. W. Huber, J. Chheda, C. B. Barrett and J. A. Dumesic,
Science, 2005, 308, 1446–1450.
J.-P. Lange, E. van der Heide, J. van Buijtenen and R. Price,
ChemSusChem, 2012, 5, 150–166.
S. G. Wettstein, D. Martin Alonso, E. I. Gürbüz and J. a.
Dumesic, Curr. Opin. Chem. Eng., 2012, 1, 218–224.
S. Van de Vyver, J. Geboers, P. A. Jacobs and B. F. Sels,
ChemCatChem, 2011, 3, 82–94.
G. Liang, C. Wu, L. He, J. Ming, H. Cheng, L. Zhuo and F.
Zhao, Green Chem., 2011, 13, 839.
P. Barbaro, F. Liguori and C. Moreno-Marrodan, Green
Chem., 2016, 18, 2935–2940.
Y. Nakagawa and K. Tomishige, Catal. Today, 2012, 195,
136–143.
2
3
4
5
6
7
8
General method of the catalyst synthesis
Sulfonation: The sulfonated carbon was produced in the way
described by Foo et al.32 Desired amount of support material
(AC Norit SX, Sigma Aldrich) was weighted and placed into a
flask followed by addition of desired amount of sulphuric acid.
The slurry was mixed for desired time at chosen temperature.
Next, the slurry was filtered and the carbon material was
washed with water until the pH of the filtrate was neutral.
Washed carbon material was then dried in an oven over-night.
Next, the carbon material was washed using Roth S30, by
placing the carbon material into a class insert with water. The
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 7
Please do not adjust margins

Pleas eG dr eo e nn o Ct ha de jmu si st tmr yargins
Page 8 of 9
ARTICLE
Journal Name
9
Y. Nakagawa, M. Tamura and K. Tomishige, ACS Catal.,
013, 3, 2655–2668.
I. Ahmed, N. A. Khan, D. K. Mishra, J. S. Lee, J. S. Hwang
and S. H. Jhung, Chem. Eng. Sci., 2013, 93, 91–95.
A. Yamaguchi, N. Hiyoshi, O. Sato and M. Shirai, Green
Chem., 2011, 13, 873.
F. Fenouillot, A. Rousseau, G. Colomines, R. Saint-Loup and
J. P. Pascault, Prog. Polym. Sci., 2010, 35, 578–622.
Y. Zhu, M. Durand, V. Molinier and J.-M. Aubry, Green
Chem., 2008, 10, 532–540.
P. Tundo, F. Aricò, G. Gauthier, L. Rossi, A. E. Rosamilia, H.
S. Bevinakatti, R. L. Sievert and C. P. Newman,
ChemSusChem, 2010, 3, 566–570.
36
37
T. G. Ros, A. J. van Dillen, J. W. Geus and D. C.View Article Online
Koningsberger, Chemistry, 2002, 8, 1151–1162.
DOI: 10.1039/C7GC01821E
2
10
11
12
13
14
C. Montassier, J. C. Ménézo, L. C. Hoang, C. Renaud and J.
Barbier, J. Mol. Catal., 1991, 70, 99–110.
B. J. Arena, Appl. Catal. A Gen., 1992, 87, 219–229.
K. Nakajima, M. Okamura, J. N. Kondo, K. Domen, T.
Tatsumi, S. Hayashi and M. Hara, Chem. Mater., 2009, 21,
186–193.
A. P. Terzyk, J. Colloid Interface Sci., 2003, 268, 301–329.
M. Yabushita, H. Kobayashi, J. Y. Hasegawa, K. Hara and A.
Fukuoka, ChemSusChem, 2014, 7, 1443–1450.
H. Kobayashi, H. Kaiki, A. Shrotri, K. Techikawara and A.
Fukuoka, Chem. Sci., 2016, 7, 692–696.
38
39
40
41
42
1
5
B. Op De Beeck, J. Geboers, S. Van De Vyver, J. Van Lishout, 43
J. Snelders, W. J. J. Huijgen, C. M. Courtin, P. A. Jacobs and
B. F. Sels, ChemSusChem, 2013, 6, 199–208.
E. A. Paoli, F. Masini, R. Frydendal, D. Deiana, C. Schlaup,
M. Malizia, T. W. Hansen, S. Horch, I. E. L. Stephens and I.
Chorkendorff, Chem. Sci., 2015, 6, 190–196.
J. Jae, W. Zheng, A. M. Karim, W. Guo, R. F. Lobo and D. G.
Vlachos, ChemCatChem, 2014, 6, 848–856.
16
17
18
19
R. W. Torget, J. S. Kim and Y. Y. Lee, Ind. Eng. Chem. Res.,
000, 39, 2817–2825.
44
2
A. Yamaguchi, O. Sato, N. Mimura and M. Shirai, Catal.
Commun., 2015, 67, 59–63.
45
P. Panagiotopoulou and D. G. Vlachos, Appl. Catal. A Gen.,
2014, 480, 17–24.
P. Sun, X. Long, H. He, C. Xia and F. Li, ChemSusChem, 2013,
6, 2190–2197.
J. Geboers, S. Van de Vyver, K. Carpentier, K. de Blochouse,
P. Jacobs and B. Sels, Chem. Commun. (Camb)., 2010, 46,
3577–9.
2
2
2
2
2
0
1
2
3
4
R. Palkovits, K. Tajvidi, J. Procelewska, R. Rinaldi and A.
Ruppert, Green Chem., 2010, 12, 972.
B. Kusserow, S. Schimpf and P. Claus, Adv. Synth. Catal.,
2
003, 345, 289–299.
T. Komanoya, H. Kobayashi, K. Hara, W. J. Chun and A.
Fukuoka, Appl. Catal. A Gen., 2011, 407, 188–194.
E. Crezee, B. W. Hoffer, R. J. Berger, M. Makkee, F. Kapteijn
and J. a. Moulijn, Appl. Catal. A Gen., 2003, 251, 1–17.
J. M. A. R. Almeida, L. Da Vià, P. Demma Carà, Y. Carvalho,
P. N. Romano, J. A. O. Peña, L. Smith, E. F. Sousa-Aguiar
and J. A. Lopez-Sanchez, Catal. Today, 2017, 279, 187–193.
M. Rose and R. Palkovits, ChemSusChem, 2012, 5, 167–176.
C. H. Bartholomew, Appl. Catal. A Gen., 2001, 212, 17–60.
I. Sádaba, M. López Granados, A. Riisager and E. Taarning,
Green Chem., 2015, 4133–4145.
25
26
27
28
29
30
31
32
J. Xi, Y. Zhang, D. Ding, Q. Xia, J. Wang, X. Liu, G. Lu and Y.
Wang, Appl. Catal. A Gen., 2014, 469, 108–115.
J. Jow, G. L. Rorrer, M. C. Hawley and D. T. A. Lamport,
Biomass, 1987, 14, 185–194.
W. Weiqi and W. Shubin, Chem. Eng. J., 2017, 307, 389–
398.
T. S. Hansen, J. M. Woodley and A. Riisager, Carbohydr.
Res., 2009, 344, 2568–2572.
G. S. Foo, A. H. Van Pelt, D. Krꢀtschel, B. F. Sauk, A. K.
Rogers, C. R. Jolly, M. M. Yung and C. Sievers, ACS Sustain.
Chem. Eng., 2015, 3, 1934–1942.
3
3
3
4
H. P. Boehm, Carbon N. Y., 1994, 32, 759–769.
M. L. Toebes, F. F. Prinsloo, J. H. Bitter, A. J. Van Dillen and
K. P. De Jong, J. Catal., 2003, 214, 78–87.
35
A. H. Van Pelt, O. A. Simakova, S. M. Schimming, J. L.
Ewbank, G. S. Foo, E. A. Pidko, E. J. M. Hensen and C.
Sievers, Carbon N. Y., 2014, 77, 143–154.
8
| J. Name., 2012, 00, 1-3
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 9 of 9
Green Chemistry
View Article Online
DOI: 10.1039/C7GC01821E
A recyclable Ru-based hydrogenation catalyst for the two-step conversion lignocellulosic substrate into
isosorbide.