Enhanced activity for gas phase propylene oxide rearrangement to allyl alcohol by Au promotion of Li3PO4 catalyst

Article abstract of DOI:10.1016/j.jssc.2019.120922

In present paper, lithium phosphate (Li₃PO₄) promoted by nano-gold (denoted as Au/Li₃PO₄) was used in gas phase propylene oxide (PO) rearrangement to produce allyl alcohol (AA). It was found that the support of gold could improve the catalytic activity as compared with pure Li₃PO₄. Au/Li₃PO₄ with gold loading of 0.042 wt% showed higher catalytic performance. The enhancement might come from the increase of the strength and amount of strong basic sites after gold deposition. Results of XPS spectra and density functional theory (DFT) verified that the charge transfer from oxygen species to gold species led to the electron-rich gold species, which acted as stronger base sites.

Full text of DOI:10.1016/j.jssc.2019.120922

Journal of Solid State Chemistry 279 (2019) 120922
Contents lists available at ScienceDirect
Journal of Solid State Chemistry
journal homepage: www.elsevier.com/locate/jssc
Enhanced activity for gas phase propylene oxide rearrangement to allyl
alcohol by Au promotion of Li3PO4 catalyst
Xiangshuai Zhu, Dongyu Wang, Zhishan Li, Weihua Ma *
School of Chemical Engineering, Nanjing University of Science and Technology, Nanjing, 210094, PR China
A R T I C L E I N F O
A B S T R A C T
Keywords:
Nano-gold
Lithium phosphate
pH
Base strength
DFT
In present paper, lithium phosphate (Li3PO4) promoted by nano-gold (denoted as Au/Li3PO4) was used in gas
phase propylene oxide (PO) rearrangement to produce allyl alcohol (AA). It was found that the support of gold
could improve the catalytic activity as compared with pure Li3PO4. Au/Li3PO4 with gold loading of 0.042 wt%
showed higher catalytic performance. The enhancement might come from the increase of the strength and amount
of strong basic sites after gold deposition. Results of XPS spectra and density functional theory (DFT) verified that
the charge transfer from oxygen species to gold species led to the electron-rich gold species, which acted as
stronger base sites.
1. Introduction
discovered to date, such as environmental control, energy processing and
chemical synthesis [12,13]. Recently, Raptis et al. [4] found a new
The rearrangement of epoxides to carbonyl compounds or allyl al-
cohols (AA) is a fundamentally important process due to their industrial
utility. For instance, the Meinwald rearrangement, which refers to rear-
rangement of epoxides to carbonyl compounds under acidic conditions, is
a fundamentally important transformation [1]. The composition of
rearrangement products is determined by the acidic-basic properties of
the catalysts [2]. Imanaka et al. have proposed that, in rearrangement of
PO, AA is produced by acid-basic bifunctional catalysts, acetone, by
basic-site only, and propylaldehyde, by acid-site only [3]. According to
previous studies [3–5], we summarized the mechanism of PO conversion
pathways for different products in Scheme 1. Therefore, in order to
obtain AA, amphoteric oxides including Al2O3, ZrO2, TiO2 [6] or Li3PO4
have been used as catalysts for epoxides rearrangement. Li3PO4 has been
a special catalyst due to high conversion and selectivity for PO rear-
rangement to AA. There are two processes, i.e., the gas phase and the
liquid phase [7]. The gas phase process is much simpler and cleaner than
the liquid phase. However, carbon deposition has always been a major
problem [8], thus Li3PO4 should be further studied and improved.
Gold has been considered as an inert metal and catalytically inactive
for a long time [9]. But it was overturned when Haruta and coworkers
[10] discovered supported gold catalysts exceedingly active for
low-temperature CO oxidation. Since then, much attention has been
attracted to gold catalysts. It was likely that a Gold Rush occurred in the
field of catalysis [11]. Various applications of gold catalysts have been
application of gold catalyst that Au/TiO2 could catalyze the rearrange-
ment reaction of epoxides to AA with high yields. But Raptis’s work is
mainly on the liquid phase process. Our work is on the gas phase process.
We supported Au on Li3PO4 to catalyze PO and results demonstrated that
the support of gold could obviously improve the catalytic activity. From
FT-IR, CO2-TPD, XPS and DFT, it could be concluded that the enhance-
ment came from the strength and amount of strong basic sites, namely,
the electron-rich gold species that charge was transferred from oxygen
species.
2. Experimental section
2.1. Preparation of Au/Li3PO4
A basic Li3PO4 was synthesized by 14.5 g lithium hydroxide (LiOH)
and 32.8 g sodium phosphate (Na3PO4), which were dissolved in
deionized water and stirred at 60 ꢀC for 2 h. The resultant was separated
by centrifugation and washed several times until the pH decreased to 13,
then dried at 100 ꢀC for 5 h and calcined at 320 ꢀC for 6 h with a heating
rate of 2 ꢀC minꢁ1. This resulting was denoted as Li3PO4-13. Li3PO4 with
different pH values, denoted as Li3PO4-x (x ¼ pH), was obtained by
adjusting the pH of the slurry of uncalcined Li3PO4 aqueous solution to
the desired value by 1 M hydrochloric acid (HCl) solution. Gold sup-
ported onto Li3PO4-x (denoted as Au/Li3PO4-x) was prepared via a
* Corresponding author.
E-mail address: maweihuacn@163.com (W. Ma).
https://doi.org/10.1016/j.jssc.2019.120922
Received 20 May 2019; Received in revised form 31 July 2019; Accepted 22 August 2019
Available online 29 August 2019
0022-4596/© 2019 Elsevier Inc. All rights reserved.

X. Zhu et al.
Journal of Solid State Chemistry 279 (2019) 120922
Scheme. 1. Schematic representation of the different mechanistic pathways for propylene oxide rearrangement. LA ¼ Lewis acid, LB ¼ Lewis base.
deposition precipitation method [14,15]. Approximately 0.1 g
HAuCl4⋅4H2O was dissolved in 50 mL deionized water and 1.0 g Li3PO4
was added with stirring. Afterwards, the pH of the slurry was adjusted to
13 by 1 M sodium hydroxide (NaOH) solution. The suspension was stir-
red at 80 ꢀC for 2 h and then centrifuged, washed and dried at 100 ꢀC for
5 h. Finally,ꢁt1he product was calcined at 300 ꢀC for 4 h with a heating rate
0.30–0.45 mm) was placed in a fixed-bed reactor with 10 mm in diam-
eter. A mixture of vaporized PO and N2 (the molar ratio is 0.59) was
introduced into the reactor, and the WHSV of propylene oxide was 33.2
h
ꢁ1. The reaction temperature was controlled at the optimal reaction
temperature of 300 ꢀC. The products were condensed via a sample col-
lector settled in ice water bath and analyzed by a SP-1000 Gas Chro-
matography equipped with PEG-20M capillary column.
of 2 ꢀC min
.
3. Results and discussions
2.2. Catalyst characterization
3.1. Characterization and catalytic performance of the catalysts
The Brunauer-Emmett-Teller (BET) surface area was determined by
N2 adsorption-desorption measurements at 77 K (Gold App V-sorb 2008).
Powder X-ray diffraction (XRD) was carried out with a general X-ray
diffraction instrument (Beijing Purkinjie XD-3). TEM experiments were
carried out on a Philips Tecnai 12 transmission electron microscope.
UV–visible spectra were recorded by a Thermo Scientific Evolution 220
spectrophotometer equipped with an ISA-220 integrating sphere. The
gold contents in catalysts were measured by Optima 7300 DV inductively
coupled plasma spectrometer (PerkinElmer). FT-IR spectra were recor-
ded by Thermo Scientific Nicolet iS10. X-ray photoelectron spectroscopy
(XPS) experiments were carried out on a RBD-upgraded PHI–5000C
Fig. S1 shows the catalytic performances of Au/Li3PO4 at different
preparation conditions, including gold loading, calcination temperature
and pH value of Li3PO4, which are of significance to the catalytic per-
formance of propylene oxide rearrangement to allyl alcohol. As can be
seen, Au/Li3PO4-13 with gold loading content of 0.042 wt% and calcined
at 300 ꢀC, showed the best catalytic performance and was chosen as the
catalyst for PO rearrangement reaction. Furthermore, the catalytic per-
formance of Au/Li3PO4 is sensitive to the pH values. As presented in
Table S1, the selectivity to AA has changed greatly at pH ¼ 9, which
verified again that higher catalytic activity and selectivity is maintained
under strong basic condition.
The physical properties of the Au/Li3PO4-13 and Li3PO4-x are listed
in Table S2. It can be seen that the surface area of Li3PO4 decreases
gradually from 26 to 14 m2 gꢁ1 as the pH value decreases from 13 to 9.
When the gold is deposited on Li3PO4-13, the surface area decreases to
20 m2 gꢁ1, which might be due to gold filling in some pores of Li3PO4
substrate. When Au/Li3PO4-13 was used for 5 h (denoted as Au/Li3PO4
5 h), the surface area decreased to 10 m2 gꢁ1. Table S3 shows the bulk
and surface composition of the Au/Li3PO4-13 samples with different
loadings of Au. It should be noted that the Au cluster tends to be located
on the outer surface of the Li3PO4 support. XRD patterns of Au/Li3PO4-13
and Li3PO4-x, as shown in Fig. S2, confirming that the structure of Li3PO4
is β form with orthorhombic symmetry [16,17], which does not change
after gold deposition or pH adjustment. Additionally, metallic gold is not
detected over Au/Li3PO4-13 and Au/Li3PO4 5 h, owing to the fact that
ESCA system (PerkinElmer) with Mg K
α radiation (hν ¼ 1253.6 eV),
calibrated by the C 1s peak at 284.6 eV. Carbon dioxide temperature-
programmed desorption (CO2-TPD) experiments were carried out
under a flow of He (70 cm3 minꢁ1) and the temperature was raised from
room temperature to 600 ꢀC at a heating rate of 10 ꢀC minꢁ1. Ammonia
gas temperature-programmed desorption (NH3-TPD) experiments were
carried out under a flow of He (70 cm3 minꢁ1) and the temperature was
raised from room temperature to 600 ꢀC at a heating rate of 10 ꢀC minꢁ1
.
The fresh and used catalysts were characterized by thermogravimetric
analysis (TG, Mettler-Toledo TGA/SDTA851e) at the rate of 10 ꢀC/min in
air (100 ml/min) atmosphere.
2.3. Measurement of catalytic activity
Catalytic performance was determined in a fixed bed reactor under
atmospheric pressure. The catalyst (0.3 g, particle diameter of
2

X. Zhu et al.
Journal of Solid State Chemistry 279 (2019) 120922
Fig. 1. TEM image of gold particles in fresh Au/Li3PO4-13.
Fig. 4. The CO2-TPD curves of Au/Li3PO4-13 and Li3PO4-x.
gold nanoparticles in the TEM image. Due to the low gold loading, only a
small amount of metal particles can be seen [21]. It can be seen that the
Au particles of Au/Li3PO4-0.042 with an average diameter of 11 nm are
dispersed on the carrier. The Au particles of Au/Li3PO4-0.055 have an
average diameter of 18 nm. However, the performance of
Au/Li3PO4-0.055 is not the best, so it can be concluded that excess Au
loading will form larger Au particles, which is detrimental to charge
transfer and ultimately leads to a decrease in catalytic performance.
The catalytic performance of PO rearrangement on Li3PO4 and Au/
Li3PO4 are shown in Fig. 2. Wang et al. reported that the AA rate of the
ꢁ1
Li3PO4 catalyst was 7.6 kgAA⋅hꢁ1⋅kg at 1 h. The AA rate of the Au/
cat
ꢁ1
Li3PO4-13 catalyst was 12.5 kgAA⋅hꢁ1⋅kg at 1 h, improved by 39%. In
cat
addition, it can be seen that the performance of catalysts becomes stable
after 5 h, and the AA rate of Au/Li3PO4-13 increases by 20% compared
with Li3PO4-13.
Fig. 2. Production yield of AA for Li3PO4-13 and Au/Li3PO4-13.
3.2. Discussion of the active sites
3.2.1. Result of FT-IR
The surface group information of the catalysts can be easily acquired
by FT-IR technique, as shown in Fig. 3. The peaks at 1100 and 600 cmꢁ1
are ascribe to the P–O antisymmetric stretching vibration and symmetric
stretching vibration, respectively. The absorption peak at 1635 cmꢁ1 is
ascribed to the scissor vibration of O–H (δOH), owing to the adsorption of
water in the air. The peaks at 1445 and 881 cmꢁ1 are ascribed to the C–O
stretching vibration and plane deformation vibration, respectively. This
result demonstrates the existence of CO23ꢁ, which is due to the reaction of
basic catalyst surface with CO2 in the atmosphere. Because the sampling
process was strictly controlled, the adsorption amount of CO2 could be
quantitatively compared. In order to quantitatively compare the amount
of base site, the area of peaks at 1445 and 881 cmꢁ1 were integrated,
respectively. The area of peak at 1445 cmꢁ1 is 24.8 for Au/Li3PO4-13,
and those of Li3PO4-x (x ¼ 13, 11 and 9) are 22.2, 14.4 and 1.6, respec-
tively. The area of peak at 881 cmꢁ1 are 1.2, 1.1, 0.8 and 0.2 for Au/
Li3PO4-13 and Li3PO4-x (x ¼ 13, 11 and 9). As the pH decreases to 9, the
two peaks attributed to CO23ꢁ almost vanish, suggesting that the basicity
of the surface is weakened. Unexpectedly, after the deposition of gold,
the FT-IR absorption of CO23ꢁ is not weakened. On the contrary, it is
reinforced. It can be deduced that the deposition of gold may increase the
base amount or the strength. In order to verify this, CO2-TPD was
employed to discuss the further information of activity enhancement.
Fig. S6 shows the FT-IR spectra of Au/Li3PO4-x with different Au
Loading. As the Au loading increases, the peak area at 1445 and
881 cmꢁ1 gradually increases. It can be inferred that the Au loading has a
positive effect on the increase of the basic site.
Fig. 3. The FT-IR spectra of (a) Au/Li3PO4-13, (b) Li3PO4-13, (c) Li3PO4-11 and
(d) Li3PO4-9.
the gold content of Au/Li3PO4 is low (0.042 wt%) or the gold particles
are well-dispersed on Li3PO4 [18].
Fig. S3 shows the UV–visible spectra of Au/Li3PO4-13 and Li3PO4-x.
The absorption of visible light at 500–600 nm is due to the surface
plasmon resonance band of gold nanoparticles [19,20]. Fig. S4 shows the
ultraviolet–visible spectra of the rest of gold catalyst. The surface plas-
mon resonance band of gold nanoparticles (530 nm band) can be seen.
Fig. 1 and Fig. S5 show the TEM images of Au/Li3PO4 samples with Au
loadings of 0.042 wt% and 0.055 wt%, respectively. Black dots represent
3

X. Zhu et al.
Journal of Solid State Chemistry 279 (2019) 120922
Fig. 5. XPS profiles of Au 4f region of Au/Li3PO4-13.
Fig. 7. (a) Surface structure of the Li3PO4 (001) surface, (b) Top view. The red
balls represent oxygen atoms, green balls represent lithium atoms, orange balls
represent phosphorus atoms. (For interpretation of the references to colour in
this figure legend, the reader is referred to the Web version of this article.)
strong acid sites, and the amount of weakly acidic sites increases slightly
with the decrease of pH, so the main reason for the change in perfor-
mance is the change of the basic sites.
3.2.3. Result of XPS
XPS measurement is used to analyze the chemical state of elements so
as to discuss the shifts of electron binding energy over Au/Li3PO4. Fig. 5
shows the XPS spectra of the Au 4f regions for Au/Li3PO4-13. There are
two broad peaks at 86.0 and 82.3 eV (corresponding to the 4f7/2 and 4f5/2
peaks) for the Au element, which is consistent with the Au0 oxidation
state [23]. The difference between these two peaks for gold (3.7 eV) is
exactly the same as that of bulk Au0 [24–26]. However, the peak posi-
tions are slightly different from that of bulk Au0. The Au 4f5/2 peak of
Au/Li3PO4 is seen to shifts to lower binding energies by 1.7 eV than that
of bulk Au0, which is reported to be at 84.0 eV [27,28] This indicates that
Au species is negatively charged after gold was deposited on Li3PO4.
Fig. 6 shows the XPS spectra of the O1s regions of Au/Li3PO4-13 and
Li3PO4-13. It can be seen that the binding energy of the O1s spectra of
Au/Li3PO4-13 relative to the O of Li3PO4-13 was increased by 0.5 eV.
This indicates that the O species loses electrons after gold is deposited on
Li3PO4. The cause of electron density changes of the O and Au should be
further discussed.
Fig. S9 shows the XPS spectrum of the Li 1s (a) and P 2p (b) regions of
Au/Li3PO4-x and Li3PO4-x. It can be seen that there is no change in the
binding energy of Li and P of these catalysts. Fig. S10 is the XPS data for
different Au loaded samples. It can be seen that the binding energy of Au
of the samples with different Au loadings is reduced. However, it is worth
noting that the binding energy of the sample with the highest Au loading
is reduced by 1.5 eV, which is lower than the reduced binding energy of
the sample with an Au loading of 0.042 wt%.
The Au 4f binding energy for supported catalysts compared with that
in bulk Au has been investigated in a number of literatures [29–31].
Yejun Park et al. [32] proposed that in the Au/AlPO4 nanocomposite, the
shift of the Au 4f peaks occurs toward lower binding energies, indicating
the presence of a negative charge on gold nanoparticles due to the
electron transfer from AlPO4. Caixia Qi et al. [33] have reported that the
majority of the gold tends to be zero valence metal, but there is a small
shift toward lower binding energy for three P-containing samples,
implying the transfer of electrons to Au and the formation of Auδꢁ atoms.
However, the decrease in the Au 4f binding energy for Au/Li3PO4 has not
been studied yet. In Fig. 5, Au 4f shifts to lower binding energy. In Fig. 6,
O 1s shifts to a higher binding energy. It can be inferred that the decrease
of the Au 4f binding energy may come from the charge transfer from the
surface of Li3PO4. The interaction between gold and Li3PO4 results in a
local perturbation of the electronic structure and a shift in the density of
Fig. 6. XPS profiles of O 1s region of Au/Li3PO4-13.
3.2.2. Result of CO2-TPD and NH3-TPD
The profiles of Au/Li3PO4-13 and Li3PO4-x are shown in Fig. 4. There
are two CO2 desorption peaks in the profiles. For the Li3PO4-13, the first
peak with Tmax ¼ 139 ꢀC represents the weak base site and the second
peak with Tmax ¼ 326 ꢀC corresponds to the strong base site. However,
for Au/Li3PO4-13, the peaks of weak and strong base sites are at 134 ꢀC
and 382 ꢀC, respectively. In other words, weak base strength is not
changed, but strong base strength is enhanced after gold deposition.
Considering the pK values of H3PO4 (2.1,7.2,12.3), the prevalent ions in
the solution changes with the decrease of pH values. For Li3PO4-11 and
Li3PO4-9, as the monohydrogen and dihydrogen phosphate ions increase,
both weak and strong base strength are weakened. In order to quanti-
tatively compare the amount of base site, the peak area of Au/Li3PO4-13
and Li3PO4-x were integrated, respectively, as shown in Table S4. For Au/
Li3PO4-13, the amount of weak base sites decreases, but the amount of
strong base sites increases. Because strong base sites are favorable to
improve the activity of basic catalysts [22], it can be concluded that the
enhancement of catalytic activity comes from the improvement of the
strength and amount of basic sites after gold deposition. As for the reason
of this enhancement, we consider that gold species act as stronger base
site and make it improve. XPS was employed to get further information.
Fig. S7 shows the CO2-TPD curves of two Au/Li3PO4 catalysts with
different calcination temperatures. It can be seen that the strong basic site
of Au/Li3PO4 with higher calcination temperature is stronger, and the
strong basic site is advantageous for catalyzing the rearrangement of
propylene oxide, so the performance of Au/Li3PO4-300 is better.
Fig. S8 shows the NH3-TPD curves for Li3PO4 catalysts of different pH
values. It can be seen that there is no great difference in the amount of
4

X. Zhu et al.
Journal of Solid State Chemistry 279 (2019) 120922
Fig. 10. PDOS of Au-6s and O-2p after Au13 adsorption.
Fig. 8. Configuration of Au6 (a) and Au13 (b) cluster adsorption on the Li3PO4
(001) surface.
Fig. 11. TG curves of Li3PO4 and Au/Li3PO4-13.
zone only at the Γ point. The structural relaxations were terminated when
the total energy of the slab converged to 0.001 eV. To better understand
the trend in the atomic charge transfer between the gold and substrate,
the net charge-transfer was calculated using Mulliken population anal-
ysis. The result is shown in Tables S5 and S6. We found that after gold
adsorption on Li3PO4 (001), the Au6 cluster gains 1.08 e. The charge of
the O atom increases from ꢁ1.22 to ꢁ0.92 e. The Au13 cluster gains 0.45
e. The charge of the O atom increases from ꢁ1.22 to ꢁ1.06 e. The PDOS
(partial density of states) of Au-6s and O-2p after Au6 and Au13 adsorp-
tion are shown in Fig. 9 and Fig. 10. After the two clusters adsorbed on
Li3PO4, the two orbitals have good overlap, which indicates that the
electron transfer trend of the Au13 clusters is consistent with that of Au6.
In Table S7, we summarize the P–O bond and O–P–O bond angle of
the surface O of the Li3PO4 unit cell before and after adsorption. Table S8
shows the distance of Au–Au before and after adsorption and the distance
from our chosen Au atom to the surface of Li3PO4 unit cell. It can be seen
from the table that the average bond length of the Au clusters before and
after adsorption increases. The reason is that electrons are transferred
from O atoms to gold. This electron transfer results in an increase in the
average bond length of the Au cluster [36].
Fig. 9. PDOS of Au-6s and O-2p after Au6 adsorption.
states as compared with bulk Au [34]. Because of electron-rich oxygen
atoms and their high content, we considered that the charge transfers,
most likely, from oxygen atoms to the gold cluster. The electron-rich Au
acts as strong base sites, which promotes the catalytic activity.
3.2.4. Result of DFT
In order to verify the conclusion, we have performed structural re-
laxations at the level of DFT using the CASTEP simulation package. The
calculations were done in the framework of the generalized gradient
approximation, using projector-augmented wave pseudopotentials and
the Perdew-Burke-Ernzerhof exchange-correlation energy [35]. We have
used 3 ꢂ 2Li3PO4 (001) surface slab with a 15 Å vacuum spacing. The
thickness of the slabs in the computational supercells is 7 layers, and the
bottom two atomic layers were kept fixed during the structural re-
laxations to simulate the underlying bulk. Fig. 7 shows the optimized
structure of the surface of Li3PO4. To describe the state of O and the
distance from Au to the surface of Li3PO4, we selected a gold atom
labeled Au1 and labeled O as O1, O2 and O3 on the surface of Li3PO4. In
order to simplify the calculation and ensure that the interactions between
the periodic images of the substrate-support, we chose a two-dimensional
Au6 cluster and a minimum cubic octahedron Au13 cluster composed of
13 atoms as a model. The model is shown in Fig. 8. In all calculations, we
used a plane-wave cutoff energy of 450 eV and sampled the Brillouin
3.2.5. Result of TG and regeneration of deactivated catalyst
Fig. 11 shows the TG curves of Li3PO4 and Au/Li3PO4. The decom-
position temperatures of the carbonaceous deposits of the two catalysts
are basically the same, indicating that their carbonaceous sediments are
similar in nature. The amount of carbonaceous deposits of Au/Li3PO4 is
lower than that of Li3PO4. For many catalysts, carbonaceous deposits can
5

X. Zhu et al.
Journal of Solid State Chemistry 279 (2019) 120922
China (No.21276127) and the Priority Academic Program Development
of Jiangsu Higher Education Institutions (PAPD).
Appendix A. Supplementary data
Supplementary data to this article can be found online at https://do
i.org/10.1016/j.jssc.2019.120922.
References
[1] A. Jamalian, B. Rathman, G.L. Borosky, K.K. Laali, Appl. Catal. Gen. 486 (2014)
1–11.
[2] G. Neri, G. Rizzo, C. Crisafulli, L. De Luca, A. Donato, M.G. Musolino,
R. Pietropaolo, Appl. Catal. Gen. 295 (2005) 116–125.
[3] T. Imanaka, Y. Okamoto, S. Teranishi, Bull. Chem. Soc. Jpn. 45 (1972) 1353–1356.
[4] C. Raptis, H. Garcia, M. Stratakis, Angew. Chem. Int. Ed. 48 (2009) 3133–3136.
[5] M. Liang, et al., Catal. Lett. 149 (2019) 942–949.
[6] E. Ruiz-Hitzky, B. Casal, J. Catal. 92 (1985) 291–295.
Fig. 12. Production yield of AA for Li3PO4-13 and Au/Li3PO4-13 after
regeneration.
́
[7] D.P. Serrano, R. van Grieken, J.A. Melero, A. Garcıa, Appl. Catal. Gen. 269 (2004)
137–146.
[8] L. Jiang, H. Li, Y. Wang, W. Ma, Q. Zhong, Catal. Commun. 64 (2015) 22–26.
[9] M. Haruta, M. Date, Appl. Catal. Gen. 222 (2001) 427–437.
[10] M. Haruta, N. Yamada, T. Kobayashi, S. Iijima, J. Catal. 115 (1989) 301–309.
[11] M. Haruta, Nature 437 (2005) 1098–1099.
[12] H.F. Yin, Z. Ma, H.G. Zhu, M.F. Chi, S. Dai, Appl. Catal. Gen. 386 (2010) 147–156.
[13] A.S.K. Hashmi, G.J. Hutchings, Angew. Chem. Int. Ed. 45 (2006) 7896–7936.
[14] M. Li, Z. Wu, Z. Ma, V. Schwartz, D.R. Mullins, S. Dai, S.H. Overbury, J. Catal. 266
(2009) 98–105.
[15] Z. Li, Y. Wang, J. Zhang, D. Wang, W. Ma, Catal. Commun. 90 (2017) 87–90.
[16] R.M. Rojas, J.L.M. De Vidales, A. Delgado, J.V. Sinisterra, J. Solid State Chem. 106
(1993) 237–252.
[17] Y.A. Du, N. Holzwarth, Phys. Rev. B 76 (2007), 174302.
[18] M. Hosseini, T. Barakat, R. Cousin, A. Aboukaïs, B.L. Su, G. De Weireld, S. Siffert,
Appl. Catal. B Environ. 111 (2012) 218–224.
be removed, at least in part, by calcination in air. We use TG to determine
the conditions under which carbonaceous deposits may be removed. It
can be seen from Fig. 11 that most of the carbonaceous deposits react
with air at a temperature of 300–400 ꢀC, and the carbonaceous deposits
are substantially decomposed at 400 ꢀC. However, according to the
previous studies, high calcination temperature would be bad for the
performance of the Au-loaded catalyst, so we calcined the catalyst in air
at 350 ꢀC for 2 h to remove most of the carbonaceous deposits. The
performance after regeneration is shown in Fig. 12. Compared to the
fresh catalyst, the performance of Li3PO4 is reduced by 14%, while that of
Au/Li3PO4 is reduced by 26%. This may be due to an increase in the
particle size of the Au particles due to an increase in temperature, which
affects the catalytic performance [37].
[19] R. Zanella, S. Giorgio, C.-H. Shin, C.R. Henry, C. Louis, J. Catal. 222 (2004)
357–367.
ꢀ
ꢀ
ꢀ
[20] S. Oros-Ruiz, R. Gomez, R. Lopez, A. Hernandez-Gordillo, J.A. Pedraza-Avella,
ꢀ
E. Moctezuma, E. Perez, Catal. Commun. 21 (2012) 72–76.
ꢀ
[21] D. Horvath, L. Toth, L. Guczi, Catal. Lett. 67 (2000) 117–128.
[22] Y. Okamoto, T. Imanaka, S. Teranishi, Bull. Chem. Soc. Jpn. 46 (1973) 4–8.
[23] M.P. Casaletto, A. Longo, A. Martorana, A. Prestianni, A.M. Venezia, Surf. Interface
Anal. 38 (2006) 215–218.
4. Conclusions
ꢀ
[24] A. Zwijnenburg, A. Goossens, W.G. Sloof, M.W.J. Craje, A.M.V.D. Kraan,
L.J.D. Jongh, M. Makkee, J.A. Moulijn, J. Phys. Chem. B 106 (2002) 9853–9862.
[25] A.M. Venezia, L.F. Liotta, G. Pantaleo, V.L. Parola, G. Deganello, A. Beck,
Au/Li3PO4 were synthesized via a deposition precipitation method.
Li3PO4 with pH ¼ 13 showed a higher catalytic performance. Compared
to Li3PO4, the support of gold could improve the catalytic activity.
Furthermore, Au/Li3PO4-13 with gold loading of 0.042 wt% which was
calcined at 300 ꢀC showed the best performance. This enhanced activity
over Au/Li3PO4-13 has been discussed by FT-IR, CO2-TPD, XPS and DFT.
From the results of FT-IR and CO2-TPD, it can be concluded that the
enhancement might come from the strength and amount of strong basic
sites after gold deposition. Results of XPS spectra and DFT further
confirmed that the charge transferred from oxygen species to gold spe-
cies, leading to the electron-rich gold species, which could act as stronger
base sites.
ꢀ
ꢀ
Z. Koppany, K. Frey, D. Horvath, L. Guczi, Appl. Catal. Gen. 251 (2003) 359–368.
[26] S.W. Han, Y. Kim, K. Kim, J. Colloid. Interf. Sci. 208 (1998) 272.
[27] K. Takahiro, S. Oizumi, A. Terai, K. Kawatsura, J. Appl. Phys. 100 (2006) 314–325.
[28] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, J. Chastain, R.C. King, Physical
Electronics USA Inc Minnesota, 1992, p. 220.
[29] P. Konova, A. Naydenov, C. Venkov, D. Mehandjiev, D. Andreeva, T. Tabakova,
J. Mol. Catal. 213 (2004) 235–240.
[30] P. Lignier, M. Comotti, F. Schüth, J.-L. Rousset, V. Caps, Catal. Today 141 (2009)
355–360.
[31] J. Moma, M. Scurrell, W. Jordaan, Top. Catal. 44 (2007) 167–172.
[32] Y. Park, B. Lee, C. Kim, J. Kim, S. Nam, Y. Oh, B. Park, J. Phys. Chem. C 114 (2010)
3688–3692.
[33] C. Qi, H. Su, R. Guan, X. Xu, J. Phys. Chem. C 116 (2012) 17492–17500.
[34] N. Kruse, S. Chenakin, Appl. Catal. Gen. 391 (2011) 367–376.
[35] J.P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 77 (1996) 3865–3868.
[36] A. Abbasi, et al., Surf. Sci. 663 (2017) 35–46.
Acknowledgements
[37] A. Tomoki, et al., Surf. Interface Anal. 31 (2001) 73–78.
The work is financed by the National Natural Science Foundation of
6