Mechanistic study on -C-O- and -C-C- hydrogenolysis over Cu catalysts: Identification of reaction pathways and key intermediates

Article abstract of DOI:10.1039/c7cy02426f

Important petro-based polyol compounds with a longer carbon chain, such as oligohydroxy hexanes (e.g. 1,2- and 1,6-hexanediol or 1,2,6-hexanetriol), require at least three to four synthesis steps. Replacing this complex chemistry by a one-pot reaction via -C-O- bond cleavage from sugars would be a significant breakthrough for the use of renewable feedstocks. Cu is known for its dehydroxylation (deoxygenation) properties, yielding the desired products from sugars. In this joint research between academic and industrial chemistry, we have identified so far unknown intermediate products and present the first mechanism that explains the selective cleavage of OH-groups over copper. Strong interactions between polyols, unsaturated species and the copper surface are observed. Stable five-membered rings are formed with Cu via two vicinal OH-groups of the polyol reactant that makes these OH-groups inert to -C-O- bond cleavage. Adjacent free OH-groups in close proximity to the catalyst are dehydroxylated (deoxygenated). We further show that degradation of polyols not only occurs via commonly cited retro-aldol reactions. The formation of acid intermediates with subsequent decarboxylation is validated as a new pathway for -C-C- bond cleavage to short-chain polyols and CO₂. The proposed mechanisms for -C-O- and -C-C- bond cleavage elucidate why hydrogenolysis reactions require high hydrogen pressure (up to 200 bar) to suppress the degradation of sugars and obtain high yields of deoxy C6 products. With this knowledge, the improvement of a standard commercial Cu-RANEY catalyst under optimized reaction conditions was shown. In contrast to alumina-supported Cu, the Cu-Al alloy in a RANEY-type catalyst shows selective -C-O- bond cleavage properties while maintaining the C6 carbon chain. These new insights into the transformation of sugars to value added commodities show the potential for new approaches in future biorefinery concepts.

Full text of DOI:10.1039/c7cy02426f

View Article Online
View Journal
Catalysis
Science &
Technology
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: B. B. Kühne, H.
Vogel, R. Meusinger, S. Kunz and M. Kunz, Catal. Sci. Technol., 2017, DOI: 10.1039/C7CY02426F.
This is an Accepted Manuscript, which has been through the
Volume 6 Number 1 7 January 2016 Pages 1–308
Royal Society of Chemistry peer review process and has been
accepted for publication.
Catalysis
Science &
Technology
Accepted Manuscripts are published online shortly after
www.rsc.org/catalysis
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
ISSN 2044-4753
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
PAPER
Qingzhu Zhang et al.
Catalytic mechanism of C–F bond cleavage: insights from QM/MM
analysis of fluoroacetate dehalogenase
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
rsc.li/catalysis

Page 1 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02426F
Mechanistic study on -C-O- and -C-C- hydrogenolysis over
Cu catalysts: identification of reaction pathways and key
intermediates
Benjamin Kühne,a Herbert Vogel,a Reinhard Meusinger,b
Sebastian Kunz,c and Markwart Kunza,*
[a] Institute for Technical and Macromolecular Chemistry, TU Darmstadt, Alarich-Weiss-Str. 8, D-64287 Darmstadt
[b] Institute for Organic Chemistry and Biochemistry, TU Darmstadt, Alarich-Weiss-Str. 4, D-64287 Darmstadt
[c] Institute for Applied and Physical Chemistry, University of Bremen, P.O. Box 330 440, D-28359 Bremen
* Corresponding Author
Prof. Dr. Markwart Kunz
Institute for Technical and Macromolecular Chemistry,
TU Darmstadt,
Alarich-Weiss-Str. 8, D-64287 Darmstadt
Fax: +49 6151 16-23703, phone: +49 6151 16-23700
Email: markwart.kunz@t-online.de
1

Catalysis Science & Technology
Page 2 of 29
View Article Online
DOI: 10.1039/C7CY02426F
Abstract
Important petro-based polyol compounds with a longer carbon chain, such as oligohydroxy
hexanes (e.g. 1,2- and 1,6-hexanediol or 1,2,6-hexanetriol), require at least three to four syn-
thesis steps. Replacing this complex chemistry by a one-pot reaction via -C-O- bond cleav-
age from sugars would be a significant breakthrough for the use of renewable feedstocks. Cu
is known for dehydroxylation (deoxygenation) properties, yielding the desired products from
sugars. In this joint research between academic and industrial chemistry we have identified
so far unknown intermediate products and present the first mechanism that explains the se-
lective cleavage of OH-groups over copper. Strong interactions between polyols, unsaturated
species and the copper surface are observed. Stable five-membered rings are formed with
Cu via two vicinal OH-groups of the polyol reactant that makes these OH-groups inert
to -C-O- bond cleavage. Adjacent free OH-groups in close proximity to the catalyst are dehy-
droxylated (deoxygenated). We further show that degradation of polyols not only occurs via
commonly cited retro aldol reactions. The formation of acid intermediates with subsequent
decarboxylation is validated as a new pathway for -C-C- bond cleavage to short-chain poly-
ols and CO2. The proposed mechanisms for -C-O- and -C-C- bond cleavage elucidate why
hydrogenolysis reactions require high hydrogen pressure (up to 200 bar) to suppress the
degradation of sugars and obtain high yields of deoxy C6 products. With this knowledge, the
improvement of a standard commercial Cu-Raney catalyst at optimized reaction conditions
was shown. In contrast to alumina-supported Cu, the Cu-Al alloy in a Raney-type catalyst
shows selective -C-O- bond cleavage properties while maintaining the C6 carbon chain.
These new insides on the transformation of sugars to value added commodities show the
potential for new approaches in future bio refinery concepts.
2

Page 3 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02426F
Graphical Abstract
Sugar alcohols are found to adsorb over Cu catalysts via two vicinal OH-groups,
forming a five-membered ring (similar to a chelate complex) with selective -C-O-
bond cleavage of adjacent free OH-groups. Different side reactions via unsaturated
intermediates on the catalyst surface yield C1 to C5 polyols.
3

Catalysis Science & Technology
Page 4 of 29
View Article Online
DOI: 10.1039/C7CY02426F
1. Introduction
The development of bio-based intermediates is a highly desired aim in order to re-
duce the reliance on fossil feedstocks [1-5]. Transformation reactions of renewable
feedstocks by hydrogenolysis have been investigated intensively, but mainly with fo-
cus on -C-C- bond cleavage [6-15]. Catalysts for selective -C-C- bond cleavage of
sugars to glycols with yields of up to 91 wt-% are well known [9,13]. In contrast, the
formation of deoxygenated products by -C-O- bond cleavage (deoxygenation) to C6
tetrols, triols and diols still requires significant improvements in order to become ap-
plicable. Thus, a better understanding of the major and side reaction pathways of
these demanding reactions is necessary. For this reason, the scope of the present
work was to explore Cu-catalyzed transformations of biomass-derived sugar alcohols
(e.g. sorbitol, mannitol, and xylitol) and according reaction intermediates. In particu-
lar, deoxygenated compounds (e.g. C4 to C6 triols and diols) and short-chain polyols
were used as reactants to better understand Cu based hydrogenolysis. The charac-
teristic properties of Cu catalysts have been studied for the conversion of sorbitol and
sugar derivatives [6,13,16]. However, reaction pathways for the formation of deoxy-
genated products are still unknown. Particularly the effect of the OH-group arrange-
ment in the reactants has not been investigated yet, except for degradation reactions
[16-24].
Based on our application of different reactants we show new mechanisms for the
formation of deoxygenated C6 pentols, tetrols, and triols. The postulated reaction
pathways are validated by several poly alcohols and reveal the presence of so far not
identified reaction intermediates. The selectivity of -C-O- bond cleavage and for-
mation of specific intermediates indicates strong adsorption effects on Cu that corre-
lates with the arrangement of OH-groups in the reactant.
4

Page 5 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02426F
2. Experimental
All reactions were performed in a stainless steel autoclave (Berghof, BR-700 series,
990 ml) in temperature, H2-pressure, and hexitol concentration dependent experi-
ments. The reaction mixture was vigorously stirred at 600 rpm in all experiments us-
ing an external gas distribution stirrer. The temperature was varied between 180 to
220 °C. For each experiment the autoclave was filled with 500 mL reactant solution
(7.5 wt-% of specific polyol in deionized water for all reactions), sealed and flushed
three times with 40 bar H2 (Linde 5.0, purity >99.999 %) to remove residual air. Sub-
sequently, a pressure of approximately 100 bar H2 was adjusted and the autoclave
was heated up to reaction temperature at a constant stirring rate of 600 rpm. A pres-
sure increase from 100 to 150 bar was obtained when heating to 180 °C. Hence, the
pressures discussed reflect the system pressure obtained under reaction conditions.
C2 to C6 polyols ranging from ethylene glycol to hexitols were analyzed using an Ag-
ilent 6890 gas (details are explained in the ESI). Mono alcohols are not detected with
this GC method and are thus quantified by headspace analysis. Specific hexane tri-
ols, tetrols and pentols were identified by NMR analysis (Bruker BioSpin GmbH,
DRX-500 NMR spectrometer). Volatile hydrocarbons, carbon dioxide, and methanol
were detected by IR spectroscopy at ambient temperature without quantification
(Bruker ALPHA FT-IR-spectrometer, spectral range from 500 to 4000 cm-1). Howev-
er, the integral of the IR bands was found to reflect the trends expected from the cal-
culated carbon yield.
Carbon yield in the liquid phase is based on the following equation:
ꢏꢐꢑꢒꢊꢌꢊꢈꢌꢊꢒ
∑
ꢇ
ꢈ
ꢅ
ꢓꢔꢕ.
ꢆ
ꢇ
ꢎ
ꢅ
ꢆ
ꢈ ꢉꢊꢋꢈꢌꢋꢍꢌ
ꢅ ꢆ
% =
ꢀ
ꢗ 100
Eq. 1
ꢖ
ꢁ ꢂꢃꢄ
ꢇ
,
ꢅ
ꢆ
ꢈ ꢉꢊꢋꢈꢌꢋꢍꢌ
Mono alcohols are not included in the carbon yield (YC,liq / %) because the GC meth-
od is optimized for the quantification of diols and polyols or disaccharide sugars with
up to eleven OH-groups.
Carbon loss is based on equation 2. It correlates to the amount of gaseous products,
mono alcohols and potentially undetected products.
ꢏꢐꢑꢒꢊꢌꢊꢈꢌꢊꢒ
∑
ꢇ
ꢎ
ꢇ
ꢅꢓꢔꢕ.ꢆ
ꢈ ꢅꢉꢊꢋꢈꢌꢋꢍꢌꢆ
ꢈ
ꢅ ꢆ
% = 100 −
ꢀ
ꢗ 100
Eq. 2
ꢖ
ꢁ ꢂꢘꢙꢙ
ꢇ
,
ꢈ ꢅꢉꢊꢋꢈꢌꢋꢍꢌꢆ
5

Catalysis Science & Technology
Page 6 of 29
View Article Online
DOI: 10.1039/C7CY02426F
Conversion of the reactants was calculated based on the following equation:
ꢇ
ꢚ % = ꢛ1 − ꢈ ꢅꢉꢊꢋꢈꢌꢋꢍꢌꢆꢜ ꢗ 100
Eq. 3
ꢅ ꢆ
ꢖ
ꢁ
ꢇ
ꢈ ꢅꢉꢊꢋꢈꢌꢋꢍꢌꢆ
All isomeric hexitols in the reaction mixture (sorbitol, mannitol, galactitol and iditol)
were considered as hexitol reactants.
The calculation of the selectivity is based on the following equation:
ꢏꢐꢑꢒꢊꢌꢊꢈꢌꢊꢒ
∑
ꢇ
ꢈ
ꢅꢓꢔꢕ.ꢆ
ꢝ ꢅ%ꢆ =
ꢗ 100
Eq. 4
ꢖ
ꢁ
ꢇ
ꢞ ꢇ
ꢈ ꢅꢉꢊꢋꢈꢌꢋꢍꢌꢆ
ꢈ ꢅꢉꢊꢋꢈꢌꢋꢍꢌꢆ
A variety of isomeric deoxy hexitols (hexane pentols, tetrols, triols and diols) was
found. For each group of deoxy hexitols we defined a cumulative selectivity in which
all isomeric hexane pentols, tetrols, triols and diols are accumulated.
Sorbitol (purity >99 %), mannitol (>99 %), and xylitol (99 %) were generously provid-
ed from Südzucker. 1,2,3,4,5-hexanepentol was obtained from hydrogenation of
rhamnose (purity >98 %, provided from Carl Roth). 1,2,5,6-hexanetetrol was isolated
(purity 75 %) from a product mixture obtained from hydrogenolysis of mannitol over
Cu-Raney. 1,2,6-hexanetriol (96 %), 1,2,4-butanetriol (>90 %), 1,2-butanediol
(>98 %), 2,3-butanediol (98 %), 1,2-propanediol (99 %), glycerol (>99 %), 1,3-
propanediol (98 %), ethanediol (>99 %), isosorbide (as dianhydro-D-glucitol, 98 %),
ethanol (99 %) and methanol (99 %) were purchased from Sigma-Aldrich. Isomaltu-
lose (purity >99 %) was also provided from Südzucker.
The heterogeneous catalysts (molded as granulate and spheres) based on Cu were
applied in a 100 ml catalyst basket. In the case of Cu-Raney 200 g granulate were
used and the amount of Cu/Al2O3 was 140 g. Only with such high catalyst quantities it
was possible to achieve the desired conversions within reasonable reaction periods.
As such high catalyst quantities are needed, no well-defined model catalyst could be
applied as sophisticated catalyst preparation methods (e.g. colloidal routes) are usu-
ally limited to rather small batch sizes and metal concentrations. Furthermore, under
applied reaction conditions supported Cu catalysts undergo considerable structural
changes due to sintering, diminishing the advantageous of using model catalysts for
the present study (chapter 3.2). All catalysts were used for hydrogenation of isomal-
6

Page 7 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02426F
tulose (at least 30 h on stream time) prior to hydrogenolysis reactions in order to re-
move soluble inorganic species that may originate from the catalyst preparation. The
catalyst preconditioning was necessary to exclude effects arising from impurities on
the fresh catalyst that catalyze undesired side reactions. Cu-Raney (SMR-type by
Grace) was provided from Südzucker AG. Cu/Al2O3 (as Cu(II)O on alumina, loading
13 wt%, LOT: MKBP4388V) was purchased from Sigma-Aldrich. An alumina support
without Cu (pure Al2O3) was also provided from Südzucker AG but did not lead to any
significant catalytic conversion under applied reaction conditions.
7

Catalysis Science & Technology
Page 8 of 29
View Article Online
DOI: 10.1039/C7CY02426F
3. Results and discussion
3.1 Hydrogenolysis of C6 sugar alcohols over Cu catalysts
The conversion of C6 sugar alcohols (hexitols, e.g. sorbitol or mannitol) over Cu-
Raney shows that selective -C-O- bond cleavage is achievable with appreciable se-
lectivity at moderate temperatures between 180 and 220 °C (Fig. 1, left). A molar
carbon selectivity of 75 % is obtained at 180 °C which decreases to about 35 % at
220 °C because of enhanced -C-C- bond cleavage. The mixture of deoxy C6 polyols
comprises of linear C6 pentols, tetrols, and triols. Interestingly, the product distribu-
tion suggests that -C-O- bond cleavage occurs in the middle of the C6 carbon chain
while terminal OH-groups are rather stable. 1,2,4,5,6-hexanepentol, 1,2,5,6-
hexanetetrol, and 1,2,5- / 1,2,6- / 1,4,5-hexanetriols are the main deoxy C6 products.
The composition of this product mixture is independent of the reactant (sorbitol or
mannitol) because tautomerization into a mixture of four C6 polyols (sorbitol, manni-
tol, galactitol, and iditol) always occurs (Fig. 2). Cu/Al2O3 also catalyzes tautomeriza-
tion into the same mixture of C6 polyols. However, for the supported catalyst primari-
ly -C-C- bond cleavage into C3 polyols is obtained, even at a moderate temperature
of 180 °C (Fig. 1, right). Glycerol is the main product, along with 1,2-propanediol. The
comparison of Cu/Al2O3 and Cu-Raney shows that degradation reactions are much
more pronounced for the alumina-supported catalyst. The formation of short-chain
products, in particular C3 polyols (glycerol and 1,2-propanediol), is predominantly
attributed to retro aldol elimination of the C6 sugar alcohols [14,16,23]. This reaction
pathway competes with the -C-O- bond cleavage and seems to be much more fa-
vored when Cu/Al2O3 is used.
8

Page 9 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02426F
Fig. 1: Conversion of mannitol and selectivity to polyols of different carbon chain length for Cu-Raney (left) and
Cu/Al2O3 (right) at different temperatures, 150 bar H2 pressure and 5 h reaction time YC: as defined in
eq. 1.
Fig. 2: Cu-catalyzed tautomerization of sorbitol or mannitol into a mixture of C6 polyols (hexitols) and -C-O-
bond cleavage to deoxy C6 polyols. The -C-C- bond cleavage to C3 polyols occurs as a non-catalytic
side reaction initiated form unsaturated intermediates.
However, it is unlikely that the Al2O3 support enhances such base-catalyzed reac-
tions because oxygen defects in the support should lead to (Lewis-) acidic properties.
In fact, pure Al2O3 without Cu showed no conversion at all. Hence, the
hanced -C-C- bond cleavage is related to the formation of unsaturated intermediates
over Cu that initiate degradation reactions. Polyols are known to undergo dehydro-
genation to aldehyde and ketone intermediates during adsorption over alumina-
supported Cu [16] and Cu-Raney [14,29,30]. These unsaturated intermediates ena-
ble the retro aldol reaction and thus, need to be re-hydrogenated as quickly as possi-
ble (Fig. 2). To validate this proposal the hydrogenation properties of both Cu cata-
lysts were compared for isomaltulose (with a fructose building block) hydrogenation
under identical reaction conditions at 70 °C and 150 bar H2. Cu-Raney was about
9

Catalysis Science & Technology
Page 10 of 29
View Article Online
DOI: 10.1039/C7CY02426F
three times more active for hydrogenation of the ketone reactant than the Cu/Al2O3
catalyst (Fig. 3).
Fig. 3: Hydrogenation rates for Cu-Raney and the Cu/Alumina for the hydrogenation of isomaltulose (fructose
building block) at 70 °C and 150 bar H2. The conversion rate is standardized per one gram catalyst.
Hence, we propose that side reactions like the retro aldol reaction are more pro-
nounced over Cu/Al2O3 because the unsaturated intermediates that initiate -C-C-
bond cleavage are less likely to be re-hydrogenated. The formation of aldehydes and
ketones during adsorption is a quasi-equilibrated step as indicated in Figure 2 for
both catalysts. Such unsaturated species degrade even in the absence of any cata-
lyst and thus, hydrogenation of these intermediates has to be improved to suppress -
C-C- bond cleavage. This shifts the reaction network towards -C-O- bond cleavage.
Cu-Raney leads to higher hydrogenation rate and thus, the selectivity to undesired
short-chain polyols is much lower. The same correlation is seen between the applied
H2-pressure and the formation of short-chain polyols over Cu-Raney. Figure 4 shows
that the selectivity for C2 to C5 polyols from -C-C- bond cleavage is enhanced when
low H2 pressures are applied. Conversely, the formation of such short-chain polyols
is greatly suppressed when the H2 pressure is increased. Consequently, the selectivi-
ty for deoxy C6 polyols steadily increases from 51 mol-% at 50 bar H2 to 76 % at
200 bar H2. These results indicate that fast hydrogenation of unsaturated intermedi-
ates under high H2 pressure is crucial in order to maintain the C6 carbon chain of
sugar alcohols which results in higher selectivities for deoxy C6 polyols.
10

Page 11 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02426F
Fig. 4: Effect of H2 pressure on the conversion of mannitol (7.5 wt-% aqueous solution) and selectivity towards
deoxy C6 polyols and short-chain polyols (C2 to C5) over Cu-raney at 180 °C and 5 h reaction time.
In addition to the comparatively low hydrogenation activity, Cu/Al2O3 shows signifi-
cant signs of deactivation at hydrogenolysis conditions (up to 220 °C at 150 bar H2).
The structure of the supported catalyst seems to collapse as the BET surface area is
strongly decreased from 176 m²/g (fresh catalyst) to 8 m²/g after 30 h reaction time at
temperatures between 180 and 220 °C. The rapid deactivation of supported Cu cata-
lysts is a well-known phenomenon at such conditions [25]. In contrast, the surface
area for Cu-Raney remains about the same with 34 m²/g before and 28 m²/g after
200 h time-on-stream at 180 to 220 °C just like it’s activity and selectivity.
11

Catalysis Science & Technology
Page 12 of 29
View Article Online
DOI: 10.1039/C7CY02426F
3.2 Effect of OH-group arrangement on hydrogenolysis activity for Cu-Raney
Due to the deactivation of the Cu/Al2O3 catalyst, the effect of OH-group arrangement
on hydrogenolysis activity was only investigated for Cu-Raney. Figures 2 and 5 show
that the steric orientation of OH-groups in different stereoisomers does not affect the
activity and selectivity for -C-O- bond cleavage reactions. Independent of the reactant
(pure mannitol or sorbitol), a similar hexitol mixture of mannitol, sorbitol, galactitol,
and iditol is obtained for reaction periods between 2,5 and 5 h (Fig. 5). Sorbitol is the
main stereoisomer in this mixture, followed by mannitol. Galactitol and iditol are
formed in smaller amounts. All four hexitols are converted at comparable rates and
the formation rates of deoxy C6 products are also similar.
Fig. 5: Isomerization and hydrogenolysis of mannitol (top) and sorbitol (bottom) over Cu-Raney at 180 °C and
150 bar H2. Independent of the reactant, the isomerization products galactitol and iditol are formed in
comparable amounts and also converted at approximately the same rate as mannitol and sorbitol.
12

Page 13 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02426F
However, the arrangement of OH-groups in the deoxygenated reaction intermediates
strongly affects the hydrogenolysis pathway and thus, the selectivity to specific deoxy
C6 polyols. In order to investigate how the selective -C-O- bond cleavage over Cu-
Raney occurs, we have categorized numerous polyols into different types based on
similarities in the molecular structure (Fig. 6) and studied the dependency of deoxy-
genation reactions from the structure of the reactant. Type A polyols consist of at
least three adjacent OH-groups with glycerol as the smallest possible compound of
this category. The type B polyols consist of two vicinal OH-groups at the 1- and 2-
position and an adjacent methyl or methylene group at the 3-position when the car-
bon chain is longer than two carbon atoms (e.g. 1,2-propanediol and 1,2-butanediol).
2,3-butanediol does not consist of terminal OH-groups but rather two adjacent OH-
groups in the middle of the molecule (type C).
Fig. 6: Classification of polyols into types A to D, based on the arrangement of OH-groups.
The type D polyols isosorbide and 1,3-propanediol are also diol compounds but con-
sist of isolated OH-groups. The application of these different reactants enables the
investigation how the number and the arrangement of OH-groups affects -C-O-
and -C-C- bond cleavage over Cu catalysts. Our study reveals that both conversion
and selectivity for Cu catalyzed -C-O- bond cleavage strongly depend on the number
and arrangement of the OH-groups within the molecule. Sugar alcohols with three or
13

Catalysis Science & Technology
Page 14 of 29
View Article Online
DOI: 10.1039/C7CY02426F
more adjacent OH-groups (e.g. glycerol or hexitols, type A) are highly reactive and
more effectively converted than compounds with only two adjacent OH-groups (type
B and C polyols) or isolated OH-groups (type D) as shown in Table 1. Comparison of
the conversion (XC / %) reveals that type B polyols are rather stable towards -C-O-
and -C-C- bond cleavage over Cu-Raney. The carbon chain length does not seem to
affect the hydrogenolysis activity as long as the reactant only comprises of vicinal
OH-groups at the 1- and 2-position. Ethanediol, 1,2-propanediol, and 1,2-butanediol
show only very low conversion (1 to 12 %).
Table 1: Reactivity and product distribution of various polyols compounds over Cu-Raney (100 ml catalyst
basked, 7.5 wt-% aqueous solution of each polyol, 180 °C, 20h reaction time, and 150 bar H2).
X /
%
Yc,liq
/ %
Yc,loss
/ %
Structure
Reactant
Reactivity
Three or more adjacent OH-groups (type A)
Sorbitol
Mannitol
1,2,3,4,5-hexanepentol
Xylitol
Glycerol
100
100
75
96
77
83
86
80
81
92
17
14
20
19
18
very high
very high
high
very high
high
Two adjacent OH-groups (terminal position, type B)
1,2,5,6-hexanetetrol
1,2,6-hexanetriol
1,2,4-butanetriol
1,2-butanediol
1,2-propanediol
Ethanediol
4
10
48
8
12
1
96
91
85
92
89
99
4
9
15
8
11
1
very low
very low
moderate
very low
very low
very low
Two adjacent OH-groups (internal position, type C)
2,3-butanediol
1
99
1
very low
Other polyols with isolated OH-groups (type D)
1,3-propanediol
Isosorbide
20
9
82
92
18
8
low
very low
When the reactant carries an additional OH-group, the distance between the third
OH-group and the vicinal OH-groups at the 1- and 2-position determines the reactivi-
ty for -C-O- bond cleavage. For example, the third OH-group in 1,2,6-hexanetriol is
separated from the OH-groups at the 1- and 2-position by three methylene groups. In
1,2,5,6-hexanetetrol the distance to the next OH-group is two methylene groups.
Both compounds are only weakly converted (10 and 4 %, respectively), similar to
ethanediol, 1,2-propanediol, and 1,2-butanediol. In contrast, the third OH-group in
1,2,4-butanetriol is only separated by one methylene group and this reactant shows
much higher conversion (48 %). Interestingly, the only deoxy product obtained from
1,2,4-butanetriol hydrogenolysis over Cu-Raney is 1,2-butanediol. Hence, the OH-
14

Page 15 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02426F
group at the 4-position is also preferably cleaved whereas the vicinal OH-groups at
the 1- and 2-position are stable towards -C-O- bond cleavage (Fig. 7). Based on this
comparison of the type B polyols we conclude that the number of OH-groups in a re-
actant only seems to affect the hydrogenolysis activity when OH-groups are in close
proximity. Furthermore, we conclude that vicinal OH-groups at the 1- and 2-position
of a reactant are more stable.
Fig. 7: Selective -C-O- bond cleavage of 1,2,4-butanetriol over Cu-Raney (180 °C, 150 bar H2).
The type C polyol 2,3-butanediol consists of vicinal OH-groups in the middle of the
molecule with adjacent methyl groups on both sides. This reactant also shows negli-
gible conversion (1 %), similar to 1,2-butanediol. The same very low reactivity is seen
for isosorbide, consisting of isolated OH-groups (type D polyol), and a conversion of
only 9 %. The conversion of 20 % for 1,3-propanediol is mainly attributed to -C-C-
bond cleavage via retro aldol elimination yielding ethanol and methanol. The carbon
loss (YC,loss / %) of 18 % as shown in Table 1 accounts for these mono alcohols. As
the GC method used to analyze di-, tri-, and tetrols does not allow to detect mono
alcohols, ethanol and methanol were detected and quantified separately by additional
headspace analysis In contrast to 1,3-propanediol, isosorbide and vicinal diols are
not able to follow the retro aldol reaction and thus, show less conversion.
Type A polyols with three or more adjacent OH-groups show much higher reactivity
than type B, C and D polyols (Table 1). Sorbitol, mannitol and xylitol are almost com-
pletely converted (96 to 100 %). The main products are deoxy C6 polyols and deoxy
C5 polyols when xylitol is used as reactant, respectively. Glycerol and the 1,2,3,4,5-
hexanepentol show conversions of 75 and 77 %, respectively. A comparison of the
pentols 1,2,3,4,5-hexanepentol and xylitol indicates that the presence of terminal me-
thyl groups has a negative effect on the hydrogenolysis activity. Both reactants com-
15

Catalysis Science & Technology
Page 16 of 29
View Article Online
DOI: 10.1039/C7CY02426F
prise of five adjacent OH-groups with one additional methyl group in 1,2,3,4,5-
hexanepentol. We attribute this methyl group is a hindrance in the interaction be-
tween OH-groups of the reactant and the Cu surface. Makkee et al. proposed that
reducing sugars such as fructose preferably adsorb on a Cu catalyst with specific
OH-groups, in particular at the 1- and 2-position (Fig. 8) [26]. Based on our results,
we assume that polyols adsorb similarly via at least two OH-groups, including the
terminal position. Consequently, a methyl group as in 1,2,3,4,5-hexanepentol has a
negative effect on the adsorption because the reactant can only adsorb from one
side. In contrast, xylitol can adsorb from both sides via two vicinal OH-groups which
results in the higher conversion compared to 1,2,3,4,5-hexanepentol.
Fig. 8: Adsorption of fructose on a Cu surface and hydrogenation to sorbitol and mannitol via hydride-like spe-
cies as proposed by Makkee et al. (left) [25]. The adapted mechanism proposes adsorption of C5 and C6
sugar alcohols via two vicinal OH-groups in a five-membered ring, similar to a chelating complex (right).
The free OH-group in closest proximity is then cleaved via substitution with a hydride-like species.
Herein, the adsorbed OH-groups seem to be inert towards -C-O- bond cleavage
which is probably attributed to the stable five-membered ring in the chelating complex
with Cu (Fig. 8). This is in agreement with a postulated mechanism from Shinmi et al.
regarding deoxygenation from glycerol to 1,2-propanediol [27], where the adsorbed
OH-groups are not cleaved and only the adjacent free OH-groups are deoxygenated.
Our proposing would also explain the negligible conversion of the type B polyols (e.g.
1,2-butanediol, 1,2,6-hexanetriol and 1,2,5,6-hexanetetrol). When these compounds
are adsorbed via the two OH-groups at the 1- and 2-position, there are no free OH-
groups in close proximity to the catalyst that could undergo -C-O- bond cleavage. All
other diols (type C and D) also do not have free OH-groups left for -C-O- bond cleav-
age and therefore are not effectively converted.
16

Page 17 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02426F
Fig. 9: IR spectra of the gas phase composition for hydrogenolysis reactions of various polyols and mono alco-
hols over Cu-Raney at 180 °C, 150 bar H2 and 5 h reaction time.
The side reactions over Cu-Raney are mainly -C-C- bond cleavage to short-chain
polyols via retro aldol reactions. Formation of gaseous C1 products, in particular CO2,
is only significant when polyols comprise of more than two adjacent OH-groups. Fig-
ure 9 shows that only traces of volatile C1 products are detected when type B, C, and
D polyols are used as reactants. In contrast, the type A compounds (e.g. glycerol)
undergo degradation to CO2 in larger quantity. We attribute this to the requirement of
a 1,3-diol arrangement of OH-groups for decarboxylation as discussed in more detail
in chapter 3.4. Type A polyols with at least three adjacent OH-groups allow such re-
actions whereas the type B, C, and D polyols are not able to follow these degradation
pathways. Hence, the OH-group arrangement in the reactant does not only affect
deoxygenation reactions but also degradation via different -C-C- bond cleavage
mechanisms.
3.3 Selective -C-O- bond cleavage mechanism
Based on the results from section 3.1, the selective -C-O- bond cleavage of the type
A polyols such as sorbitol and mannitol was studied for the Cu-Raney catalyst. Our
analysis of reaction intermediates reveals that the hydrogenolysis of hexitols to deoxy
products (hexanepentols, -tetrols, and -triols) occurs via multiple consecutive -C-O-
bond cleavage steps. Hexanepentols (monodeoxy compounds) are the first transient
product while hexanetetrols (dideoxy) and hexanetriols (trideoxy polyols) are pro-
duced from these intermediates as shown in Figure 10.
17

Catalysis Science & Technology
Page 18 of 29
View Article Online
DOI: 10.1039/C7CY02426F
Fig. 10: Product distribution for hydrogenolysis of mannitol (7.5 wt-% aqueous solution) over Cu-Raney as func-
tion of reaction time. The experiment was performed at 180 °C and 150 H2. The successive formation of
hexanepentols, -tetrols, and -triols shows that multiple consecutive -C-O- bond cleavage steps occur.
The liquid phase composition after 40 h reaction time shows that the main product is
1,2,5,6-hexanetetrol. This isomer is the only hexanetetrol obtained in large amounts.
Similarly, the mixture of hexanetriols only comprises of three isomers (1,2,5- / 1,2,6- /
1,4,5-isomers). It is remarkable that only four deoxy C6 polyols account for more
than 60 mol-% of the applied carbon from mannitol. These products have to derive
from selective -C-O- bond cleavage pathways. Combined GC- and NMR-analysis
shows that only two hexanepentols are formed by the initial -C-O- bond cleavage
step over Cu-Raney. The first deoxygenation predominantly occurs at the 3-position
(Fig. 11, Fig. S1, S2, S3, and S5). 6-deoxy hexitols (1,2,3,4,5-hexanepentols, Fig. S3
and S4) are also identified but only as a minor side product in small amounts. These
are the same stereoisomers as obtained from hydrogenation of the 6-deoxy hexose
rhamnose. In a recent study by Palkovitz et al. it was shown that the deoxygenation
of terminal OH-groups is not acid-catalyzed but rather a metal-catalyzed reaction via
the E2 mechanism [28]. This reaction was initially proposed for Ru catalysts but it
seems also plausible for Cu. However, the 3-deoxy and 6-deoxy hexitols are formed
in a ratio of about 8:1, which means that the E2 mechanism is not the
nant -C-O- bond cleavage pathway over Cu-Raney. 2-deoxy hexitols (or 5-deoxy,
respectively) are not detected which reflects the special catalytic properties of Cu.
18

Page 19 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02426F
We attribute this large excess of 3-deoxy products to the favored adsorption
via the vicinal OH-groups at the 1- and 2-position with -C-O- bond cleavage of the
adjacent OH-group (see Fig. 8 and Fig. 11). Four stereoisomers of 3-deoxy hexitols
are detected with GC analysis which corresponds to the four detected hexitols: sorbi-
tol, mannitol, galactitol, and iditol. The next -C-O- bond cleavage step should yield a
3,4-dideoxy hexitol as the main product (1,2,5,6-hexanetetrol, see Fig. 12, Fig. S6,
S7, and S8). According to our mechanism, adsorption can now occur at the 5- and 6-
position with -C-O- bond cleavage of the adjacent OH-group at the 4-position. This
reaction is basically the same as the first -C-O- cleavage step, just repeated from the
other side of the molecule. This proposed mechanism would explain why 1,2,5,6-
hexanetetrol is the main product in the liquid phase as shown in Figure 9. In a side
reaction, 3-deoxy hexitols may also undergo direct -C-O- bond cleavage at the
4-position, when adsorbed via the 1- and 2-position, as was shown by hydrogenoly-
sis of 1,2,4-butanetriol (Fig. 7).
Fig. 11: Selective -C-O- bond cleavage of hexitols over Raney-Cu. Only two constitutional isomers are detected:
3-deoxy hexitols as the main products and 6-deoxy hexitols as side products. The formation of 6-deoxy
products via the E2 mechanism is adapted from Palkovitz et al. [28].
The further conversion of the 3-deoxy and 6-deoxy hexitols via the E2 mechanism
and our proposed mechanism for the cleavage of internal OH-groups should yield
three different dideoxy hexitols (see Fig. 12). This is in line with combined GC-MS
and NMR analysis which shows that only three constitutional isomers are formed.
19

Catalysis Science & Technology
Page 20 of 29
View Article Online
DOI: 10.1039/C7CY02426F
The 3,4-deoxy hexitol (1,2,5,6-hexanetetrol) is the main product and the 3,6-dideoxy
hexitol (1,2,4,5-hexanetetrol) is the major side product (Fig. S9, S10 and S11). The
structure of the third isomer could not be identified with NMR analysis, but it is ex-
pected to be the 2,3-dideoxy hexitol (1,4,5,6-hexanetetrol).
Fig. 12: Selective -C-O- bond cleavage of 3-deoxy and 6-deoxy hexitols over Raney-Cu. Our proposed
mechanism invovles adsorption via vicinal OH-groups with cleavage of an adjacent OH-group or an OH-
group with one methylene group distance. This leads to only three possible dideoxy products
(hexanetetrols).
The 3,6-dideoxy hexitol may be formed from the 3-deoxy intermediate along the E2
mechanism. A second route to the 3,6-dideoxy hexitol could be the adsorption via the
4- and 5-position with -C-O- bond cleavage of the adjacent terminal OH-group at the
6-position. When the OH-group at the 2-position is cleaved instead, the 2,3-dideoxy
hexitol is obtained. Figure 13 shows that the 3,6-dideoxy and 2,3-deoxy product are
20

Page 21 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02426F
only intermediates and further converted to hexanetriols. The 3,6-dideoxy hexitol
leads to 1,2,5- or 1,4,5-hexanetriol, whereas the 2,3-dideoxy hexitol yields the 1,4,5-
or 1,2,6-hexanetriol. Other hexanetriols are not detected. The formation of 1,2,6-
hexanetriol is also possible from ring-opening hydrogenolysis of cyclic ethers such as
isosorbide. The product mixture was checked for these intermediates, but not even
trace amounts were detected. Hence, the selective -C-O- bond cleavage over Cu-
Raney only seems to occur via linear deoxy C6 polyols.
Fig. 13: Selective -C-O- bond cleavage of 3,6-dideoxy and 2,3-dideoxy hexitols over Raney-Cu. Our proposed
mechanistic model leads to three hexanetriols. 3,4-deoxy hexitols are stable and not further converted.
The 1,2,5,6-hexanetetrol and the three hexanetriols (1,2,5- / 1,2,6- / 1,4,5-isomers,
see Fig. S11 to S15) consist of vicinal OH-groups and two adjacent methylene
groups, similar to the OH-group arrangement of the type B polyols shown Figure 5
and Table 1. Thus, we conclude that this is a stable configuration over Cu and such
polyols do not undergo further -C-O- bond cleavage.
21

Catalysis Science & Technology
Page 22 of 29
View Article Online
DOI: 10.1039/C7CY02426F
Likewise, the deoxygenation of xylitol should yield 1,2,5-pentanetriol (Fig. 14). A de-
tailed analysis of tautomerization products (arabitol and ribitol) and the respective
diastereomers after -C-O- bond cleavage showed exactly the same results as de-
scribed for the C6 and C4 polyols previously. Two stereoisomers are obtained from
xylitol, ribitol and arabitol after cleavage of the OH-group at the 3-position. The se-
cond -C-O- bond cleavage step either eliminates the OH-group at the 2-position or
the 4-position. Both pathways lead to the same stereo isomer (a pair of enantiomers)
13
as the final product, which is validated by GC and C-NMR analysis. These investi-
gations underline the selectivity of -C-O- bond cleavage over Cu-Raney, again.
Fig. 14: Selective -C-O- bond cleavage of xylitol over Raney-Cu. Formation of tautomerization and deoxygena-
tion products, with separation of stereo isomers, is validated with combined GC and 13C-NMR analysis.
The amount of the predicted possible diastereomers is in line with the detected products and yield the fi-
nal product 1,2,5-pentanetriol.
We conclude that stable final products consist of two vicinal OH-groups with two
neighboring methylene groups. Products with this arrangement of OH-groups are not
further converted via -C-O- or -C-O- bond cleavage (Fig. 15). We attribute this special
catalytic property of Cu-Raney to preferred adsorption of vicinal OH-groups in poly-
ols. We further propose that the reactivity of -C-O- hydrogenolysis in additional free
OH-groups decreases in the following order:
22

Page 23 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02426F
1.
2.
3.
Cleavage of an adjacent OH-group is the most likely reaction
Cleavage with one methylene group distance is less likely than 1.
Cleavage with two or three methylene groups in between OH-groups is very
unlikely to occur
Fig. 15: Reactivity of OH-groups over Cu catalysts for -C-O- bond cleavage. The final products consist of two
vicinal OH-groups and two adjacent methylene groups, independent of the carbon chain length of the re-
actant.
23

Catalysis Science & Technology
Page 24 of 29
View Article Online
DOI: 10.1039/C7CY02426F
3.4 Degradation of polyols via CO2 formation over Cu
The formation of CO2 is another -C-C- bond cleavage reaction that needs to be re-
duced in order to obtain high yields of deoxygenated C6 polyols from sugars. Signifi-
cant amounts of CO2 are only detected when polyols with a minimum of three adja-
cent OH-groups are used as reactant (type A). Based on these results we compared
different possible routes for CO2 formation that are initiated by dehydrogenation of
the reactant. The presence of unsaturated products was validated by GC-analysis as
well as isomerization of polyols that occurs via these sp2-hybridized intermediates.
CO2 may result from one of three possible reaction pathways:
1. Decarboxylation from unstable 3-keto-acids [31]; we detected several 3-
hydroxy-acids (aldonic acids) that potentially convert to 3-keto-acids via ad-
sorption at the catalyst.
2. Decarbonylation from aldehydes followed by water-gas shift reaction to CO2.
3. Methanol reforming after C6 polyols are converted to methanol via -C-C- bond
cleavage reactions, presumably by retroaldolization [16].
Multiple routes for the formation of 3-hydroxy-acids are possible in aqueous solution
at temperatures between 180 °C and 220 °C (Fig. 16). Although Cu-Raney is a hy-
drogenation catalyst, it also possesses dehydrogenation properties [29,30]. The al-
dehyde intermediates on the catalyst surface are possibly hydrated to an acetal in
aqueous solution with subsequent dehydrogenation similar to the methanol reforming
mechanism (path 1). The conversion of aldohexoses into sugar acids in the presence
of hydroxide anions is also well-known (path 2) [32-34]. The aldehyde intermediates
show the properties of reducing sugars and thus, are also possibly converted into
sugar acids via this route. The dissociation of water at high temperatures (220 °C)
automatically leads to comparatively large amounts of hydroxide species. Further-
more carbonyl compounds are able to follow the Cannizzaro reaction to form carbox-
ylic acids (path 3) [14]. Intramolecular reactions from
α-keto-aldehydes are known
and give lactic acid [35] or saccharinic acids (mono-deoxy aldonic acids) [36]. These
reactions may also take place under high H2 pressures (150 bar) because the oxida-
tion of the C1 carbon occurs via dehydrogenation over the Cu catalyst. In fact, glu-
conic acid is a known side product in glucose hydrogenation reactions [37] and was
also detected in our experiments along with saccharinic acids (Fig. S16). Further evi-
24

Page 25 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02426F
dence that acid-formation occurs is the detection of formic acid, acetic acid, and lac-
tic acid (Fig. S17). Decarbonylation of aldehydes over Cu can be excluded as a deg-
radation mechanism because the aldehydes, which are formed during adsorption,
would give much higher yields of CO2. Especially ethanediol and ethanol would show
larger amounts of CO2 because adsorption always leads to an aldehyde intermedi-
ate. However, for these compounds the yield of gaseous products is negligible
(Fig. 9). The same is observed for methanol; barely any CO2 is detected. Conse-
quently, methanol reforming to CO2 does not seem to occur over Cu as previously
proposed by Palkovitz et al. [16]. The degradation of sorbitol to CO2 rather seems to
follow one of the routes shown in Figure 16 via an acid intermediate.
Decarboxylation is most likely to occur via 3-hydroxy or rather 3-keto acids
that are obtained from an additional dehydrogenation step over Cu. The mechanism
for the cleavage of the terminal -C-C- bond resembles the retro aldol reaction as
shown in Figure 2. Experiments with gluconic acid as reactant confirm significant
carbon loss due to formation of large amounts of CO2 (Fig. 17). The acid seems to be
more reactive towards degradation. In comparison to mannitol and sorbitol much
more CO2 is formed after 5 h reaction time (220 °C and 150 bar H2).
Based on these results we conclude that CO2 formation from sugar alcohols and
glycerol over Cu based catalysts occurs via decarboxylation from sugar acids.
25

Catalysis Science & Technology
Page 26 of 29
View Article Online
DOI: 10.1039/C7CY02426F
Fig. 16: Formation of 2-hydroxy and 3-hydroxy acids and decarboxylation over Cu.
Fig. 17: IR spectra of the gas phase composition for hydrogenolysis reactions of mannitol and gluconic acid on
Cu/Al2O3 at 220 °C, 150 bar H2 and 5 h reaction time.
26

Page 27 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02426F
4. Summary
A mechanistic study has been presented on the selective -C-O- bond cleavage of
polyols over a Raney-type Cu catalyst. Our proposed mechanistic model predicts
adsorption of polyols occurs via two OH-groups, preferably at the 1- and 2-position,
with cleavage of OH-groups in close proximity. The reactivity for deoxygenation of a
third OH-group decreases in the following order: adjacent OH-group > one methylene
group distance > two methylene groups distance. Hydroxy compounds with at least
two methylene groups between the OH-groups are rather stable under the chosen
reaction conditions. These special catalytic properties of Cu result in the formation of
specific 3-deoxy and 3,4-dideoxy compounds such as 1,2,5,6-hexanetetrol or 1,2,5-
and 1,2,6-hexanetriol.
In contrast, a supported Cu catalyst yields larger amounts of short-chain polyols by
degradation via unsaturated intermediates. These undesired side reactions are pre-
sumably increased because of weak hydrogenation properties of the catalyst (due to
strong deactivation) that allows unsaturated intermediates on the catalyst surface to
exist longer and initiate -C-C- bond cleavage. We can further show that CO2 for-
mation as a major side reaction over Cu is probably attributed to decarboxylation
from a 3-hydroxy or rather 3-keto acid. The proposed mechanisms for the formation
of acid intermediates again suggest that dehydrogenation initiates the side reaction
pathways. Hence, to improve catalyst selectivity and maintain the carbon chain, bet-
ter hydrogenation properties are required. With this mechanistic understanding a ra-
tional approach for the production of specific C6 tetrols and triols from sugars can be
developed.
27

Catalysis Science & Technology
Page 28 of 29
View Article Online
DOI: 10.1039/C7CY02426F
5. Conflict of interest statement
There are no conflicts to declare.
6. Acknowledgements
The authors kindly thank Willi Kundel and the laboratory group for GC, GC-MS, and
HPLC analytics at Südzucker’s Central Department for Research, Development, and
Services (CRDS) in Offstein.
7. References
[1]
J. Q. Bond, D. M. Alonso, J. A. Dumesic, Aqueous Pretreatment of Plant Bio-
mass for Biological and Chemical Conversion to Fuels and Chemicals, Wiley-
VCH, 2013, 61-102
[2]
[3]
[4]
[5]
P. Claus, G. H. Vogel, Chem. Ing. Tech., 2006, 8, 991-1012
A. Corma, S. Iborra, A. Velty, Chem. Rev., 2007, 107, 2411-2502
G. H. Vogel, Chem. Ing. Tech., 2014, 86, 1-16
T. Werpy, G. Peterson, Top Value Added Chemicals from Biomass. Results of
Screening for Potential Candidates from Sugars and Synthesis Gas
,
Vol. 1
,
US Department of Energy, 2004.
[6]
[7]
B. Blanc, A. Bourrel, P. Gellezot, T. Haas, P. Taylor, Green Chem., 2000, 2
89-91
,
I. T. Clark, Ind. Eng. Chem., 1958, Vol. 50 (8), 1125-1126
[8]
[9]
K. L. Deutsch, D. G. Lahr, B. H. Shanks, Green Chem., 2012, 14, 1635-1642
M. Dubeck, G. G. Knapp, Patent US4476331, 1984
[10] H. R. Goldschmid, A. S. Perlin, Can. J. Chem., 1960, 38, 2178-2186
[11] D. K. Sohounloue, C. Montassier, J. Barbier, React. Kin. Catal. Let., 1983,22
,
391-397
[12] R. Palkovitz, A. M. Ruppert, K. Weinberg, Angew. Chem., 2012, 51, 2564-
2601
[13] L. Schuster, W. Himmele, BASF Patent DE3818198 A1, 1989
[14] K. Wang, M. C. Hawley, T. D. Furney, Ind. Eng. Chem. Res., 1995, 34 (11),
3766-3770
[15] W. H. Zartman, H. Adkins, J. Am. Chem. Soc., 1933, 55 (11), 4559-4563
28

Page 29 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY02426F
[16] K. Tajvidi, P. J. C. Hausoul, R. Palkovitz, ChemSusChem, 2014, 7, 1311-1317
[17] M. A. Dasari, P.-P. Kiatsimkul, W. R. Sutterlin, G. J. Suppes, Appl. Catal. A
2005, 281, 225-231
,
[18] E. P. Maris, R. J. David, J. Catal., 2007, 249, 328-337
[19] E. P. Maris, W. C. Ketchie, M. Murayama, R. J. Davis, J. Catal.,2007, 251
,
281-294
[20] L. Ma, D. He, Z. Li, Catal. Comm., 2008, 9, 2489-2495
[21] L. Ma, D. He, Top. Catal., 2009, 52, 834-844
[22] T. Miyazawa, Y. Kusunoki, K. Kunimori, K. Tomishige, J. Catal., 2006, 260
,
213-221
[23] C. Montassier, D. Giraud, J. Barbier, Polyol Conversion by Liquid Phase Het-
erogenous Catalysis Over Metals Elsevier, Amsterdam, 1988, 165-170
[24] C. Montassier, J.C. Ménézo, L.C. Hoang, C. Renaud, J. Barbier, J. Mol. Catal.
,
1991, 70 (1), 99-110
[25] C. Montassier, J. M. Dumas, P. Granger, J. Barbier, Appl. Catal. A, 1995, 121
(2), 231-244
[26] M. Makkee, A. P. G. Kieboom, H. van Bekkum, Carb. Res., 1985, 138 (2)
,
225-236
[27] Y. Shinmi, S. Koso, T. Kubota, Y. Nakagawa, K. Tomishige, Appl. Catal. B
2010, 94 (3-4), 318-326
,
[28] P. J. C. Hausoul, L. Negahdar, K. Schute, R. Palkovitz, ChemSusChem, 2015,
, 3323-3330
8
[29] M. Berweiler, D. Ostgard, K. Seelbach, Degussa AG Patent EP1127613 A1
,
2001
[30] M. Berweiler, M. Höpp, R. Vanheertum, W. Girke, Degussa AG Patent
EP1382389 A1, 2004
[31] A. Kornberg, S. Ochoa, A. H. Mehler, J. Biol. Chem., 1948, 174, 159-172
[32] G. de Wit, J. J. de Vlieger, A. C. Kock-van Dalen, A. P. G. Kieboom, H. van
Bekkum, Tetrahedron Lett., 1978, 15, 1327-1330
[33] G. de Wit, J. J. de Vlieger, A. C. Kock-van Dalen, R. Heus, R. Laroy, A. J. van
Hengstum, A. P. G. Kieboom, H. van Bekkum, Carb. Res., 1981, 91, 125-138
[34] H. J. Isbell, Carb. Res., 1969, 9, 163-175
[35] M. G. Manas, J. Campos, L. S. Sharninghausen, E. Lin, R. H. Crabtree, Green
Chem., 2015, 17, 594-600
[36] H. J. Isbell, J. Res. Natl. Bur. Stand., 1944, 32, 45
[37] B. J. Arena, Deactivation of ruthenium catalysts in continuous glucose hydro-
genation, Applied Catalysis A: General, 1992, 87 (2), 219-229
29