Quantum yields for sacrificial hydrogen generation from saccharides over a Pt-loaded TiO2 photocatalyst

Article abstract of DOI:10.1007/s11164-015-2253-5

Much attention is paid to sacrificial hydrogen generation from saccharides over a Pt-loaded titanium oxide (Pt/TiO₂) from the viewpoint of being a renewable energy alternative to petroleum-based fuels. Here, monosaccharides such as glucose, gluconic acid, galactose, sorbose, fructose, mannose, and disaccharides such as sucrose and cellobiose as well as related compounds such as ascorbic acid and sugar alcohols (xylitol and sorbitol) were used as sacrificial agents (R-H) in order to determine quantum yields (Pdbl) for hydrogen evolution. From double-reciprocal plots of Pdbl to the concentration of R-H, we obtained the limiting quantum yield in an infinite concentration of R-H. The limiting quantum yields were determined to be of similar values (0.058-0.168) for monosaccharides. Moreover, the rate constant (k _R) for the hydrogen abstraction by hydroxyl radical from saccharides were obtained from the double reciprocal plots. It was found that the k _R value of glucose was larger compared to galactose, mannose, sorbose, and fructose, which were the stereo and structural isomers of glucose.

Full text of DOI:10.1007/s11164-015-2253-5

Res Chem Intermed
DOI 10.1007/s11164-015-2253-5
Quantum yields for sacrificial hydrogen generation
from saccharides over a Pt-loaded TiO photocatalyst
2
1
1
2
Masahide Yasuda
•
Ryo Kurogi
•
Tomoko Matsumoto
Received: 30 May 2015 / Accepted: 28 August 2015
Ó Springer Science+Business Media Dordrecht 2015
Abstract Much attention is paid to sacrificial hydrogen generation from saccha-
rides over a Pt-loaded titanium oxide (Pt/TiO ) from the viewpoint of being a
2
renewable energy alternative to petroleum-based fuels. Here, monosaccharides such
as glucose, gluconic acid, galactose, sorbose, fructose, mannose, and disaccharides
such as sucrose and cellobiose as well as related compounds such as ascorbic acid
and sugar alcohols (xylitol and sorbitol) were used as sacrificial agents (R–H) in
order to determine quantum yields (U) for hydrogen evolution. From double-re-
ciprocal plots of U to the concentration of R–H, we obtained the limiting quantum
yield in an infinite concentration of R–H. The limiting quantum yields were
determined to be of similar values (0.058–0.168) for monosaccharides. Moreover,
the rate constant (k ) for the hydrogen abstraction by hydroxyl radical from sac-
R
charides were obtained from the double reciprocal plots. It was found that the kR
value of glucose was larger compared to galactose, mannose, sorbose, and fructose,
which were the stereo and structural isomers of glucose.
Keywords Hydrogen evolution Á Saccharides Á Pt/TiO Á Quantum yields Á Rate
2
constant for hydrogen abstraction by hydroxyl radical
Electronic supplementary material The online version of this article (doi:10.1007/s11164-015-
2253-5) contains supplementary material, which is available to authorized users.
&
Masahide Yasuda
yasuda@cc.miyazaki-u.ac.jp
1
2
Department of Applied Chemistry, Faculty of Engineering, University of Miyazaki, 1-1,
Gakuen-Kibanadai-Nishi, Miyazaki 889-2192, Japan
Center for Collaborative Research and Community Cooperation, University of Miyazaki, 1-1,
Gakuen-Kibanadai-Nishi, Miyazaki 889-2192, Japan
123

M. Yasuda et al.
Introduction
Saccharides are energy-storage substances. Plants collect sunlight energy through
photosynthesis and store it as a variety of polymeric saccharides. Polymeric
saccharides are easily converted into monomeric saccharides, which are then
changed into energy in all living organisms. Moreover, polymeric saccharides are
raw materials to produce ethanol in the brewing industry. Recently, bio-ethanol
productions from a variety of biomasses such as starch, lignocellulose, and algae
have received a great deal of interest as potential renewable energy alternatives to
petroleum-based fuels [1]. Bio-ethanol can be produced easily from glucose through
fermentation using yeast (Saccharomyces cerevisiae). However, it is not so easy to
produce ethanol from saccharides other than glucose, since they are not fermented
by yeast.
On the other hand, hydrogen production from biomasses through thermochem-
ical, biological, and photocatalytic processes are some of the most promising
approaches to bio-fuels [2, 3]. We have been interested in the photocatalytic
reforming of lignocellulosic rice straw [4], Napier grass, and Italian ryegrass [5]
through the enzymatic saccharification followed by sacrificial hydrogen generation
from saccharides over a Pt-loaded titanium oxide (Pt/TiO ). The photocatalytic
2
hydrogen generation from H O by Pt/TiO is initiated by charge-separation on TiO
2
2
2
under photoexcitation [6]. The electron reduced water to generate H on Pt while
2
hole were used to oxidize hydroxide to generate hydroxyl radicals. Recently, it has
been reported that the charge-separation is enhanced by plasmonic near-fields on
Au–TiO and TiO -coated metal nanoparticles even under visible-light irradiation
2
2
[
(
7, 8]. Moreover, it is well known that the use of electron-donating sacrificial agents
hole scavengers) remarkably accelerates the TiO -photocatalyzed hydrogen
2
evolution since the hydroxyl radical is consumed by the sacrificial agents [9].
Recently, we have found that sacrificial agents with all of the carbon attached
oxygen atoms such as glucose, xylose, and glycerol continued to serve as electron
sources until their sacrificial ability was exhausted [10]. All saccharides meet this
requirement. Therefore, the chemical yields for the H evolution from saccharides
2
should be almost 100 %, providing an efficient route to bio-fuels from saccharides.
However, quantum yields (U) were still unclear. Here we measured U for the
sacrificial hydrogen generation from a variety of saccharides (1) in order to evaluate
the activity of saccharides as sacrificial agents.
Experimental
Preparation of photocatalyst
According to the literature [11], a Pt-loading to an anatase-type of TiO (ST-01,
2
Ishihara Sangyo, Japan) was performed as follows: An aqueous solution (50 ml)
containing TiO (1.0 g), K PtCl (50 mg), and 2-propanol (0.38 ml) was introduced
into a reaction vessel. Since the band gap of anatase-type of TiO is known to be
2
2
6
2
1
23

Quantum yields for sacrificial hydrogen generation from...
3
.20 eV, which corresponded to 385 nm, the TiO can be excited by a 366-nm
2
emission from a high-pressure mercury lamp (100 W, UVL-100HA, Riko, Japan).
The lamp was inserted into the reaction vessel, which was set in a water bath at
room temperature. After the oxygen was purged by N gas, the solution was
2
irradiated for 24 h with stirring to produce the Pt-loaded TiO catalyst (Pt/TiO ).
2
2
After the irradiation, water was entirely removed from the photolysate by the
evaporator. The resulting precipitate was washed with water in a filter and then
dried under reduced pressure to produce Pt/TiO . The optimized Pt-content on TiO
2
2
was determined to be 2.0 wt% by the comparison of the amounts of H -evolution by
2
the Pt/TiO2 (100 mg) under irradiation for 6 h using glucose (100 mg) as a
sacrificial reagent. Details for the optimization are described in our previous
literature [12].
The structure of Pt/TiO was analyzed by a Shimadzu XRD 7000 diffractometer
2
using CuKa radiation (40 kV, 20 mA) from 5° to 90° of 2h in 0.02 steps [13].
Transmission electron microscope (TEM) images were taken with a JEOL-2010
microscope operated at 200 keV. TEM samples were placed on a micro-grid
(Nisshin EM, Co., 200 mesh). XRD pattern and TEM images were shown in Figs. 1
and 2, respectively.
Photocatalytic H generation
2
Saccharides (1) were purchased from Wako Chemicals as D-forms except for L-
sorbose. The Pt/TiO (100 mg, 1.25 mmol) was introduced to the reaction vessel,
2
which was attached to a mess-cylinder with a gas-impermeable tube to collect the
evolved gas. Aqueous solutions of 1 and alcohols (2) were added in reaction vessel
so that the amounts of 1 and 2 became a range of 0.086–0.389 mmol and then the
volume of the solution was adjusted to 150 ml by adding water. Initial pH became
slightly acidic (pH 4.5). The lamp was inserted into the reaction vessel and set in a
water bath to keep it at a constant temperature (usually 20 °C). After the oxygen
was purged by N gas, irradiation was performed with vigorous stirring using a
2
Fig. 1 X-ray diffraction of a Pt-
loaded TiO (2.0 wt% of Pt
2
content). Mark asterisk was the
peak for Pt
123

M. Yasuda et al.
2 2
Fig. 2 TEM images of a Pt–TiO (2.0 wt% of Pt content) and b anatase TiO (ST-01)
magnetic stirrer. The evolved gas was collected with a mess-cylinder to measure the
total volume of the evolved gas. The irradiation was performed until gas evolution
had ceased. The quantitative analysis of H , CO , and N were analyzed on a
2
2
2
Shimadzu GC-8A equipped with a TCD detector at temperature raised from 40 to
180 °C using a stainless column (3 mm U 9 6 m) packed with a SHINCARBON
ST (Shimadzu).
Measurement of quantum yield
Irradiation was performed for an aqueous solution (150 ml) containing 1
(
was attached to a gas-bullet. It was confirmed that only H had evolved, CO did not
8–40 mM) and the Pt/TiO (100 mg, 1.25 mmol) in the reaction vessel, which
2
2
2
evolve at low conversion. The evolved H per hour (mmol/h) was measured from
2
the slope of the H time-conversion plots until the conversion reached about 30 %.
2
Salas et al. [14] have reported that the quantum yield (U) for the H evolution was
2
determined to be 0.057 in the photocatalytic reaction by Pt-TiO (1.0 wt% of Pt
2
content) at pH 10.0 using ethanol (0.434 M) as sacrificial agent. Using this value, U
was determined at given concentrations of 1 and 2.
Results and discussion
Saccharides as sacrificial agents
Sacrificial hydrogen generation from saccharides over a Pt/TiO photocatalyst has
2
been investigated using glucose [15–17], xylose [12], fructose [18], and sucrose [19]
so far, revealing that these saccharides functioned well as sacrificial agents.
However, the U was still not determined. In order to determine U for the sacrificial
hydrogen generation over Pt/TiO , we used pyranoses such as glucose (1e),
2
galactose (1f), mannose (1g), sorbose (1h), and fructose (1i), and disaccharides such
as sucrose (1j) and cellobiose (1k) as well as related compounds such as ascorbic
acid (1a), sugar alcohols such as xylitol (1b) and sorbitol (1c) and gluconic acid
1
23

Quantum yields for sacrificial hydrogen generation from...
(
1d) as sacrificial agents. Scheme 1 shows the structures of 1a–k, which are the
most highly populated in aqueous solutions. Moreover, methanol (2a) [20], ethanol
2b), propanol (2c), 1,2-ethanediol (2d), glycerol (2e) [21], and 1-hydroxy-2-
(
propanone (2f) were used as reference compounds.
Chemical yield of hydrogen generation
The Pt/TiO (100 mg, 2 wt% of Pt content) was suspended in an aqueous solution
2
(
using N gas, UV irradiation was performed under vigorous stirring until gas
150 ml) of 1 (0.25–1.25 mmol). After O was purged from the reaction vessel
2
2
evolution had ceased. In the GLC analysis of the evolved gas, H and CO were
2
2
detected in addition to N as the purging gas. It was confirmed that the irradiation of
2
Pt/TiO in the absence of sacrificial reagents only formed small amounts of H .
2
2
Scheme 1 Ascorbic acid (1a), sugar alcohols (1b, 1c), gluconic acid (1d) as well as pyranoses (1e–1i)
and disaccharides (1j–k) acting as sacrificial agents (R–H)
123

M. Yasuda et al.
The amounts of the evolved H and CO from the photocatalytic H generation of
2
2
2
various amounts of 1 and 2 were summarized in Table S1 (supplementary
information). Figure 3a shows an example of the H and CO evolutions in the
2
2
photocatalytic H generation using ascorbic acid (1a). The evolved H increased as
2
2
the amounts of 1a used increased. However, the molar ratio of H to 1a (H /1a) was
2
2
dependent on the amount of 1a used; as the 1a/catalyst decreased, the H /1a values
2
increased. Therefore, we plotted the H /1a against the molar ratio of 1a to the
2
catalyst (1a/catalyst), as shown in Fig. 3b. The intercept of the plots represented
max
H2 which was the limiting mole of H obtained from one mole of 1a at an infinite
2
max
amount of the catalyst. In the case of 1a, the H2 value with an actual hydrogen
amount was 9.8, reaching the theoretical amounts of H evolved (10.0). Also, the
2
max
plots of the molar ratio of CO to 1a (CO /1a) against 1a/catalyst gave CO
2
,
max
2
2
which was the limiting amount of CO obtained from one mole of 1a. The CO
2
2
value was determined to be 5.87, resulting in the occurrence of complete
max
max
decarboxylation from 1a. The H2 and CO2 values are listed in Table 1.
According to Eq. (1), the theoretical hydrogen amount (P) will become (n ? m/
2
) mol when 1 mol of 1 is completely decomposed. The yields of H generation
2
max
were calculated by 100 9 H2 /P. The yields were nearly 100 % in the case of 1a–
k, as shown in Table 1.
hv
CnHmOn þ nH2O À! nCO2 þ ðn + m/2ÞH2
ð1Þ
PtÀTiO2
Fig. 3 a Plots of H
(
2
(filled circle) and CO
2 2
(open diamond) gas volumes versus [1a] and b plots of H /1a
filled circle) and CO
2
/1a (open diamond) versus 1a/catalyst
1
23

Quantum yields for sacrificial hydrogen generation from...
Table 1 Chemical yields, limiting quantum yields (U?), and the I/S values in the sacrificial H
gen-
2
eration from saccharides (1) and alkanols (2) over a Pt-loaded TiO
2
max
max
2
-1
c
?d
e
I/S
Sacrificial agents
Compounds
H
mol
2
/mol
CO
/mol
Yield/ %
U
-
1
mol
a
b
P
f
Ascorbic acid (1a)
Xylitol (1b)
8
10
11
13
11
12
12
12
12
12
24
24
3
9.79
12.48
13.34
10.71
11.80
11.48
12.24
11.90
12.46
22.67
24.78
3.0
5.87
98
113
103
97
0.070
0.086
0.066
0.067
0.058
0.064
0.168
0.064
0.098
0.070
0.201
0.057
0.044
0.069
0.058
0.078
0.045
20.6
15.2
11.3
8.5
12
14
12
12
12
12
12
12
22
22
4
5.90
6.16
5.84
5.90
5.73
6.11
5.95
6.23
11.33
12.39
1.0
Sorbitol (1c)
Gluconic acid (1d)
Glucose (1e)
98
11.9
4.0
Galactose (1f)
Mannose (1g)
Sorbose (1h)
96
102
99
0.9
4.9
Fructose (1i)
104
94
3.8
Sucrose (1j)
2.8
Cellobiose (1k)
Methanol (2a)
103
100
67
1.1
44.8
31.7
20.8
16.1
15.1
19.5
f
Ethanol (2b)
3
6
4.0
1.2
Propanol (2c)
3
9
4.1
1.0
46
1,2-Ethanediol (2d)
6
5
5.1
2.0
102
103
67
Glycerol (2e)
8
7
7.2
3.1
f
Hydroxyacetone (2f)
3
7
4.7
2.5
a
Number of reaction sites
b
c
d
e
f
Theoretical amounts of hydrogen
max
2
Chemical yield = 100 9 H
Limiting quantum yield (U?) for H
The I/S value equaled the relative k
Methane was formed accompanying the formation of H
/P
2
evolution at infinite amounts of 1 and 2
R
2
and CO
2
Quantum yields (U) for H evolution
2
The amounts of H evolved per unit irradiation time were measured from the slope
2
of the time-conversion plots of hydrogen at a low conversion where the evolved gas
involved only H and not CO . The U for H evolution in the photocatalytic H
2
2
2
2
production from ethanol (0.434 M) using Pt-TiO (Pt content 1.0 wt%) at pH 10.0
2
was determined to be 0.057 [14]. Using this reaction as an actinometer, the amounts
of H evolved per unit irradiation time was transformed into U. The U at various
2
concentration of 1 and 2 (8–40 mM) were determined and summarized in Table S2
(
supplementary information).
The decomposition of saccharides is initiated by hydrogen atom abstraction by
HOÁ radicals [15]. Therefore, we postulated Scheme 2 to be the reaction scheme for
the photocatalytic H generation where R–H denotes the sacrificial agents of 1 and
2
2
. The k and k were the rate constants for the hydrogen atom abstraction from R–
R D
H by the HOÁ radical and the recombination of electrons with HOÁ radical,
123

M. Yasuda et al.
Scheme 2 Reaction scheme
respectively. Equation (1) can be derived from Scheme 2 and was modified into
Eq. (2) where the f value is the number of hydrogen atoms such as H–C–O and C–
O–H that are capable of being abstracted by the HOÁ radical. In order to determine
?
the limiting quantum yields (U ) at an infinite concentration of R–H, the double
reciprocal plots of U vs. [f 9 R–H] were performed (Fig. 4). The inverse of the
?
intercept (I) of the plots was equated to the ab (= U ) which were determined to be
0
.058–0.201 for 1–k and 0.044-0.078 for 2a–f, as shown in Table 1.
ꢀ
ꢁ
bk ½f  RÀHŠ
1
1
kD
R
U ¼ a
¼
1 þ
ð2Þ
kD þ kR½f  RÀHŠ
U
ab
kR½f  RÀHŠ
Rate constant (k ) for the hydrogen-abstraction by HOÁ radicals
R
According to Eq. (2), the intercept-to-slope ratio (I/S) of the double reciprocal plots
was equated to k /k . Since the k values are thought to be constant irrespective of
the kinds of 1 and 2, the I/S values showed the relative k values. The I/S value of
R
D
D
R
methanol (2a) was determined to be 44.8, while the I/S values of the other alcoholic
Fig. 4 Double reciprocal plots of U versus [1] in sacrificial hydrogen generation from a sorbitol (1c,
filled triangle), gluconic acid (1d, open diamond), glucose (1e, ?), and b 1e (?), galactose (1f, filled
square), and mannose (1g, open circle)
1
23

Quantum yields for sacrificial hydrogen generation from...
agents (2b–f) were 15.1–31.7. As the steric hindrance around the H–C–O moiety
increased, the I/S values grew smaller.
Ascorbic acid (1a), which is a typical electron donor, had large I/S values (20.6).
Furthermore, sugar alcohols such as xylitol (1b) and sorbitol (1c) had large I/
S values (15.2 and 11.3) which were similar values to those of the simple
polyalcohols such as 2d and 2e (16.1 and 15.1). This shows that the HOÁ radicals
could efficiently abstract hydrogen atom from sugar alcohols. In the cases where
saccharides had cyclic pyranose and furanose forms, on the other hand, the I/
S values showed a structural dependence. For example, the I/S value (11.9) of
glucose (1e) showed no large difference with those of 1c and gluconic acid (1d),
both of which had different structures at C1 carbon. However, the I/S values of
galactose (1f) and mannose (1g), which were the stereo isomers of 1e at C2 and C4
carbons, were smaller than 1e. Thus the configuration changes at C2 and C4 carbons
strongly affected the I/S values. The configuration at C2 strongly affected the I/S
values, especially. All hydrogens at C2, C3, and C4 of 1e located axial positions.
Therefore, the HOÁ radicals approached the H–C–O located at axial positions more
easily compared with those at the equatorial positions. These observations suggested
that initial hydrogen abstraction occurred at the axial positions of C2 and/or C4
carbons. In the cases of furanoses such as sorbose (1h) and fructose (1i), the I/
S values was little, since there were few axial hydrogen atom in 1h and 1i. In the
cases of sucrose (1j) and cellobiose (1k), the I/S values were small, although they
were disaccharides which have glucose structure. Steric hindrance around C2
carbons might disturb the hydrogen abstraction.
Conclusions
In this work, we have elucidated that photocatalytic hydrogen production from
saccharides proceeded in 0.058–0.201 of quantum yields and almost 100 % of
chemical yields. Therefore, sacrificial hydrogen production from saccharides
becomes one approach to developing renewable energy. It was found that the
hydrogen abstraction from glucose by HOÁ radical was faster compared with those
from galactose, mannose, sorbose, and fructose, which were the stereo and
structural isomers of glucose. Thus, the hydrogen abstraction from saccharides by
HOÁ radical were sterically affected, depending the structure of saccharides.
Acknowledgments This study was supported by a Grant-in-Aid for Scientific Research (C) No
24610055 from the Ministry of Education, Culture, Sports, Science, and Technology of Japan.
References
1
2
3
4
. O.P. Ward, A. Singh, Adv. Appl. Microbiol. 51, 53 (2002)
. M. Ni, D.Y. Leung, M.K. Leung, K. Sumathy, Fuel Process. Technol. 87, 461 (2006)
. R.M. Navarro, M.A. Pe n˜ a, J.L.G. Fierro, Chem. Rev. 107, 3952 (2007)
. M. Yasuda, R. Kurogi, H. Tsumagari, T. Shiragami, T. Matsumoto, Energies 7, 4087 (2014)
123

M. Yasuda et al.
5. M. Yasuda, Misriyani, Y. Takenouchi, R. Kurogi, S. Uehara, T. Shiragami, J. Sustain. Bioenergy
Syst. 5, 1 (2015)
. A. Fujishima, T.N. Rao, D.A. Tryk, J. Photochem. Photobiol. C 1, 1 (2000)
. Z.W. Seh, S. Liu, M.-Y. Han, Chem. Asian J. 7, 2174 (2012)
. Z. Seh, S. Liu, M. Low, S.-Y. Zhang, Z. Liu, A. Mlayah, M.-Y. Han, Adv. Mater. 24, 2310 (2012)
. A. Galinska, J. Walendziewski, Energy Fuels 19, 1143 (2005)
6
7
8
9
1
1
1
1
0. T. Shiragami, T. Tomo, T. Matsumoto, M. Yasuda, Bull. Chem. Soc. Jpn 86, 382 (2013)
1. J.C. Kennedy III, A.K. Datye, J. Catal. 179, 375 (1998)
2. M. Yasuda, T. Tomo, S. Hirata, T. Shiragami, T. Matsumoto, Catalysts 4, 162–173 (2014)
3. T. Shiragami, T. Tomo, H. Tsumagari, R. Yuki, T. Yamashita, M. Yasuda, Chem. Lett. 41, 29–31
(2012)
14. S.E. Salas, B.S. Rosales, H. de Lasa, Appl. Catal. B Environ. 140–141, 523 (2013)
15. D.I. Kondarides, A. Patsoura, X.E. Verykios, J. Adv. Oxid. Technol. 13, 116 (2010)
16. X. Fu, J. Long, X. Wang, Y. Leung, Z. Ding, L. Wu, Z. Zhang, Z. Li, X. Fu, Int. J. Hydrog. Energy
3
3, 6484 (2008)
7. P. Gomathisankar, D. Yamamoto, H. Katsumata, T. Suzuki, S. Kaneco, Int. J. Hydrog. Energy 38,
517 (2013)
1
5
18. S. Deguchi, N. Shibata, T. Takeichi, Y. Furukawa, N. Isu, J. Jpn. Petrol. Inst. 53, 95 (2010)
19. N. Luo, Z. Jiang, H. Shi, F. Cao, T. Xiao, P.P. Edwards, Int. J. Hydrog Energy 34, 125 (2009)
20. L.S. Al-Mazroai, M. Bowker, P. Davies, A. Dickinson, J. Greaves, D. James, L. Millard, Catal.
Today 122, 46 (2007)
21. V.D. Daskalaki, D.I. Kondarides, Catal. Today 144, 75 (2009)
1
23