A One-Step Synthesis of C6 Sugar Alcohols from Levoglucosan and Disaccharides Using a Ru/CMK-3 Catalyst

Article abstract of DOI:10.1021/acscatal.6b00296

Sorbitol is an important commercially available chemical with a broad application range and is typically made by the catalytic hydrogenation of glucose. Here we report a high-yield synthesis of sorbitol from levoglucosan (1,6-anhydro-β-d-glucopyranose) and cellobiose, two sugars present in pyrolysis liquids, using a mesoporous carbon-supported Ru catalyst (Ru/CMK-3). The hydrogenation reactions were performed in a batch autoclave setup under a hydrogen pressure of 50 bar and temperatures ranging from 120 to 180 °C in water. The hydrogenation of levoglucosan gave essentially quantitative yields of sugar alcohols, composed of 96.2 wt % of sorbitol and 3.8 wt % of mannitol (180 °C, 5 h). Ru/CMK-3 shows superior catalytic performance compared to a commercial Ru/C catalyst. A reaction pathway involving glucose as an intermediate and subsequent (hydrogenolysis) reactions of the desired sorbitol is proposed. Reactions with glucose and sorbitol were performed to define the reaction pathways and to highlight the differences between Ru/C and Ru/CMK-3. Disaccharides including cellobiose and sucrose were also tested, yielding up to 95 wt % of C6 sugar alcohols at 180 °C in 5 h for both substrates. Detailed catalyst characterization studies (N₂ physisorption, TEM, XRD, NH₃-TPD, H₂-TPD) revealed that Ru/CMK-3 contains considerable amounts of strong acid sites (NH₃-TPD). Catalyst stability was tested by catalyst recycling experiments using levoglucosan in batch. After three successive runs, the rate of the hydrolysis reaction of LG to glucose was about constant, though the subsequent hydrogenation reaction to sorbitol/mannitol was slightly retarded as evidenced from a slight increase in the remaining amounts of glucose at the end of reaction.

Full text of DOI:10.1021/acscatal.6b00296

Subscriber access provided by The Library | University of Bath
Article
A One Step Synthesis of C6 Sugar Alcohols from
Levoglucosan and Disaccharides using a Ru/CMK-3 Catalyst
Wang Yin, Zhenchen Tang, Robertus Hendrikus Venderbosch, Zheng Zhang,
Catia Cannilla, Giuseppe Bonura, Francesco Frusteri, and Hero Jan Heeres
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.6b00296 • Publication Date (Web): 26 May 2016
Downloaded from http://pubs.acs.org on June 4, 2016
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a free service to the research community to expedite the
dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts
appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been
fully peer reviewed, but should not be considered the official version of record. They are accessible to all
readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered
to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published
in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just
Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor
changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers
and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors
or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
ACS Catalysis is published by the American Chemical Society. 1155 Sixteenth Street
N.W., Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 34
ACS Catalysis
1
2
3
4
5
6
A One Step Synthesis Of C6 Sugar Alcohols From Levoglucosan And
Disaccharides Using A Ru/CMK-3 Catalyst
7
8
9
Wang Yin1, Zhenchen Tang1, Robertus Hendrikus Venderbosch2, Zheng Zhang1, Catia
Cannilla3, Giuseppe Bonura3, Francesco Frusteri3, Hero Jan Heeres*1
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
Department of Chemical Engineering, University of Groningen, Nijenborgh 4, 9747 AG,
Groningen, The Netherlands
2 Biomass Technology Group BV, Josink Esweg 34, 7545 PN Enschede, The Netherlands
3 CNR-ITAE, Istituto di Tecnologie Avanzate per l’Energia “Nicola Giordano”, Via S. Lucia
sopra Contesse, 5-98126 Messina, Italy.
(*) corresponding author: h.j.heeres@rug.nl, tel.: +31 503634484; fax: +31 503634479.
1
ACS Paragon Plus Environment

ACS Catalysis
Page 2 of 34
1
2
3
Abstract
4
5
6
7
8
Sorbitol is an important commercially available chemical with a broad application range and
is typically made by the catalytic hydrogenation of glucose. Here we report a high yield
synthesis of sorbitol from levoglucosan (1,6-anhydro-β-D-glucopyranose) and cellobiose, two
sugars present in pyrolysis liquids, using a mesoporous carbon supported Ru catalyst
(Ru/CMK-3). The hydrogenation reactions were performed in a batch autoclave set-up under
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
o
a hydrogen pressure of 50 bar and temperatures ranging from 120 to 180 C in water. The
hydrogenation of levoglucosan gave essentially quantitative yields of sugar alcohols,
composing of 96.2 wt.% of sorbitol and 3.8 wt.% of mannitol (180 oC, 5 h). Ru/CMK-3 shows
superior catalytic performance compared to a commercial Ru/C catalyst. A reaction pathway
involving glucose as an intermediate and subsequent (hydrogenolysis) reactions of the
desired sorbitol is proposed. Reactions with glucose and sorbitol were performed to define
the reaction pathways and to highlight the differences between Ru/C and Ru/CMK-3.
Disaccharides including cellobiose and sucrose were also tested, yielding up to 95 wt.% of C6
sugar alcohols at 180 oC in 5 h for both substrates. Detailed catalyst characterization studies
(N2 physisorption, TEM, XRD, NH3-TPD, H2-TPD) revealed that Ru/CMK-3 contains
considerable amounts of strong acid sites (NH3-TPD). Catalyst stability was tested by catalyst
recycling experiments using levoglucosan in batch. After 3 successive runs, the rate of the
hydrolysis reaction of LG to glucose was about constant, though the subsequent
hydrogenation reaction to sorbitol/mannitol was slightly retarded as evidenced from a slight
increase in the remaining amounts of glucose at the end of reaction.
Keywords: levoglucosan, pyrolysis liquids, sorbitol, Ru/CMK-3, hydrogenation
2
ACS Paragon Plus Environment

Page 3 of 34
ACS Catalysis
1
2
3
1.Introduction
4
5
6
7
8
Sorbitol is listed as one of the top 12 bio-based building blocks by the U.S. Department of
Energy1. It is widely used as an additive in many industrial products, particularly in the food,
cosmetic, and paper industry, but also as a building block for the synthesis of various fine
chemicals including vitamin C2. The global production of sorbitol was 700,000 t/a in 20073 to
over 1,000,000 t/a in 20094. The overall capacity of sorbitol was 1,700,000 t/a in 2011 with
an estimated growth rate of approximately 30 % per year5.
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
In the last decade, sorbitol has also been proposed as a starting material for biofuels and
hydrogen production using aqueous phase reforming6. Another example is the conversion of
sorbitol to isosorbide, an interesting biobased diol for the synthesis of polyesters (e.g. PET)
and a range of other polymers1. The synthesis of isosorbide from sorbitol is typically carried
out at temperatures between 160−180 °C using strong mineral or organic Brønsted acids
such as sulfuric acid or p-toluenesulfonic acid in aqueous solutions.7
Nowadays, the production of sorbitol is solely accomplished by the catalytic hydrogenation
of glucose8-15. The latter is obtained by hydrolysis of various starches like corn, rice, potato,
and so forth. However, there is a strong incentive to use non-food glucose precursors for the
production of sorbitol. An example is the use of cellulose, which can be converted to glucose
(and subsequently to sorbitol) by a number of technologies.16-29 Disadvantageously, yields
are not on par and major breakthroughs are required before large scale commercial
operation become economically viable.
Fast pyrolysis30 is an alternative method for the depolymerization of lignocellulosic biomass
and offers good prospective for the production of inexpensive low molecular weight sugars.
It involves heating a solid biomass in the absence of oxygen, at atmospheric pressure and
temperatures ranging from 300 to 600°C to yield a condensable liquid product referred to as
pyrolysis liquids with yields up to 70 wt.%. Fast pyrolysis liquids contain considerable
amounts of monomeric and oligomeric (anhydro)-sugars such as levoglucosan (the most
abundant one), cellobiosan and glucose. The exact amounts are depending on the reactor
configuration, feed, and process conditions, but are as high as 20.5 wt.% on biomass feed for
red oak, see Table 1 for details. By using isolated cellulose as the feed, levoglucosan is the
major product with yields up to 60 wt.%.31, 32
3
ACS Paragon Plus Environment

ACS Catalysis
Page 4 of 34
1
2
3
4
5
6
Table 1 Overview of reported amounts of low molecular weight sugars in pyrolysis liquids and sugar fractions derived thereof
7
8
9
Entry Feed
Reactor Type
Fluidized bed reactor
Analytical method
GC/MS, HPLC
Sugar Content
Ref.
33
1
2
Red oak
20.62 ±2.52 wt.% (levoglucosan,
cellobiosan, glucose, etc) in pyrolysis liquids
based on biomass feed
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
34
Birch wood
sawdust
CDS Pyroprobe 100 combined
with a gas chromatograph (CP
9000)
GC/MS
15.3 wt.% of levoglucosan in pyrolysis
liquids based on dry biomass feed
35
36
3
4
Pine wood
Red oak
Auger reactor
GC/MS
HPLC
16.43±0.71 wt.% of levoglucosan in organic
fraction of the pyrolysis liquids
Free fall reactor
8 wt.% (levoglucosan, cellobiosan, xylose,
etc) in pyrolysis liquids based on feed
6 wt.% levoglucosan in pyrolysis liquids
23.45 wt.% of levoglucosan in pyrolysis
liquids based on biomass feed
37
38
5
6
Mallee wood Fluidized bed reactor
GC–MS
GC-MS
Sugarcane
bagasse
Switch grass
A CDS 5200 micropyrolyzer
coupled to a GC-MS
39
7
A Frontier Lab Double Shot
Micropyrolyzer 2020iS coupled
to a Varian 450 GC
GC-MS
15 wt.% of levoglucosan in pyrolysis liquids
based on biomass feed
40
41
42
8
9
Red oak
Twin-screw auger reactor
HPLC
16 wt.% of xylose and glucose hydrolyzable
sugars in the pyrolysis liquids
35 wt.% of levoglucosan in condenser 1
Pine wood
and bagasse
Pine wood
Fluidized bed reactor
Transport bed reactor
GC-MS
10
Brix method
35 wt.% ether-insoluble sugar fraction in
pine derived pyrolysis liquids
4
ACS Paragon Plus Environment

Page 5 of 34
ACS Catalysis
1
2
3
4
5
6
7
8
9
Pyrolytic sugars, isolated fractions of pyrolysis liquids enriched in sugars, can be obtained
using staged condensation of the vapors formed during fast pyrolysis43 (Table 1, Entry 1, 8, 9)
or solvent fractionation of isolated pyrolysis liquids44. In the latter case, water is added to
pyrolysis liquids and this results in a water soluble and water insoluble phase. The water
soluble phase is comprised of sugars, organic acids, small aldehydes, ketones and water
soluble lignin monomers, which can be separated using solvent extraction schemes (Table 1,
Entry 10).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The application of pyrolytic sugars as the precursors for bio-ethanol by hydrolysis (for
example using sulfuric acid) followed by fermentation has been studied before45-49. Besides,
their conversion to platform chemicals such as 5-HMF, levulinic acid and levulinic esters in
various solvents has also been investigated50, 51. Another example involves the use of the
aqueous sugar fraction of pyrolysis liquids for the synthesis of hydrogen, alkanes and polyols
by aqueous phase reforming52. The sugars were readily hydrogenated, however levoglucosan
o
showed a very low rate of hydrogenation at 125 C using Ru/C as the catalyst. At elevated
o
temperatures (170 C in 9 h), 28 mol.% of sorbitol (based on levoglucosan and sugars) was
formed (68.9 bar H2). Sanna et al.53 recently reported the hydrogenation of aqueous
o
fractions of pyrolysis liquids with Ru/C at 125 C under 50 bar (H2 flow rate 150 ml/min) in a
continuous set-up. The sorbitol yield was about 75 mol.% (based on levoglucosan and
glucose), whereas substantial amounts of C1-C4 alcohols were formed as well. Li et al.54
reported the electrocatalytic hydrogenation of the aqueous phase of pyrolysis liquids using
Ru supported on activated carbon cloth at 27 oC under atmospheric pressure and found that
alcohols and glycols are the main products after 6.5 h reaction time. However, information
regarding the sorbitol yield was not reported. Thus, sorbitol formation from the sugar
fractions of pyrolysis liquids is feasible though the yield is still far from quantitative.
In this paper, we report (model component) studies aiming at the efficient conversion of
levoglucosan and other low molecular weight sugars present in pyrolysis liquids to sugar
alcohols and particularly sorbitol (Scheme 1), preferably in a single pot reaction.
5
ACS Paragon Plus Environment

ACS Catalysis
Page 6 of 34
1
2
3
4
5
6
7
8
9
Scheme 1 Proposed strategy for conversion of levoglucosan to sorbitol.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
For an efficient conversion of levoglucosan to sorbitol, two reactions are of relevance, viz. i)
the hydrolysis of levolglucosan to glucose, followed by the hydrogenation of glucose to
sorbitol (Scheme 1). Hydrogenolysis and the formation of smaller diols such as glycerol and
propylene glycol from glucose must be avoided, as well as subsequent acid catalyzed
conversions of glucose to 5-HMF and levulinic acid. Proof of concept for the hydrolysis of
levoglucosan is available in the literature using sulfuric acid as a Brønsted acid catalyst45, 46
.
Both supported Ru (e.g. Ru/C) and Ni catalysts (e.g. Raney Ni) are known catalysts for
glucose hydrogenation13, 14. However, the use of inorganic acids for the hydrolysis reaction is
not preferred as these are highly corrosive and neutralization is required to dispose of the
acid after reaction. Thus, it is imperative to find catalysts for one step conversion of
levoglucosan to sorbitol without the use of a homogeneous acid.
So far, only one study on the hydrogenation of levoglucosan has been reported. Bindwal et
al.55 reported that ethylene glycol (4.7 wt.%), 1,2-propanediol (28.6 wt.%) and 1,4-butanediol
(1.8 wt.%) are the main products for the hydrogenation of levoglucosan in water (140 oC, 20
bar H2 pressure, 2 h batch time) using a conventional Ru/C catalyst. Sorbitol was not
detected, indicating that hydrogenolysis occurs to a large extent. A reaction pathway was
proposed involving initial hydrolysis of levoglucosan to glucose, hydrogenation to sorbitol
and subsequent hydrogenolysis reactions of sorbitol to smaller diols. The reactions were
carried out in the absence of a Brønsted acid, which may affect the product portfolio to a
significant extent.
We anticipated that the use of supported bifunctional catalysts which possess acid sites for
the hydrolysis reaction and an active (noble) metal for the hydrogenation reaction could be
advantageous to achieve high C6 sugar alcohol yields from levoglucosan. Such supported
bifunctional catalysts have also been applied for the conversion of cellulose to sugar alcohols.
In this case, cellulose first undergoes hydrolysis to give oligomers and ultimately glucose
over acidic sites, and the glucose is subsequently hydrogenated over metal catalysts to
6
ACS Paragon Plus Environment

Page 7 of 34
ACS Catalysis
1
2
3
4
5
sorbitol and mannitol.23, 28, 56 As such, there is a strong analogy with the proposed route for
sorbitol from levoglucosan.
6
7
8
9
Here, we report studies on the hydrogenation of levoglucosan in water to C6 sugar alcohols
and particularly sorbitol using a bifunctional Ru catalyst on a mesoporous carbon support
(CMK-3). This support is characterized by a large surface area and a highly oxygen-
functionalized surface containing acid sites57. For instance, Ru/CMK-3 (in the absence of
hydrogen) was reported to be a water-tolerant and reusable catalyst for the hydrolysis of
cellulose58 to low molecular weight sugars, indicating the presence of acid sites. For
comparison, hydrogenation of levoglucosan using a commercial Ru/C catalyst was also
carried out. The proposed reaction pathway was confirmed by performing catalytic
hydrogenation experiments with glucose and sorbitol using both catalysts to identify the
origin of the superior performance of Ru/CMK-3. In addition, the scope of the catalytic
reaction was investigated by performing reactions with disaccharides like cellobiose and
sucrose. Various techniques (nitrogen physisorption, X-ray diffraction (XRD), transmission
electron microscopy (TEM), NH3-temperature programmed desorption (NH3-TPD) and H2-
temperature programmed desorption (H2-TPD)) were used to gain insights in the texture and
acidity of the Ru/CMK-3 catalyst. Finally, catalyst stability was investigated by performing a
number of batch experiments with recycled catalyst.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2. Experimental procedures
2.1 Materials and methods
D-glucose (≥ 99.5 %), D-sorbitol (99 %), D-mannitol (≥ 98 %), glycerol (≥ 99.5 %), 1,2-
propanediol (≥ 99.5 %), 2-propanol (anhydrous, 99.5 %), sucrose (≥ 99.5 %), RuCl3 hydrate
(Ruthenium content, 40.0 - 49.0 %) were purchased from Sigma-Aldrich and levoglucosan
(1,6-anhydro-β-D-glucopyranose), cellobiose (both from from Carbosynth, UK) were used as
received. Polyvinylalcohol (PVA, 1 wt.% aqueous solution, Mw= 10000, 80 % hydrolyzed) was
obtained from Aldrich. Ordered mesoporous carbon (CMK-3) with a BET surface area around
1000 m2/g, was supplied by ACS material, LLC. The elemental composition was determined
by elemental analysis (C, 90.66 wt.%; H, 0.55 wt.%; O, 8.79 wt.%). Ru/C (5 wt.% Ruthenium
loading) was obtained from KaiDa Technology Limited, UK, and was crushed and sieved to
25-75 µm before use. TEM images (Supplementary information, Figure S1) show the
7
ACS Paragon Plus Environment

ACS Catalysis
Page 8 of 34
1
2
3
4
5
6
7
8
9
presence of Ru particles, that are not homogeneously distributed on the carbon surface. The
particle sizes range from 20 to 50 nm. The particles are mainly in spherical form; however
even Ru clusters with different shapes are present. Syringe filters (pore size 0.2 µm,
diameter 13 mm) were purchased from VWR international.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
2.2 Preparation of Ru/CMK-3
Ru/CMK-3 (2 wt.% Ru on support), was prepared according to a published procedure59. RuCl3
hydrate (45 mg) and a PVA solution (PVA to Ru weight ratio of 1 to 1.2) were dissolved in
water (30 mL) and the mixture was diluted to 40 ml and stirred for 30 min at room
temperature. 8 ml of a NaBH4 solution in water (0.1 M, NaBH4 to Ru molar ratio of 5) was
added to the mixture and further stirred for 30 min, then acidified to pH 1 by the addition of
a solution of sulfuric acid in water (1 M). Subsequently, the appropriate amount of CMK-3 to
achieve a metal loading of 2 wt.% was added under vigorous stirring conditions. After 2 h,
the slurry was filtered and the catalyst was washed thoroughly with distilled water and dried
at 100 oC overnight.
2.3 Catalyst Characterization
o
Gas physisorption: Nitrogen physisorption analyses (-196.2 C) were carried out in a
o
Micromeritics ASAP 2020. The samples were degassed in vacuum at 350 C for 10 h. The
surface area was calculated using the standard BET and Langmuir methods.
X-Ray diffraction (XRD): XRD analysis was performed using a Philips X-Pert diffractometer
equipped with a Ni β-filtered Cu-Kα radiation (λ= 1.5406 Å) at 40 kV and 30 mA. Low angle
data were collected over 2θ angles from 0.8-5, with a step size of 0.005° at a scan speed of
0.0025°·s-1. Then, data were also collected over a 2θ angles from 5–80°, with a step size of
0.04° at a speed of 0.013°·s-1. Identification of the diffraction peaks was performed on the
basis of the JCPDS database of reference compounds.
Transmission Electron Microscopy (TEM): A Philips CM12 instrument equipped with a high-
resolution camera was used to acquire and elaborate TEM images. Powdered samples were
dispersed in 2-propanol under ultrasound irradiation and the resulting suspension put drop-
wise on a holey carbon-coated support grid.
8
ACS Paragon Plus Environment

Page 9 of 34
ACS Catalysis
1
2
3
4
5
6
7
8
9
NH3 Temperature Programmed Desorption (NH3-TPD): Surface concentration of acidic sites
was determined by using a conventional flow apparatus operating both in continuous and
pulse mode. Before TPD experiments, the catalyst samples (∼50 mg) were reduced, at
atmospheric pressure, by flowing hydrogen (60 STP ml/min) in a linear quartz micro-reactor
(l, 200 mm; i.d., 4 mm) from room temperature to 350 °C at a heating rate of 10 °C/min.
Then, the samples were saturated for 30 min in flow of a gas mixture containing 5 vol.% of
NH3/He at 150°C (flow rate of 25 ml/min). After purging in helium flow until a constant
baseline level was attained, TPD measurements were performed in the temperature range
100–800 °C at a rate of 10 °C/min using helium (25 STP ml/min) as carrier flow. The evolved
ammonia was detected by an on-line thermal-conductivity detector, calibrated by the peak
area of known pulses of NH3.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
H2 Temperature Programmed Desorption (H2-TPD): Hydrogen temperature programmed
desorption (H2-TPD) measurements were performed by using a U-shaped quartz micro-
reactor (i.d., 4 mm) and reducing the catalyst (50-100 mg) in hydrogen flow (60 STP ml/min)
at 350 °C. After the reduction treatment, the sample was cooled in flowing H2 to room
temperature; afterwards H2 was shut off and the sample was purged with argon flow (50 STP
ml/min) for 10 min. After purging, the desorption process was started by heating the sample
up to 700 °C in the argon carrier, with a heating rate of 10 °C/min. The desorption process
was monitored and quantified by a thermal conductivity detector (TCD) connected to a TCD.
Calibration test was performed by injecting in the carrier gas a known amount of H2 in order
to obtain reliable quantitative measurements. Metal dispersion (D) was calculated by
assuming a chemisorption stoichiometry H/Rusurface=1. Metal surface area (MSA) was
calculated assuming a site density of 8.2 Å2/at for Ru, while the Ru crystallite diameter (d)
was assuming a spherical shape of the metal particle: d (nm)=n/D (%), where “n” is 101 for
Ru60.
2.4 Reaction procedures
Catalysts reduction: The Ru/CMK-3 catalyst was used as prepared. In case of Ru/C, a pre-
reduction step was applied in the batch reactor. This involved charging the reactor (Parr
o
autoclave, 100 ml, max. 350 C and 350 bar) with 0.2 g of catalyst. The reactor was
9
ACS Paragon Plus Environment

ACS Catalysis
Page 10 of 34
1
2
3
4
5
6
7
8
9
pressurized with 100 bar of N2 to check for leakage, and then the reactor was depressurized.
The reactor was flushed three times with 20 bar of H2 to remove air and the Ru/C was pre-
reduced at 350 oC under 2-30 bar H2 for 1 h at a stirring speed of 800 rpm, before cooling to
room temperature.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Hydrogenation reactions: A solution of levoglucosan (20 mmol in 40 mL, 0.5 M) or the
disaccharides (10 mmol in 40 mL, 0.25 M) was injected into the reactor containing the
catalyst from a feed vessel using pressurized N2. Then the reactor was flushed 3 times with
H2 (20 bar). The reactor was pressurized to 50 bar with additional H2 and then heated to the
desired temperature with a heating rate of around 10 oC/min at a stirring speed of 1400 rpm.
Both reactor pressure and temperature were monitored using sensors and stored on a
computer. As the desired temperature was reached, the time was set as t = 0 h, the typical
reaction time was 0.5-5 h. After reaction, a sample of the liquid phase was filtered using 0.2
µm filters, and then analyzed using HPLC. Hydrogenation experiments with sorbitol and
glucose were carried out in a similar manner.
Hydrolysis reactions: The experimental procedure for the hydrolysis reactions of
levoglucosan is the same as for the hydrogenation reactions, the only exception is the use of
50 bar of N2 instead of H2.
Recycling tests: A typical levoglucosan hydrogenation experiment was performed with a 3 h
batchtime. Subsequently, the spent catalyst was separated from the liquid phase by
centrifugation (4500 rpm, 20 min). The spent catalyst was washed 3 times with 40 ml mili-Q
o
water and then dried in an oven at 70 C overnight. A second experiment was performed
using the spent catalyst using the same procedure and conditions as for the first run.
2.5 Analytical methods
HPLC analyses of the liquid phase: The liquid phase after reaction was analyzed using HPLC.
A HPLC apparatus equipped with a Hewlett Packard 1050 pump, a Bio-Rad organic acid
column (Aminex HPX-87H) and a differential refractometer was used. The mobile phase
consisted of an aqueous solution of sulfuric acid (5 mmol/l) operated at a flow rate of 0.55
o
cm3/min. The column was operated at 60 C. The amount of the products was calculated
using calibration curves obtained from standard solutions of known concentrations.
10
ACS Paragon Plus Environment

Page 11 of 34
ACS Catalysis
1
2
3
3 Results and discussions
4
5
3.1 Characterization of Ru/CMK-3 and Ru/C
6
7
8
9
A detailed catalyst characterization study was performed to gain insights in the structural
features (texture and acid sites) of Ru/CMK-3 as well as the Ru/C benchmark catalyst. The
well-ordered mesoporous structure of CMK-3 and Ru/CMK-3 was confirmed by N2
physisorption. Both the CMK-3 support and the Ru/CMK-3 catalyst display a type IV isotherm
with a characteristic H2 hysteresis loop in the relative pressure range from 0.4 to 0.9 (Figure
S2, supporting information). The pore size distributions are provided in Figure S3 (Supporting
information).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 2 Physicochemical properties of CMK-3 and Ru/CMK-3
Sample
SBET (m2/g)
990
Vp (cm3/g)a
1.20
Dp (nm)b
4.7
APD c (nm)
4.8
CMK-3
Ru/CMK-3
Ru/CMK-3 (spent)
Ru/C
913
900
745d
1.09
1.10
0.27
4.6
4.8
3.1
4.7
4.9
1.5
a Single point desorption total pore volume of pores less than 840.779 Å width at p/p° = 0.97 b BJH Desorption
average pore width (4V/A). c Average Pore Diameter evaluated as 4PV/SA d SA Langmuir
The BET surface area of the CMK-3 used in this study was about 990 m2/g, as shown in Table
2, which is slightly lower than the value reported in the literature (1120 m2/g)58. Differences
in synthetic procedures for CMK-3 are the plausible cause for this observation. The average
pore diameter, as measured using the BJH method, was 5.0 nm, which is slightly higher than
reported in the literature 58. Upon introduction of Ru, the surface area decreases from 990 to
913 m2/g and the pore diameter reduces from 4.7 to 4.6 nm, most likely due to the
incorporation of the small Ru nanoparticles in the pores or pore openings (vide infra).
11
ACS Paragon Plus Environment

ACS Catalysis
Page 12 of 34
1
2
3
4
5
6
7
8
9
150000
100000
50000
0
50000
<100>
CMK_3
CMK_3
RuCMK_3
40000
30000
20000
10000
0
RuCMK_3
<002>
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
<101>
<100>
<110>
<200>
<110>
<112>
0.8 1.2 1.6 2.0 2.4 2.8 3.2 3.6 4.0 4.4 4.8
10
20
30
40
50
60
70
80
90
100
2
θ
2
θ
Figure 1 Small and wide angle XRD patterns of CMK-3 and Ru/CMK-3
The well-ordered mesoporous structure of CMK-3 and Ru/CMK-3 was also confirmed by
small angle X-ray diffraction (SAXRD), see Figure 1. The CMK-3 sample shows a very intense
diffraction peak (100) and two low intense peaks (110) and (200), which are characteristic of
a two-dimensional hexagonal structure 61. The intensities of the XRD peaks for the Ru/CMK-3
sample are slightly reduced compared to the CMK-3 precursor, though the pattern is similar,
indicating that the ordered structure is retained after introduction of Ru and the subsequent
reduction step.
The wide-angle XRD spectra of CMK-3 and Ru/CMK-3 (Figure 1) show two intense diffraction
peaks located at around 23 and 43°, corresponding to (002) and (101) diffractions of typical
turbostratic carbon, respectively62. Clear diffraction peaks from Ru nanoparticles could not
be detected, owing to the small dimensions of the Ru-particles (vide infra).
TEM images of CMK-3 and Ru/CMK-3 were recorded to gain insights in the position and the
size of the Ru-nanoparticles (Figure 2). For both CMK-3 and Ru/CMK-3, uniform channels can
be observed, confirming the structured mesoporous character of the support. Ru
nanoparticles are present as clusters of different sizes (Figure 2b), and seem to be located
preferentially on the CMK-3 edges. These TEM results are not in full agreement with the
characterization study on Ru/CMK-3 performed by Komanoya et al. 62 where the presence of
small Ru particles was observed with an average particle diameter of 1.1 nm. In our case, the
particles are slightly larger (and agglomerated) and between 1.7-2.3 nm (Figure 2d). These
differences are likely due to the catalyst preparation procedures, viz. a colloidal precipitation
12
ACS Paragon Plus Environment

Page 13 of 34
ACS Catalysis
1
2
3
and chemical reduction procedure with NaBH4 at low temperature in this study versus a wet
4
5
impregnation and a reduction with hydrogen at elevated temperatures by Komanoya et al. 62
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Figure 2 TEM images of a) CMK-3; b-d) Ru/CMK-3
The total amount of acid sites was determined using NH3-TPD (Table 3 and Figure 3). A large
peak was observed for both samples with a maximum of 435°C (Ru/CMK-3) and 483°C (CMK-
3), indicative for the presence of strong acid sites. Ru/CMK-3 also shows weak acid sites as
o
indicated by the peak at 267 C. This peak is likely due to the presence of Brønsted acidic
62
RuO2·2H2O sites, as proposed by Komanoya et al. As such, the main source for the acid
sites is the support, though some additional acidity is generated by oxidized Ru nanoparticles.
However, the latter species are likely reduced under reaction conditions by molecular
hydrogen and as such likely do not contribute during the actual hydrolysis/hydrogenation
reactions.
13
ACS Paragon Plus Environment

ACS Catalysis
Page 14 of 34
1
2
3
4
5
Ru/CMK-3
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
CMK-3
100 200 300 400 500 600 700 800
Temperature (°C)
Figure 3 NH3-temperature programmed desorption profiles of CMK-3 and Ru/CMK-3
Table 3 NH3-temperature programmed desorption of CMK-3, Ru/CMK-3, spent Ru/CMK-3
and fresh Ru/C
[a]
[a]
NH3-uptake
Td1
Td2
[b]
[b]
SAMPLE
x1
x2
[°C]
[°C]
[ꢀmol/gcat
158
]
CMK-3
-
-
483
435
392
607
1.0
Ru/CMK-3
Ru/CMK-3 (spent)
Ru/C
180
267
223
257
0.10
0.06
0.12
0.90
0.94
0.88
110
122
a Temperature of maximum desorption of NH3: Td1, 100-300°C; Td2, 300-800°C; bFractional population of sites: x1, between
100 and 300°C; x2, between 300 and 800°C
The acidity value for CMK-3 (158
ꢀ
mol/gcat) is in the range for carbon nanotubes activated
with nitric acid using various synthetic procedures (120-250
ꢀmol/gcat). CMK-3 is typically
prepared by using SBA-15 silica as the template and sucrose as the carbon source. After
sucrose impregnation, the suspension is treated with sulfuric acid, and carbonized at
elevated temperatures, followed by SBA-15 removal using a basic treatment. As a result,
14
ACS Paragon Plus Environment

Page 15 of 34
ACS Catalysis
1
2
3
4
5
CMK-3 has oxygenated functional groups on the surface, examples are carboxylic, lactonic,
and phenolic groups57. For comparison, the acidity of the commercial Ru/C catalyst was also
6
7
determined (122
ꢀmol/gcat) and the values is considerably lower than for Ru/CMK-3 (158
8
9
ꢀmol/gcat), which is likely related to the preparation procedure.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The metal dispersion for both Ru/CMK-3 and Ru/C was determined by H2 chemisorption
measurements (see Table 4 for details). The average particle diameter for the Ru/CMK-3
catalyst was 3.3 nm, in line with TEM data. The dispersion was calculated to be close to 31%.
The average Ru particle diameter for the commercial Ru/C catalyst is by far larger (12.4 nm),
which is expected to have an effect on catalyst activity (vide infra). These findings are in line
with the TEM images (Figure S1, supplementary information) that evidenced the presence of
Ru clusters containing Ru particles of 5-10 nm in size, together with single Ru particles with a
regular spherical form with size ranging from 5 to 50 nm.
Table 4 Catalyst data from H2 chemisorption measurements
H2 uptake
[µmol/gcat]
20.1
MSAa
[m2/gcat]
2.0
D
db
[nm]
12.4
3.3
Sample
[%]
Ru/C
8.1
Ru/CMK-3
Ru/CMK-3 (spent)
30.2
3.0
30.6
21.8
21.6
2.1
4.6
a Assuming a surface concentration of metal atoms of 1.22·1019 at/m2 (from the arithmetic average of planar
densities for the three lowest index surface planes), b Crystallites diameter assuming a spherical shape of
particles
3.2 Hydrogenation of levoglucosan using Ru/CMK-3 and Ru/C
The hydrogenation of levoglucosan (LG) in water using Ru/CMK-3 (2 wt.% Ru on support)
was investigated as a function of the temperature (120-180 oC) in a batch hydrogenation set-
up at 50 bar hydrogen pressure for a batch time of 5 h (0.5 M LG, 6 wt.% catalyst on
substrate). After reaction, the liquid phase was analysed by HPLC to identify and quantify the
main reaction products. The conversion of levoglucosan and the selectivity towards the main
products versus the reaction temperature are shown in Figure 4.
15
ACS Paragon Plus Environment

ACS Catalysis
Page 16 of 34
1
2
3
4
5
6
7
8
9
Glucose
Mannitol
Sorbitol
100
80
60
40
20
0
100
80
60
40
20
0
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
120
150
180
Reaction Temperature, °C
o
Figure 4 Hydrogenation of levoglucosan using Ru/CMK-3 at three temperatures (120-180 C,
5 h, 0.5 M LG, 6 wt.% catalyst on LG)
o
At 120 C, the conversion of levoglucosan is limited (3 wt.%), and sorbitol is the main
product. Substantial amounts of glucose are also present, indicating that hydrolysis of LG to
glucose indeed occurs under the reaction conditions and that glucose is the intermediate
product. At 150 oC, the conversion of levoglucosan increased to 31 wt.%. Main products are
again sorbitol and glucose. The selectivity to sorbitol increased from 75 to 91 wt.%,
indicating that the rate of the hydrogenation reaction is more enhanced at elevated
temperatures than the hydrolysis reaction of LG to glucose. Full conversion of levoglucosan
is achieved at 180 oC in 5 h, with a sorbitol selectivity as high as 96.2 wt.%, together with 3.8
wt.% of mannitol. The presence of mannitol indicates the occurrence of an isomerization
reaction of glucose to fructose or a subsequent isomeriation of sorbitol to mannitol. Possible
other by-products, like smaller diols from hydrogenolysis reactions and conversion products
from glucose like HMF and levulinic acid, were not observed.
For comparsion, levoglucosan hydrogenations in water using a commercial Ru/C (5 wt.% Ru
o
loading) catalyst were also performed in the temperature range 120-180 C. The conversion
of levoglucosan and the product selectivity are shown in Table 5.
16
ACS Paragon Plus Environment

Page 17 of 34
ACS Catalysis
1
2
3
Table 5 Results for the catalytic conversion of levoglucosan using Ru/Ca
4
5
6
7
8
Temperature Conversion,
wt.%
Sorbitol,
wt.%
100
Mannitol, Glycerol, wt.% 1,2-Propanediol,
wt.%
0
wt.%
0
120 oC
150 oC
180 oC
2.0
0
0
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
14.7
59.7
22.4
0
17.3
2.3
41.1
8.5
10.3
a
6 wt.% catalyst on substrate, 0.5 M LG in water, 50 bar, 5 h, batch. The mass balance is not closed due to the
presence of other unidentified components (HPLC).
It is evident that the Ru/C catalyst is by far less active than Ru/CMK-3, despite the fact that
the Ru-loading is 2.5 times higher (2 versus 5 wt.%). At 180°C, the LG conversion was 60 %,
compared to close to quantitative conversion for Ru/CMK-3. In addition, the product
portfolio differs considerably from that for Ru/CMK-3 and substantial amounts of glycerol
and propanediol are formed besides sorbitol and small amounts of mannitol. The presence
of the former compounds are indicative for the occurrence of hydrogenolysis reactions, in
line with research from Bindwal et al. using Ru/C.55 For Ru/C, glucose was not detected in
any of the product samples.
3.3 Mechanistic pathways
The proposed mechanistic pathway for the hydrogenation of levoglucosan to C6-polyols is
shown in Scheme 2. The desired pathway involves the hydrolysis of levoglucosan to glucose
followed by the hydrogenation of glucose to sorbitol. The former is expected to be catalyzed
by the acid sites on the catalysts, while the latter is Ru catalysed. However, the formation of
C-C splitting products was also observed for Ru/C, indicating the occurrence of catalytic
hydrogenolysis reactions of intermediates/products.
To gain insights in the reaction pathways and to identify the differences in performance
between Ru/CMK-3 and Ru/C, a number of additional reactions were performed. These
involved i) the determination of concentration-time profiles for a number of reactions,
allowing calculation of initial rates and turnover frequencies (TOF) and ii) individual reactions
with intermediate/products (glucose/sorbitol), both in the presence and absence of
hydrogen.
17
ACS Paragon Plus Environment

ACS Catalysis
Page 18 of 34
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 2 Proposed reaction pathways for hydrogenation of levoglucosan
Initial experiments involved determination of the initial rate for the catalytic hydrogenation
reactions of LG with both Ru/CMK-3 and Ru/C. The initial rates were determined from the
slope of the concentration-time profiles for a batch reaction of 5 h with intermediate
sampling (see Figures S4 and S5 in the supplementary information). The data are compiled in
Table 6.
Table 6 Initial reaction rates and TOF’s for levoglucosan and glucose hydrogenation using
Ru/CMK-3 and Ru/Ca
Entry
Catalysts
Substrate
Initial rate
(mmol
substrate/h)
TOF
(mmol/(g
cat·h))
TOF
TOF
TOF
(mol/mol (mol/(mol
(mol/(mol
Ru at
acid
Ru
sites.h)b
187
intake·h)c
surface·h)d
1
2
3
4
Ru/CMK-3 Levoglucosan
Ru/C Levoglucosan
Ru/CMK-3 Glucose
Ru/C Glucose
7.0
4.1
98
34.1
20.4
487
495
167
2460
1000
8040
12350
101
a 6 wt.% catalyst on substrate, 0.5 M substrate in water, 70 bar total pressure at 180 oC, 5 h, batch, b acidity as
determined by ammonia-TPD, c Total amount of Ru on the catalysts, d Ru on the surface of the catalysts as
determined by hydrogen chemisorption.
18
ACS Paragon Plus Environment

Page 19 of 34
ACS Catalysis
1
2
3
4
5
6
7
8
9
The hydrogenation reaction of LG with Ru/CMK-3 gave sorbitol/mannitol as the main
product, with glucose as the intermediate, in line with a consecutive mechanism involving
the hydrolysis of LG to glucose followed by hydrogenation to C6-polyols (Scheme 2). For
Ru/C, glucose was not detected in significant amounts and sorbitol was the initial product
whereas glycerol and 1,2-propanediol were formed after longer batch times. This pattern of
reactivity implies that sorbitol is initially formed and subsequently converted to glycerol and
1,2-propanediol (vide infra).
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
The initial rate for the conversion of LG was considerably higher for Ru/CMK-3 (34.1 mmol/g
cat·h) than for Ru/C (20.4 mmol/g cat·h). A likely explanation for the higher conversion rate
of levoglucosan in the case of Ru/CMK-3 is a higher reaction rate of the acid catalysed
hydrolysis of LG to glucose. These findings are in line with the acidity measurements by NH3-
TPD, showing that Ru/CMK-3 is more acidic in nature than Ru/C. The initial activity,
expressed in mol per mol acid sites per h is about equal for both catalyst (167-187 mol/mol
acid sites·h, see Table 6), an indication for a first order dependency in acid sites. It is also of
interest to note that neither 5-hydroxymethylfurfural (HMF) nor levulinic acid were detected
in the reaction mixtures by acid catalysed reactions of glucose (Scheme 2). As such, the rate
of hydrogenation of glucose under the prevailing conditions to sorbitol is much faster than
the acid catalysed conversions.
The higher activity for the hydrolysis reaction of LG for Ru/CMK-3 was supported by
performing experiments with Ru/CMK-3 and Ru/C using LG in the presence of the catalyst
but by replacement of the hydrogen atmosphere by an inert N2 atmosphere. The results are
compiled in Table 7 and show a much higher conversion rate for Ru/CMK-3, confirming its
higher acidity.
Table 7 Hydrolysis reactions of levoglucosan using Ru/CMK-3 and Ru/Ca
Entry
Catalysts
Ru/CMK-3
Ru/C
Conversion of levoglucosan, wt.% Selectivity to glucose, wt.%
1
2
98.2
64.9
34.0
54.9
a
6 wt.% catalyst on substrate, 0.5 M LG in water, 50 bar N2, 180 oC, 5 h, batch
The performance of the catalysts for the hydrogenation of the intermediate glucose was also
tested independently by performing hydrogenation reactions with glucose in a batch set-up
19
ACS Paragon Plus Environment

ACS Catalysis
Page 20 of 34
1
2
3
4
5
6
7
8
9
for both catalysts with product sampling at various batch times. The results are given in
Table 6, the individual concentration profiles are given in the supplementary information
(Figures S6-S7). For both catalyst, glucose is converted quantitatively within about 1 h. The
initial activity for both catalyst is about equal (490 mmol/(g cat·h)). When compensating for
the differences in Ru loading (2 wt.% for Ru/CMK-3 versus 5 wt.% for Ru/C), the Ru/CMK-3
catalyst is more active (2460 versus 1000 mol/mol Ru intake·h). This difference in activity is
mainly due to the differences in the average Ru nanoparticle sizes according to H2
chemisorption and supported by TEM, which were shown to be larger for Ru/C than for
Ru/CMK-3. When considering the activity on the amount of Ru sites on the surface of the
nanoparticles, the activity of the Ru/C is actually higher than for Ru/CMK-3 (Table 6). Of
interest is the product portfolio for both catalyst and particularly the evolution of
intermediates/products in time. It is evident that sorbitol is the main product. However,
particularly for Ru/C, the amount of sorbitol shows a maximum and after a certain batch
time, considerable amounts of glycerol are formed (Figure S6, supporting information). This
indicates that sorbitol is, particularly for Ru/C, not inert under the given reaction conditions
and reacts further to smaller polyols by hydrogenolysis reactions, lowering the selectivity to
the desired sorbitol.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Further proof that C3 products are indeed formed from sorbitol was obtained by performing
batch experiments with sorbitol as the substrate at 180°C with both Ru/CMK-3 and Ru/C as
the catalyst (Table 8). Indeed, sorbitol is not inert under these conditions, though
conversions were below 22 %. In addition, Ru/C was by far more reactive than Ru/CMK-3.
The main products were mannitol and hydrogenolysis products. These findings indicate that
sorbitol is the source for the smaller polyols like glycerol. In addition, it also suggests that
sorbitol may be isomerized to mannitol by the Ru catalyst, in line with literature data63 .
Table 8 Sorbitol hydrogenolysis using Ru/C and Ru/CMK-3a
Catalysts Reaction Sorbitol Conversion, Mannitol Selectivity, Glycerol Selectivity,
time, h
wt.%
21.5
9.8
wt.%
39.1
35.6
wt.%
60.9
64.4
Ru/C
5
5
Ru/CMK-3
a
6 wt.% catalyst on substrate, 0.5 M sorbitol in water, 50 bar, 180 oC, 5 h, batch
20
ACS Paragon Plus Environment

Page 21 of 34
ACS Catalysis
1
2
3
4
5
6
7
8
9
Thus, from the data above, we can conclude that sorbitol is formed from LG by a two step
hydrolysis-hydrogenation sequence. However, the sorbitol is not inert under the conditions
applied and may be converted to smaller polyols, the result of C-C scission products by
hydrogenolysis. Hydrogenolysis reactivity is by far more pronounced for Ru/C than Ru/CMK-
3 and this is the explanation for the superior perfromance of Ru/CMK-3 in the reaction of LG
to C6-polyols. So far, we do not have a convincing explanation for this difference in reactivity
between Ru/CMK-3 and Ru/C for the hydrogenolysis reaction. Recently, Triantafyllidis et al.8
reported experimental work on the hydrogenation of glucose to sorbitol using Ru on
activated carbon (Norit, SAE, SUPER) and showed that the amount of byproducts from
hydrogenolysis of sorbitol was considerable (7-45%) and a function of process conditions,
metal loading and catalyst reduction procedure (H2 versus NaBH4). XPS and HRTEM
measurements for both types of activated catalysts showed a difference in the amounts of
zerovalent (crystalline) and charged Ru (non-crystalline) nanoparticles and this was put
forward as an explanation for the differences in the rate of hydrogenolysis of sorbitol for the
various catalysts.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
3.4 Catalyst stability
To gain insights in the stability of the Ru/CMK-3 catalyst, recycling studies for levoglucosan
hydrogenation were carried out and the results are given in Figure 5. After 3 successive runs
o
at 180 C for 3 h batch times, around 92 wt.% of levoglucosan conversion was observed for
all the 3 runs. However, the selectivity to sorbitol decreased from 94 to 88 wt.% whereas
glucose selectivity increased from 2 to 7 wt.%. As such it implies that the acid sites on
Ru/CMK-3 necessary for the hydrolysis of LG are largely preserved under the selected
reaction condition, while the hydrogenation activity of Ru/CMK-3 slightly decreased.
21
ACS Paragon Plus Environment

ACS Catalysis
Page 22 of 34
1
2
3
4
5
Glucose
Mannitol
Sorbitol
100
80
60
40
20
0
100
80
60
40
20
0
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
1
2
3
Number of Runs
Figure 5 Recycling tests for Ru/CMK-3 for levoglucosan hydrogenation (5 h, 0.5 M LG, 6 wt.%
catalyst on LG)
The spent Ru/CMK-3 catalyst after 1st run was characterized by TEM (Figure 6), ammonia-
TPD (Table 3) and hydrogen chemisorption (Table 4). The textural appearance as determined
by TEM showed strong resemblance with the fresh sample. Ammonia TPD measurements
indicated that the acidity is slightly reduced after an experiment, whereas the mean
diameter of the Ru nanoparticles increased slightly (from 3.3 to 4.6 nm, H2 chemisorption
data, see Table 4). As such, the observed reduction in hydrogenation activity after one
recycle is likely due to a slight increase in the Ru nanoparticle size.
Figure 6 TEM images of spent Ru/CMK-3
22
ACS Paragon Plus Environment

Page 23 of 34
ACS Catalysis
1
2
3
3.5 Hydrogenation of disaccharides using Ru/CMK-3
4
5
6
7
8
9
To extend the scope for the use of Ru/CMK-3 for combined hydrolysis-hydrogenation
reactions, the catalyst was also tested for two sugar oligomers, cellobiose and sucrose.
Cellobiose is the simplest sugar oligomers containing a 1, 4-β-glucoside bond connecting two
glucose monomers. In sucrose, glucose and fructose are linked via an ether bond between
C1 on the glucosyl unit and C2 on the fructosyl unit. The results for the conversion of
cellobiose and sucrose and the product selectivity using Ru/CMK-3 in water at 180 oC for 5 h
are shown in Table 9.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Table 9 Conversion and product selectivity for the hydrogenation of disaccharides using
Ru/CMK-3a
Substrate
Conversion,
wt. %
Sorbitol,
wt.%
Mannitol,
wt.%
Others,
wt.%
4.4
Total C6 alcohols,
wt.%
95.6
97.9
Cellobiose
Sucrose
100
91.1
4.5
100
61.6
36.3
2.1
a
6 wt.% catalyst on substrate, 0.25 M of disaccharides in water, 50 bar, 5 h, batch, 180°C
Full conversion was observed for both substrates after a reaction of 5 h at 180 oC. In the case
of cellobiose, the total sugar alcohol yield was 95.6 % with 91.1 % of sorbitol and 4.5 % of
mannitol (HPLC). Mannitol is known to be formed by the hydrogenation of fructose, the
isomerization product of glucose. Glucose was not detected in any of the product mixtures.
Two reaction pathways can be envisaged for the reaction, viz i) a hydrolysis-hydrogenation
pathway involving hydrolysis of the 1,4-β-glucoside bond of cellobiose forming glucose,
which is further hydrogenated to sorbitol or ii) an initial hydrogenation of the reducing sugar
end to cellobitol (3-β-D-glucopyranosyl-D-glucitol) followed by hydrolysis to sorbitol and
glucose and a second hydrogenation (see Scheme 3).
Recent kinetic studies by Palkovits et al.64 on cellobiose and trisaccharides using Ru/C in
combination with a heteropoly acid showed that the latter pathway is taking place to a large
extent. The hydrogenation followed by hydrolysis sequence was shown to be preferred at
low reaction temperatures and for longer oligomers. Similar findings were reported by Wang
et al. 56 for bifunctional Ru catalysts supported on Ru/Cs2HPW12O40.
23
ACS Paragon Plus Environment

ACS Catalysis
Page 24 of 34
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Scheme 3 Proposed reaction pathways for hydrogenation of cellobiose over Ru/CMK-3
catalyst
In the product mixtures of the experiments reported in Table 9, cellobiotol, was not
observed, likely due to the long reaction time of 5 h. A number of experiments was
performed using cellobiose as the starting material at shorter reaction times and the results
are provided in Figure 7. Indeed, after 0.5 h, significant amounts of cellobitol (41 %
selectivity) were present in the reaction mixture, beside only minor amounts of glucose. In
addition, all cellobiose was quantitative converted after 30 min. As such, reaction pathway ii)
(hydrogenation followed by hydrolysis) seems to be favored with the Ru/CMK-3 catalyst
under these conditions.
Others
Glucose
Cellobitol
Mannitol
Sorbitol
100
80
60
40
20
0
100
80
60
40
20
0
0.5
5.0
3.0
Reaction Time, h
24
ACS Paragon Plus Environment

Page 25 of 34
ACS Catalysis
1
2
3
4
5
Figure 7 Product composition versus time for the hydrogenation of cellobiose using Ru/CMK-
3 (180 oC, 0.25 M cellobiose, 6 wt.% catalyst on cellobiose)
6
7
8
9
Catalytic cellobiose hydrogenations have been reported in the literature and the results for
Ru based catalysts56, 65-67, of relevance for this paper, are summarized in Table 10. Best results
with quantitative sorbitol yield at 100 % cellobiose conversion were obtained using Ru
nanoparticles in combination with a Brønsted acid (HCl) at 120°C, 40 bar hydrogen and a
batch time of 12 h65. For bifunctional hydrolysis-hydrogenation catalyst, the best sorbitol
yield was 93 wt.%56 (Table 10, entries 8-10) using a Ru/Cs2HPW12O40 catalyst. As such, the
performance of the Ru/CMK-3 catalyst is about similar to the best bifunctional catalyst
reported to date for the conversion of cellobiose to sorbitol.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
25
ACS Paragon Plus Environment

ACS Catalysis
Page 26 of 34
1
2
3
4
5
6
7
8
Table 10 Overview of hydrogenations of cellobiose using supported Ru catalysts
Entry
Substrate
Cellobiose
Cellobiose
Temperature
120 oC
H2
Reaction time
12 h
Catalysts
Ru/C
Conversion (mol.%)
100 %
Product selectivity (mol.%)
>99 % of cellobitol
Ref.
65
1
2
40 bar
40 bar
9
120 oC
12 h
Ru nanoparticles with 100 %
HCl (pH=2)
100 % of sorbitol
65
65
65
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
3
4
Cellobiose
Cellobiose
120 oC
120 oC
40 bar
40 bar
12 h
12 h
Ru
nanoparticles, 87.8 %
64.8 % of cellobitol, 26.4 % of sorbitol,
1.6 % of glucose, 7.2 % of polyols
55.7 % of cellobitol, 24.0 % of sorbitol,
3.2 % of glucose, 17.1 % of polyols
99 % of cellobitol
pH=7
Ru
nanoparticles, 75.6 %
with NaOH (pH=10)
Ru/C
66
66
5
6
Cellobiose
Cellobiose
160 oC
120 oC
50 bar
50 bar
2 h
100 %
1.8 h
Ru/C + silicotungstic 83 %
81 % of cellobitol, 18 % of sorbitol,
trace amount of glucose
acid
66
7
Cellobiose
170 oC
50 bar
1.25 h
Ru/C + silicotungstic 100 %
acid
<1 % of cellobitol, 75 % of sorbitol,
trace amount of glucose
56
56
67
8
Cellobiose
Cellobiose
Cellobiose
140 oC
140 oC
180 oC
20 bar
20 bar
50 bar
6 h
6 h
3 h
Ru/Cs2HPW12O40
Ru/Cs3PW12O40
Ru/CNT
100 %
100 %
100 %
trace amount of cellobitol, 93 % of sorbitol
10 % of cellobitol, 86 % of sorbitol
5 % of cellobitol, 88 % of sorbitol, 2 % of
mannitol
9
10
26
ACS Paragon Plus Environment

Page 27 of 34
ACS Catalysis
1
2
3
4
5
6
7
8
9
In the case of sucrose, quantitative conversion was observed after 5 h at 180°C, with a C6
sugar alcohol selectivity of close to 98 %. The major difference when comparing with the
results for cellobiose is the sorbitol to mannitol ratio, and for sucrose, the mannitol yield is
as high as 36.3 wt.%. This may be explained by considering that the hydrogenation of
fructose is known to give mannitol and sorbitol in a close to one to one ratio, while
hydrogenation of glucose gives mainly sorbitol with trace amounts of mannitol (vide supra).
Literature precedents for the hydrogenation of sucrose to C6-sugar alcohols in a single step
are rare.68 In most cases, sucrose is first hydrolysed to glucose and fructose and then
hydrogenated In a second step.69 Thus, the concept of the use of a bifunctional Ru/CMK-3
catalyst for disaccharides also is highly successful for sucrose.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Conclusions
C6-sugar alcohols and particularly sorbitol are attainable from LG in excellent yields (> 96%)
using a mesoporous carbon supported Ru/CMK-3 as the catalyst and water as the solvent.
Catalyst performance is superior to Ru/C, for which significant amounts of smaller diols were
obtained, indicative for the occurrence of hydrogenolysis reactions leading to C-C bond
cleavage. The excellent performance of the Ru/CMK-3 catalyst, both in terms of activity and
selectivity, is due to the low rate of subsequent hydrogenolysis reactions of sorbitol to
smaller polyols, as was shown by separate reactions with sorbitol. The bifunctional character
of the catalyst (hydrolysis/hydrogenation capabilities) was also shown for the hydrolysis-
hydrogenation of disaccharides (cellobiose and surcose), again giving excellent yields of C6
sugar alcohols. In addition, recycling studies in batch with LG as the starting material reveal a
small reduction in the rate of glucose hydrogenation for 3 successive runs, rationalised by a
slight increase in the Ru nanoparticle size (TEM, H2 chemisorption). These findings indicate
that Ru/CMK-3 is a potentially very attractive catalyst for the conversion of sugar fractions of
pyrolysis liquids, known to contain considerable amounts of LG and cellobiose, to sugar
alcohols. These studies are in progress and will be reported in due course.
Acknowledgement
Financial support from Agentschap NL (Groene aardolie via pyrolyse, GAP) is gratefully
acknowledged. We thank Jan Henk Marsman, Leon Rohrbach, Erwin Wilbers, Marcel de Vries
and Anne Appeldoorn for technical support. Financial support from Italian Research Fund
27
ACS Paragon Plus Environment

ACS Catalysis
Page 28 of 34
1
2
3
4
5
6
(PON R&C 2007–2013, DD 713/Ric. – PON02_00451_3362376) through the project “BIO4BIO
– Biomolecular and Energy valorization of residual biomass from Agroindustry and Fishing
Industry” led by the Cluster Sicily Agrobio and Fishing Industry is also acknowledged.
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
Supporting Information
TEM images of the Ru/C catalyst, N2 adsorption isotherms, pore size distributions of CMK-3
and the Ru analogue and concentration-time profiles of LG and glucose hydrogenation using
Ru/CMK-3 and Ru/C.
References
(1) Werpy, T.; Petersen, G.; Aden, A.; Bozell, J.; Holladay, J.; Manheim, A.; Elliott, D. C.;
Lasure, L.; Jones. S. Top Value Added Chemicals from Biomass; U.S. Department of Energy:
Oak Ridge, TN, 2004; Vol. 1.
(2) Besson, M.; Gallezot, P.; Pinel, C. Chem. Rev. 2013, 114, 1827-1870.
(3) Corma, A.; Iborra, S.; Velty, A. Chem. Rev. 2007, 107, 2411-2502.
(4) Eisenbeis, C.; Guettel, R.; Kunz, U.; Turek, T. Catalysis Today 2009, 147, S342-S346.
(5) Delidovich, I.; Hausoul, P. J.; Deng, L.; Pfützenreuter, R.; Rose, M.; Palkovits, R. Chem. Rev. 2016
,
116, 1540-1599.
(6) Huber, G. W.; Cortright, R. D.; Dumesic, J. A. Angewandte Chemie International Edition 2004, 43,
1549-1551.
(7) Dabbawala, A. A.; Mishra, D. K.; Huber, G. W.; Hwang, J. Applied Catalysis A: General 2015, 492,
252-261.
28
ACS Paragon Plus Environment

Page 29 of 34
ACS Catalysis
1
2
3
(8) Lazaridis, P. A.; Karakoulia, S.; Delimitis, A.; Coman, S. M.; Parvulescu, V. I.; Triantafyllidis, K. S.
4
5
6
Catalysis Today 2015, 257, 281-290
.
7
8
9
(9) Wang, S.; Wei, W.; Zhao, Y.; Li, H.; Li, H. Catalysis Today 2015, 258, 327-336
.
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
(10) Aho, A.; Roggan, S.; Simakova, O. A.; Salmi, T.; Murzin, D. Y. Catalysis Today 2015, 241, 195-199.
(11) Perrard, A.; Gallezot, P.; Joly, J.; Durand, R.; Baljou, C.; Coq, B.; Trens, P. Applied Catalysis A:
General 2007, 331, 100-104.
(12) Crezee, E.; Hoffer, B. W.; Berger, R. J.; Makkee, M.; Kapteijn, F.; Moulijn, J. A. Applied Catalysis A:
General 2003, 251, 1-17.
(13) Hoffer, B.; Crezee, E.; Devred, F.; Mooijman, P.; Sloof, W.; Kooyman, P.; Van Langeveld, A.;
Kapteijn, F.; Moulijn, J. Applied Catalysis A: General 2003, 253, 437-452.
(14) Hoffer, B.; Crezee, E.; Mooijman, P.; Van Langeveld, A.; Kapteijn, F.; Moulijn, J. Catalysis Today
2003, 79, 35-41.
(15) Makkee, M.; Kieboom, A. P.; van Bekkum, H. Carbohydr. Res. 1985, 138, 225-236.
(16) Kobayashi, H.; Ohta, H.; Fukuoka, A. Catalysis Science & Technology 2012, 2, 869-883.
(17) Kobayashi, H.; Matsuhashi, H.; Komanoya, T.; Hara, K.; Fukuoka, A. Chemical Communications
2011, 47, 2366-2368.
(18) Kobayashi, H.; Ito, Y.; Komanoya, T.; Hosaka, Y.; Dhepe, P. L.; Kasai, K.; Hara, K.; Fukuoka, A.
Green Chem. 2011, 13, 326-333.
(19) Kobayashi, H.; Hosaka, Y.; Hara, K.; Feng, B.; Hirosaki, Y.; Fukuoka, A. Green Chem. 2014, 16, 637-
644.
29
ACS Paragon Plus Environment