Mechanistic Studies of the Cu(OH)+-Catalyzed Isomerization of Glucose into Fructose in Water

Article abstract of DOI:10.1002/cssc.201800483

The isomerization of glucose to fructose is a crucial interim step in the processing of biomass to renewable fuels and chemicals. This study investigates the copper-catalyzed glucose–fructose isomerization in water, focusing on insights into the roles of the dissolved copper species. Depending on the pH, the thermodynamic equilibrium shifted towards one or a few copper species, namely Cu²⁺, Cu(OH)⁺, and Cu(OH)₂. According to thermodynamics, the highest concentration of Cu(OH)⁺ is at pH 5.3, at which the highest fructose yield of 16 % is achieved. The obtained fructose yields strongly correlate with the concentration of Cu(OH)⁺. A pH decrease of 2–3 units was observed during the reaction, resulting in the deactivation of the catalyst through hydrolysis in acidic media. Based on the results of the catalytic experiments, as well as spectroscopic and spectrometric studies, we propose Cu(OH)⁺ as an active Lewis-acidic species following an intramolecular 1,2-hydride shift.

Full text of DOI:10.1002/cssc.201800483

Accepted Article
Title: Mechanistic Studies of the Cu(OH)+-Catalyzed Isomerization of
Glucose into Fructose in Water
Authors: Joel Mensah, Irina Delidovich, Peter J.C. Hausoul, Laurent
Weisgerber, Wolfgang Schrader, and Regina Palkovits
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: ChemSusChem 10.1002/cssc.201800483
Link to VoR: http://dx.doi.org/10.1002/cssc.201800483
A Journal of
www.chemsuschem.org

ChemSusChem
10.1002/cssc.201800483
FULL PAPER
+
Mechanistic Studies of the Cu(OH) -Catalyzed Isomerization of
Glucose into Fructose in Water
[
a]
[a]
[a]
[a]
[b]
Joel B. Mensah , Irina Delidovich , Peter J. C. Hausoul , Laurent Weisgerber , Wolfgang Schrader ,
[
a]
and Regina Palkovits*
Abstract: The isomerization of glucose to fructose is a crucial
interim step in the processing of biomass to renewable fuels and
chemicals. This study investigates the copper catalyzed glucose-
fructose-isomerization in water focusing on insights into the roles of
dissolved copper species. Depending on the pH, the thermodynamic
equilibrium is shifted towards one or a few copper species namely
of fructose remains challenging as several drawbacks result
from the use of an immobilized D-xylose keto-isomerase.
Application of this bio-catalyst requires a highly pure feed, the
employment of buffer solutions, careful temperature control, and
periodic catalyst replacement making the process uneconomic in
[
11,12]
terms of biomass-valorization.
In this regard, more stable
2+
+
Cu , Cu(OH) , and Cu(OH)
2
. According to thermodynamics, the
chemo-catalysts can give a remedy as they potentially function
over a greater range of reactant purity, temperature, and pH.
This enables application of non-ambient conditions which are
thermodynamically more favorable for higher fructose-yields.
Thus, recently a great number of reports have focused on
utilizing non-enzymatic catalysts for the production of fructose
+
highest concentration of Cu(OH) is found at pH 5.3 for which also
the highest fructose yield of 16% is achieved. The obtained fructose
+
yields strongly correlate with the concentration of Cu(OH) . A pH
decrease by 2-3 units was observed during the reaction resulting in
the deactivation of the catalyst through hydrolysis in acidic media.
Based on the results of the catalytic experiments, spectroscopic and
[10,13–24]
starting from glucose.
The catalytic potential of Lewis acids in solely aqueous phase
+
spectrometric studies we propose Cu(OH) as active Lewis acidic
[
25]
species following an intramolecular 1,2-hydride shift.
was discovered in 2001 by Vaccaro et al.
Lewis acids
comprise one class of catalysts which have shown high activity
for aldose-isomerization and ever since various studies for
glucose-isomerization over soluble and solid Lewis acid
Introduction
[18,26–30]
catalysts in water have followed.
One huge asset is the
potential compatibility of Lewis acids with their Brønsted
counterparts. It allows for a process design in which glucose is
isomerized to fructose first by the Lewis acid upon further
fructose-dehydration to HMF by a Brønsted acid.
Focusing on soluble Lewis acids, metal chlorides are the most
In order to access the full potential of biomass as renewable
carbon source thoroughly, efficient ways, involving the use of
suitable catalysts, are required. A key-intermediate step in
biomass-valorization is the isomerization of glucose into
[5]
[
1]
fructose. Fructose is a principal substrate for the production of
-hydroxymethylfurfural (HMF) and further upgrading into
platform chemicals that can serve as basis for fuels,
prevalent class owing to their catalytic activity of their metal
5
[28,29,31,32]
cations which can host electron pairs.
Recently, Tang et
, CrCl and SnCl
al. have investigated the performance of AlCl
3
3
4
[2–7]
commodities, and fine chemicals.
Remarkably, fructose, in
in the production of fructose starting from aqueous solutions of
aqueous solution, is present as a mixture of pyranose and
[29]
4
.5 wt.% glucose at 120 °C. These metal chlorides were found
[
8]
furanose forms.
The latter are believed to be of great
to be catalytically active reaching up to 26% fructose yield in the
importance in the synthesis of HMF. A much higher share of the
furanose forms for fructose than for glucose (ca. 30% vs. <1% in
aqueous solutions) ensures significantly higher yields of HMF
+
case of AlCl
3 2
, where [Al(OH) ] was identified as active species
for glucose isomerization. In a previous study, Sandler et al.
2+
2 5
have reported mono-hydrolyzed [Cr(H O) (OH)] as catalytic
[
4,8]
when using fructose over glucose as substrate.
[32]
center for glucose conversion in aqueous medium.
Currently, the glucose-fructose isomerization is already of great
industrial importance in the food sector, for the production of
Besides water, several studies for glucose-isomerization in
organic solvents e.g. in methanol, DMF, DMSO and mixtures of
[
9,10]
high fructose corn syrup as sweetener or additive.
The
the latter with ethylene- and propylene glycol have been carried
industrial bio-technical process provides a fructose yield of 42%
[17,19,24,33]
out.
Moreover, catalytic activity of various metal
[11]
at a temperature of 45 °C. However, the enzymatic production
chlorides (most notably chromium) in different 1-alkyl-3-
methylimidazolium-based ionic liquids was shown for the
efficient isomerization of glucose to fructose and subsequent
[
a]
J. B. Mensah, Dr. I. Delidovich, Dr. P. J. C. Hausoul, L. Weisgerber,
Prof. Dr. R. Palkovits
Institut für Technische und Makromolekulare Chemie (ITMC)
RWTH Aachen University, 52074 Aachen (Germany)
Worringerweg 2
E-mail: palkovits@itmc.rwth-aachen.de
Prof. Dr. W. Schrader
Max-Planck-Institut für Kohlenforschung
Kaiser-Wilhelm-Platz 1, 45470 Mülheim an der Ruhr (Germany)
[34]
dehydration to HMF.
High activity of copper acting as Lewis
acid in glucose isomerization in ionic liquids was carefully
[
35,36]
studied by Pidko et al.
More recently, Yepez et al. reported
catalytic activity of copper-containing solid catalysts for the
microwave assisted production of HMF and 5-methylfurfuryl
alcohol in aqueous phase via glucose-fructose isomerization as
[
b]
[
37]
initial step. However, the mechanism of the isomerization over
2
+
soluble Cu , specifically the nature of the active species in
aqueous media, remains an open question.
Supporting information for this article is given via a link at the end of
the document.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800483
FULL PAPER
The reported fructose yields in aqueous solutions typically do
[
10,16]
not exceed 30%.
The yields are limited by the reversibility of
the isomerization reaction, the reduced product stability, and
other peculiarities of the reaction kinetics which have not been
well understood yet. Thus, further research on novel catalytic
systems as well as the prevailing mechanisms is required in
order to promote a more economic production of fructose for
non-food usage based on abundantly available bio-derived
[
38]
glucose.
As water is a green solvent and the most common medium for
the production of sugars starting from biomass, herein the
copper catalyzed isomerization of glucose into fructose in
aqueous solutions is studied focusing on insights into the roles
of dissolved active copper species.
Results and Discussion
The isomerization of glucose to fructose was studied over
various copper-containing solid and molecular catalysts in
aqueous phase. The tested solid materials were commercially
available Raney-copper, CuO, and Cu
exchanged Cu-SAPO-34 zeolite. 5 wt.% CuO and Cu supported
on Al , ZrO , and 25 wt.%-Al /75 wt.%-SiO were prepared
2
O along with an ion-
2
O
3
2
O
2 3
2
via impregnation and characterized by means of nitrogen
physisorption and TPR (Figure S1 and Table S2, Supporting
Info). Further, Cu(NO ) ·3H O, hereafter referred to as Cu(NO ) ,
3 2 2 3 2
was employed as molecular catalyst.
2
A blank reaction performed at 110 °C and 30 bar N without the
Figure 1. a) Fructose yield over various copper catalysts with different
oxidation states; b) Fructose yields over Cu concentration in the liquid phase
after the reaction determined via ICP. 5 mL 10 wt.% aq. glucose, catalytic
use of a catalyst resulted in almost no glucose conversion (<1%),
which clearly shows the need for an efficient catalyst in the
production of fructose from glucose. The temperature of 110 °C
as well as a substrate solution of 5 mL and a catalyst mass of
2
amount equivalent to 4.2 mg Cu, 110 °C, 1.5 h, 30 bar N , 500 rpm.
1
00 mg, respectively, were chosen for an easier comparison of
Role of soluble Copper Species
[
39,40]
the results with the available literature data.
Preliminary
experiments were performed under autogenous pressure, but
evaporation of water out of the inlet and its re-condensation on
the upper part of the autoclave were observed. Application of 30
bars N enabled keeping water in the liquid state and avoiding
2
partial evaporation.
An ICP-OES-analysis of the respective reaction mixtures
presented in Figure 1a revealed the presence of copper in the
liquid phase after the reaction. The amounts of leached copper
were correlated to the obtained fructose yields as shown in
Figure 1b. It is demonstrated that catalytic activity is rather a
function of leached copper ions than of the oxidation state.
Increased catalytic activity was observed with higher amounts of
leached copper.
Studying the solid copper catalysts with different oxidation states
0
I
II
(
Cu , Cu, and Cu ) did not reveal any clear correlation between
the oxidation state and catalyst performance, as presented in
Figure 1a. Similar yields were found for Cu(NO , Cu O, and
3
)
2
2
Further, hot filtration tests were performed where the catalyst
was removed from the reaction mixture by filtration after 20 min
and the filtrate was monitored for further catalytic activity. The
results displayed in Figure 2 demonstrate continued catalytic
activity of the filtrate indicating the presence of leached Cu.
These findings in addition to the increased catalytic activity with
higher amounts of leached copper clearly suggest that the
reaction is catalyzed by soluble Cu species, which led us to
investigate the catalytic potential of molecular Cu-catalysts more
closely.
Raney which all contain Cu in different oxidation states.
Moreover, no consistency can be verified for only considering
one oxidation state. Here, very different catalytic activities were
found for Cu(NO
3
)
2
, CuO, and Cu-SAPO-34 which all feature
II
Cu .
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800483
FULL PAPER
Table 1. Kinetic data of the glucose-fructose-isomerization and final pH values
(
pH
f
). Starting pH 5.3, 5 mL, 1 wt.% aq. glucose, 16 mg Cu(NO
, 500 rpm. *: no glucose, only Cu(NO
3 2
) , 110 °C, 30
bar N
2
3 2
) .
Entry
Time / min
Fructose yield / %
pH
f
1
2
3
4
5
6
0
20
45
90
180
90
0
5
5.3
4.1
3.4
3.1
3.1
4.8
12
14
15
*
2
Figure 2. Hot filtration test for Raney-Cu and CuO/SiO . Catalyst filtered off
after 20 min and reaction continued for further 25 min. 5 mL 1 wt.% aq.
glucose, catalytic amount equivalent to 4.2 mg Cu, 110 °C, 30 bar N , 500 rpm.
Recent literature reports have shown that the catalytic activity of
certain molecular Lewis acids is strongly pH-dependent as they
2
3
+
are present as partially hydrolyzed metal cations (e.g. Al or
Table 1 summarizes the kinetic data of the glucose-fructose
isomerization catalyzed by soluble Cu(NO . The concentration
of Cu(NO was chosen in such a way that the charged amount
of copper corresponded to that of CuO/SiO described in the
3+ [29,32]
Cr ).
3 2
)
3
)
2
2+
+
Distribution of Cu , Cu(OH) , and Cu(OH)
2
2
previous section. The data show an increasing fructose yield up
to 90 min of reaction time. The prolongation of the reaction time
from 90 to 180 min did not result in a further increase of neither
fructose yield nor higher by-product formation was observed.
Thus, the deactivation of the active Cu species during the
reaction is suggested, as the thermodynamic equilibrium
2+
Copper is represented in an aqueous solution as [Cu(H
2
2+
6
O) ]
+
and [Cu(H
2
O)
Cu(OH) , respectively) as well as Cu(OH)
thermodynamic equilibrium and described by Equation (1) and
2).
5
(OH)] (further are referred to as Cu
and
+
2
. All the species are in
(
[
41]
predicts fructose yields around 50%. Noteworthy, the reaction
was accompanied by a decrease of the pH in the reaction
(
(
1)
2)
mixture to final pH values (pH
f
) as low as 3.1. This decrease of
the pH during the reaction can be partially explained by the
2
+
+
hydrolysis of Cu leading to the formation of H as shown in
Equations (1) and (2). The pH of the solution directly after
The respective equilibrium distribution graphically presented in
Figure 3 (detailed calculation provided in the Supporting Info).
Cu predominates at pH<4, whereas the highest concentrations
dissolution of Cu(NO
3
)
2
was 5.3 (Table 1, Entry 1). As recently
2+
reported by Norton et al., hydrolysis of salts can take a few
+
of Cu(OH) are at a pH of ca. 5. At pH>8 copper species nearly
hours at room temperature and is accelerated at elevated
2
completely precipitate as Cu(OH) as major species.
[
42]
temperature.
3 2
Heating the aqueous solution of Cu(NO )
without adding glucose indeed brought about a decrease of the
pH value to 4.8 (see Table 1, Entry 6). A more significant pH
lowering to ca. 3, observed upon glucose isomerization in the
3 2
presence of Cu(NO ) (Table 1, Entries 2-5), can be explained
by formation of by-products. The known decomposition
pathways of saccharides cover retroaldol cleavage, dehydration
[43]
and benzylic-acid rearrangement-like reactions.
Moreover,
dicarbonyl compounds formed upon decomposition are highly
reactive and are believed to be intermediates for acidic
[
44]
oligomeric by-products.
Products of retroaldolization such as
glycol aldehyde, glyceraldehyde and 1,3-dihydroxyacetone were
indeed detected in minor amounts (Table S5, Supporting Info).
Most probably, the same species as for glucose-fructose
isomerization are active for retroaldol cleavage or benzylic-acid
rearrangement-like reactions, since these processes are known
2+
+
Figure 3. Equilibrium distribution for Cu
2
, Cu(OH) , and Cu(OH) as a
[
45,46]
to be catalyzed by Lewis acids.
However, a significance of
function of pH in an aqueous Cu(NO solution.
3 2
)
+
[45]
H , e.g. for dehydration processes, cannot be excluded.
For the catalysis, the initial total concentration of copper species
was 13.4 mM and the pH value was changing during the
experiment.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800483
FULL PAPER
These estimations show that the total concentration of soluble
copper species in solution significantly decreases at pH higher
than 5 due to precipitation as Cu(OH) .
2
Decreasing the pH
0
from 5.3 to 1.0 leads to a significant loss in
the yield of fructose from 14 to 0%. This finding can be
explained by the corresponding lower initial concentration of
+
Cu(OH) which is the proposed active species. Noteworthy, a
change of the pH occurred during the reaction as the pH
f
at the
Influence of starting pH on fructose yield
end of the reaction was found to be lower than pH (see Table 2).
0
Furthermore, as presented in Table 1 it could be verified that the
pH in the reaction solution is a function of the applied reaction
time. Hence, the Cu(OH) concentration does not stay constant
Thus, in order to gain further insight into the nature of the
catalytically active Cu species, besides potential catalyst
deactivation in acidic milieu, the influence of the pH was
+
throughout the reaction time. Thus, in order to account for the
examined more closely. For this purpose, the influence of pH
the final fructose yield was examined, by adjusting the pH
dropwise addition of either HNO
0
on
by
observed pH decrease during the reaction using an average
+
Cu(OH) concentration (based on the mean of pH
0
and pH
f
) is
0
more accurate for the further evaluation. Figure 4 correlates the
obtained fructose yields with the averaged concentration of
3
or NaOH solutions to the
reaction solutions prior to the reaction. The catalytic activity of
Brønsted species for aldose-ketose-isomerization has been
widely researched and only Brønsted bases were found
+
Cu(OH) . The data emphasize that increased fructose yields
+
agree well with higher concentrations of Cu(OH) . This direct
[
16,21,47,48]
+
active;
accordingly no catalytic activity of HNO
3
is
relation between the amount of Cu(OH) and the increased
expected. A catalytic activity of NaOH is also rather unlikely
since it was instantly neutralized upon addition to the solution
catalytic activity strongly supports its proposed role as active site.
For average Cu(OH) concentrations below 2 mM a linear
+
containing Cu(NO
acidic (pH 1.0) to slightly acidic (pH 5.5).
3
)
2
. The initial pH value was thus varied from
dependence could be verified. For concentrations higher than 2
mM the change in the fructose yield is less distinct. However,
The experimental results summarized in Table 2 show a
dependence of the glucose conversion and fructose yields on
here it should be noted that the difference between pH
0 f
and pH
tends to be more pronounced and the applicability of the
+
the starting pH. The experiments were conducted for 90 minutes, estimation for an averaged Cu(OH) concentration potentially
since the first results summarized in Table 1 showed no
significant improvement in fructose yield for longer reactions.
Nearly all the experiments were repeated twice. The pH values
of the solution after the reaction deviated by 0.1 pH units at
reaches its limit. Noteworthy, a rather high selectivity to fructose
mainly in the range of 70-80% was observed (Table 2). Though
pH decreased in this series of experiments, no by-products
could be detected by HPLC. Most probably, a rapid formation of
the aforementioned oligomeric acidic by-products took place.
2
+
maximum. The data reveal that lower pH values, i.e. when Cu
species predominate, lead to a decreased fructose yield. The
best results were obtained for initial pH of ca. 5, which suggests
+
Cu(OH) as catalytically active species.
+
Table 2. Fructose yields over varied pH
(
0
and initial Cu(OH) concentrations (c
0
+
CuOH) ) accordingly. 5 mL 1 wt.% aq. glucose, 16 mg Cu(NO
3 2
) , 110 °C, 30
bar N
, 1.5 h, 500 rpm. pH adjusted with NaOH (Entry 1), HNO (Entries 3-11).
2 0 3
+
Entry
c
0
(CuOH) ) /
mM
Glucose
conversion / %
Fructose
yield / %
pH
0
pH
f
1
2
3
4
5
6
7
8
9
11.0
11.3
11.2
9.21
5.66
2.54
18
15
17
14
14
12
9
16
14
12
11
13
11
6
5.5
5.3
5.0
4.5
4.0
3.5
3.0
2.5
2.0
1.5
1.0
2.6
3.1
3.0
2.9
2.7
2.9
2.9
2.4
1.9
1.3
1.0
+
Figure 4. Fructose yield over averaged Cu(OH) concentration (mean of c
0
+
+
(
Cu(OH) ) and cfinal (Cu(OH) )); 5 mL 1 wt.% aq. glucose, 16 mg Cu(NO
3 2
) ,
1
10 °C, 30 bar N , 1.5 h, 500 rpm.
2
ESI-MS was employed in order to investigate the interactions
between different Cu species and glucose. The resulting
spectrum of glucose and Cu(NO ) dissolved in water is
3 2
displayed in Figure 5 (for magnifications see Figure S5 and S6,
Supporting Info). It reveals the presence of the single charged
ion (m/z = 99). Moreover, the signal at m/z = 260
confirms coordination of glucose to mono-hydrolyzed Cu(OH)
6 12 6
leading to the formation of a [Cu(OH)(C H O )] -complex.
Further single charged copper-glucose complexes appear at m/z
-
-
-
-
-
1
1
2
2
3
9.24 × 10
3.07 × 10
9.86 × 10
3.13 × 10
9.93 × 10
4
3
3
1
+
[
Cu(H
2
O)
2
]
+
1
0
1
1
0
+
1
1
0
ratios of 242, 422, and 602 which can be assigned to
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800483
FULL PAPER
+
[
Cu(C
6
H
12
O
6
)
x
-H] (x = 1-3). Additionally, the presence of double
charged copper-glucose complexes of the structure
2
+
[
Cu(C
6
H
12
O
6
)
x
]
(with x = 3-6) is evidenced. The corresponding
signals are found at m/z = 301, 391, 481, and 571 respectively.
Figure 6. Blue spectrum shows the reaction of glucose-2d to fructose-1d in
D
2
O. Red spectrum shows the reaction of D-glucose to D-fructose in D
2
O. 10
mL of 10 wt.% aq. glucose, 21 mg CuO, 140 °C, 30 bar N , 1 h, 750 rpm.
2
The results suggest that no deuterium incorporation takes place
1
in this case, since no triplet signal was observed in the
H
1
3
Figure 5. ESI-MS spectrum (positive mode) of 1 wt.% aqueous glucose and
decoupled C spectrum. The last observation also enables
exclusion of a mechanism through an ene-diol formation, since
3 2
13.4 mM Cu(NO ) .
+
deuterium incorporates from D
2
O into the product in this
Our results suggest that Cu(OH) is the catalytically active
species for the isomerization with deactivation occurring during
the reaction due to lowering of the pH caused by the formation
of acidic by-products.
[
48,49]
case.
Thus, glucose isomerizes into fructose via an
intramolecular hydride shift from C-2 to C-1 atom in the
presence of copper catalysts.
The proposed overall mechanism for the copper-catalyzed
isomerization over Lewis acid sites is shown in Scheme 1. At
Proposed Mechanism
2
+
first, [Cu(H
species [Cu(H
2
O)
6
]
is transformed into the catalytically active
(OH)] , by the exchange of one water ligand
+
2
O)
5
Two mechanisms of glucose-fructose isomerization in the
presence of chemical catalysts were proposed in the literature,
namely a mechanism via an ene-diol formation
The former is typical of base
catalysts, whereas the latter is observed for the isomerization
catalyzed by Lewis acids. We assumed Cu(OH) to catalyze the
reaction as Lewis acidic species. This hypothesis was verified
using a labeled glucose-2d. A H-decoupled C NMR spectrum
of this substrate revealed characteristic triplets corresponding to
C-2 atoms of -D-glucopyranoside and -D-glucopyranoside at
with a hydroxide counterpart. Subsequently, glucose with its
carbonyl group at C-1 coordinates to the metal center replacing
one water ligand. Further, the deprotonation of the C-2 hydroxyl-
group takes place which is catalyzed by the hydroxide-ligand of
the active center. In the following the C-2-C-1-hydride shift takes
place and the reaction is completed by protonation of the oxygen
at C-1 by a water ligand on the metal center, releasing the
isomerized fructose molecule and the mono-hydrolyzed copper
complex. As the lowering of the pH during the reaction proceeds,
[
48,49]
and an
[
27,29]
intramolecular hydride shift.
+
1
13
+
the active Cu(OH) species is deactivated and forms the inactive
7
1.9 and 74.8 ppm respectively (blue spectrum, Figure 6).
These triplets were observed due to a one-bond C- D coupling
in the glucose molecules. After the isomerization of glucose-2d
1
3
2
hexa-aqua complex.
in D
2
O, a signal corresponding to C-1 of -D-fructopyranoside at
64.7 ppm appeared as a triplet (blue spectrum, Figure 6),
whereas all other signals of fructose were present as singlets. In
addition, a very low intensity of the C-1 -D-fructopyranoside
2
signal provides another evidence of its bonding to a D atom.
The signal intensity of the not deuterated fructose is rather high
owing to nuclear Overhauser effect, for example a resonance
signal at 64.5 ppm corresponding to
a C-6 atom -D-
fructopyranoside. This can be observed in the red spectrum in
Figure 6. This spectrum was obtained using unlabeled glucose
as a substrate in D O as a solvent.
2
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800483
FULL PAPER
Experimental Section
Materials
All chemicals were used without any further purification. Copper
nitrate (Cu(NO ·3H O, ≥98%), CuO, (99.99%), D O (99.9
atom%), D-(-)-fructose (Ph.Eur.), and D-(+)-glucose-2d were
purchased from Sigma Aldrich. HNO (65 wt.%), and NaOH,
8.8%) were obtained from GeyerChemSolute. Further, Cu
99.9%) was supplied by Alfa Aesar, D-(+)-glucose (Reag.
3
)
2
2
2
3
9
2
O
(
Ph.Eur.) purchased from Merck, and Raney-copper (aqueous
copper-slurry, 50 wt.%) obtained from abcr. Cu-SAPO-34 was
kindly supplied by Clariant.
Catalytic Tests
Catalytic tests were carried out in an autoclave with a glass inlet
at 110 °C using 5 mL of aqueous 1-10 wt.% glucose and the
respective amount of
autoclave was charged with 30 bar of N
a
copper-containing catalyst. The
at room temperature
2
prior to being heated to 110 °C. The reaction mixture was stirred
for 20-180 min at 500 rpm. After the reaction the 1 mL of the
mixture was diluted with water (1:10) and analyzed by HPLC.
pH-measurements were performed using a pH-meter by Hanna
Instruments (model HI9216, accuracy ±0.03 pH). Calibrations
were done with two buffer solutions (pH = 1.68 and 7.00) by
Hamilton. Adjustments of the starting pH were achieved by using
Scheme 1. Proposed catalytic cycle for the copper-catalyzed glucose
fructose-isomerization following a 1,2-hydride shift.
-
3 0 0
HNO in case of pH 5.0 and below and NaOH for pH 5.5.
HPLC
The HPLC method used in this work was similar to a procedure
Conclusions
[39,50]
described in earlier literature with some modifications.
The
In summary, we have studied the isomerization of glucose into
fructose over Cu(NO in aqueous phase. The catalyst was
reaction mixtures were analyzed using a Shimadzu Prominence
LC-2020 system equipped with two Organic Acid Resin columns
3 2
)
active for the isomerization, giving 14% fructose yield after 1.5 h
at 110 °C. However, deactivation of the copper catalyst under
the applied reaction conditions was observed: the fructose yield
was limited, whereas as high as 50% yield is thermodynamically
predicted Due to the formation of acidic by-products, a pH
decrease by 2-3 pH units was observed during the reaction.
Thus, the influence of the pH in the reaction was carefully
studied. Depending on the pH value, the thermodynamic
equilibrium is shifted towards one or a few copper species
(
CS-Chromatography, 100 mm × 8.0 mm and 300 mm × 8.0
mm) connected in series via a 6 port switching valve. The eluent
2 mM trifluoroacetic acid) was supplied at 40 °C with a flow rate
(
-1
of 1 mL min . The first 7 minutes after probe injection, the eluent
was supplied with pump A through two successively connected
columns and a refractive index (RI) detector. Subsequently, the
valve position was switched and elution continued for 18
minutes through the columns separately. The eluent was
pumped using pump A through the first column (100 mm × 8.0
mm), and a photodiode array (PDA) detector with a detection
wave length of 270 nm was used for the analysis of possible
furan-derivatives as by-products. Simultaneously, the eluent was
pumped through the second column (300 mm × 8.0 mm) and the
RI detector using pump B, in order to analyze the
monosaccharides and by-products, namely glycolaldehyde,
glyceraldehyde, 1,3-dihydroxyacetone, and lactic acid. The
detection of possible mannose formation was achieved using the
following HPLC method: the samples were injected into a
2
+
+
2
namely Cu , Cu(OH) , and Cu(OH) . The obtained fructose
+
yield correlates with the concentration of Cu(OH) , suggesting
the latter as active Lewis acidic species. According to the
thermodynamics, the highest concentration of Cu(OH) is
achieved at pH ca. 5.3-5.5 which also gave the highest fructose
yield.
+
With the herein conducted catalytic, spectrometric, and
spectroscopic experiments we could show that only copper in
+
form of Cu(OH) is catalytically active in the isomerization of
glucose to fructose. We conclude that copper follows the same
catalytic mechanism as tin and chromium catalyzed reactions,
namely an intramolecular 1,2-hydride shift.
2
Phenomenex 5μ Luna-NH column (250 mm × 4.6 mm) and
eluted at 35 °C with a mixture of 80 vol.% acetonitrile and 20
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800483
FULL PAPER
-
1
vol.% water, feed with a flow rate of 1 mL min . The system was
equipped with an RI detector.
Alexander von Humboldt Foundation and the Bayer Foundation
is acknowledged by I. Delidovich. We thank Noah Avraham and
Heike Bergstein for performing HPLC and ICP-OES analyses as
well as Dirk Kampen for conducting the ESI-MS measurements.
Dr. Peirong Chen (Chair of Inorganic Chemistry and
Electrochemistry at RWTH Aachen University) and Clariant are
being thanked for supplying the Cu-zeolite material. We are
grateful to Mohammad G. Al-Shaal, Dr. Jonathan David, F.
Joschka Holzhäuser, and Marie M. Döring for their valuable time
for comments, discussions and input.
Electrospray Ionization-Mass Spectrometry (ESI-MS)
Measurements were carried out using the Q Exactive Plus
Thermo Fisher Scientific, Dreieich, Germany) equipped with an
electrospray ionization (ESI) source. The measured sample
contained a 1 wt.% aqueous glucose solution with 13.4 mM of
3 2
Cu(NO ) and was diluted with methanol. Data were recorded at
(
a resolving power of 280,000 at m/z 200, scanning a mass
range of 50 to 750. ESI voltage was set at 3.5 kV with a transfer
capillary temperature of 200 °C.
Keywords: • biomass conversion • catalysis • copper • glucose
isomerization • Lewis acid
Temperature-programmed reduction (TPR)
TPR was performed with a BET Pulsar TPR/TPD automated References
chemisorption analyzer (Quantachrome) equipped with
a
thermal conductivity detector (TCD). Approximately 0.05
g
[1]
C. Chatterjee, F. Pong, A. Sen, Green Chem. 2015, 17, 40–
71.
catalyst was used in an analysis and first pretreated at 300 °C
-1
(
10 °C min heating ramp) for 100 min under a constant helium
[
[
2]
3]
A. H. Jadhav, A. Chinnappan, R. H. Patil, S. V. Kostjuk, H.
Kim, Chem. Eng. J. 2014, 243, 92–98.
-1
flow of 40 mL min in order to remove any contaminant e.g.
adsorbed water. A cooling trap filled with ice was placed
between the sample cell with and the TCD to ensure no water
produced during the reduction distorts the results. After
subsequent cooling down to room temperature the feed gas was
Y. Román-Leshkov, J. N. Chheda, J. A. Dumesic, Science
2006, 312, 1933–7.
[
[
4]
5]
B. F. M. Kuster, Starch - Stärke 1990, 42, 314–321.
A. A. Rosatella, S. P. Simeonov, R. F. M. Frade, C. A. M.
Afonso, Green Chem. 2011, 13, 754–793.
-
1
switched to 5% hydrogen in argon supplied at 23.8 mL min .
Before the TPR program was initiated the establishment of a
stable TCD-signal was ensured. TPR curves were taken in a
[
[
6]
7]
A. B. Rao, E. S. Rubin, Environ. Sci. Technol. 2002, 36,
4467–4475.
-1
range from 40 to 500 °C using a 10 °C min heating ramp.
I. Delidovich, K. Leonhard, R. Palkovits, Energy Environ.
Sci. 2014, 7, 2803–2830.
ICP-OES
[8]
[9]
H. E. van Dam, A. P. G. Kieboom, H. van Bekkum, Starch -
Stärke 1986, 38, 95–101.
ICP-OES analyses were carried out using the ICP Spectroflame
D from Spectro. Liquid samples were acidified prior to detection.
Solid samples (approx. 30 mg) dissolved in a mixture of 2 mL
conc. sulfuric acid, 8 mL 40 wt.% hydrofluoric acid and 40 mL
water.
J. N. Chheda, G. W. Huber, J. A. Dumesic, Angew. Chem.
2007, 119, 7298–7318; J. N. Chheda, G. W. Huber, J. A.
Dumesic, Angew. Chem. Int. Ed. Engl. 2007, 46, 7164–83.
H. Li, S. Yang, S. Saravanamurugan, A. Riisager, ACS
Catal. 2017, 7, 3010–3029.
[
[
[
10]
11]
12]
S. H. Bhosale, M. B. Rao, V. V Deshpande, Microbiol. Rev.
1996, 60, 280–300.
NMR
K. N. Allen, A. Lavie, G. K. Farber, A. Glasfeld, G. A.
Petsko, D. Ringe, Biochemistry 1994, 33, 1481–1487.
M. Liu, S. Jia, C. Li, A. Zhang, C. Song, X. Guo, Chinese J.
Catal. 2014, 35, 723–732.
1
3
The NMR spectra were obtained with a quantitative C-NMR
using a BRUKER Avant 600 spectrometer (600 MHz) with a BBI-
probe head.
[13]
14]
[
J. Lecomte, A. Finiels, C. Moreau, Starch - Stärke 2002, 54,
7
5–79.
I. Delidovich, R. Palkovits, Green Chem. 2016, 18, 5822–
830.
Acknowledgements
[15]
[16]
[17]
[18]
5
J. B. Mensah and R. Palkovits acknowledge financial support by
the European Regional Development Fund (ERDF) and the
state of North-Rhine Westphalia, Germany under the operational
program ‘’Regional Competitiveness and Employment’’ Project
I. Delidovich, R. Palkovits, ChemSusChem 2016, 9, 547–
561.
S. Yu, E. Kim, S. Park, I. K. Song, J. C. Jung, Catal.
Commun. 2012, 29, 63–67.
‘
’Sustainable Chemical Synthesis’’. Financial support of the
M. Moliner, Y. Roman-Leshkov, M. E. Davis, Proc. Natl.
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800483
FULL PAPER
Acad. Sci. 2010, 107, 6164–6168.
ChemSusChem 2014, 7, 1833–1846.
[
[
19]
20]
S. Saravanamurugan, M. Paniagua, J. A. Melero, A.
Riisager, J. Am. Chem. Soc. 2013, 135, 5246–5249.
W. R. Gunther, Y. Wang, Y. Ji, V. K. Michaelis, S. T. Hunt,
R. G. Griffin, Y. Román-Leshkov, Nat. Commun. 2012, 3,
[44]
[45]
J. M. De Bruijn, A. P. G. Kieboom, H. Van Bekkum, J.
Carbohydr. Chem. 1986, 5, 561–569.
J. Dijkmans, M. Dusselier, D. Gabriëls, K. Houthoofd, P. C.
M. M. Magusin, S. Huang, Y. Pontikes, M. Trekels, A.
Vantomme, L. Giebeler, et al., ACS Catal. 2015, 5, 928–
940.
1109.
[
21]
S. J. Angyal, in Glycosci. Epimerisation, Isomerisation
Rearrange. React. Carbohydrates (Ed.: A.E. Stütz),
Springer, Berlin/Heidelberg, Germany, 2001, pp. 1–14.
Q. Yang, S. Zhou, T. Runge, J. Catal. 2015, 330, 474–484.
C. Moreau, R. Durand, A. Roux, D. Tichit, Appl. Catal. A
Gen. 2000, 193, 257–264.
[46]
Q. Guo, F. Fan, E. A. Pidko, W. N. P. Van Der Graaff, Z.
Feng, C. Li, E. J. M. Hensen, ChemSusChem 2013, 6,
1352–1356.
[
[
22]
23]
[47]
[48]
[49]
[50]
C. Kooyman, K. Vellenga, H. G. J. De Wilt, Carbohydr. Res.
1977, 54, 33–44.
[
[
[
[
24]
25]
26]
27]
G. Lee, Y. Jeong, A. Takagaki, J. C. Jung, J. Mol. Catal. A
Chem. 2014, 393, 289–295.
G. De Wit, A. P. G. Kieboom, H. van Bekkum, Carbohydr.
Res. 1979, 74, 157–175.
F. Fringuelli, F. Pizzo, L. Vaccaro, J. Org. Chem. 2001, 66,
J. M. Carraher, C. N. Fleitman, J.-P. Tessonnier, ACS Catal.
2015, 5, 3162–3173.
4719–4722.
C.-C. Chang, Z. Wang, P. Dornath, H. Je Cho, W. Fan,
RSC Adv. 2012, 2, 10475–10477.
I. Delidovich, R. Palkovits, Catal. Sci. Technol. 2014, 4,
4322–4329.
Y. Román-Leshkov, M. Moliner, J. A. Labinger, M. E. Davis,
Angew. Chem. 2010, 122, 9138–9141; Y. Román-Leshkov,
M. Moliner, J. A. Labinger, M. E. Davis, Angew. Chem. Int.
Ed. Engl. 2010, 49, 8954–8957.
[
28]
V. Choudhary, S. H. Mushrif, C. Ho, A. Anderko, V.
Nikolakis, N. S. Marinkovic, A. I. Frenkel, S. I. Sandler, D. G.
Vlachos, J. Am. Chem. Soc. 2013, 135, 3997–4006.
J. Tang, X. Guo, L. Zhu, C. Hu, ACS Catal. 2015, 5, 5097–
[
[
[
[
[
[
[
[
[
[
29]
30]
31]
32]
33]
34]
35]
36]
37]
38]
5
103.
I. Delidovich, A. Hoffmann, A. Willms, M. Rose, ACS Catal.
017, 7, 3792–3798.
Y. Román-Leshkov, M. E. Davis, ACS Catal. 2011, 1,
566–1580.
2
1
V. Choudhary, A. B. Pinar, R. F. Lobo, D. G. Vlachos, S. I.
Sandler, ChemSusChem 2013, 6, 2369–2376.
S. Despax, B. Estrine, N. Hoffmann, J. Le Bras, S.
Marinkovic, J. Muzart, Catal. Commun. 2013, 39, 35–38.
H. Zhao, J. E. Holladay, H. Brown, Z. C. Zhang, Science
2007, 316, 1597–1600.
E. A. Pidko, V. Degirmenci, R. A. van Santen, E. J. M.
Hensen, Inorg. Chem. 2010, 49, 10081–10091.
E. A. Pidko, V. Degirmenci, E. J. M. Hensen,
ChemCatChem 2012, 4, 1263–1271.
A. Yepez, A. Pineda, A. Garcia, A. a. Romero, R. Luque,
Phys. Chem. Chem. Phys. 2013, 15, 12165–12172.
F. W. Lichtenthaler, in Biorefineries-Industrial Process.
Prod., Wiley-VCH, Weinheim, Germany, 2008, pp. 2–59.
I. Delidovich, R. Palkovits, J. Catal. 2015, 327, 1–9.
J. Dijkmans, D. Gabriëls, M. Dusselier, F. de Clippel, P.
Vanelderen, K. Houthoofd, A. Malfliet, Y. Pontikes, B. F.
Sels, Green Chem. 2013, 15, 2777–2785.
[
[
39]
40]
[
[
[
41]
42]
43]
K. Buchholz, J. Seibel, Carbohydr. Res. 2008, 343, 1966–
1979.
A. M. Norton, H. Nguyen, N. L. Xiao, D. G. Vlachos, RSC
Adv. 2018, 8, 17101–17109.
I. Delidovich, A. N. Simonov, O. P. Taran, V. N. Parmon,
This article is protected by copyright. All rights reserved.

ChemSusChem
10.1002/cssc.201800483
FULL PAPER
Entry for the Table of Contents (Please choose one layout)
Layout 1:
FULL PAPER
Brønsted-pHobia:
J. B. Mensah, I. Delidovich, P. J. C.
Hausoul, L. Weisgerber, W. Schrader,
R. Palkovits*
Catalytic experiments with different
starting pH reveal the influence of
the presence of Brønsted acid
species on the activity of Lewis
Page No. – Page No.
+
acidic Cu(OH) for glucose-fructose-
+
isomerization in aqueous medium. A
strong decline in activity is observed
for low pH due to the hydrolysis of
Cu(OH) to inactive Cu while a high
pH leads to precipitation of inactive
Mechanistic Studies of the Cu(OH) -
Catalyzed Isomerization of Glucose
into Fructose in Water
+
2+
2
Cu(OH) .
This article is protected by copyright. All rights reserved.