Study of the active site for acetylene hydrochlorination in AuCl3/C catalysts

Article abstract of DOI:10.1016/j.jcat.2015.07.008

AuCl₃/C catalysts were prepared using activated carbon pretreated at different temperatures. The set of catalysts was evaluated for acetylene hydrochlorination, and the most stable catalyst exhibited the weakest acetylene adsorption capacity, as characterized by the method of temperature-programmed desorption (TPD). The catalysts were also investigated using X-ray diffraction, X-ray photoelectron spectroscopy, Fourier transform infrared spectroscopy, and other methods. It is shown that the activated carbon works as a constituent part of the active site that is responsible for the activation of acetylene. The result provides evidence for the assumption that the Au³⁺ at the AuCl₃/C interface of the catalyst is the active site of the classical AuCl₃/C catalyst.

Full text of DOI:10.1016/j.jcat.2015.07.008

Journal of Catalysis 330 (2015) 273–279
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Study of the active site for acetylene hydrochlorination in AuCl3/C
catalysts
Songlin Chao a, Qingxin Guan a, Wei Li a,b,
⇑
a Key Laboratory of Advanced Energy Materials Chemistry (Ministry of Education), College of Chemistry, Nankai University, Tianjin 300071, China
b Collaborative Innovation Center of Chemical Science and Engineering (Tianjin), Nankai University, Tianjin 300071, China
a r t i c l e i n f o
a b s t r a c t
Article history:
AuCl3/C catalysts were prepared using activated carbon pretreated at different temperatures. The set of
catalysts was evaluated for acetylene hydrochlorination, and the most stable catalyst exhibited the weak-
est acetylene adsorption capacity, as characterized by the method of temperature-programmed desorp-
tion (TPD). The catalysts were also investigated using X-ray diffraction, X-ray photoelectron
spectroscopy, Fourier transform infrared spectroscopy, and other methods. It is shown that the activated
carbon works as a constituent part of the active site that is responsible for the activation of acetylene. The
result provides evidence for the assumption that the Au3+ at the AuCl3/C interface of the catalyst is the
active site of the classical AuCl3/C catalyst.
Received 4 April 2015
Revised 26 June 2015
Accepted 2 July 2015
Keywords:
Acetylene hydrochlorination
Active site
Temperature-programmed desorption
Ó 2015 Elsevier Inc. All rights reserved.
Au
Activated carbon
1. Introduction
by growing an N-doped carbon layer out of preshaped silicon car-
bide granules as the catalyst for acetylene hydrochlorination, and
Vinyl chloride monomer (VCM) is an important chemical raw
material that is used to manufacture polyvinylchloride. Because
of China’s abundant coal reserves, the manufacture of VCM is pri-
marily performed via acetylene hydrochlorination, by means of
HgCl2 supported on activated carbon as the catalyst [1,2].
Because HgCl2 is toxic and apt to sublime, HgCl2 supported cata-
lysts have a relatively short lifetime and are environmentally haz-
ardous [3].
The pioneering work and systematic studies carried out by
Hutchings and co-workers showed that supported AuCl3 catalysts
would be viable catalysts for acetylene hydrochlorination [3–6].
Hutchings et al. demonstrated that Au3+ played a crucial role in
the reaction, and the reduction of Au3+ to Au0 during the reaction
caused deactivation of the catalyst [4,7–10]. In addition, recent
research has shown that an excess of Au3+ does not contribute to
catalytic activity, and the location of Au3+ at the AuCl3/C interface
of the catalyst is likely to be related to the efficiency of the catalyst
[7,11].
demonstrated that the carbon atoms bonded to pyrrolic nitrogen
atoms are the active sites [13]. Wei et al. reported that N-doped
carbon nanotubes are activated for the reaction, based on the con-
sideration that the electron affinity of carbon can be tuned as the
active metal cation for acetylene hydrochlorination by doping [15].
In spite of the outstanding work that has been reported, the
active site of the classic supported gold catalyst is still not clear.
The previous work mentioned above inspired us consider that
the activated carbon may play a central role in the overall reaction
mechanism of AuCl3/C catalysts. Activated carbon has abundant
surface functional groups that could affect the reaction mechanism
and are easily modified by suitable thermal or chemical
post-treatments [16,17]. This prompted us to investigate the effect
of activated carbon’s surface functional groups by modifying acti-
vated carbon using a thermal treatment that will not add new
elements.
Therefore, we modified the surface functional groups on acti-
vated carbon to achieve a change in catalytic performance. The
activity was examined using X-ray photoelectron spectroscopy
(XPS), X-ray diffraction (XRD), temperature-programmed desorp-
tion (TPD), and several other characterization techniques. The
results of TPD of acetylene from the catalyst indicated that Au3+
has a negative influence on acetylene activation, but the surface
functional groups play an important role. This could provide a
new perspective on the reaction mechanism. Hutchings et al. have
Moreover, several metal-free catalysts have been reported
recently [12–14]. Bao et al. reported a nanocomposite obtained
⇑
Corresponding author at: Key Laboratory of Advanced Energy Materials
Chemistry (Ministry of Education), College of Chemistry, Nankai University, Tianjin
300071, China. Fax: +86 22 23508662.
E-mail address: weili@nankai.edu.cn (W. Li).
http://dx.doi.org/10.1016/j.jcat.2015.07.008
0021-9517/Ó 2015 Elsevier Inc. All rights reserved.

274
S. Chao et al. / Journal of Catalysis 330 (2015) 273–279
proposed that the efficiency of the catalyst is probably related to
the location of Au3+ at the Au/C interface, and this is an important
aspect so far neglected for the catalysts [11]. Thus, we demon-
strated that the surface functional groups are a constituent part
of the active site and provide a preliminary discussion on the role
of each part.
2.2. Catalytic activity test
Catalysts used for acetylene hydrochlorination were tested in a
fixed-bed microreactor. Ceramic rings (diameter 4 mm and length
4 mm) were used to extend the bed length above the catalysts
(3.2 g, 10 mL), which could mix and preheat the reactant at the
same time. Prior to the reaction, the catalysts were pretreated
in situ with HCl (30 mL minÀ1) at 180 °C for 1 h [9]. After that,
HCl (66 mL minÀ1) and C2H2 (60 mL minÀ1) were fed to the heated
reactor (180 °C) via calibrated mass flow controllers, with a C2H2
gas hourly space velocity (GHSV) of 360 hÀ1. A blank experiment
was carried out using an empty reactor filled with ceramic rings
under the same conditions, and the ceramic rings did not show
any catalytic activity. The gas phase products were analyzed on
line using a gas chromatograph (GC) equipped with a thermal con-
ductivity detector.
2. Experimental
2.1. Catalyst preparation
A commercial activated carbon (extruded coconut carbon of
diameter 1.5 mm and length 3–5 mm) was used as the starting
material. The activated carbon was modified by thermal and chem-
ical treatments in order to obtain supports with different surface
chemical properties. Then all catalysts were prepared using an
incipient wetness impregnation method.
2.3. Characterization of catalysts
TPD analysis was carried out using a Micromeritics Chemisorb
2750 gas adsorption instrument equipped with a thermal conduc-
tivity detector. The sample (ca. 200 mg) was pretreated under N2 at
130 °C for 2 h. After cooling in N2, the sample adsorbed a probe gas
for 1 h. Subsequently, the sample was purged for 30 min under N2.
Then the sample was treated under He for 1 h in order to replace
N2 in the instrumentation system. After that, the sample was
heated to 900 °C at a rate of 10 °C minÀ1 under He at a gas flow rate
2.1.1. Preparation of modified activated carbon
Pretreatment of activated carbon: Initially, activated carbon
was crushed and the crushed particles between 20 and 60 mesh
were sieved out for use. The selected materials were washed with
distilled water at 60 °C for 3 h to remove the carbon dust attached
to the activated carbon particles. Then the mixture was filtered,
and then dried at 130 °C for 12 h (sample ACoriginal).
Thermal treatments of activated carbon: Supported ACoriginal
was treated in an argon flow for 3 h at different temperatures
(300, 600, and 900 °C) and then cooled to ambient temperature
in situ. The obtained samples are referred to as AC300, AC600, and
AC900, respectively. Thermal treatments modified the activated car-
bon surface functional groups because no ions are introduced to
the sample using this technique [17].
Acid treatment of activated carbon: Supported ACoriginal was
soaked in HCl (2 M, 10 mL per 1 g activated carbon) or HF (2 M,
10 mL per 1 g activated carbon) at ambient temperature for 48 h,
and then the obtained materials were filtered and washed with
distilled water (2 L per 1 g activated carbon) until the water was
near neutral pH. The obtained materials were dried at 130 °C for
of 25 mL minÀ1
.
The Brunauer–Emmett–Teller (BET) specific surface area data
were obtained using nitrogen adsorption/desorption measure-
ments at 77 K with a BELSORP-Mini instrument. Powder XRD
was performed on a Bruker D8 focus diffractometer with CuK
radiation at 40 kV and 40 mA. XPS was carried out using a PHI
1600 ESCA spectrometer equipped with an AlK X-ray source
a
a
(250 W) with an analyzer pass energy of 188 eV for survey scans
and 30 eV for detailed elemental scans. The Fourier transform
infrared radiation (FTIR) spectra were obtained using a Bio-Rad
FTS 6000 spectrometer with 16 scans and a resolution of 8 cmÀ1
.
The samples were prepared in KBr pellets. Atomic absorption spec-
troscopy (AAS) was performed using a TAS-990 instrument with an
air–acetylene flame.
10 h and are referred to as ACHCl and ACHF
.
Oxidation and reduction of activated carbon: Supported
ACoriginal was soaked in H2O2 (10 wt.%, 10 mL per 1 g activated car-
bon) at ambient temperature for 3 h, and then the obtained mate-
rials were filtered and washed with distilled water (2 L per 1 g
activated carbon). The obtained materials were dried at 130 °C
for 10 h and are referred to as ACH2O2. Supported ACoriginal was trea-
ted in a H2 or C2H2 flow for 3 h at 180 °C and then cooled to ambi-
ent temperature in an argon flow in situ. The obtained samples are
referred to as ACH2 and ACC2H2, respectively.
3. Results and discussion
3.1. Correlation between catalytic activity and C2H2 temperature-
programmed desorption
The activity of the set of AuCl3/C catalysts (Catoriginal, Cat300
,
Cat600, and Cat900) is reported in Fig. 1. The C2H2 GHSV and the feed
volume ratio between HCl and C2H2 were chosen to test catalysts
under mild conditions. It should be mentioned that for all of the
activity tests carried out in the current study, the selectivity to
VCM was virtually 100%, with only trace amounts of
1,2-dichloroethane and chlorinated oligomers [9].
It can be seen that all four AuCl3/C catalysts exhibited unstable
acetylene conversion (ca. 92%) in the first hour, and the activity
eventually increased to about 97.5% in the second hour.
However, the catalysts showed a decrease in C2H2 conversion
(C2H2 conversion by Catoriginal, Cat300, Cat600, and Cat900 decreased
ca. 5.67%, 5.79%, 1.70%, and 4.04%, respectively) after being fol-
lowed for 7 h on stream, and Cat600 exhibited the best stability
(stability: Cat600 > Cat900 > Cat300 ꢀ Catoriginal). The deactivation of
the AuCl3/C catalysts tested at the reaction temperature (180 °C)
was mostly due to reduction of the active gold species Au3+ to
Au0, which is caused by the negative effects of C2H2 [8,10,19,20].
2.1.2. Preparation of catalysts
Dilute aqua regia was used as the solvent for the catalyst prepa-
ration to reduce the effect of the concentrated aqua regia in oxidiz-
ing the surface functional groups on activated carbon. The gold
precursor, HAuCl4Á4H2O, was dissolved in dilute aqua regia (1:9
aqua regia:H2O by volume, 16.4 mL), and then the solution was
added dropwise with stirring to the pretreated support in order
to obtain the catalyst with a final AuCl3 loading of 1 wt.%.
Stirring was continued at ambient temperature for half an hour
and then the product was dried for 24 h at 130 °C [11,18]. The
obtained catalysts are referred to as Catoriginal, Cat300, Cat600
,
Cat900, CatHCl, CatHF, CatH2O2, CatH2, and CatC2H2 according to the
supports used in the preparation process.

S. Chao et al. / Journal of Catalysis 330 (2015) 273–279
275
Fig. 3. C2H2 TPD and He TPD of the Coriginal
.
Fig. 1. Acetylene conversion by AuCl3/C catalysts. Reaction conditions:
temperature = 180 °C, C2H2 GHSV = 360 hÀ1, feed volume ratio between HCl and
C2H2 = 1.1.
between C2H2 and the surface functional groups changed C2H2 into
other compounds, and this would not help the activation of C2H2.
Considering that active carbons have strong physisorption, the
amount of adsorbed C2H2 may be affected by the texture of the
support. The BET surface area data are shown in Table 1. The sur-
face area exhibited a decline with the increase in the pretreatment
temperature, and this was caused by the decomposition of the sur-
face functional groups under high temperatures [16]. But the
amount of C2H2 adsorbed has no relationship to the surface area.
For example, Cat300 and Cat600 are similar in surface area, pore vol-
ume, and pore diameter, but have obvious differences in C2H2 des-
orption peak area. Thus, we can conclude that the change of
texture of activated carbon makes little contribution to the adsorp-
tion of C2H2 mentioned above.
In general, the desorption temperatures in the C2H2 TPD profiles
reflect the binding strength of the adsorbed species on the catalyst
surface, and the peak area correlates with the amount of adsorbed
species. The C2H2 desorption peak is generally attributable to the
ability of the catalyst to activate C2H2 directly [12,13,21,22].
Thus, we can conclude that the peaks below 200 °C represent the
adsorption and activation of C2H2. This can explain the reverse
order between catalyst stability and C2H2 adsorption capacity,
because C2H2 leads to deactivation [9].
In order to gain insight into the interaction between the catalyst
and reactant, C2H2 TPD of the catalysts was conducted (Fig. 2). It
is worth noting that desorption peaks before 200 °C have different
areas in the order Cat600 < Cat900 < Cat300 ꢀ Catoriginal, which is the
opposite of the stability order.
To further identify the C2H2 desorption peak and investigate the
effects between C2H2 and the surface functional groups, C2H2 TPD
and a blank TPD of ACoriginal were carried out (Fig. 3). In the blank
TPD of ACoriginal, surface functional groups on activated carbon
decompose upon heating by releasing CO and CO2 at different tem-
peratures. In general, a CO2 peak results from carboxylic acids at
low temperatures, or lactones at higher temperatures; carboxylic
anhydrides produce both CO and CO2 peaks; phenols, ethers, and
carbonyls produce a CO peak [17]. Comparing these two TPD
curves of ACoriginal in Fig. 3, there are two obvious points that
should be noted: (1) C2H2 desorption take place before the CO2
or CO peaks increase and (2) in the curve of ACoriginal–C2H2, the
amount of CO2 and CO produced by decomposition of surface func-
tional groups shows a significant decrease between 200 and
800 °C, and the decomposition peak shifts toward high tempera-
ture. These results indicate that there are two interaction mecha-
nisms between C2H2 and activated carbon: (1) the C2H2 desorbed
before 200 °C was adsorbed by activated carbon with weak interac-
tion and (2) the decrease of CO2 and CO may be because of the
reaction between C2H2 and surface functional groups, which could
inhibit the decomposition of those functional groups. The reaction
Otherwise, according to the C2H2 TPD of ACoriginal and Catoriginal
,
it seems that the addition of gold suppresses the two kinds of
adsorption of acetylene. At the same time, we could find differ-
ences between the C2H2 TPD of the set of catalysts shown in
Fig. 2. It is obvious that the peak area between 200 and 800 °C
decreased significantly with the increase in pretreatment temper-
ature. This is mainly because the surface functional groups had
decomposed in the pretreatment program, and the higher the pre-
treatment temperatures, the more decomposition of the functional
groups.
For the peak below 200 °C in Fig. 2, the desorption temperature
and the peak area obtained from the set of catalysts are displayed
in Table 2. A counterintuitive observation should be highlighted
that the adsorption capacity of the active sites for activating C2H2
has a negative correlation with the catalyst stability. In other
Table 1
Surface area data for different AuCl3/C catalysts.
Sample
Surface area
(m2 gÀ1
Total pore volume
Average pore
diameter (nm)
)
(cm3 gÀ1
)
Catoriginal 1006
0.79
0.85
0.81
0.73
3.13
2.96
2.95
2.93
Cat300
Cat600
Cat900
1141
1098
997
Fig. 2. C2H2 TPD on different AuCl3/C catalysts.

276
S. Chao et al. / Journal of Catalysis 330 (2015) 273–279
Table 2
Desorption temperature and the peak area of the TCD signal of the set of catalysts.
Sample
Quality (g)
Desorption temperature (°C)
Peak area
(TCD signal)
Catoriginal
Cat300
Cat600
0.2003
0.2000
0.2003
0.2000
105.3
104.6
100.9
100.0
33.32
32.14
22.65
28.10
Cat900
words, we can assume that the activated carbon has outstanding
ability to activate C2H2 and the addition of gold suppresses the
adsorption and activation of acetylene.
In summary, activated carbon has the ability to activate C2H2
and work as a constituent part of the active site.
3.1.1. Existence of gold in the AuCl3/C catalysts
Fig. 5. XRD patterns of different catalysts.
Because the existence of Au3+ will determine the activity of the
supported gold catalyst [11,23], the Au3+ may affect the adsorption
of reactants for acetylene hydrochlorination. To test the existence
of gold in the catalysts, XPS was systematically carried out for
the catalysts (Fig. 4) (wide-scan spectra are shown in Fig. S1 in
the Supplementary Material). It is evident that Cat600 presented
the highest signal intensity of Au, Cat900 exhibited a lower signal
intensity, and Catoriginal and Cat300 had the lowest intensity, which
may represent a noisy signal. For these four catalysts, the loading
of AuCl3 is 1 wt.%, which means that the surface atomic percentage
of Au may be less than 1% (the detection limit of XPS). This may be
the main reason for the weak signals. However, Cat600 presented
the strongest signal, which means there are more gold atoms
anchored on the surface of catalyst. The amount of Au0 is greater
than that of Au3+. This may be because the Au3+ was undergoing
reduction during XPS analysis.
indicating that the particles were smaller than 4 nm or the mate-
rial had a large amount of Au0 and Au3+ distributed on the surface
[11].
Considering all of the results of XPS, AAS, and XRD, it can be
assumed the set of catalysts all have the same loading of Au.
However, for Catoriginal and Cat300, sintering of gold was observed
because the precursor, HAuCl4, was reduced by the functional
groups on the activated carbon and grew into nanoparticles in
the absence of concentrated aqua regia as the solvent [11]. For
Cat600 and Cat900, the functional groups were modified by a ther-
mal method; consequently, less precursor was stirred into the Au
nanoparticles and more Au0 and Au3+ were anchored on the surface
of the active carbon, which may contribute to the activity and sta-
bility of the catalyst.
According to the report of Conte et al., an excess of Au3+ does
not contribute to catalyst activity [11], which could explain the
set of catalysts evaluated above having the same initial activity.
However, combined with the results of C2H2 TPD and the deactiva-
tion rate of the catalysts, it is obvious that the greater the content
of Au3+ or Au0 anchored on the surface, the less C2H2 is adsorbed
and activated by the catalysts. This phenomenon decreased the
reduction of Au3+ by C2H2 during the reaction, and thus improved
the stability of the catalysts.
Because XPS is a surface technique, we determined the gold
loading using AAS, which showed that all of the catalysts have a
total AuCl3 loading of approximately 1 wt.%. In addition, XRD,
which is also a bulk technique, was carried out in order to detect
the gold content in the catalysts (Fig. 5).
According to the results of AAS (AuCl3 content of Catoriginal
,
Cat300, Cat600, and Cat900: 0.97, 0.98, 0.96, and 0.98 wt.%), the con-
tent of gold in the AuCl3/C catalysts is basically the same as the
AuCl3 loading of 1 wt.%. The catalysts were prepared using the
incipient wetness impregnation method, which is known to be
the most effective preparation method and guarantees that all of
the gold precursor can be loaded onto the supports [3–
5,10,11,18,24]. However, the XRD patterns show that Catoriginal
and Cat300 have a stronger gold signal, which indicates larger gold
particles. As for Cat600, no discernible gold reflection was detected,
3.1.2. Effect of surface functional groups in the AuCl3/C catalysts
According to these results, the gold existing on the surface of
the catalysts suppresses the adsorption of C2H2, which means that
gold and C2H2 may located in the same sites on the surface of cat-
alysts. In order to investigate the effect of surface chemistry on
C2H2 TPD, the catalysts were also characterized using FTIR and XPS.
FTIR was employed to explore the change in functional groups
of different catalysts (Fig. 6). The relative intensities of all of the
broad peaks decreased with increased treatment temperatures.
At 900 °C there is a significant decrease in all of the relevant peaks,
which indicates the decomposition of all the functional groups.
This phenomenon could also be demonstrated by the C2H2 TPD
results (Fig. 2), where the surface area between 200 and 850 °C
decreased with increased treatment temperature.
Because the Cat900 reveals inconspicuous peaks in its FTIR spec-
trum, XPS analysis was obtained only for the other three catalysts.
Atomic ratios of the three catalysts are shown in Table 3. The
atomic ratios of O1s exhibited a decline with increase in the pre-
treatment temperature, and this indicated the decomposition of
the surface functional groups at high temperatures. Cat600 had
the highest atomic ratios of Au4f, which is consistent with the dis-
cussion in Section 3.1.1.
XPS spectra of the C1s region of the catalysts are shown in Fig. 7.
Carbon atoms differ in their binding energy depending on whether
Fig. 4. XPS spectra of different AuCl3/C catalysts.

S. Chao et al. / Journal of Catalysis 330 (2015) 273–279
277
Fig. 6. FTIR spectra of different catalysts.
Table 3
Atomic ratios of C1s, O1s, and Au4f of different catalysts.
Sample
C1s (%)
O1s (%)
Au4f (%)
Catoriginal
Cat300
Cat600
82.57
84.16
90.30
17.41
15.83
9.66
0.02
0.01
0.04
they are linked to one O atom by a single bond or a double bond or
to two or three oxygen atoms [25]. Thus, deconvolution of the C1s
spectra gives five individual component peaks [26,27]: peak I
(284.6 eV, which represents graphitic carbon), peak II (286.1–
286.3 eV, representing carbon present in phenolic or alcohol
groups), peak III (287.3–287.6 eV, representing carbonyl or qui-
none groups), peak IV (288.4–288.9 eV, representing carboxyl or
ester groups), and peak V (290.4–290.8, carbon present in carbon-
ate groups and/or adsorbed CO or CO2).
Comparing the spectrum of Cat600 with the spectra of Catoriginal
and Cat300, there was a significant decrease in the relative content
of phenolic or alcohol groups (peak II) and carbonyl or quinone
groups (peak III), and there was an increase in peak I. This may
be explained by assuming that as the treatment temperature
increased, the surface functional groups decomposed into CO or
CO2. But the amount of carboxyl (peak V) decreased only slightly
from 4.19% to 3.75%. This result may be because of using dilute
aqua regia in the process of preparation of catalysts, by which part
of other function groups (peak II or peak III) oxidized into carboxyl
(peak V).
Fig. 8 shows the O1s spectra fitted to three component peaks
[26,28]: peak I (531.1–531.7 eV, representing C@O groups: ketone,
lactone, and/or carbonyl); peak II (533.1–533.4 eV, representing
CAOH and/or CAOAC groups); peak III (535.5–535.9 eV, represent-
ing chemisorbed oxygen and perhaps some adsorbed water).
Regardless of peak III, the content of peak I and peak II in each cat-
alyst is listed: Catoriginal (peak I, 30.46%, peak II, 69.54%), Cat300
(peak I, 32.82%, peak II, 67.18%), Cat600 (peak I, 46.72%, peak II,
53.28%). Although XPS is a semiquantitative technique, by combin-
ing spectra of C1s, we can find out that Cat600 possesses the highest
relative content of ketone, lactone, and/or carbonyl (peak I in spec-
tra of O1s) and the lowest relative content of phenolic and/or alco-
Fig. 7. XPS C1s spectra of the different catalysts.
useful for the reaction, while phenolic and/or alcohol groups easily
reduce Au3+ into Au0 in the form of nanoparticles, which is useless
for the reaction [28]. It also demonstrated that the efficiency of the
catalyst is likely to be related to the location of Au3+ at the gold/-
carbon interface of the catalyst [11].
It has been reported that the interaction between a hydrogen
atom of C2H2 and an oxygen atom of microporous materials is
based not only on electrostatic attraction but also on the electron
delocalization effect [29]. Combined with the results of C2H2 TPD
and the stability of catalysts, it may be assumed that groups such
as ketone, lactone, or carbonyl play the main role in the adsorption
of C2H2. It can be further demonstrated that the surface functional
groups in the AuCl3/C catalyst are a constituent part of the active
site, having a role in adsorbing and activating one of the reactants.
However, because the functional groups have effects on gold
hol groups (peak II in spectra of C1s). At the same time, for Cat600
,
less precursor, HAuCl4, was stirred into the Au nanoparticles and
more gold atoms were anchored on the surface of the active carbon
(part of Au0 may be due to the Au3+ undergoing reduction during
XPS analysis). These results are consistent with the theory first
reported by Davies et al. that ketone, lactone, and/or carbonyl tend
to anchor gold in a surface with a good dispersion state that is

278
S. Chao et al. / Journal of Catalysis 330 (2015) 273–279
[9]. Considering this, pretreatment with ClÀ may have a positive
effect on the catalyst performance.
Catalysts prepared using activated carbon pretreated with HCl
and HF were evaluated (Fig. 9), and the C2H2 TPD of the catalysts
was characterized. Compared with Catuntreated and CatHF, the initial
activity of CatHCl increased dramatically and the stability was basi-
cally the same. From the results of C2H2 TPD (Fig. 10), CatHCl pos-
sessed the maximum peak area among those spectra.
Treatment of activated carbon with HCl and HF resulted in the
removal of inorganic constituents that may poison the catalyst. It
is obvious from the texture of CatHCl and CatHF (Table 4) that there
is a significant reduction in ash content and an enhancement of
BET surface area and total pore volume of the activated carbon. It
is reported that HCl generally increases the surface oxygen content,
whereas HF decreases the surface oxygen content [30]. This may
explain the maximum C2H2 adsorption peak area of CatHCl
.
However, the observation that CatHCl possesses the highest initial
activity may be ascribed to the effect of the ClÀ, and this was con-
sistent with the assumption.
3.2.2. Controllable modulation of the induction period of the catalyst
In acetylene hydrochlorination, there will be an induction per-
iod before the catalyst reaches the highest activity [7,18,23,31].
The induction period may be related to adsorption and diffusion
of the reactants, the formation of the active sites, etc. For those
AuCl3/C catalysts, we conjectured that the induction period
Fig. 9. Acetylene conversion by AuCl3/C catalysts. Reaction conditions:
temperature = 180 °C, C2H2 GHSV = 360 hÀ1, feed volume ratio between HCl and
C2H2 = 1.1.
Fig. 8. XPS O1s spectra of the different catalysts.
desorption [28], and because of the limitations of the characteriza-
tion techniques, a detailed mechanism for activating C2H2 is still
unclear.
3.2. Other evidence for the assumption of the active site
3.2.1. Effect of ClÀ
Activated carbon has excellent ability to activate C2H2, and it
could be further predicted that improving the ability of HCl to
adsorb and activate will improve the promote performance of the
AuCl3/C catalyst. It has been reported that preactivating the cata-
lyst with HCl before the reaction and improving the feed propor-
tion of HCl can improve the performance of the AuCl3/C catalyst
Fig. 10. C2H2 TPD on different AuCl3/C catalysts.

S. Chao et al. / Journal of Catalysis 330 (2015) 273–279
279
Table 4
to anchor gold on the surface with a good dispersion state. The
phenolic and/or alcohol groups can easily reduce Au3+ to Au0 in
the form of nanoparticles. The catalyst with the best dispersibility
of gold exhibits the best stability, but the adsorption capacity of
the active sites for activating C2H2 has a negative correlation with
the catalyst stability and gold dispersibility. This phenomenon
indicates that the gold dispersed on the surface of activated carbon
is not the active site for activating C2H2.
Textural properties of different AuCl3/C catalysts.
Sample
Surface area
(m2 gÀ1
Total pore volume
Average pore
diameter (nm)
Ash
(%)
)
(cm3 gÀ1
)
Catoriginal 1006
CatHCl
CatHF
0.79
0.82
0.84
3.13
3.03
3.08
5.24
3.71
2.56
1030
1037
It seems that the surface functional groups may be a constituent
part of the active sites adsorbing and activating C2H2. The support
of Au on the activated carbon has a negative effect on the adsorp-
tion of C2H2. This may indicate that Au3+ and the surface functional
groups work in synergy as the active sites. Thus, we can change our
understanding of the reaction mechanism of acetylene hydrochlo-
rination and change the design of the catalyst.
Acknowledgments
This work was financially supported by the NSFC – China
(21376123, U1403293), MOE – China (IRT-13R30 and 113016A),
and the Research Fund for 111 Project (B12015).
Appendix A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.jcat.2015.07.008.
Fig. 11. Induction period of AuCl3/C catalysts for acetylene hydrochlorination.
Reaction conditions: temperature = 180 °C, C2H2 GHSV = 480 hÀ1, feed volume ratio
between HCl and C2H2 = 1.
References
[1] X. Li, M. Zhu, B. Dai, Appl. Catal. B 142–143 (2013) 234.
[2] K. Zhou, J.C. Jia, X.G. Li, X.D. Pang, C.H. Li, J. Zhou, G.H. Luo, F. Wei, Fuel Process.
Technol. 108 (2013) 12.
[3] B. Nkosi, N.J. Coville, G.J. Hutchings, Chem. Commun. (1988) 71.
[4] B. Nkosi, M.D. Adams, N.J. Coville, G.J. Hutchings, J. Catal. 128 (1991) 378.
[5] B. Nkosi, N.J. Coville, G.J. Hutchings, Appl. Catal. 43 (1988) 33.
[6] G. Hutchings, J. Catal. 96 (1985) 292.
[7] M. Conte, C.J. Davies, D.J. Morgan, T.E. Davies, A.F. Carley, P. Johnston, G.J.
Hutchings, Catal. Sci. Technol. 3 (2013) 128.
[8] M. Conte, A.F. Carley, G.J. Hutchings, Catal. Lett. 124 (2008) 165.
[9] M. Conte, A.F. Carley, C. Heirene, D.J. Willock, P. Johnston, A.A. Herzing, C.J.
Kiely, G.J. Hutchings, J. Catal. 250 (2007) 231.
[10] B. Nkosi, N.J. Coville, G.J. Hutchings, M.D. Adams, J. Friedl, F.E. Wagner, J. Catal.
128 (1991) 366.
[11] M. Conte, C.J. Davies, D.J. Morgan, T.E. Davies, D.J. Elias, A.F. Carley, P. Johnston,
G.J. Hutchings, J. Catal. 297 (2013) 128.
[12] X. Li, Y. Wang, L. Kang, M. Zhu, B. Dai, J. Catal. 311 (2014) 288.
[13] X. Li, X. Pan, L. Yu, P. Ren, X. Wu, L. Sun, F. Jiao, X. Bao, Nat. Commun. 5 (2014)
3688.
[14] X. Li, X. Pan, X. Bao, J. Energy Chem. 23 (2014) 131.
[15] K. Zhou, B. Li, Q. Zhang, J.Q. Huang, G.L. Tian, J.C. Jia, M.Q. Zhao, G.H. Luo, D.S.
Su, F. Wei, ChemSusChem 7 (2014) 723.
[16] E.G. Rodrigues, M.F.R. Pereira, X.W. Chen, J.J. Delgado, J.J.M. Orfao, J. Catal. 281
(2011) 119.
[17] J.L. Figueiredo, M.F.R. Pereira, M.M.A. Freitas, J.J.M. Orfao, Carbon 37 (1999)
1379.
[18] W. Wittanadecha, N. Laosiripojana, A. Ketcong, N. Ningnuek, P. Praserthdam,
J.R. Monnier, S. Assabumrungrat, Appl. Catal. A 475 (2014) 292.
[19] J.M. Ma, S.J. Wang, B.X. Shen, React. Kinet. Mech. Catal. 110 (2013) 177.
[20] S.J. Wang, B.X. Shen, Q.L. Song, Catal. Lett. 134 (2010) 102.
[21] J. Zhao, J. Xu, J. Xu, J. Ni, T. Zhang, X. Xu, X. Li, ChemPlusChem 80 (2015) 196.
[22] C. Zhang, L. Kang, M. Zhu, B. Dai, RSC Adv. 5 (2015) 7461.
[23] M. Conte, C.J. Davies, D.J. Morgan, A.F. Carley, P. Johnston, G.J. Hutchings, Catal.
Lett. 144 (2014) 1.
[24] M. Conte, A.F. Carley, G. Attard, A.A. Herzing, C.J. Kiely, G.J. Hutchings, J. Catal.
257 (2008) 190.
[25] H.P. Boehm, Carbon 40 (2002) 145.
[26] Z.R. Yue, W. Jiang, L. Wang, S.D. Gardner, C.U. Pittman Jr., Carbon 37 (1999)
1785.
[27] S.D. Gardner, C.S.K. Singamsetty, G.L. Booth, G.-R. He, C.U. Pittman Jr., Carbon
33 (1995) 587.
represents a process of formation of active sites. We assumed that
the formation of active sites is due to the modification of the func-
tional groups by reducibility of C2H2. Thus, we pretreated activated
carbon with H2, C2H2, and H2O2 in order to modify surface func-
tional groups and modulate the induction period of the catalyst.
To reduce the influence of the catalyst filling amount on the
induction period, the set of catalysts (Catoriginal, CatH2O2, CatH2,
and CatC2H2) were tested in a smaller fixed-bed microreactor
(the reactors are shown in Fig. S2 in the Supplementary
Material). Silica sand was used to extend the bed length above
the catalysts (0.4 g, 1.25 mL), which could mix and preheat the
reactant at the same time. Prior to the reaction, the catalysts were
pretreated in situ with HCl (5 mL minÀ1) at 180 °C for 1 h. After
that, HCl (10 mL minÀ1) and C2H2 (10 mL minÀ1) were fed into
the heated reactor (180 °C) via calibrated mass flow controllers,
with a C2H2 GHSV of 480 hÀ1. A blank experiment was carried
out using an empty reactor filled with silica sand under the same
conditions, and the ceramic rings did not show any catalytic activ-
ity. The gas phase products were analyzed on line using a GC
equipped with a thermal conductivity detector.
The results are shown in Fig. 11. It was obvious that the reduc-
tion treatment shortened the induction period and the oxidation
treatment increased the induction period. This result suggested
that the reduction treatment could form the functional groups,
which were part of the active sites before the reaction.
Because of the complexity and particularity of activated carbon
materials, and the limitations of the characterization techniques,
the study on the effect of ClÀ and the induction period of the cat-
alysts needs more effort. But it could provide circumstantial evi-
dence for the assumption that the surface functional groups were
a constituent part of the active site adsorbing and activating C2H2.
[28] R. Burgess, C. Buono, P.R. Davies, R.J. Davies, T. Legge, A. Lai, R. Lewis, D.J.
Morgan, N. Robinson, D.J. Willock, J. Catal. 323 (2015) 10.
[29] R. Matsuda, R. Kitaura, S. Kitagawa, Y. Kubota, R.V. Belosludov, T.C. Kobayashi,
H. Sakamoto, T. Chiba, M. Takata, Y. Kawazoe, Y. Mita, Nature 436 (2005) 238.
[30] S.B. Wang, G.Q. Lu, Carbon 36 (1998) 283.
4. Conclusions
Thermal treatments would change the surface functional
groups on activated carbon. Ketone, lactone, and/or carbonyl tend
[31] C. Huang, M. Zhu, L. Kang, X. Li, B. Dai, Chem. Eng. J. 242 (2014) 69.