Selective synthesis of 1,3-propanediol from glycidol over a carbon film encapsulated Co catalyst

Article abstract of DOI:10.1039/c9cy01162e

1,3-Propanediol (1,3-PDO) is an important chemical and feedstock in the synthesis of polytrimethylene terephthalate (PTT) resin, but the catalytic production of 1,3-PDO is difficult. In this work, a carbon film encapsulated Co nanoparticle catalyst (Co?NC) was synthesized via pyrolysis of Co₂(1,4-benzenedicarboxylic acid)₂(triethylenediamine) and tested in the synthesis of 1,3-PDO from glycidol. It was found that this Co?NC catalyst was highly active, selective and stable for this reaction under mild conditions. Characterization results indicated that the formation of 1,3-PDO was sensitive to the adsorbed hydrogen. The selectivity to 1,3-PDO reached its maximum over the Co?NC catalyst as carbon film encapsulation can inhibit the excessive adsorption and activation of H₂ molecules, while the adsorbed dissociated hydrogen on naked Co NPs and/or N atoms (in the carbon shell) promoted the formation of propanol.

Full text of DOI:10.1039/c9cy01162e

Catalysis
Science &
Technology
View Article Online
View Journal
PAPER
Selective synthesis of 1,3-propanediol from
glycidol over a carbon film encapsulated Co
Cite this: DOI: 10.1039/c9cy01162e
catalyst†
ab
Yanyan Sun,a Zhongshun Cai,a Xuewen Li,a Ping Chena and Zhaoyin Hou
*
1,3-Propanediol (1,3-PDO) is an important chemical and feedstock in the synthesis of polytrimethylene
terephthalate (PTT) resin, but the catalytic production of 1,3-PDO is difficult. In this work, a carbon film en-
capsulated Co nanoparticle catalyst (Co@NC) was synthesized via pyrolysis of Co2IJ1,4-benzenedicarboxylic
acid)2IJtriethylenediamine) and tested in the synthesis of 1,3-PDO from glycidol. It was found that this
Co@NC catalyst was highly active, selective and stable for this reaction under mild conditions. Characteri-
zation results indicated that the formation of 1,3-PDO was sensitive to the adsorbed hydrogen. The selec-
tivity to 1,3-PDO reached its maximum over the Co@NC catalyst as carbon film encapsulation can inhibit
the excessive adsorption and activation of H2 molecules, while the adsorbed dissociated hydrogen on na-
ked Co NPs and/or N atoms (in the carbon shell) promoted the formation of propanol.
Received 14th June 2019,
Accepted 11th August 2019
DOI: 10.1039/c9cy01162e
rsc.li/catalysis
Nevertheless, the biological fermentation has a low metabolic
efficiency because the reaction proceeds slowly in two
1 Introduction
1,3-Propanediol (1,3-PDO) is an important chemical for its ap-
plication in pharmaceuticals, plasticizers, surfactants, emulsi-
fiers and demulsifiers. More importantly, 1,3-PDO is an indis-
pensable monomer for production of polytrimethylene
terephthalate (PTT) resin, a new type of polyester which ex-
hibits good light stability, softness, bulkiness, pollution resis-
tance, and dyeability,1,2 and much better resilience and
stress/recovery properties than polyethylene terephthalate
(PET) and polybutylene terephthalate (PBT).3 These excellent
properties of PTT make it more competitive in the production
of new textiles, carpets, fibers, films, and thermoplastics.4 As
one of the feedstocks of PTT, the potential market capacity of
1,3-PDO was estimated to be more than 2.27 million tons in
2020 and the synthesis of 1,3-PDO has attracted great atten-
tion in recent years.
1,3-PDO can be chemically synthesized via the hydro-
formylation of ethylene oxide followed by hydrogenation, or
via the hydration of acrolein to 3-hydroxypropionaldehyde
followed by aldehyde hydrogenation.5 But the main drawback
of these petroleum-based routines lies in the low selectivity.
Currently, bacterial fermentation of glycerol is a relatively
well-developed approach to obtain 1,3-PDO in high yield.
steps.4,6 In addition, the catalytic selective hydrogenolysis of
glycerol to 1,3-PDO over a heterogeneous catalyst has become
a hot topic in the past decade.
Glycerol, an excessively generated by-product during the
production of soap and biodiesel,7,8 is one of the top twelve
biomass-derived building blocks.9 In published studies, many
efforts were reported on the catalytic hydrogenolysis of the
C–O bond in glycerol, and the obtained products included
1,2-propanediol (1,2-PDO), 1,3-PDO, 1-propanol (1-PO),
2-propanol (2-PO), ethanol and propane.10–13 However, a
large number of catalytic systems gave high yields of 1,2-PDO
and propanol as the major products,14–16 and considerable
studies have been carried out in order to increase the yield of
1,3-PDO via the direct hydrogenolysis of glycerol. At the mo-
ment, the most effective strategy is bifunctional catalytic sys-
tems consisting of an active noble metal (Ir, Rh, or Pt) and
oxophilic additives (mainly MoOx, ReOx or WOx).2,17–22
Among which, it was found that the selectivity toward
1,3-PDO was 51.6% (with an 81.4% conversion of glycerol)
over 0.10AuPt/WOx under relatively mild conditions (140 °C,
1 MPa H2).23 On the surface of a WOx-supported single/
pseudo-single atom Pt catalyst, the space–time yield toward
−1
1,3-PDO reached 3.78 g gPt h−1.24 The yield of 1,3-PDO over
a Key Laboratory of Biomass Chemical Engineering of Ministry of Education,
Department of Chemistry, Zhejiang University, Hangzhou 310028, PR China.
E-mail: zyhou@zju.edu.cn
Ir–ReOx/SiO2 reached 38% with 81% conversion of glycerol at
8 MPa H2 and 120 °C.2 At a higher temperature (200 °C), it
was reported that the yield of 1,3-PDO could reach 38.5%
over Pt/WOx/Al2O3 (at 9 MPa H2 and 4 h).25 Until now, the
highest yield of 1,3-PDO in the literature reached 66% (with a
complete conversion of glycerol, over Pt/WOx/AlOOH at 5
b Center of Chemistry for Frontier Technologies, Department of Chemistry,
Zhejiang University, Hangzhou 310028, PR China
† Electronic supplementary information (ESI) available: Additional experimental
results. See DOI: 10.1039/c9cy01162e
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
MPa H2, 180 °C and Pt/glycerol = 0.9%).26 Therefore, the di-
rect hydrogenolysis of glycerol to 1,3-PDO is still a challeng-
ing task, and there is a strong motivation to explore new al-
ternative methods to obtain 1,3-PDO in high yield.
It has been reported that cyclic ethers with an adjacent
–CH2OH group, such as tetrahydrofurfuryl alcohol and
2-(hydroxymethyl)tetrahydropyran, can be selectively hydroge-
nated to their corresponding α–ω diols, i.e. 1,5-pentanediol
and 1,6-hexanediol over Rh–ReOx/C, Rh–ReOx/SiO2 or Ir–ReOx/
SiO2 catalysts.17,27–29 And it was proposed that this route was
performed in a consecutive ring opening at the C–O bond
neighboring –CH2OH group, followed by hydrogenation to
form the α–ω diols.
Fig. 1 Schematic illustration of the synthesis of Co@NC.
Glycidol, a derivative of glycerol, can be easily prepared
from glycerol over mixed oxides of Ba, Ce, and/or ionic liq-
uid catalysts, and the reported yield of glycidol was higher
than 80%.30–34 Moreover, glycidol can also be synthesized
easily from 2-chloro-1,3-propanediol (β-MCH, a by-product
during the production of epichlorohydrin from glycerol),
and the selectivity of glycidol reached 99% within 30 mi-
nutes even at room temperature.35 Glycidol could be selec-
tively converted to 1,2-PDO via a hydrogenolysis reaction
over noble metal-based Pd/C or Pd/C + Amberlyst-15 cata-
lysts.36,37 In 2017, Gebretsadik found that 1,3-PDO could be
synthesized via the hydrogenation of glycidol, and the
highest yield of 1,3-PDO over a saponite supported Ni cata-
lyst reached 29% at 180 °C.38 More recently, it was found
that the yield of 1,3-PDO could be further increased to
46.1% over a ReOx modified Ni catalyst.39 These pioneering
studies paved a new strategy for the synthesis of 1,3-PDO
from renewable feedstocks.
2 Experimental
2.1 Catalyst preparation
Synthesis of Co@NC. Typically, 6.0 g of CoIJNO3)2·6H2O,
3.5 g of 1,4-benzenedicarboxylic acid and 1.12 g of triethyl-
enediamine
were
dispersed
in
100
mL
N,N-
dimethylmethanamide (DMF) by stirring. Subsequently, the
solution was sealed in a Teflon-lined autoclave and subjected
to a solvothermal reaction at 120 °C in an oil bath and kept
stirring for 2 days. After that, the autoclave was cooled to
room temperature and the resultant precipitates were col-
lected by filtration. The solid precipitates were immersed in
100 mL anhydrous methanol to undergo solvent exchange at
70 °C for 3 h. Finally, the solid sample was dried in a vacuum
at 60 °C for 20 h, and denoted as Co2IJbdc)2IJted).
A carbon film encapsulated Co NP catalyst (identified as
Co@NC) was prepared by pyrolyzing the Co2IJbdc)2IJted) pre-
cursor in a N2 atmosphere at 700 °C for 2 h with a heating
rate of 2 °C min−1.
Recently, carbon-based materials become
a research
hotspot due to their unique properties, such as large specific
surface area, superior electrical conductivity, extraordinary
mechanical flexibility, chemical stability and easy
functionalization. These carbon materials have been widely
used in many fields such as catalysis,40,41 lithium–sulfur bat-
teries,42 lithium ion batteries,43 electromagnetic wave
Synthesis of Co@NG-ZIF. Co@NG-ZIF, as a reference cata-
lyst with a higher N content in its framework, was prepared
in a similar procedure to that of Co@NC. Typically, 4.47 g of
CoIJNO3)2·6H2O and 9.84
g of 2-methylimidazole were
dissolved in 100 mL anhydrous methanol. The solution was
subjected to stirring for 10 h at room temperature and then
kept static for 12 h. The purple solid precursor was obtained
by centrifugation and dried in a vacuum at 60 °C for 20 h.
Co@NG-ZIF was prepared via pyrolyzing this precursor at 700
°C for 2 h with a heating rate of 2 °C min−1.
Synthesis of other reference catalysts. Other cobalt-based
catalysts supported on active carbon (AC), Al2O3, MgO, ZnO,
MCM-41, H-BEA and multiwall carbon nanotubes (MWCNTs)
were prepared via a wetness impregnation method using
CoIJNO3)2·6H2O as a precursor. Typically, these supports were
impregnated with appropriate amounts of precursor solution
(mass ratio of Co/support was controlled at 15%). At the
same time, a series of reference catalysts, such as MWCNT
supported Ni, Cu, Pt, Pd, and Ru catalysts, were prepared via
shielding,44
microwave-adsorbing
multifunctional
ceramic
polymer
composites,47
composites,45,46
meta-
composites,48,49 and sizing agents.50 Among which, graphene
encapsulated metal catalysts exhibit high stability under
harsh conditions due to their unique core–shell structure,
and the synthesis and application of this type of catalyst has
attracted extensive attention.
In this work, a carbon film encapsulated Co nanoparticle
(NP) catalyst (Co@NC) was synthesized via the pyrolysis of
Co2IJbdc)2IJted) (bdc: 1,4-benzenedicarboxylic acid; ted: triethyl-
enediamine) (see Fig. 1), and it was tested in the synthesis of
1,3-PDO from glycidol. It was found that the Co@NC catalyst
was highly active, selective and stable for the formation of
1,3-PDO among the tested Pt-, Pd-, Co-, Ni-, Cu- and Ru-
based catalysts under mild conditions. The reaction mecha-
nism and structure–performance relationship of this Co@NC
catalyst were discussed on the basis of a series of characteri-
zation and controlled experiments.
a wetness impregnation method using NiIJNO3)2, CuIJNO3)
,
2
H2PtCl6, PdCl2, and RuCl3 as precursors (mass ratio of Ni and
Cu/support = 15%, and Pt, Pd and Ru/support = 4%). Before
the catalytic reaction, the impregnated samples were dried at
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
80 °C overnight, calcined in N2 or air (for carbon supports
and other inorganic supports, respectively), and then reduced
in H2.
the mass ratio of Co, Ni, and Cu/glycidol = 9.8 wt%, Pt, Pd,
and Ru/glycidol = 0.66 wt%. After the reactor was sealed, air
was purged by flushing three times with 2.0 MPa H2, then
the autoclave was pressurized with 2.0 MPa H2, and stirred at
120 °C. After the reaction, the reactor was cooled to room
temperature and depressurized carefully. The gas phase was
2.2 Characterization
Power X-ray diffraction (XRD) patterns of the parent
Co2IJbdc)2IJted) and other Co-based catalysts were detected at
room temperature on a Rigaku D/WAX-2500 diffractometer
using Cu Kα radiation (λ = 1.5406 Å) with a 2θ step of 0.02°.
N2 adsorption was measured at its normal boiling point
using an ASAP 2010 analyzer (Micromeritics). Before the N2
adsorption test, the parent Co2IJbdc)2IJted) and the Co@NC
catalyst were pretreated at 100 °C and 150 °C for 4 h under
vacuum, respectively. The specific surface area was calculated
using the Brunauer–Emmett–Teller (BET) method, and the
pore volume and pore size distribution were calculated by the
Barrett–Joyner–Halenda (BJH) method. Thermogravimetric
analysis (TGA) of Co2IJbdc)2IJted) was carried out on a Netzsch
STA 409 thermobalance from room temperature to 800 °C at
a heating rete of 5 °C min−1 in a N2 flow. Scanning electron
microscopy (SEM) images were detected on a Leo Evo Series
SEM (VP 1430, Germany). Transmission electron microscopy
(TEM, JEOL-2010F) was employed to observe the morphol-
ogies and dimensions of the catalysts by using an accelerat-
ing voltage of 200 kV.
collected and analyzed using
a
gas chromatograph
(Shimadzu, 8A) equipped with a TCD detector. n-Butanol was
added as an internal standard to the reaction liquid. The
mixture was quantitatively analyzed with a gas chromato-
graph (Shimadzu, 14B) which was equipped with a 30 m cap-
illary column (DB-WAX 52 CB, USA) and flame ionization de-
tector. At the same time, GC-MS analysis confirmed that the
reaction mixture contained glycidol, 1-PO, 1,2-PDO, 1,3-PDO
and ring-opening products via the alcoholysis reaction be-
tween glycidol and solvent (ethanol), i.e., 3-ethoxy-1,2-
propanediol (RO1) and 2-ethoxy-1,3-propanediol (RO2). No
condensation by-product of glycidol was detected in the reac-
tion mixture. In view of the irreversible formation of RO1 and
RO2 even in a blank experiment (without a catalyst), RO1 and
RO2 were regarded as the unreacted glycidol when calculat-
ing the conversion of glycidol and the selectivity to products.
In this case, the conversion of glycidol was calculated on the
basis of the formed 1-PO, 1,2-PDO and 1,3-PDO, as follows:
Conversion = (total moles of formed 1-PO, 1,2-PDO, and
X-ray photoelectron spectra (XPS) were recorded on a
Perkin-Elmer PHI ESCA system. The X-ray source was a Mg
standard anode 146 (1253.6 eV) at 12 kV and 300 W. The ac-
tual content of Co and other metals in the prepared catalysts
was detected on inductively coupled plasma atomic emission
spectroscopy equipment (ICP, plasma-Spec-II spectrometer,
Perkin-Elmer Optima 2000 instrument) according to the fol-
lowing procedure. 0.1 g of sample was pretreated in air from
room temperature to 900 °C, then the residual solids were
dissolved with 50 mL aqua regia solution. Metal ions in the
above solution were detected via ICP, and these data are sum-
marized in Table S1.†
1,3-PDO)/(moles of glycidol added) × 100%.
Selectivity = (moles of one product)/(moles of glycidol
converted) × 100%.
The specific activity of the added catalyst, turnover fre-
quency (TOF), was calculated as follows:
TOF = (moles of glycidol converted)/(moles of total metal
atoms)/(reaction time).
H2 activation activity of the prepared Co@NC, Co/AC, and
Co@NG-ZIF catalysts was characterized via temperature-
programmed desorption of H2 (H2-TPD), which was
performed in a home-made microreactor equipped with a
mass spectrometer (OmniStar GSD301, Switzerland). 0.1 g of
sample was first pretreated at 450 °C for 1 h in an Ar atmo-
sphere and then cooled to 30 °C. After that, the sample was
exposed to H2 for 1 h. Then, the reactor was purged by Ar at
30 °C for 1.5 h to eliminate the physically adsorbed H2. H2-
TPD was conducted by ramping to 350 °C at 10 °C min−1.
The concentration of H2 (m/z = 2) in the effluent was
recorded using a mass spectrometer.
For each successive use, the catalyst was washed with eth-
anol three times, followed by drying in a vacuum at 60 °C
overnight. The procedure of the controlled experiments, such
as hydrogenation of 1,2-PDO, 1,3-PDO and propylene oxide,
was similar to that above, but using 1,2-PDO, 1,3-PDO and
propylene oxide as the starting substrates.
3 Results
3.1 The structure of the Co2IJbdc)2IJted) precursor
At first, Co2IJbdc)2ted was fabricated by traditional
solvothermal methods according to ref. 51. XRD analysis
found that the diffraction peaks of the synthesized
Co2IJbdc)2IJted) matched well with the simulated pattern given
in the literature, confirming its well crystallized structure
(Fig. 2).52,53 The coordination environment of Co2IJbdc)2IJted)
is shown in Fig. S1,† that is, four oxygen atoms from bdc and
2.3 Catalytic reaction
The selective hydrogenation of glycidol over different cata-
lysts was carried out in a 100 mL Teflon-lined stainless auto-
clave. In a typical procedure, 0.3 g of glycidol and a certain
amount of the catalyst were dispersed in 10 mL ethanol, and
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Fig. 2 XRD patterns of simulated and as-synthesized Co2IJbdc)2IJted).
Fig. 3 XRD pattern of Co@NC (top).
one nitrogen atom from ted are coordinated with each Co
atom. Moreover, it was noteworthy that all peaks (except for
(001)) shifted to lower 2θ values, which should be attributed
to the unusual guest-dependent dynamic behavior of
Co2IJbdc)2ted: the framework expands upon guest (DMF) re-
lease and shrinks upon guest inclusion.52 In this work, the
synthesized Co2IJbdc)2IJted) was subjected to solvent exchange
with methanol and drying in a vacuum, which resulted in the
expansion of the framework and the smaller diffraction an-
gles as DMF molecules were removed.
The quadrangular prism-shaped structure of Co2IJbdc)2-
IJted) was confirmed in its typical SEM image (Fig. S2†). The
N2 adsorption–desorption isotherm of Co2IJbdc)2IJted) was a
type I model with a rapid increase of N2 adsorption quantity
at the P/P0 range of 0–0.01 (Fig. S3†), which corresponded to
its microporous structure. In addition, the calculated BET
surface area of Co2IJbdc)2IJted) was 1700 m2 g−1.
Thermogravimetric analyses (TGA) indicated that there were
two major mass losses during the themolysis of
Co2IJbdc)2IJted) (Fig. S4†). The first mass loss (23.6%) at 25–
400 °C was attributed to the volatilization of water,54,55 and
the other abrupt mass loss (about 44.7%) was observed be-
tween 400 and 505 °C, indicating that the entire structure of
Co2IJbdc)2IJted) underwent drastic thermal decomposition.
And the mass of the remaining catalyst changed slightly after
700 °C.
Nitrogen adsorption–desorption analysis (Fig. 4) showed
that Co@NC had a typical type IV isotherm with a H4-type
hysteresis loop at the P/P0 range of 0.5–1.0. These results con-
firmed that Co@NC was a mesoporous material. The specific
surface area of Co@NC was 130 m2 g−1, and its cumulative
mesopore volume reached 0.1802 cm3 g−1. The calculated
mesopore size in Co@NC was 3.9 nm (Fig. 4, inset).
The morphology of the Co@NC catalyst was observed by
SEM (Fig. 5a). It can be found that Co@NC maintained the
quadrangular prism-shaped structure similar to the parent
Co2IJbdc)2IJted) (Fig. S2†). TEM analysis disclosed that the
Co@NC catalyst was composed of spherical Co NPs embed-
ded in the thick carbon matrix (Fig. 5b and c), in which Co
NPs have sizes of 13–35 nm. HRTEM images further
disclosed that those Co NPs were encapsulated in a layer of
carbon film (around 1.0–1.2 nm) (Fig. 5d). The average size
of Co NPs in Co@NC was estimated to be 25.6 nm from the
TEM analysis of over 300 randomly selected particles.
3.3 Hydrogenation of glycidol over different catalysts
Table
1 summarizes the performance of Co@NC and
MWCNT supported Pt-, Pd-, Co-, Ni-, Cu-, Ru-based catalysts
3.2 The fine structure of the Co@NC catalyst
Fig. 3 shows the XRD pattern of the Co@NC catalyst. It was
found that mainly metallic Co was detected in Co@NC, and
the characteristic diffraction peaks at 44.5, 51.8 and 76.4°
corresponded to the (111), (200) and (220) crystalline plane
reflections of Co0,56,57 respectively. These results demon-
strated that Co2+ could be reduced to the metallic phase via
carbothermic reduction under the pyrolysis conditions. The
average crystalline size of Co in Co@NC was calculated to be
23.1 nm according to the Scherrer equation. The content as
well as the particle size of Co in Co@NC and other reference
catalysts is summarized in Table S1.†
Fig. 4 N2 adsorption–desorption isotherm and pore size distribution
(inset) of the Co@NC catalyst.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
3.4 Hydrogenation of glycidol over Co-based catalysts
Table 2 compares the performance of Co-based catalysts that
were supported on active carbon (AC), and traditional oxides
or zeolites (ZnO, Al2O3, MgO, MCM-41, and H-BEA) for the se-
lective hydrogenation of glycidol. Once again, it was con-
firmed that the Co-based catalyst was selective for the forma-
tion of 1,3-PDO from glycidol (except Co/AC and Co@NG-
ZIF). And carbon material supported Co catalysts, such as
Co@NC, Co/AC and Co/MWCNTs, were more active than ox-
ide supported Co NPs. The carbon film encapsulated Co cata-
lyst (Co@NC) still exhibited the best performance as it pro-
vided a high activity (61.0% conversion) and the highest
selectivity toward 1,3-PDO (59.0%) (entry 1). On the other
hand, the conversion of glycidol decreased to 53.5% over
commercial AC supported Co catalysts (entry 2, Co/AC), and
the selectivity toward 1,3-PDO was only 37.8%, the selectivity
to less valuable and undesired 1-PO increased to 53.6%.
What's more, the selectivity to 1,3-PDO decreased sharply to
11.0% over Co@NG-ZIF, while the selectivity to 1-PO was as
high as 85.9% despite its highest activity (entry 3, 65.0% con-
version at 2 h). Traditional oxide (ZnO, Al2O3, and MgO) or
zeolite (MCM-41 and H-BEA)-supported Co-based catalysts
can catalyze the hydrogenation of glycidol to 1,3-PDO with a
selectivity of 40–56%, but the corresponding hydrogenation
rates were quite low. As listed in entries 4–8, the conversion
of glycidol over these supported Co catalysts was less than
28.4%. These results demonstrated that the nature of the
support played a crucial role in the hydrogenation of glycidol.
In order to disclose the different product distributions
during the hydrogenation of glycidol, the time courses over
Co@NC, Co/AC and Co@NG-ZIF are compared in Fig. S5.† It
was interesting to note that 1,3-PDO was the major product
over Co@NC from the beginning of the reaction (Fig. S5(a)†).
On the other hand, mainly 1-PO formed over Co/AC and
Co@NG-ZIF (Fig. S5(b) and (c)†). These results indicated that
1-PO was the main by-product, and the challenge in the selec-
tive hydrogenation of glycidol to 1,3-PDO over Co-based cata-
lysts lies in suppressing the formation of 1-PO.
Fig. 5 (a) SEM and (b–d) TEM images of Co@NC (the scale bar in Fig. d
is 10 nm).
for the hydrogenation of glycidol at 2.0 MPa H2 and 120 °C.
It can be found that Pt- and Pd-based catalysts were active for
the hydrogenation of glycidol, and the highest conversion of
glycidol (in 5 h) reached 69.6% over Pt/MWCNTs, but mainly
1,2-PDO formed in the reaction mixture. On the other hand,
Ni-, Cu- and Ru-based catalysts were less active under the
same reaction conditions. It is quite interesting to note that
Co/MWCNTs were selective for the formation of 1,3-PDO
from glycidol, and the selectivity to 1,3-PDO was 50.4% (with
a 43.4% conversion of glycidol). Furthermore, the carbon film
encapsulated Co catalyst (Co@NC) was more active and selec-
tive for this reaction, and the selectivity to 1,3-PDO increased
to 59.0% (with a 61.0% conversion of glycidol). These results
indicated that the Co-based catalyst was promising for the se-
lective hydrogenation of glycidol to 1,3-PDO. A blank experi-
ment without a catalyst was carried out under the same con-
ditions (Table 1, entry 1). No hydrogenation products (1,3-
PDO, 1,2-PDO and 1-PO) were detected except the alcoholysis
products (RO1 and RO2), which indicated that those carbon
film encapsulated Co NPs were the indispensable active sites
in the hydrogenation of glycidol. The alcoholysis reaction of
glycidol could occur without a catalyst at 120 °C, and it seems
that the formation of RO1 and RO2 is spontaneous.
Subsequently, the hydrogenation of glycidol under various
H2 pressures, temperatures, prolonged reaction times, and
Table 1 Selective hydrogenation of glycidol over different metal catalystsa
Selectivity (%)
Entry
Catalyst
Conv. (%)
TOFb (h−1
)
1,3-PDO
1,2-PDO
1-PO
1
2
3
4
5
6
7
8
—
0
0
—
0.9
5.7
50.4
28.1
2.2
40.4
59.0
—
—
24.8
6.3
37.6
4.0
93.6
36.2
31.7
Pt/MWCNTs
Pd/MWCNTs
Co/MWCNTs
Ni/MWCNTs
Cu/MWCNTs
Ru/MWCNTs
Co@NC
69.6
49.6
43.4
23.9
14.5
7.9
55.53
21.59
0.7046
0.3864
0.2539
3.266
0.9903
74.3
88.0
12.0
67.9
4.2
23.4
9.3
61.0
a
b
Reaction conditions: 0.3 g glycidol in 10.0 mL ethanol with 9.8 wt% of Co, Ni, Cu, or 0.66 wt% of Pt, Pd, Ru, 2.0 MPa H2, 120 °C, 5 h. TOF
was calculated on the basis of total metal atoms: TOF = (moles of glycidol converted)/(moles of total metal atoms)/(5 h).
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Table 2 Selective hydrogenation of glycidol over different Co-based catalystsa
Selectivity (%)
1,3-PDO
Conv.
(%)
TOF
Entry
Catalyst
(h−1 c
)
1,2-PDO
1-PO
1
2
3
4
5
6
7
8
Co@NC
61.0
53.5
65.0
28.4
11.9
7.5
0.9903
0.8685
2.6381
0.4611
0.1932
0.1218
0.0584
0.0097
59.0
37.8
11.0
55.2
47.3
54.6
56.1
41.4
9.3
8.6
3.1
17.8
6.1
7.1
31.7
53.6
85.9
27.0
46.6
38.3
38.6
15.2
Co/AC
Co@NG-ZIFb
Co/ZnO
Co/MCM-41
Co/Al2O3
Co/H-BEA
Co/MgO
3.6
0.6
5.3
43.4
a
b
c
Reaction conditions: 0.3 g glycidol in 10.0 mL ethanol with 9.8 wt% of Co, 2.0 MPa H2, 120 °C, 5 h. 2 h. TOF was calculated on the basis
of total Co: TOF = (moles of glycidol converted)/(moles of total Co atoms)/(reaction time).
different catalyst loadings was investigated in detail. Fig. S6†
shows that the conversion of glycidol and the yield of
1,3-PDO increased obviously from 46.5% and 27.8% (at 1
MPa) to 67.3% and 41% (at 3 MPa), respectively while the
yield of 1,2-PDO and 1-PO increased slightly from 4.9% and
13.8% (at 1 MPa) to 6.3% and 20% (at 3 MPa). Then the ex-
cellent performance of Co@NC was further confirmed under
various reaction times at 100 °C, 120 °C and 140 °C (Fig.
S7†). It can be found that 1,3-PDO was the dominant product
over the Co@NC catalyst in all experiments and the yield of
1,3-PDO increased continuously with the reaction time. The
detected conversion of glycidol was as high as 83.6% and the
best yield of 1,3-PDO reached 50.4% at 140 °C, 4 h (Fig.
S7(c)†). Regrettably, the calculated conversion of glycidol and
the yield of 1,3-PDO cannot be increased with the prolonged
reaction time (above 4 h) because of the competitive forma-
tion of 3-ethoxy-1,2-propanediol (RO1) and 2-ethoxy-1,3-
propanediol (RO2) between glycidol and solvent. These re-
sults meant that the reaction ended because there was no
free glycidol in the reaction mixture. Fig. S8† shows that the
conversion of glycidol increased rapidly from 24.4% to
83.6%, and the yield of all the products increased simulta-
neously as the loading of Co increased from 2.0 wt% to 9.8
wt%. However, these data changed slightly with excess cata-
lysts, which might be attributed to the competitive
alcoholysis reaction of glycidol.
ous loss of activity and selectivity toward 1,3-PDO was ob-
served during the successive seven runs. The excellent stabil-
ity of the Co@NC catalyst could be ascribed to the fact that
Co NPs were well encapsulated in the carbon film, which
protected the Co NPs from sintering, leaching or oxidation
during the reaction process. The spent catalyst after 7 cycles
was characterized, and the morphology of recycled Co@NC is
shown in Fig. S9.† It was confirmed that the core–shell struc-
ture in Co@NC was well preserved. XRD analysis also indi-
cated that metallic Co was still the dominant phase in this
spent Co@NC (Fig. S10†).
4 Discussion
The above experiments indicated that Co-based catalysts were
selective for the formation of 1,3-PDO from glycidol. Among
which, the best yield of 1,3-PDO reached 50.4% over the car-
bon film encapsulated Co NP catalyst (Co@NC). At the same
time, it was found that the fine structure of the Co based cat-
alysts played a crucial role for the selective formation of
1,3-PDO. It is of great importance to disclose the real reason
for the opposite product distribution over Co@NC (mainly
1,3-PDO formed) and Co@NG-ZIF (mainly 1-PO formed).
Table S2† compares the previously published results with
those obtained in this work. First, it was found that 1,2-PDO
was the major product when the hydrogenation reaction was
catalyzed by noble metal-based Pd/C or Pd/C-A15 catalysts,
which matched well with our conclusion (Table 1, entry 3).
The selectivity to 1,3-PDO reached 26% over the saponite-
supported Ni catalyst (NiIJ40)/Sap). More recently, it was found
that the selectivity to 1,3-PDO increased to 47% over a ReOx-
modified Ni catalyst (NiReIJ7)/MS). In this work, the selectivity
to 1,3-PDO reached up to 60.3% over Co@NC, and the best
yield of 1,3-PDO was 50.4%, which was better than the results
in published studies.
The stability of the Co@NC catalyst was investigated to as-
sess its potential in the hydrogenation of glycidol, and the
performance of recycled Co@NC is shown in Fig. 6. No obvi-
Fig.
6 Recycling of Co@NC for the selective hydrogenation of
glycidol. Reaction conditions: 0.3 g glycidol in 10.0 mL ethanol with
9.8 wt% of Co, 2.0 MPa H2, 120 °C, 5 h.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
Undoubtedly, PDOs were derived from the cleavage and
hydrogenation of one C–O bond in the glycidol. However, the
controversy was that 1-PO may be generated via the hydro-
genolysis of PDOs and/or the simultaneous cleavage and hy-
drogenation of two C–O bonds in glycidol. In this case, a se-
ries of controlled experiments were carried out in order to
disclose the formation pathway of 1-PO. At first, 0.3 g of
1,3-PDO or 0.3 g of 1,2-PDO was separately used as the
starting material to be subjected to hydrogenation over
Co@NC, Co/AC, and Co@NG-ZIF catalysts for 4 h under 2.0
MPa H2 and 140 °C, and the results are summarized in Table
S3.† It was found that no 1-PO was detected in the reaction
mixture (Table S3,† entry 1–6). Therefore, it could be con-
cluded that 1,3-PDO and 1,2-PDO were stable under the reac-
tion conditions (140 °C), and all three catalysts did not have
the ability to catalyze the deoxidation reaction of propanediol
(or propanol). Therefore, it was reasonable to speculate that
1-PO was generated via the simultaneous cleavage and hydro-
genation of two C–O bonds in glycidol. The results of the hy-
drogenation of propylene oxide gave an irrefutable support
for the above suggestion. It was found that mainly propane
formed over Co@NG-ZIF (with a 87.0% selectivity at a 95.0%
conversion of propylene oxide) (Table S3,† entry 8).
Given the fact that different selectivities to 1-PO were
detected over Co@NC, Co/AC, and Co@NG-ZIF in the hydro-
genation of glycidol (31.7, 53.6, and 85.9%, respectively), the
remaining question is what promoted the simultaneous
cleavage and hydrogenation of two C–O bonds in glycidol.
First, the morphology of Co/AC and Co@NG-ZIF was charac-
terized in detail by TEM (Fig. S11†) and compared with that
of Co@NC (Fig. 5). It was found that mainly naked Co NPs
formed in the Co/AC catalyst (Fig. S11a and b†), while the
Co@NG-ZIF catalyst consisted of a thick carbon matrix deco-
rated with graphene shell encapsulated Co NPs (Fig. S11c
and d†). It seemed that the core–shell structure of Co@NG-
ZIF was similar to that of Co@NC (Fig. 5). But XPS analysis
further indicated that the content of the N element in
Co@NG-ZIF (5.8%) was much higher than that in Co@NC
(1.1%, Table S4†), demonstrating that much more N atoms
were incorporated into the graphene shell in Co@NG-ZIF.
H2-TPD experiments further disclosed that these structure
differences among Co/AC (naked Co NPs), Co@NC (thin car-
bon film with low content of N (1.1%)) and Co@NG-ZIF
(thick graphene shell with high content of N (5.8%)), played
a crucial role on the adsorption and activation of H2 (Fig. 7
and 8). On the surface of those core–shell structured cata-
lysts, it was disclosed that electrons can transfer from the
metal core to the graphene shell, which increases the
electron density of the graphene surface and therefore pro-
motes the adsorption and activation of reactants on the
graphene shell to enhance the catalytic activity.58,59 There-
fore, we speculated that H2 adsorption over Co@NC and
Co@NG-ZIF mainly occurred at the carbon shell outside the
Co NPs, rather than on the carbon support. The two weak
peaks of Co@NC at 68 and 155 °C could be assigned to the
desorption of molecular hydrogen from the thin carbon film
Fig. 7 H2-TPD profiles of (a) Co@NC, (b) Co/AC and (c) Co@NG-ZIF
catalysts.
outside Co NPs,15,16,60,61 and of dissociated hydrogen from
small amounts of N atoms doped in this carbon film,62 re-
spectively (Fig. 7(a) and 8A). On the surface of Co@NG-ZIF,
large amounts of dissociated hydrogen were desorbed at 150
°C (Fig. 7(c)), indicating that the abundant N atoms in the
graphene shell greatly boosted the adsorption and activation
of H2 on both the graphene shell and the N atoms in it
(Fig. 8C).62 As for Co/AC, large amounts of dissociated hydro-
gen were desorbed from the naked Co NPs at 115 °C (Fig. 7b
and 8B).63
The above analysis confirmed that H2 molecules were ex-
cessively adsorbed and dissociated (or activated) into H
atoms on the naked Co NPs (in Co/AC)64,65 and the N-rich
graphene shell outside Co NPs (in Co@NG-ZIF)62 at room
temperature. And these highly activated H atoms promoted
the simultaneous cleavage and hydrogenation of two C–O
bonds in glycidol, thus, giving high selectivity to 1-PO during
the hydrogenation of glycidol. Recently, Zeng et al. also
disclosed that the N atoms incorporated into Co nanosheets
could provide distinct hydrogen adsorption and activation
sites to highly boost the catalytic activity toward CO2 hydroge-
nation; moreover, the hydrogen species adsorbed on N atoms
were more active than those on Co NPs.62 Therefore, it was
reasonable to conclude that the abundant N atoms in
Co@NG-ZIF (5.8%) led to the excessive adsorption and activa-
tion of H2 molecules, providing high hydrogenation activity
Fig. 8 Schematic diagram of H2 adsorption on different Co catalysts.
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
but at the cost of the high selectivity of undesired 1-PO in the
hydrogenation of glycidol. Similarly, the excessively activated
hydrogen species on Co@NG-ZIF also stimulated the simulta-
neous hydrogenation of two C–O bonds in propylene oxide to
propane (Table S3,† entry 8). Meanwhile the thin carbon film
outside Co NPs would inhibit the excessive adsorption and
activation of H2 molecules, and the moderate amount of hy-
drogen species on Co@NC contributed to the selective cleav-
age and hydrogenation of the C–O bond of neighboring –CH2
OH to form 1,3-PDO. The conclusion was also confirmed by
the hydrogenation of propylene oxide over Co@NC, that is,
the C–O bond of neighboring –CH3 was selectively hydroge-
nated, and the obtained selectivity to 1-propanol was 86.8%
(Table S3,† entry 7).
5 G. A. Kraus, Clean, 2008, 36, 648–651.
6 T. Kurosaka, H. Maruyama, I. Naribayashi and Y. Sasaki,
Catal. Commun., 2008, 9, 1360–1363.
7 Y. Nakagawa, M. Tamura and K. Tomishige, J. Mater. Chem.,
2014, 2, 6688–6702.
8 H. Zhang, Z. Hu, L. Huang, H. Zhang, K. Song, L. Wang, Z.
Shi, J. Ma, Y. Zhuang and W. Shen, ACS Catal., 2015, 5,
2548–2558.
9 P. Gallezot, Chem. Soc. Rev., 2012, 41, 1538–1558.
10 J. T. Dam and U. Hanefeld, ChemSusChem, 2011, 4,
1017–1034.
11 L. Ma and D. He, Top. Catal., 2009, 52, 834–844.
12 O. M. Daniel, A. DeLaRiva, E. L. Kunkes, A. K. Datye, J. A.
Dumesic and R. J. Davis, ChemCatChem, 2010, 2, 1107–1114.
13 H. Zhao, Y. Jiang, P. Chen, J. Fu, X. Lu and Z. Hou, Green
Chem., 2018, 20, 4299–4307.
14 L. Zheng, X. Li, W. Du, D. Shi, W. Ning, X. Lu and Z. Hou,
Appl. Catal., B, 2017, 203, 146–153.
15 S. Xia, R. Nie, X. Lu, L. Wang, P. Chen and Z. Hou, J. Catal.,
2012, 296, 1–11.
16 S. Xia, L. Zheng, W. Ning, L. Wang, P. Chen and Z. Hou,
J. Mater. Chem. A, 2013, 1, 11548–11552.
17 M. Chia, Y. J. Pagán-Torres, D. Hibbitts, Q. Tan, H. N. Pham,
A. K. Datye, M. Neurock, R. J. Davis and J. A. Dumesic, J. Am.
Chem. Soc., 2011, 133, 12675–12689.
18 Y. Amada, Y. Shinmi, S. Koso, T. Kubota, Y. Nakagawa and
K. Tomishige, Appl. Catal., B, 2011, 105, 117–127.
19 M. Edake, M. Dalil, M. J. D. Mahboub, J.-L. Dubois and G. S.
Patience, RSC Adv., 2017, 7, 3853–3860.
20 L.-Z. Qin, M.-J. Song and C.-L. Chen, Green Chem., 2010, 12,
1466–1472.
5 Conclusions
In summary, a carbon film encapsulated Co NP catalyst
(Co@NC) was synthesized via the direct pyrolysis of
Co2IJbdc)2IJted) in a N2 atmosphere. And it was confirmed that
Co@NC was highly active, selective and stable for the hydro-
genation of glycidol to 1,3-PDO among the tested Pt-, Pd-, Co-
, Ni-, Cu- and Ru-based catalysts under mild conditions. The
best yield of 1,3-PDO over the Co@NC catalyst reached 50.4%
at 140 °C, 2.0 MPa H2 and 4 h. A series of characterization re-
sults disclosed that the carbon film could inhibit the exces-
sive adsorption and activation of H2 molecules on the naked
Co NPs, and a moderate amount of activated hydrogen spe-
cies on Co@NC contributed to the selective hydrogenation of
the C–O bond of neighboring –CH2OH to form 1,3-PDO.
Meanwhile the excessively dissociated H atoms on naked Co
NPs (in Co/AC) and abundant N atoms (in Co@NG-ZIF) stim-
ulated the simultaneous cleavage and hydrogenation of both
C–O bonds in glycidol to undesired 1-PO.
21 Y. Fan, S. Cheng, H. Wang, J. Tian, S. Xie, Y. Pei, M. Qiao
and B. Zong, Appl. Catal., B, 2017, 217, 331–341.
22 F. Cai, X. Song, Y. Wu, J. Zhang and G. Xiao, ACS Sustainable
Chem. Eng., 2018, 6, 110–118.
Conflicts of interest
23 X. Zhao, J. Wang, M. Yang, N. Lei, L. Li, B. Hou, S. Miao, X.
Pan, A. Wang and T. Zhang, ChemSusChem, 2017, 10,
819–824.
There are no conflicts to declare.
24 J. Wang, X. Zhao, N. Lei, L. Li, L. Zhang, S. Xu, S. Miao, X.
Pan, A. Wang and T. Zhang, ChemSusChem, 2016, 9,
784–790.
25 S. García-Fernández, I. Gandarias, J. Requies, F. Soulimani,
P. L. Arias and B. M. Weckhuysen, Appl. Catal., B, 2017, 204,
260–272.
Acknowledgements
This work was financially supported by the National Natural
Science Foundation of China (Contract No. 21773206,
21473155) and the Natural Science Foundation of Zhejiang
Province (Contract No. L12B03001).
26 R. Arundhathi, T. Mizugaki, T. Mitsudome, K. Jitsukawa and
K. Kaneda, ChemSusChem, 2013, 6, 1345–1347.
27 S. Koso, Y. Nakagawa and K. Tomishige, J. Catal., 2011, 280,
221–229.
28 K. Chen, K. Mori, H. Watanabe, Y. Nakagawa and K.
Tomishige, J. Catal., 2012, 294, 171–183.
29 K. Chen, S. Koso, T. Kubota, Y. Nakagawa and K. Tomishige,
ChemCatChem, 2010, 2, 547–555.
30 S. E. Kondawar, C. R. Patil and C. V. Rode, ACS Sustainable
Chem. Eng., 2017, 5, 1763–1774.
References
1 C.-H. Zhou, J. N. Beltramini, Y.-X. Fan and G. Q. Lu, Chem.
Soc. Rev., 2008, 37, 527–549.
2 Y. Nakagawa, Y. Shinmi, S. Koso and K. Tomishige, J. Catal.,
2010, 272, 191–194.
3 H. Liu, X. Ou, S. Zhou and D. Liu, Plastics from Bacteria:
Natural Functions and Applications, 2010, vol. 14, pp. 405–
425.
4 M. Pagliaro, R. Ciriminna, H. Kimura, M. Rossi and C. D.
Pina, Angew. Chem., Int. Ed., 2007, 46, 4434–4440.
31 Y. Zhou, F. Ouyang, Z.-B. Song, Z. Yang and D.-J. Tao, Catal.
Commun., 2015, 66, 25–29.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2019

View Article Online
Catalysis Science & Technology
Paper
32 Y. T. Algoufi, U. G. Akpan, M. Asif and B. H. Hameed, Appl.
Catal., A, 2014, 487, 181–188.
33 M. K. Munshi, S. M. Gade, V. H. Rane and A. A. Kelkar, RSC
Adv., 2014, 4, 32127–32133.
34 S. M. Gade, M. K. Munshi, B. M. Chherawalla, V. H. Rane
and A. A. Kelkar, Catal. Commun., 2012, 27, 184–188.
35 D. Cespi, R. Cucciniello, M. Ricciardi, C. Capacchione, I.
Vassura, F. Passarini and A. Proto, Green Chem., 2016, 18,
4559–4570.
48 H. Gu, X. Xu, M. Dong, P. Xie, Q. Shao, R. Fan, C. Liu, S.
Wu, R. Wei and Z. Guo, Carbon, 2019, 147, 550–558.
49 Z. Zhao, P. Bai, R. D. K. Misra, M. Dong, R. Guan, Y. Li, J.
Zhang, L. Tan, J. Gao, T. Ding, W. Du and Z. Guo, J. Alloys
Compd., 2019, 792, 203–214.
50 Z. Wu, H. Cui, L. Chen, D. Jiang, L. Weng, Y. Ma, X. Li, X.
Zhang, H. Liu, N. Wang, J. Zhang, Y. Ma, M. Zhang, Y.
Huang and Z. Guo, Compos. Sci. Technol., 2018, 164,
195–203.
36 R. Cucciniello, C. Pironti, C. Capacchione, A. Proto and M.
Di Serio, Catal. Commun., 2016, 77, 98–102.
51 Y. Xu, W. Tu, B. Zhang, S. Yin, Y. Huang, M. Kraft and R.
Xu, Adv. Mater., 2017, 29, 1605957–1605964.
37 M. Ricciardi, D. Cespi, M. Celentano, A. Genga, C. Malitesta,
A. Proto, C. Capacchione and R. Cucciniello, Sustainable
Chem. Pharm., 2017, 6, 10–13.
38 F. B. Gebretsadik, J. Ruiz-Martinez, P. Salagre and Y.
Cesteros, Appl. Catal., A, 2017, 538, 91–98.
52 D. N. Dybtsev, H. Chun and K. Kim, Angew. Chem.,
2004, 116, 5143–5146.
53 J. Y. Lee, D. H. Olson, L. Pan, T. J. Emge and J. Li, Adv.
Funct. Mater., 2007, 17, 1255–1262.
54 C. Hou, Z. Tai, L. Zhao, Y. Zhai, Y. Hou, Y. Fan, F.
Dang, J. Wang and H. Liu, J. Mater. Chem. A, 2018, 6,
9723–9736.
39 F. B. Gebretsadik, J. Llorca, P. Salagre and Y. Cesteros,
ChemCatChem, 2017, 9, 3670–3680.
40 P. Yang, L. Yang, Q. Gao, Q. Luo, X. Zhao, X. Mai, Q. Fu, M.
Dong, J. Wang, Y. Hao, R. Yang, X. Lai, S. Wu, Q. Shao, T.
Ding, J. Lin and Z. Guo, Chem. Commun., 2019, 55,
9011–9014.
41 Q. Chen, Q. Yin, A. Dong, Y. Gao, Y. Qian, D. Wang, M.
Dong, Q. Shao, H. Liu, B.-H. Han, T. Ding, Z. Guo and N.
Wang, Polymer, 2019, 169, 255–262.
55 C. Hou, J. Wang, W. Du, J. Wang, Y. Du, C. Liu, J. Zhang, H.
Hou, F. Dang, L. Zhao and Z. Guo, J. Mater. Chem. A,
2019, 7, 13460–13472.
56 K. Shen, L. Chen, J. Long, W. Zhong and Y. Li, ACS Catal.,
2015, 5, 5264–5271.
57 R. Das, P. Pachfule, R. Banerjee and P. Poddar, Nanoscale,
2012, 4, 591–599.
42 M. Liu, Q. Meng, Z. Yang, X. Zhao and T. Liu, Chem.
Commun., 2018, 54, 5090–5093.
43 M. Idrees, S. Batool, J. Kong, Q. Zhuang, H. Liu, Q. Shao, N.
Lu, Y. Feng, E. K. Wujcik, Q. Gao, T. Ding, R. Wei and Z.
Guo, Electrochim. Acta, 2019, 296, 925–937.
44 C. Wang, V. Murugadoss, J. Kong, Z. He, X. Mai, Q. Shao, Y.
Chen, L. Guo, C. Liu, S. Angaiah and Z. Guo, Carbon,
2018, 140, 696–733.
45 H. Gu, H. Zhang, C. Ma, X. Xu, Y. Wang, Z. Wang, R. Wei,
H. Liu, C. Liu, Q. Shao, X. Mai and Z. Guo, Carbon,
2019, 142, 131–140.
58 J. Deng, D. Deng and X. Bao, Adv. Mater., 2017, 29, 1606967.
59 D. Deng, L. Yu, X. Chen, G. Wang, L. Jin, X. Pan, J. Deng, G.
Sun and X. Bao, Angew. Chem., Int. Ed., 2013, 52, 371–375.
60 J.-J. Wang, J.-R. Xie, Y.-H. Huang, B.-H. Chen, G.-D. Lin and
H.-B. Zhang, Appl. Catal., A, 2013, 468, 44–51.
61 M.-H. Zhang, Z.-M. Liu, G.-D. Lin and H.-B. Zhang, Appl.
Catal., A, 2013, 451, 28–35.
62 L. Wang, W. Zhang, X. Zheng, Y. Chen, W. Wu, J. Qiu, X.
Zhao, X. Zhao, Y. Dai and J. Zeng, Nat. Energy, 2017, 2,
869–876.
63 S. Xia, Z. Yuan, L. Wang, P. Chen and Z. Hou, Appl. Catal.,
A, 2011, 403, 173–182.
64 M. Pozzo and D. Alfè, Int. J. Hydrogen Energy, 2009, 34,
1922–1930.
46 M. Dong, C. Wang, H. Liu, C. Liu, C. Shen, J. Zhang, C. Jia,
T. Ding and Z. Guo, Macromol. Mater. Eng., 2019, 304,
1900010.
47 L. Chen, J. Zhao, L. Wang, F. Peng, H. Liu, J. Zhang, J. Gu
and Z. Guo, Ceram. Int., 2019, 45, 11756–11764.
65 T. Gomez, E. Florez, J. A. Rodriguez and F. Illas, J. Phys.
Chem. C, 2011, 115, 11666–11672.
This journal is © The Royal Society of Chemistry 2019
Catal. Sci. Technol.