Selective Glucose-to-Fructose Isomerization in Ethanol Catalyzed by Hydrotalcites

Article abstract of DOI:10.1021/acscatal.8b05145

A combination of ethanol solvent and hydrotalcite (HT) with a Mg/Al ratio of 3:1 was effective for the isomerization of glucose to fructose in up to 56% yield with a high selectivity of 80%. The ethanol solvent shifts the isomerization equilibrium between glucose and fructose, which allows the fructose yield to exceed the upper limit of enzymatic isomerization conducted in water at modest temperatures (i.e., 50%). The HT catalyst maintained such high catalytic performance in repeated use at least three times via calcination and subsequent reconstruction of the original layered structure by the memory effect, which was operated in an aqueous solution containing ammonium carbonate. The results of catalyst screening, characterization, and isotope tracer analysis have revealed that the base sites of HT mainly contribute to its high catalytic activity for glucose-to-fructose isomerization.

Full text of DOI:10.1021/acscatal.8b05145

Subscriber access provided by OPEN UNIV OF HONG KONG
Article
Selective Glucose-to-Fructose Isomerization
in Ethanol Catalyzed by Hydrotalcites
Mizuho Yabushita, Natsumi Shibayama, Kiyotaka Nakajima, and Atsushi Fukuoka
ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.8b05145 • Publication Date (Web): 24 Jan 2019
Downloaded from http://pubs.acs.org on January 26, 2019
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 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 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.
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 42
ACS Catalysis
1
2
3
4
5
6
7
8
9
Selective Glucose-to-Fructose Isomerization in
Ethanol Catalyzed by Hydrotalcites
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
Mizuho Yabushita,†,§,‖ Natsumi Shibayama,†,‡,‖ Kiyotaka Nakajima,*,† and Atsushi Fukuoka*,†
†Institute for Catalysis, Hokkaido University, Sapporo, Hokkaido 001-0021, Japan
‡Division of Chemical Sciences and Engineering, Graduate School of Chemical Sciences and
Engineering, Hokkaido University, Sapporo, Hokkaido 060-8628, Japan
KEYWORDS. glucose isomerization, fructose, hydrotalcites, memory effect, solid base
catalysis, equilibrium, isotope tracer analysis
ACS Paragon Plus Environment
1

ACS Catalysis
Page 2 of 42
1
2
3
4
ABSTRACT
5
6
7
8
9
A combination of ethanol solvent and hydrotalcite (HT) with a Mg:Al ratio of 3:1 was effective
for the isomerization of glucose to fructose in up to 56% yield with a high selectivity of 80%.
The ethanol solvent shifts the isomerization equilibrium between glucose and fructose, which
allows the fructose yield to exceed the upper limit of enzymatic isomerization conducted in water
at modest temperatures (i.e., 50%). The HT catalyst maintained such high catalytic performance
in repeated use at least three times via calcination and subsequent reconstruction of the original
layered structure by the memory effect, which was operated in an aqueous solution containing
ammonium carbonate. The results of catalyst screening, characterization, and isotope tracer
analysis have revealed that the base sites of HT mainly contribute to its high catalytic activity for
glucose-to-fructose isomerization.
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. INTRODUCTION
Fructose is a well-known sweetener used as an alternative to conventional table sugar (sucrose)
because it is 1.8–2-fold sweeter than the latter,1–3 and it has also emerged as a key starting
compound in biorefineries for the synthesis of value-added chemicals such as 5-
hydroxymethylfurfural, 2,5-furandicarboxylic acid, levulinic acid, and lactic acid.4–7 The
industrial production of fructose has thus far relied on the enzymatic isomerization of glucose,
which is an abundant and inexpensive sugar that can be obtained via the hydrolysis of starch and
cellulose,8,9 to produce an aqueous mixture of glucose and fructose with their ratio of 1:1; this
sugar aqueous mixture has already been commercialized as so-called high-fructose corn syrup
ACS Paragon Plus Environment
2

Page 3 of 42
ACS Catalysis
1
2
3
4
5
6
7
8
9
(HFCS).1–3 However, enzymatic systems involve inevitable issues such as short lifetime of
enzymes and the requirement of strictly-controlled conditions including solvent and
temperature.2,10 In this context, the development of a fast and selective process for glucose-to-
fructose isomerization with an easily-separable and reusable heterogeneous catalyst is required.
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 1. Isomerization of glucose to fructose, along with side reactions typically observed in
the presence of base catalysts.
Glucose-to-fructose isomerization proceeds via the Lobry de Bruyn–Alberda van Ekenstein
(LdB–AvE) transformation initiated from α-hydrogen abstraction by a base.11 Homogeneous
bases such as alkali hydroxides have been reported to be effective for this reaction,11–13 while the
desire to achieve ease of catalyst separation and consecutive reuse has led to the development of
heterogeneous base catalysts such as alkaline-modified zeolites, Mg–Al mixed oxides,
metallosilicates, and silica-supported N-containing compounds.13,14 Although these reported
heterogeneous bases exhibit good fructose selectivity at glucose conversion below 40%, none of
them are comparable to enzymatic isomerization that selectively synthesizes fructose with an
ACS Paragon Plus Environment
3

ACS Catalysis
Page 4 of 42
1
2
3
4
5
6
7
8
9
almost quantitative yield (ca. 50%).2,15 The decrease in fructose selectivity with an increase in
glucose conversion is probably due to undesired side reactions, including retro-aldol reactions
triggered by bases (see Scheme 1).9,16,17
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 this study, we demonstrate a combination of ethanol solvent and hydrotalcite (HT), which
is a layered double hydroxide consisting of Mg and Al, to selectively accelerate glucose-to-
fructose isomerization, even at high glucose conversion. The motivation to use ethanol as a
reaction solvent originated from a previous report where the addition of ethanol into water was
demonstrated to shift the isomerization equilibrium between glucose and fructose to the latter.18
This change in equilibrium can be rationalized by solvation effect: given the fact that
solubility—which reflects solvation enthalpy—of fructose in ethanol at 293 K is ca. 2.8-fold
higher than that of glucose,19 fructose more readily undergoes the solvation by ethanol molecules
to be stabilized, compared to glucose. Such a shift in equilibrium could increase the maximum
fructose yield to over 50%, which is the upper limit under aqueous conditions at modest
temperatures.2,20 We also address methods to reuse HT by the memory effect for the
reconstruction of active HT for glucose-to-fructose isomerization.21 A mechanistic study was
also conducted using isotope tracer analysis to elucidate whether base sites in the HT catalysts
are responsible for the isomerization of glucose in ethanol.
2. EXPERIMENTAL SECTION
2.1. Synthesis of HTs. HTs consisting of Mg and Al with various molar ratios (i.e., 12:1,
7:1, 5:1, 4:1, 3:1, and 1:1) were synthesized by following the reported co-precipitation method.22
ACS Paragon Plus Environment
4

Page 5 of 42
ACS Catalysis
1
2
3
4
5
6
7
8
9
50 mL of aqueous solution A containing both x mol of Mg(NO3)2·6H2O (Kanto Chemical) and y
mol of Al(NO3)3·9H2O (Kanto Chemical), where x + y = 0.075, and 50 mL of aqueous solution
B containing 0.2 mol of (NH4)2CO3 (Wako Pure Chemical Industries) were prepared and charged
into separate syringes. Both the aqueous mixtures A and B were simultaneously added from each
syringe in a dropwise manner at a rate of 0.83 mL min-1 to 25 mL of distilled water heated at 338
K under vigorous stirring. 28.0–30.0 wt% of NH3 aqueous solution (Wako Pure Chemical
Industries) was added to the resulting white suspension until the pH became 7.6–8.0 (monitored
with a pH meter), and the suspension was further stirred vigorously at 338 K for 3 h. The
resulting mixture was filtered using a Büchner funnel with filter paper 4A. The white solid was
washed with at least 500 mL of distilled water and was subsequently dried in an oven at 373 K
for 18 h. The HTs thus synthesized were denoted as HT(x:y), where x:y represents the initial
molar ratio of Mg(NO3)2·6H2O to Al(NO3)3·9H2O.
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. Characterization of HTs. The synthesized HT materials were characterized using
powder X-ray diffraction (PXRD; Rigaku, Ultima IV, Cu Kα radiation). Acid-base titration was
also conducted to quantify base sites of the HTs that can function in ethanol. Briefly, 50 mg of
HT was dispersed in 1 mL of ethanol containing a small amount of phenolphthalein color
indicator and the suspension was stirred at room temperature for 1 h. The mixture was then
titrated with 3.5 mM of p-toluenesulfonic acid in ethanol until the color of suspension changed
from pink to white, the latter of which is the original color of the HTs.
2.3. Glucose Isomerization. Glucose isomerization was conducted in a pressure-resistant
glass vessel (Aceglass, 15 mL inner volume), into which 50 mg of glucose (Kanto Chemical), 50
mg of catalyst (descriptions of each catalyst tested in this study are summarized in Table S1 in
ACS Paragon Plus Environment
5

ACS Catalysis
Page 6 of 42
1
2
3
4
5
6
7
8
9
the Supporting Information), 7.0 mL of ethanol (Wako Pure Chemical Industries), and a stirring
bar were charged. For isotope tracer studies, the reactant used was glucose-2-d1 (Sigma-Aldrich).
After purging with Ar, the reactor was immersed in an oil bath equipped with a magnetic stirrer
under prescribed conditions, after which the reactor was cooled to room temperature in a cold
water bath. The supernatant was obtained via centrifugation and decantation. The solid residue
was washed with 2 mL of distilled water three times to extract any water-soluble compounds.
The washing solvent was separated from the solid via centrifugation and decantation and was
then mixed with the reaction mixture. The amounts of residual glucose and products in the
reaction mixture were quantified using high-performance liquid chromatography (HPLC;
Shimadzu, Prominence HPLC System, refractive index detector) with a SUGAR SH1011
column (Shodex, ø8.0 × 300 mm, mobile phase of distilled water at a flow rate of 0.5 mL min-1,
column temperature of 323 K), where an absolute calibration method was employed.
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 isotope tracer studies using glucose-2-d1 as a reactant, after the reaction, the solvent was
removed by rotary evaporation and the products were dissolved into 1 mL of distilled water, and
sodium acetate-d3 (Toronto Research Chemicals) was then added to the mixture as an internal
standard. In addition to HPLC analysis, the products with sodium acetate-d3 in water were
2
2
analyzed using H nuclear magnetic resonance spectroscopy (NMR; JEOL, JNM-ECX600, H
92.1 MHz). The reaction mixture dissolved in deuterium oxide (Acros Organics) was also
analyzed using proton-decoupled 13C NMR spectroscopy (13C 100 MHz) and distortionless
enhancement by the polarization transfer technique (DEPT).
2.4. Catalyst Regeneration via Memory Effect of HT and Reuse. After washing of the
solid residue after the reaction, the solid was calcined at 723 K for 7 h in air to remove organic
ACS Paragon Plus Environment
6

Page 7 of 42
ACS Catalysis
1
2
3
4
5
6
7
8
9
compounds deposited on the surface. The calcined material was treated at room temperature for
8 h in 1 mL of an aqueous solution containing 5 mM of ammonium salt ((NH4)2SO4,
(NH4)2HPO4, (NH4)2CO3, NH4NO3, or NH4Cl). When synthesizing HT that involves hydroxide
ions in the interlayer spaces, the calcined material was treated in 1 mL of water without
ammonium salt. The sample was then completely dried on a hot plate at 383 K. The dried
powder was employed for subsequent runs of glucose isomerization by the same procedure
described previously. Each regenerated catalyst was characterized by PXRD, high-precision
field-emission scanning electron microscopy (SEM; JEOL, JSM-7400F), and N2 sorption
measurement at 77 K (BEL Japan, BELSORP-mini, the range of relative pressure is from 0.05 to
0.995)
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. RESULTS AND DISCUSSION
3.1. Textural Properties of HTs. The crystalline structures of the HTs with different Mg:Al
ratios were characterized by PXRD measurement. The PXRD pattern of HT(3:1) in Figure 1
presents prominent diffraction peaks at 2θ of 11.3°, 22.8°, 34.9°, 39.5°, 46.7°, 60.8°, and 62.0°,
which are identical to those of typical HT materials previously reported.22–24 The interlayer
distance of HT(3:1) can be calculated from the diffraction at 11.3°, which originates from the
(003) plane of HT,22,23 to be 7.8 Å based on Bragg’s law. In the cases of higher Mg:Al ratio, (i.e.,
12:1, 7:1, 5:1, and 4:1), new peaks (e.g., 13.7°, 15.2°, 21.1°, and 30.7°) belonging to
Mg5(CO3)4(OH)2·4H2O (hydromagnesite)22,25 appeared along with the same peaks observed for
HT(3:1), which indicates that these HTs involve both HT and Mg5(CO3)4(OH)2·4H2O phases.
HT(1:1) has diffraction peaks at 2θ of 18.4° and 20.2° that originate from the MgAl2(OH)8
ACS Paragon Plus Environment
7

ACS Catalysis
Page 8 of 42
1
2
3
4
5
6
7
8
9
phase,22,26 in addition to those from the HT phase; HT(1:1) is thus composed of HT and
MgAl2(OH)8. This comparison indicates that a ratio of 3:1 is suitable for the synthesis of a well-
ordered HT material without impurity phases. Calcination of HT(3:1) at 803 K for 5 h (this
calcined material is denoted as HT(3:1)cal hereafter) resulted in the loss of the peaks derived from
HT(3:1), and instead afforded new and broad diffraction peaks centered at 35.4°, 43.2°, and
62.8°, which are attributed to a phase transition of the original HT phase to the MgO (periclase,
43.0° and 62.4°)27 and/or monoclinic MgAl2O4 spinel (36.6°, 44.5°, and 64.8°)28 phases. This
phenomenon is always observed for HT; calcination of HT above 673 K destroys its layered
structure and produces a mixture of MgO, Al2O3, and/or a Mg–Al mixed oxide, whereas the
original crystalline structure can be recovered with the resulting oxide in water, which is known
as the memory effect, as discussed later.29 Here, HT(3:1)cal was used as a reference catalyst for
glucose-to-fructose isomerization (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
ACS Paragon Plus Environment
8

Page 9 of 42
ACS Catalysis
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
Figure 1. PXRD patterns of various HT materials. Legends: black diamonds = HT; red stars =
Mg5(CO3)4(OH)2·4H2O (hydromagnesite); blue triangles = MgO (periclase) and/or monoclinic
MgAl2O4 spinel; purple circles = MgAl2(OH)8.
To elucidate the base sites of each synthesized HT that possibly function as active sites for
glucose isomerization, acid-base titration in ethanol was conducted with several color indicators.
None of the HTs changed the color of 2,4-dinitroaniline from yellow to violet, but all of them
including HT(3:1)cal changed the color of phenolphthalein from colorless to pink. These results
demonstrate that the pKa value of the base sites present in all the HTs is in the range of 9.3 to
15.0.30 The base site density was then quantified via titration using phenolphthalein (see Table 1),
whereby most of the base sites that phenolphthalein and p-toluenesulfonic acid molecules can
ACS Paragon Plus Environment
9

ACS Catalysis
Page 10 of 42
1
2
3
4
access in ethanol are estimated. The density increases with the Mg:Al ratio from 1:1 (68 µmol g-
5
6
7
8
1) to 3:1 (178 µmol g-1), but in turn decreases to 102 µmol g-1 upon further increase of the ratio to
12:1. This observed decrease in the density of base sites with the Mg:Al ratio can be rationalized
by the co-presence of the Mg5(CO3)4(OH)2·4H2O phase in HT(12:1), HT(7:1), HT(5:1), and
HT(4:1), because the control acid-base titration indicated that Mg5(CO3)4(OH)2·4H2O does not
contain base sites that can be detected by phenolphthalein in ethanol. Despite a high density of
HT(3:1)cal (207 µmol g-1), these base sites formed on MgO and MgAl2O4 are distinct from those
formed on the original HT(3:1) phase.
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 1. Textural Properties of Various HT Materials
Mg:Al ratio Involved phase(s)a
Base siteb /µmol g-1
12:1
7:1
5:1
4:1
3:1
3:1c
1:1
HT and Mg5(CO3)4(OH)2·4H2O (hydromagnesite)
102
156
122
109
178
HT and Mg5(CO3)4(OH)2·4H2O (hydromagnesite)
HT and Mg5(CO3)4(OH)2·4H2O (hydromagnesite)
HT and Mg5(CO3)4(OH)2·4H2O (hydromagnesite)
HT
MgO (periclase) and/or monoclinic MgAl2O4 spinel 207
HT and MgAl2(OH)8 68
b
aDetermined from PXRD patterns in Figure 1. Quantified via acid-base titration using
phenolphthalein as a color indicator. cHT(3:1) calcined at 803 K for 5 h (i.e., HT(3:1)cal).
3.2. Glucose Isomerization over HTs. We initially investigated catalytic activity of
HT(3:1), which consists of only the HT phase as indicated by PXRD measurement (see Figure 1
ACS Paragon Plus Environment
10

Page 11 of 42
ACS Catalysis
1
2
3
4
5
6
7
8
9
and Table 1), for glucose isomerization in ethanol under Ar at 363 K for 2 h (see Table 2). The
HT(3:1) catalyst consumed 61% of glucose and fructose was synthesized in 50% yield (entry 2),
the latter of which is 38-fold higher than that for the blank reaction (1.3%, entry 1). We simply
postulate that the base sites in HT(3:1) function as active sites for glucose isomerization, so that
the turnover number (TON) for fructose production can be estimated to be 16 on the basis of the
base sites shown in Table 1. The calculated fructose selectivity of HT(3:1) was 83%. Such high
fructose selectivity has been achieved probably due to mild reaction temperature (363 K) and a
lack of strong base sites in the HT material (i.e., the pKa value of base sites involved in HTs is
9.3–15.0, see Section 3.1), since undesired retro-aldol reaction of fructose is known to proceed
either at reaction temperatures of typically over 423 K31–33 or in the presence of strong base
catalysts such as calcined MgO,34 the pKa value of which is higher than 15.0.30 It should be noted
that, to the best of our knowledge, no chemical catalysts have been reported to show such a high
fructose selectivity at a high glucose conversion to date; in contrast to the results in this work,
high fructose selectivity can be achieved only when the glucose conversion is carefully
controlled below 40% in previous reports.12,13,35 The present results thus demonstrate that the
combination of HT(3:1) and ethanol solvent realizes a very efficient catalytic system, where the
choice of both catalyst and solvent is the key to simultaneous achievement of high fructose yield
and selectivity (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
In control reactions using solely MgO and γ-Al2O3 as catalysts, in contrast to HT(3:1),
neither MgO (entry 3) nor γ-Al2O3 (entry 4) produced a high yield of fructose (≤17%), and a
physical mixture of both also lacked catalytic activity (7.4% of fructose yield, entry 5). γ-Al2O3
has exhibited catalytic activity identical to another form of aluminum oxide, boehmite (γ-
AlO(OH)), as tested in the previous report, where the Lewis acid sites have been invoked as
ACS Paragon Plus Environment
11

ACS Catalysis
Page 12 of 42
1
2
3
4
5
6
7
8
9
active sites for the isomerization.36 Another Mg-containing material, Mg5(CO3)4(OH)2·4H2O,
which was detected in the HTs with the Mg:Al ratio of larger than 3:1 and did not have
detectable base sites, was almost inactive (entry 6). HT(3:1)cal consisting of MgO and/or
MgAl2O4 (see Figure 1 and Table 1) gave 23% yield of fructose (entry 7), regardless of an
almost 1.2-fold higher content of base sites compared to HT(3:1). These data indicate that
HT(3:1) is superior to MgO, Al2O3, and Mg–Al mixed oxide with respect to glucose-to-fructose
isomerization. Heterogeneous and homogeneous base catalysts were also examined for
comparison. Typical metal oxides known to exhibit basic (CaO) and amphoteric (TiO2, ZrO2,
and CeO2) properties were poor and non-selective catalysts (fructose yield and selectivity were at
most 23% and 59%, respectively, see entries 8–11). The same reaction conditions in the presence
of a homogeneous base consisting of either NaOH or NaOC2H5 converted ≥66% of glucose;
however, the fructose yield was less than 1%, possibly due to side reactions caused by their
excessively strong basicity (entries 12 and 13).
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
ACS Paragon Plus Environment
12

Page 13 of 42
ACS Catalysis
1
2
3
4
Table 2. Glucose Isomerization in Ethanol over Various Catalystsa
5
6
7
8
Product yieldd /%C
Conv.c
/%
Entry Catalystb
Frce
GLAf DHAg PVAh EtGlci
9
1
2
None
16
1.3 (8.0) 0.0
0.0
1.4
0.1
0.3
0.2
0.4
0.9
1.0
0.0
0.1
0.0
0.0
0.0
0.0
1.2
0.2
0.4
0.0
0.4
0.8
1.1
0.8
0.2
0.3
0.0
0.0
0.0
0.4
0.1
0.1
0.1
0.1
0.3
0.4
4.8
0.0
0.0
0.0
0.0
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
HT(3:1)
MgO
61
50 (83)
4.5 (25)
17 (60)
7.4 (46)
7.3 (51)
23 (61)
23 (59)
9.8 (28)
5.5 (30)
0.8
0.0
0.7
0.0
0.2
0.5
0.4
0.9
0.5
3
18
4
29
γ-Al2O3
MgO + γ-Al2O3
j
5
16
6
Mg5(CO3)4(OH)2·4H2O 14
k
7
HT(3:1)cal
CaO
38
38
35
19
16
66
98
8
9
TiO2
10
11
12
13
ZrO2
CeO2
1.3 (8.0) 0.0
0.7 (1.1) 0.0
0.0 (0.0) 0.0
NaOH
NaOC2H5
aReaction conditions: glucose 50 mg; catalyst 50 mg; ethanol 7.0 mL; Ar 1 atm; 363 K;
2 h. Descriptions of each catalyst are summarized in Table S1 in the Supporting
Information. Glucose conversion. Carbon-based yield. Fructose, where the values in
b
c
d
e
parentheses
represent
selectivity.
fGlyceraldehyde.
g1,3-Dihydroxyacetone.
hPyruvaldehyde. Ethyl α-D-glucopyranoside. Physical mixture of MgO and γ-Al2O3
i
j
(25 mg each). kHT(3:1) calcined at 803 K for 5 h.
Given the results that HT(3:1) exhibited the highest catalytic performance for glucose
isomerization (Table 2), the impact of the Mg:Al ratio of HT on the activity was then examined.
Under the same reaction conditions as those given in Table 2 (i.e., 363 K, 2 h), all HTs afforded
similar conversion of glucose (54–64%, entries 14–19, Table 3) as well as yield and selectivity
ACS Paragon Plus Environment
13

ACS Catalysis
Page 14 of 42
1
2
3
4
5
6
7
8
9
of fructose (44–50% and 67–84%, respectively), because all cases should reach the equilibrium
of isomerization between glucose and fructose. Therefore, the initial isomerization rates far away
from the isomerization equilibrium were investigated to make a fair comparison of the catalytic
activity of the HTs. As shown in Figure S1 in the Supporting Information, three HTs (i.e.,
HT(5:1), HT(4:1), and HT(3:1)) exhibited the steeper slopes, which correspond to initial
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
isomerization rates of 239, 226, and 203 µmol gcatalyst min-1 (Table 3), respectively, and the
-1
other three HTs gave similar reaction rates to each other (106–121 µmol gcatalyst min-1). While
the calculated turnover frequency (TOF) values of HT(5:1) and HT(4:1) are superior to that of
HT(3:1) as shown in Table 3, the weight-basis activity (= reaction rate) of HT(5:1) and HT(4:1)
is almost the same with that of HT(3:1). We can therefore simply rationalize that the HT phase
involved in HTs mainly contributes to their catalytic performance, because
Mg5(CO3)4(OH)2·4H2O as the impurity phase of HT(5:1) and HT(4:1) has no detectable base
sites and thus does not function as a catalyst or inhibitor for glucose isomerization (see entry 6 in
Table 2). Hereafter, we chose HT(3:1) as one of the most active catalysts for further study, in
order to simplify the catalytic system by using single-phase material.
ACS Paragon Plus Environment
14

Page 15 of 42
ACS Catalysis
1
2
3
4
Table 3. Glucose Isomerization in Ethanol over Various HT Materialsa
5
6
7
8
Initial isomerization ratee
TOFf
/h-1
71
Conv.b,c
/%
Frcb,d
Entry Catalyst
/µmol gcatalyst-1 min-1
/%C
9
14
15
16
17
18
19
HT(12:1)
HT(7:1)
HT(5:1)
HT(4:1)
HT(3:1)
HT(1:1)
54
46 (84)
50 (78)
48 (67)
44 (79)
50 (83)
46 (76)
121
113
239
226
203
106
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
64
43
61
118
124
68
66
61
60
94
aReaction conditions: glucose 50 mg; catalyst 50 mg; ethanol 7.0 mL; Ar 1 atm; 363
K. bReaction time was 2 h. cGlucose conversion. dFructose yield, where the values in
parentheses represent selectivity. Determined from the linear slope of each time
course for glucose isomerization, as shown in Figure S1 in the Supporting
Information. Turnover frequency calculated from initial isomerization rate divided
e
f
by base site density shown in Table 1.
The solvent effects on glucose isomerization were next investigated for the lead catalyst,
HT(3:1), and the results are summarized in Table 4. The reaction conducted in water gave a
lower fructose yield of 16% and selectivity of 57% (entry 20), relative to ethanol (see entry 2 in
Table 2). Other alcohols such as methanol (entry 21) and 1-propanol (entry 22) were also
examined; both alcohols gave better results than water, but slightly lower fructose yields than
ethanol. Conventional organic solvents used in the conversion of biomass-related compounds37
such as dimethylsulfoxide (entry 23) and N,N-dimethylformamide (entry 24) were not as good as
the alcohols investigated here. There could be multiple possibilities to rationalize the observed
solvent effects: (i) as previously reported for ethanol,18 alcohols shift the isomerization
equilibrium between glucose and fructose to the latter sugar, and such a shift in equilibrium
ACS Paragon Plus Environment
15

ACS Catalysis
Page 16 of 42
1
2
3
4
5
6
7
8
9
allows the fructose yield to be easily increased, and (ii) alcohols stabilize a transition-state
species, which leads to acceleration of the desired glucose-to-fructose isomerization with the
suppression of undesired side reactions. With respect to the latter possibility, in related work
investigating glucose isomerization via electronic structure calculations,38 OH moieties involved
in water molecules or at metal oxide surfaces have been reported to compensate and stabilize a
negatively-charged species formed in the transition state (see Section 3.4), which synergistically
promotes the reaction along with active sites.
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 4. Solvent Effect on Glucose Isomerization over HT(3:1)a
Product yieldc /%C
Conv.b
/%
Entry Solvent
Frcd
GLAe DHAf PVAg
20
21
22
23
24
Water
28
16 (57) 0.3
43 (78) 0.7
48 (86) 1.0
16 (65) 0.1
35 (73) 0.0
1.1
1.0
1.0
0.2
1.1
0.6
1.2
0.9
0.2
0.0
Methanol
55
1-Propanol
Dimethylsulfoxide
56
24
N,N-Dimethylformamide 48
aReaction conditions: glucose 50 mg; HT(3:1) 50 mg; solvent 7.0 mL; Ar 1
b
c
d
atm; 363 K; 2 h. Glucose conversion. Carbon-based yield. Fructose, where
e
f
the values in parentheses represent fructose selectivity. Glyceraldehyde. 1,3-
Dihydroxyacetone. gPyruvaldehyde.
The effects of both reaction time and temperature were explored further with the HT(3:1)
catalyst in the best solvent, ethanol (see Tables 2 and 4). Figure 2 shows time courses for glucose
isomerization conducted at 333, 363, and 393 K. The fructose yield at 363 K gradually increased
ACS Paragon Plus Environment
16

Page 17 of 42
ACS Catalysis
1
2
3
4
5
6
7
8
9
with the reaction time from 26% at 15 min to 50% at 2 h and then became constant (Figure 2B),
which indicates that the glucose-to-fructose isomerization reached its equilibrium at 2 h. At a
lower temperature of 333 K (Figure 2A), isomerization proceeded slowly and the fructose yield
was still 39% even at a reaction time of 24 h. In stark contrast, the time course at 393 K (Figure
2C) shows that a fructose yield as high as 56% with 80% selectivity can be achieved within 30
min, which is higher than 50% of the fructose yield obtained at 363 K for 2 h (Figure 2B). Upon
continuing the reaction longer than 30 min at 393 K, both the fructose yield and selectivity
decreased to ca. 50% and 60%, respectively, probably due to degradation of the produced
fructose. These data indicate that high-yielding, selective synthesis of fructose from glucose can
be accomplished within a short reaction time of 30 min at high 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
Figure 2. Time courses for glucose isomerization over HT(3:1) in ethanol at (A) 333 K, (B) 363
K, and (C) 393 K. Glc, Frc, and EtGlc represent yields of glucose, fructose, and ethyl α-
glucopyranoside, respectively, and C3 represents the total yield of glyceraldehyde, 1,3-
dihydroxyacetone, and pyruvaldehyde. Reaction conditions: glucose 50 mg; HT(3:1) 50 mg;
ethanol 7.0 mL; Ar 1 atm.
ACS Paragon Plus Environment
17

ACS Catalysis
Page 18 of 42
1
2
3
4
5
6
7
8
9
At the very end of this section, we also investigated the effect of initial glucose
concentration on the isomerization rate at 15 min, which is far from the time reaching
equilibrium (see Figure 2B), at 363 K by using the HT(3:1) catalyst in ethanol (Table S2 in the
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
Supporting Information). The rate of fructose production increased from 38 µmol gcatalyst min-1
to 291 µmol gcatalyst-1 min-1 upon an increase of initial glucose concentration from 7.9 mM to 198
mM. These results indicate that mass transfer limitation is not the case of this glucose-to-fructose
isomerization using HT(3:1) in ethanol.
3.3. Repeated Use of HT(3:1) via Regeneration Using Memory Effect. After any
reactions using the lead catalyst HT(3:1), the color of the solid changed from white to yellowish
or even brownish. This color change implies the formation and deposition of insoluble humin-
like compounds over the HT(3:1) surface, which often behave as catalyst inhibitors.39–41 Such
deposited organic species can be removed by simple calcination in air; however, in the present
case, the heat-treatment process is accompanied by phase transition of the highly active HT(3:1)
catalyst to the less active HT(3:1)cal (see Figure 1 and Tables 1 and 2). Given the fact that
calcined HT, which lacks a layered structure, returns to its original layered phase (i.e., HT) via
treatment in water in the presence/absence of salts (so-called memory effect),29 a two-step
regeneration process was employed for the repeated use of spent HT(3:1) in the isomerization of
glucose, which consisted of calcination and subsequent reconstruction via the memory effect.
Before the repeated use of HT(3:1) via regeneration, the effects of anion species employed
for the reconstruction process were examined. Anion species derived from salts are taken into the
ACS Paragon Plus Environment
18

Page 19 of 42
ACS Catalysis
1
2
3
4
5
6
7
8
9
interlayer spaces of the resultant HT to compensate the positive charges of double hydroxide
layers consisting of Mg and Al,42 and such dosed anions would affect the catalytic activity of the
reconstructed HTs. To this end, the HT(3:1) material calcined at 723 K for 7 h was treated in
aqueous solutions containing a variety of ammonium salts (i.e., (NH4)2SO4, (NH4)2HPO4,
(NH4)2CO3, NH4NO3, or NH4Cl).28,43 For the synthesis of HT containing hydroxide ions, the
calcined material was simply treated in pure water without ammonium salts.43 Although the dose
of OH– species actually requires strictly-controlled experimental procedure that completely keeps
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–
away the material from air in order to avoid CO3 contamination,44 the data of both
characterization and glucose isomerization indicate that even using our experimental procedure,
we can still see the difference between the HT reconstructed in (NH4)2CO3 aqueous solution and
that in pure water (vide infra). The PXRD patterns shown in Figure 3A indicate that all resultant
materials, except for the case of (NH4)2HPO4, have layered structures that are similar to that of
bare HT(3:1), while the interlayer distance of these materials depends on a class of anion dosed
(see Table S3 in the Supporting Information). Besides, in the PXRD patterns of the materials
reconstructed in (NH4)2SO4 and NH4NO3, the broad peak centered at ca. 20° was also observed
possibly as a result of the formation of MgAl2(OH)8 phase.22,26 Figure 3B summarizes the
isomerization results by using these reconstructed HTs and demonstrates their catalytic activity
to be significantly dependent on the type of anion species; among the ammonium salts tested, the
reconstruction of HT in an aqueous solution containing (NH4)2CO3 generates the most active
material that is comparable to bare HT(3:1).
ACS Paragon Plus Environment
19

ACS Catalysis
Page 20 of 42
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
Figure 3. Effects of anion species used for the reconstruction process of calcined HT(3:1). (A)
PXRD patterns of resultant materials treated in aqueous solutions containing each ammonium
salt and (B) catalytic performance of the materials for glucose-to-fructose isomerization. Frc
shows the yield of fructose and Others represents the total yield of by-products. Reaction
conditions: glucose 50 mg; catalyst 50 mg; ethanol 7.0 mL; Ar 1 atm; 363 K; 2 h.
The effects of anion species on catalytic activity could be rationalized by the textural
2–
2–
–
properties of each reconstructed HT material. Three poor catalysts (i.e., SO4 , HPO4 , and NO3
) consist of HT and other phases, with no observable phases for the case of HPO42– (Table S3 in
the Supporting Information), which leads to a decrease of fructose selectivity (≤52%). In contrast,
2–
the other three selective catalysts (i.e., CO3 , Cl–, and OH–) involve HT phase only. Although
the order of base site density is OH– > CO32– > Cl– (see Table S3 in the Supporting Information),
2–
that of catalytic activity based on fructose yield is CO3 > OH– > Cl– (Figure 3B). This
contradiction suggests that the catalytic activity of HTs is changed not only by base site density,
but also is influenced directly by anion species themselves taken into the interlayer of HTs. We
ACS Paragon Plus Environment
20

Page 21 of 42
ACS Catalysis
1
2
3
4
5
6
7
8
9
hypothesize that anion species might show different degree of interaction with transition-state
species as is the case for cellulose dissolution in ionic liquids, where the choice of anions
impacts on cellulose solubility due to different degree of interaction between anion species and
glucose unit involved in cellulose.45,46 Such interaction would lead to stabilization of transition-
state species, which results in promotion of glucose isomerization.
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
These results indicated the (NH4)2CO3-treatment should be employed for repeated use of
HT(3:1); therefore, two-step regeneration consisting of calcination and subsequent reconstruction
in (NH4)2CO3 aqueous solution was conducted for spent HT(3:1). The PXRD measurement
confirmed that reconstruction of the original layered structure of HT(3:1) was completed after
the two-step regeneration process (Figure 4A), and the morphology looked similar before and
after the regeneration process (SEM images are shown in Figure S2 in the Supporting
Information), while the Brunauer-Emmett-Teller (BET) surface area determined from N2
adsorption-desorption measurement gradually decreased upon repeating reaction and
regeneration (Table S4 in the Supporting Information). The reconstructed HT(3:1) maintained its
catalytic performance as almost constant at least three times (Figure 4B), in which the total TON
for fructose production was calculated to be as high as 46. HT(3:1) can be repeatedly used for
glucose isomerization via the two-step regeneration process without a decrease in activity. These
results also suggest that the isomerization mainly takes place between layers inside HT particles.
ACS Paragon Plus Environment
21

ACS Catalysis
Page 22 of 42
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
Figure 4. Repeated use of HT(3:1) for glucose isomerization after regeneration via calcination
and subsequent treatment in (NH4)2CO3 aqueous solution. (A) PXRD patterns of bare and
reconstructed HT(3:1) recorded before each reaction, and (B) results of glucose isomerization.
Frc shows the yield of fructose and Others represents the total yield of by-products. Reaction
conditions: glucose 50 mg; catalyst 50 mg; ethanol 7.0 mL; Ar 1 atm; 363 K; 2 h.
3.4. Mechanistic Study and Contribution of Base Sites to Glucose Isomerization. Since
HT(3:1) exhibits high catalytic performance in glucose isomerization (see Table 2), we examined
whether the base sites of HT(3:1) function as active sites for the reaction and compared the
reaction mechanism with those for control solid catalysts. Base-catalyzed glucose isomerization
is known to proceed via the LdB–AvE mechanism involving α-hydrogen abstraction from open-
chain glucose, which is known to be present in various solvents due to anomerization
equilibrium47,48 and has been widely accepted as a starting form in previous reports,11 as an
initial step (Figure 5A). This mechanistic insight motivated us to conduct an isotope tracer study
using glucose-2-d1 as a reactant, to elucidate whether base sites of HT(3:1) abstract deuterium
ACS Paragon Plus Environment
22

Page 23 of 42
ACS Catalysis
1
2
3
4
5
6
7
8
9
atom from the substrate. In the 2H NMR spectrum of the reaction mixture obtained from glucose-
2-d1 with the HT(3:1) catalyst (Figure 5B), only peaks attributed to glucose-2-d1 were observed,
while the formation of fructose was confirmed by separate HPLC analysis for the same mixture
(we observed 24% of fructose yield and 40% of glucose conversion (Table S5 in the Supporting
Information), which indicates that C2–H bond cleavage is the rate-determining step in this
reaction catalyzed by HT(3:1) in ethanol). Note that if fructose-1-d1 was present in the reaction
mixture as a product, its signal should be observed at 3.58–3.70 ppm.49 The absence of this
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
signal in the H NMR spectrum (Figure 5B) indicates that the produced fructose molecules lack
deuterium as a result of its abstraction by HT(3:1) and subsequent diffusion of abstracted
deuterium from the catalyst surface to the protic solvent (ethanol, in this study) via proton
exchange. The 13C NMR and DEPT spectra for the same reaction mixture dissolved in deuterium
oxide also led to the same conclusion (see Figure S3 in the Supporting Information). These
results clearly demonstrate that the HT(3:1) catalyst uses its base sites to promote glucose-to-
fructose isomerization. Note that another mechanism invoked for the glucose isomerization—
intramolecular 1,2-hydride shift driven by Lewis acids (see Scheme S1 in the Supporting
Information)15,50,51—can be ruled out; this mechanism transforms glucose-2-d1 into a mixture of
fructose-1-d1, glucose-1-d1, and unreacted glucose-2-d1 (see Scheme S2 in the Supporting
Information); however, such a mixture was not observed and only glucose-2-d1 was detected by
2H NMR spectroscopy (see Figure 5B).
ACS Paragon Plus Environment
23

ACS Catalysis
Page 24 of 42
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
Figure 5. (A) Glucose-to-fructose isomerization via the LdB–AvE mechanism catalyzed by
2
bases (:B) and (B) H NMR spectrum of reaction mixture synthesized from glucose-2-d1, along
with the spectrum for glucose-2-d1 dissolved in water (i.e., standard). It should be noted that the
proton (or deuterium) abstracted by base is diffused from the catalyst surface to protic solvent (in
this study, ethanol) via proton exchange, and therefore, the return of such abstracted species to
the substrate and/or product molecules hardly occurs. Reaction conditions: glucose-2-d1 50 mg;
HT(3:1) 50 mg; ethanol 7.0 mL; Ar 1 atm; 363 K; 2 h. The two NMR peaks at 3.16 ppm and
3.45 ppm appeared due to the presence of glucose anomers (i.e., α-glucose-2-d1 and β-glucose-2-
d1).
The reaction mechanism was in turn investigated in the same manner using solely MgO and
2
γ-Al2O3 as control catalysts, and the H NMR spectra for each reaction mixture are shown in
Figure S4 in the Supporting Information. Consistent with the case of HT(3:1) shown in Figure
5B, only peaks assigned to glucose-2-d1 were observed when using either MgO or γ-Al2O3. This
indicates that both catalysts did not synthesize fructose-1-d1 during the reaction, and furthermore,
that the base sites of these oxides abstracted deuterium from glucose-2-d1 to initiate
ACS Paragon Plus Environment
24

Page 25 of 42
ACS Catalysis
1
2
3
4
5
6
7
8
9
isomerization through the LdB–AvE mechanism, which is in contrast to the previous research
where Lewis-acid sites involved in another class of aluminum oxide (boehmite) were reported as
active sites for the isomerization.36 The spectroscopic data thus demonstrate that base sites are
solely responsible for glucose-to-fructose isomerization in all cases (i.e., HT(3:1), MgO, and γ-
Al2O3), while the performance is significantly dependent upon the nature of the base sites (see
Table 2)—base sites formed in the HT phase are the most suitable for the synthesis of fructose
from glucose.
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
4. CONCLUSIONS
We have successfully demonstrated the combination of ethanol solvent and HT to selectively
catalyze glucose-to-fructose isomerization, which enables to simultaneously achieve high
selectivity and yield of fructose. The use of ethanol as a reaction solvent allows the fructose yield
to be increased beyond 50%, which is the upper limit with enzymatic isomerization conducted at
modest temperatures in water.2,15 In addition to environmentally preferable property of ethanol,52
its use as a solvent is also possibly beneficial with respect to solvent removal after the reaction
due to its lower boiling point than water, which has been used for the conventional HFCS
production.1–3 Such easier purification process of fructose in the case of ethanol solvent would
enable simple reaction systems requiring single catalyst to further convert fructose to value-
added compounds; we note that although the direct production of value-added chemicals from
glucose has been investigated so far by using multiple catalysts at the same time,53 such
complicated catalytic systems sometimes conflict with each other and suffer from undesired side
ACS Paragon Plus Environment
25

ACS Catalysis
Page 26 of 42
1
2
3
4
5
6
7
8
9
reactions, to decrease both selectivity and yield of desired products. The HT(3:1) catalyst can be
used repeatedly for the isomerization via a two-step regeneration process consisting of
calcination to remove humin compounds deposited on the catalyst and successive treatment in an
(NH4)2CO3 aqueous solution to reconstruct the layered structure of HT(3:1). Our mechanistic
study has revealed that base sites involved in all three materials (i.e., HT(3:1), MgO, and γ-
Al2O3) function as active sites to accelerate glucose isomerization through the LdB–AvE
mechanism, where HT(3:1) exhibits the highest catalytic performance due to its high density of
base sites. Further precise control of the strength and density of base sites could lead to the
development of an active and selective catalyst suitable for the large-scale production of fructose
as a renewable feedstock in the future chemical industry.
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
ASSOCIATED CONTENT
The Supporting Information is available free of charge on the ACS Publications website at DOI:
10.1021/acscatal.XXXXXXX.
Descriptions of catalysts tested for glucose isomerization, time courses of glucose
isomerization over various HTs at 363 K, effect of initial glucose concentration on
isomerization rate, textural properties of regenerated HT materials, SEM images of
regenerated HT(3:1), BET surface area of reused HT(3:1), isotope effect on glucose
isomerization by HT(3:1), 13C NMR and DEPT spectra for the reaction mixtures
synthesized with HT(3:1), isomerization mechanism via intramolecular 1,2-hydride shift
2
catalyzed by Lewis acid, and H NMR spectra for the reaction mixtures synthesized with
MgO and γ-Al2O3 (PDF)
ACS Paragon Plus Environment
26

Page 27 of 42
ACS Catalysis
1
2
3
4
5
6
7
8
AUTHOR INFORMATION
Corresponding Authors
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
*E-mail: nakajima@cat.hokudai.ac.jp (K.N.).
*E-mail: fukuoka@cat.hokudai.ac.jp (A.F.).
Present Address
§Institute of Multidisciplinary Research for Advanced Materials, Tohoku University, Sendai,
Miyagi 980-8577, Japan.
Author Contributions
‖These authors contributed equally to this work.
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
This work was supported by JST ALCA Grant Number JPMJAL1309, Japan. We are grateful for
Open Facility of Hokkaido University, to allow us to conduct the SEM measurement on HT
materials.
REFERENCES
ACS Paragon Plus Environment
27

ACS Catalysis
Page 28 of 42
1
2
3
4
5
6
(1) Casey, J. P. High Fructose Corn Syrup. A Case History of Innovation. Starch/Stärke 1977,
29, 196–204.
7
8
9
(2) Bhosale, S. H.; Rao, M. B.; Deshpande, V. V. Molecular and industrial aspects of glucose
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
isomerase. Microbiol. Mol. Biol. Rev. 1996, 60, 280–300.
(3) White, J. S. Sucrose, HFCS, and Fructose: History, Manufacture, Composition,
Applications, and Production. In Fructose, High Fructose Corn Syrup, Sucrose and Health;
Rippe, J., Eds.; Humana Press: New York, 2014; pp 13–33.
(4) Corma, A.; Iborra, S.; Velty, A. Chemical Routes for the Transformation of Biomass into
Chemicals. Chem. Rev. 2007, 107, 2411–2502.
(5) Alonso, D. M.; Bond, J. Q.; Dumesic, J. A. Catalytic conversion of biomass to biofuels.
Green Chem. 2010, 12, 1493–1513.
(6) Dusselier, M.; Van Wouwe, P.; Dewaele, A.; Makshina, E.; Sels, B. F. Lactic acid as a
platform chemical in the biobased economy: the role of chemocatalysis. Energy Environ. Sci.
2013, 6, 1415–1442.
(7) Besson, M.; Gallezot, P.; Pinel, C. Conversion of Biomass into Chemicals over Metal
Catalysts. Chem. Rev. 2014, 114, 1827–1870.
(8) Suganuma, S.; Nakajima, K.; Kitano, M.; Yamaguchi, D.; Kato, H.; Hayashi, S.; Hara, M.
Hydrolysis of Cellulose by Amorphous Carbon Bearing SO3H, COOH, and OH Groups. J. Am.
Chem. Soc. 2008, 130, 12787–12793.
ACS Paragon Plus Environment
28

Page 29 of 42
ACS Catalysis
1
2
3
4
5
6
7
8
9
(9) Kobayashi, H.; Yabushita, M.; Komanoya, T.; Hara, K.; Fujita, I.; Fukuoka, A. High-
Yielding One-Pot Synthesis of Glucose from Cellulose Using Simple Activated Carbons and
Trace Hydrochloric Acid. ACS Catal. 2013, 3, 581–587.
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) Lee, G.; Jeong, Y.; Takagaki, A.; Jung, J. C. Sonication assisted rehydration of hydrotalcite
catalyst for isomerization of glucose to fructose. J. Mol. Catal. A: Chem. 2014, 393, 289–295.
(11) Speck, J. C., Jr. The Lobry De Bruyn-Alberda Van Ekenstein Transformation. In Advances
in carbohydrate chemistry; Wolfrom, M. L., Eds.; Academic Press: New York, 1958; Vol. 13, pp
63–103.
(12) Takagaki, A.; Nishimura, S.; Ebitani, K. Catalytic Transformations of Biomass-Derived
Materials into Value-Added Chemicals. Catal. Surv. Asia 2012, 16, 164–182.
(13) Delidovich, I.; Palkovits, R. Catalytic Isomerization of Biomass-Derived Aldoses: A
Review. ChemSusChem 2016, 9, 547–561.
(14) Lima, S.; Dias, A. S.; Lin, Z.; Brandão, P.; Ferreira, P.; Pillinger, M.; Rocha, J.; Calvino-
Casilda, V.; Valente, A. A. Isomerization of D-glucose to D-fructose over metallosilicate solid
bases. Appl. Catal. A: Gen. 2008, 339, 21–27.
(15) Moliner, M.; Román-Leshkov, Y.; Davis, M. E. Tin-containing zeolites are highly active
catalysts for the isomerization of glucose in water. Proc. Natl. Acad. Sci. U.S.A. 2010, 107,
6164–6168.
(16) De Wit, G.; Kieboom, A. P. G.; van Bekkum, H. Enolisation and isomerisation of
monosaccharides in aqueous, alkaline solution. Carbohydr. Res. 1979, 74, 157–175.
ACS Paragon Plus Environment
29