Acetylene hydrochlorination over supported ionic liquid phase (SILP) gold-based catalyst: Stabilization of cationic Au species via chemical activation of hydrogen chloride and corresponding mechanisms

Chinese Journal of Catalysis 42 (2021) 334–346

a v a i l a b l e a t w w w . s c i e n c e d i r e c t . c o m

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / c h n j c

Article

Acetylene hydrochlorination over supported ionic liquid phase

(SILP) gold‐based catalyst: Stabilization of cationic Au species via

chemical activation of hydrogen chloride and corresponding

mechanisms

Jia Zhao^a,*, Saisai Wang ^a, Bolin Wang ^a,b, Yuxue Yue ^a, Chunxiao Jin ^a, Jinyue Lu ^a, Zheng Fang ^a,

Xiangxue Pang ^a, Feng Feng ^a, Lingling Guo ^a, Zhiyan Pan ^b, Xiaonian Li ^a,#

^aIndustrial Catalysis Institute of Zhejiang University of Technology, Hangzhou 310014, Zhejiang, China

^bDepartment of Environmental Engineering, Zhejiang University of Technology, Hangzhou 310014, Zhejiang, China

A R T I C L E I N F O

A B S T R A C T

Article history:

The activation of HCl by cationic Au in the presence of C₂H₂is important for the construction of

active Au sites and in acetylene hydrochlorination. Here, we report a strategy for activating HCl by

Received 21 March 2020

Accepted 10 May 2020

Published 5 February 2021

–

the Au‐based supported ionic liquid phase (Au–SILP) technology with the [N(CN)₂] anion. This

–

strategy enables HCl to accept electrons from [N(CN)₂] anions in Au–[N(CN)₂] complexes rather

than from pure [Bmim][N(CN)₂], leading to notable improvement in both the reaction path and the

stability of the catalyst without changing the reaction triggered by acetylene adsorption. Further‐

more, the induction period of the Au–SILP catalyst was shown to be absent in the reaction process

due to the high Au(III) content in the Au(III)/Au(I) site and the high substrate diffusion rate in the

ionic liquid layer. This work provides a facile method to improve the stability of Au‐based catalysts

for acetylene hydrochlorination.

Keywords:

Acetylene hydrochlorination

Electron density

Hydrogen chloride activation

Stabilization mechanism

Gold‐based supported ionic liquid

phase catalyst

1. Introduction

[16–20]. However, the HgCl₂catalyst may cause environmental

pollution because of its high volatility and toxicity [17–24].

Polyvinyl chloride (PVC) is widely used in various fields due

to its outstanding properties, such as excellent mechanical and

electrical properties and easy processing [1–8]. Currently, PVC

is mainly produced using vinyl chloride monomer (VCM)

through ethylene oxychlorination or acetylene hydrochlorina‐

tion [9–15]. Compared with ethylene oxychlorination, acety‐

lene hydrochlorination has the advantages of a single reaction

step and low operating costs, but the reaction has to be cata‐

lyzed by mercury chloride (HgCl₂) with activated carbon (AC)

Thus, intensive efforts have been devoted to developing

Hg‐free catalysts for acetylene hydrochlorination [25–35].

Since the seminal work of Hutchings in 1972 [1], Au cata‐

lysts have attracted intensive research interest, and the AuCl₃

catalyst has been shown to have high activity for acetylene

hydrochlorination, with the Au(III)/Au(I) pair acting as the

active redox site [36,37]. Even today, the Au‐catalyzed hydro‐

chlorination reaction is still considered as the most promising

non‐Hg hydrochlorination technology [38–42]. However, Au

* Corresponding author. Tel: +86‐571‐88871656; E‐mail: jiazhao@zjut.edu.cn

^#Corresponding author. Tel: +86‐571‐88320002; E‐mail: xnli@zjut.edu.cn

Financial support by the National Natural Science Foundation of China (21606199) and Science and Technology Department of Zhejiang Province

(LGG20B060004) is gratefully acknowledged.

DOI: 10.1016/S1872‐2067(20)63617‐8 | http://www.sciencedirect.com/science/journal/18722067 | Chin. J. Catal., Vol. 42, No. 2, February 2021

Jia Zhao et al. / Chinese Journal of Catalysis 42 (2021) 334–346

335

catalysts suffer from several intrinsic defects: (i) reduction of

cationic Au species (Au(III) and Au(I)) to metallic Au(0) species

in the presence of an electron‐rich substrate (such as C₂H₂); (ii)

dependence of activity on the dispersion of Au (particularly,

atomic dispersion of Au in carbon), which will require exact

control of particle size and cause poor reproducibility of cata‐

lyst performance; and (iii) aggregation of Au(0) species or Au

nanoparticles (NPs) because of the high surface energy, leading

to the deactivation of the catalyst. Previous studies have shown

that the oxidation of Au can be protected and regenerated

through on‐line or off‐line treatments in the presence of inor‐

N‐doped carbon materials [73], leading to the formation of a

new N–H bond, as well as either the adsorption of Cl ion on the

cavity of N‐doped carbon or its motion on the surface. Inspired

–

by this, we developed [N(CN)₂] as functional inorganic moie‐

ties for the synthesis of a coordinated [Bmim][N(CN)₂] IL to

stabilize high‐valance Au species and investigate the catalytic

performance of the Au‐N(CN)₂/AC catalyst (Scheme 1) in acet‐

ylene hydrochlorination. The experimental and theoretical

calculation results indicate that the dissociation of HCl on

–

[N(CN)₂] to form free Cl^–appears to be the main contributor to

the stabilization of cationic Au species. These results provide

some clues as to the design of stable high‐valent Au‐based cat‐

alysts for acetylene hydrochlorination.

ganic moieties, i.e., CN₂, S₂O₃^2–, SCN^–, I^–, or Br^–[43–45]. There‐

–

in, Hutchings et al. revealed excellent catalytic performance and

obtained significant insights on the active site through the pro‐

tective effect of S₂O₃^2–; furthermore, they have successfully

carried out a pilot‐scale evaluation of the functional protected

Au/C‐based catalyst in China. Moreover, halohydrocarbons

(C–X, X=Cl, I, Br) were extensively used in the redispersion of

sintered Au/AC catalysts via the regeneration of oxidative Au

species [44,45], which revealed a strategy and further indicated

the particular importance of inorganic moieties in protecting

high‐valance Au species. Currently, the stability challenge has

been mainly addressed by the (i) adaptation of chelating soft

donor atoms and (ii) stably anchored catalytic sites, suggesting

the necessity for research on alternative stabilization methods

to reduce the progressive deactivation.

Supported ionic liquid phase (SILP) catalysts, a special var‐

iant of supported liquid phase catalysts, feature the advantages

of heterogeneous and molecular catalysis due to the immobili‐

zation of the catalytically active components in the ionic liquid

(IL) layer [46–53]. Previous reports have shown the successful

application of IL‐immobilized catalysts in reactions such as

hydroformylation of olefins [54,55], hydrogenation [56,57],

gas‐shift reaction [58,59], and hydrosilylation [60,61]. In our

previous work, the Au(III) complex supported on AC promoted

with the 1‐propyl‐3‐methylimidazolium chloride IL

([Prmim][Cl]) was tested in the gas‐phase acetylene hydro‐

chlorination [62]. The prepared Au‐IL/AC catalyst showed high

activity and selectivity (turnover frequency (TOF) = 79.2 h^‒1,

and selectivity for VCM formation > 99.8%); however, deacti‐

vation after prolonged use still occurred due to the reduction of

cationic Au species to metallic Au(0). While CuCl₂modification

can self‐regenerate the deactivated Au‐IL/AC catalyst via a

redox mechanism [18], the addition of a Cu promoter may

cause the problem of noble metal recycling. To solve these

problems, we need to further explore the chemistry of

high‐valent late transition metals and fabricate more stable,

efficient, low‐cost, and environment‐friendly catalysts for acet‐

ylene hydrochlorination.

2. Experimental procedures and theoretical investigation

2.1. Catalyst preparation

The Au‐based SILP catalysts were prepared using the wet

impregnation method. Briefly, HAuCl₄·xH₂O (Sigma‐Aldrich)

was dissolved in an HCl acid solution, followed by the addition

of certain amounts of ILs. After stirring for 1 h, the mixed solu‐

tion was dropwise added to AC under stirring. The AC (Norit

ROX 0.8) was pre‐treated with aqua regia to remove impurities

(such as residual metal ions). Subsequently, the pre‐prepared

wet catalysts were maintained at 60 °C for 12 h, followed by

drying at 110 °C for another 12 h. Finally, the dried products

were collected as catalysts and termed as Au‐X/AC, where X

represents the ILs used. For example, Au‐Cl/AC and

Au‐N(CN)₂/AC

catalysts

were

prepared

with

1‐butyl‐3‐methylimidazolium chloride ([Bmim][Cl]) and

1‐butyl‐3‐methylimidazolium dicyanamide ([Bmim][N(CN)₂])

(Lanzhou Greenchem. ILS, LICP. CAS.), respectively. The Au

content in all the catalysts was fixed at 0.1 wt%, and the con‐

tent of ILs was maintained at 10 wt% unless otherwise stated.

2.2. Catalyst characterization

A Micromeritics ASAP 2020 instrument was used to inves‐

tigate the physical structure of the catalysts in terms of pore

structure and specific surface. Energy dispersive X‐ray (EDX)

Compared with acetylene activation, the activation of hy‐

drogen chloride (HCl) has not received enough attention in the

construction of stable Au‐based catalysts for acetylene hydro‐

chlorination. Despite the lowest unoccupied molecular orbital

(LUMO) for σ*_H–Cl[63], HCl is generally activated by oxidation

rather than reduction over catalytic metals, particularly for late

transition metal cations. Recent research has indicated that HCl

adsorption may proceed via a dissociative mechanism on

N

H

C

Au

Cl

Au-IL complexes

Scheme 1. Schematic representation of the Au‐N(CN)₂/AC catalyst. The

image is not drawn to scale.

336

Jia Zhao et al. / Chinese Journal of Catalysis 42 (2021) 334–346

spectroscopy was used to analyze the distribution and atomic

structure of the active components on a Tecnai G2 F30 S‐Twin.

The aberration‐corrected scanning transmission electron mi‐

croscopy (AC‐STEM) was used for direct analysis of the distri‐

bution of isolated single Au atoms. The X‐ray diffraction (XRD)

spectra were recorded on a PANalytical‐X’Pert PRO instrument.

The valance state of the Au sites was investigated by X‐ray

photoelectron spectroscopy (XPS) using a Kratos AXIS Ultra

DLD spectrometer. Synchrotron‐radiated X‐ray absorption fine

structure (XAFS) measurement is known as a powerful tool for

investigating the electronic structure and coordination envi‐

ronment of catalysts. In this study, the relevant XAFS meas‐

urements were carried out at the Shanghai Synchrotron Radia‐

tion Facility (SSRF). Herein, the X‐ray absorption near‐edge

structure (XANES) was recorded to analyze the orbital occu‐

pancy and charge state of the Au atoms in the investigated cat‐

alysts. The extended X‐ray absorption fine structure (EXAFS) in

the prepared catalysts was evaluated via Fourier‐transform

(FT) analysis to determine the coordination structure in terms

of coordination atoms, number, and distance between the Au

atoms and neighboring atoms.

as implemented in the Amsterdam density functional

band‐structure (ADF‐BAND) package, coupled with the

Perdew‐Burke‐Ernzerhof (PBE) functional and scalar relativ‐

istic corrections. A DZP basis set was used for the determina‐

tion of valence orbitals, and a TZP basis set was used for the

determination of Pd. The self‐consistent field was converged to

a value of 10^–5in Hartree energy units. Gradients were con‐

verged to 0.001 Hartree/Å. All other measurements were per‐

formed under the default setting.

3. Results and discussion

3.1. Catalyst characterization

A homogeneous solution containing the IL and the Au pre‐

cursor was obtained by dissolving the Au precursor,

HAuCl₄·3H₂O, separately in the [Bmim][N(CN)₂] (Figure S2a)

and [Bmim][Cl] (Figure S2b) ILs, followed by the preparation of

the Au‐SILP catalysts of Au‐N(CN)₂/AC and Au‐Cl/AC using the

conventional impregnation method [1] and vacuum removal of

the added deionized water. The physical structures of AC be‐

fore and after catalyst preparation were investigated using the

Brunauer‐Emmett‐Teller (BET) method, and the textural pa‐

rameters are shown in Table 1. The surface area and the pore

volume were 722 m²g^‒1and 0.39 cm³g^‒1(Figures 1a and S3)

for Au‐N(CN)₂/AC, and 737 m²g^‒1and 0.40 cm³g^‒1for

Au‐Cl/AC, respectively. The decreased surface area and pore

volume of AC (Table 1) were ascribed to the [Bmim][N(CN)₂]

and [Bmim][Cl] IL layers inside the pore. Similar results have

been reported in the literature [18,46,62]. Previously, we

demonstrated that the immobilization of ILs on the AC support

is attributed to the high free p‐electron density of carbon sup‐

ports, leading to the formation of more CH‐π bonds with H at‐

oms at the C2 position of the acidic cation [64]. Thermogravi‐

metric (TG) analysis showed that the supported

[Bmim][N(CN)₂] and [Bmim][Cl] ILs were remarkably stable

and remained undecomposed at temperatures below 240 °C

(Figure S4), suggesting the stability of the ILs under the reac‐

tion conditions of acetylene hydrochlorination.

2.3. Catalyst tests

The activity and stability of the as‐prepared Au‐based cata‐

lysts were evaluated in a typical fixed‐bed reactor (Figure S1).

First, a certain amount of the catalyst was loaded in the reactor,

followed by sweeping with N₂gas at 120 °C for at least 0.5 h to

remove impurities, such as water and air. Thereafter, a C₂H₂

and HCl mixed gas (V(C₂H₂):V(HCl) = 1:1.2) was added, which

was controlled using a mass flowmeter at a gas hourly space

velocity (GHSV, C₂H₂) of 1000 h^‒1. The mixed gas obtained from

the reactor outlet was washed by a NaOH solution and then

analyzed by gas chromatography using a Porapak N packed

column (6 ft × 1/800 stainless steel). The peak areas were

normalized to investigate the gas‐phase products. Given the

complete absorption of HCl, the volume of the reaction system

can be considered constant for the calculations. The amount of

C₂H₄Cl₂and chlorinated oligomeric by‐products was deter‐

mined by introducing the reactor outflow into an

N‐methylpyrrolidone solution and analyzing the obtained

N‐methylpyrrolidone solution using a Waters GCT Premier

chromatograph equipped with an HP‐5 capillary column. The

conversion was defined as the amount of C₂H₂reacted divided

by the amount of C₂H₂introduced. The selectivity toward VCM

was calculated as the amount of VCM formed divided by the

amount of C₂H₂reacted, and the carbon balance based on these

products was nearly 100%. Adopting the calculation method in

the literature [36], we define the productivity calculated in this

work as follows:

The morphology and the dispersion of Au‐SILP catalysts

were investigated, and Figure 1b shows the STEM image and

elemental mapping of the Au‐N(CN)₂/AC catalyst. The STEM

data revealed that most of the Au species were homogeneously

and highly dispersed on the carrier surface, as evidenced by the

absence of crystalline NPs. Elemental mapping analysis indi‐

cated a homogeneous distribution of N, Au, and Cl atoms (Fig‐

ure S5). STEM morphology and the elemental mapping of

Au‐Cl/AC confirmed the immobilization of homogeneously

Table 1

Textural properties of the AC support and Au‐SILP catalysts.

Productivity = Amount of VCM produced/Amount of Au

(kg_VCMkg_Au^‒1min^‒1)

Sample

S_BET(m²g^‒1) ^aVolume (cm³g^‒1) ^bDiameter (nm) ^c

Au‐N(CN)₂/AC

Au‐Cl/AC

AC

722

737

1135

0.39

0.40

0.64

2.01

2.04

2.20

2.4. Computational details

a

b

Measured using the Brunauer‐Emmett‐Teller (BET) method; calcu‐

c

All energies and structures were calculated using the densi‐

ty functional theory (DFT) under periodic boundary conditions

lated based on the adsorbed N₂volume; determined by the Bar‐

rett‐Joyner‐Halenda (BJH) method.

Jia Zhao et al. / Chinese Journal of Catalysis 42 (2021) 334–346

337

a

150

135

120

105

90

b

Adsorption Isotherm

Desorption Isotherm

75

0.0 0.2 0.4 0.6 0.8 1.0

Relative Pressure( P/P₀

)

c

d

e

AC

Au(III)

Au-N(CN) /AC

2

Au(III)

4f_7/2

4f_5/2

Au(I)

5 nm

10 20 30 40 50 60 70 80

94 92 90 88 86 84 82

2 Theta (degree)

Binding Energy (eV)

Fig. 1. Characterizations of the Au‐N(CN)₂/AC catalyst. (a) N₂gas adsorption/desorption isotherms; (b) STEM image and elemental mapping analysis;

(c) XRD patterns; (d) representative HAADF‐STEM image and (e) Au 4f XPS spectra.

dispersed Au species and the IL phase on the AC support. In‐

ductively coupled plasma mass spectrometry (ICP‐MS) analysis

verified the consistency of the Au metal loading with the nomi‐

nal amount of metal impregnated onto the support (Table S1).

No reflections associated with the metallic Au species were

observed in the XRD spectra of Au‐N(CN)₂/AC (Figure 1c) and

Au‐Cl/AC (Figure S6), implying the existence of Au on the AC

surface either in an amorphous state or with a particle size

below the XRD detection limit [65–67]. Based on this specula‐

tion, we characterized the HAADF‐STEM micrographs of the

Au‐N(CN)₂/AC catalyst (Figure 1d) and detected predominant‐

ly isolated Au sites on the AC surface. For Au‐Cl/AC, a consid‐

erably large fraction of isolated Au sites coupled with a rela‐

tively minor component of Au clusters were observed on the

AC support (Figure S7a). The obtained results suggested that

single‐atom Au catalysts can be easily synthesized via confining

metal into the IL layer. Additionally, the Au valence states in

Au‐N(CN)₂/AC and Au‐Cl/AC were detected by XPS analysis.

Au‐N(CN)₂/AC showed two Au states characterized by Au 4f_7/2

binding energies (BEs) of 86.7–86.9 and 85.1–85.5 eV [68,69],

which can be attributed to the high‐valent Au(III) and Au(I)

species, respectively (Figure 1e). Furthermore, the peaks at

84.0–84.2 eV [70,71] were assigned as the metallic Au(0) in

Au‐Cl/AC (Figure S7b), which was in accord with our previous

work [18,62]. The XPS data were consistent with the

HAADF‐STEM micrographs of Au‐N(CN)₂/AC (Figure 1d) and

Au‐Cl/AC (Figure S7a). The existence of Au(0) with a small size

in the Au‐Cl/AC catalyst may be attributed to the reduction of

Au(III) by the AC support [72] or the beam‐induced photoreac‐

tion of Au(III) salts [36,62]. XPS fitting parameters for various

samples are listed in Table S2. The high proportion of cationic

Au species in the Au‐N(CN)₂/AC catalyst suggested that the

[Bmim][N(CN)₂] IL can stabilize Au(III) species more effectively

than the [Bmim][Cl] IL.

X‐ray absorption spectroscopy (XAS) was employed to

evaluate the electronic structure and coordination environ‐

ment of the Au‐N(CN)₂/AC and Au‐Cl/AC catalysts by compar‐

ing the white‐line intensity values (the magnitude of the elec‐

tronic transition from the 2p_3/2core‐level state to a vacant 5d

state) and other spectral features with those of the reference

samples (Au foil, KAuCl₄, and (PPh₃)AuCH₃). Table 2 shows the

EXAFS fitting parameters at the Au L3‐edge for various sam‐

ples. The white‐line intensity values of the Au standards were

0.76, 1.14, and 0.81 for Au foil (Au(0)), KAuCl₄(Au(III)), and

(PPh₃)AuCH₃(Au(I)), respectively, which are consistent with

the measured values in literature [36,37,73]. The typical nor‐

malized white‐line intensity values of Au‐N(CN)₂/AC and

Au‐Cl/AC were 1.04 and 0.85 (Figure 2), which were closer to

the cationic Au(III) and Au(I) standards, respectively. To the

best of our knowledge, the intensity of the white line increases

as the oxidation state increases [36,37,73]. Therefore, the

measured results indicated the presence of cationic Au species

in the form of mixed Au(III)/Au(I) on the Au‐N(CN)₂/AC and

Au‐Cl/AC catalysts. In addition, more Au(III) species were gen‐

erated on Au‐N(CN)₂/AC than on Au‐Cl/AC due to the former’s

higher white‐line intensity values. This difference in Au species

on the surface is probably caused by the coordination structure

of cationic Au species. Therefore, the local structure evolution

around Au atoms was studied by FT‐EXAFS. In Figure 2b, we

can observe the presence of Au‐Cl entities at about 2.27 Å and

Au‐N entities at about 3.22 Å in the Au‐N(CN)₂/AC sample with

average coordination numbers (CNs) of 1.4 and 5.3 for Au‐Cl

and Au‐N, respectively, in contrast to the Au‐Cl entities (CN =

3.0) at about 2.27 Å and Au‐Au entities (CN = 0.5) at about 2.86

Å for the Au‐Cl/AC catalyst. These results demonstrate that

cationic Au species can be well stabilized during the synthesis

338

Jia Zhao et al. / Chinese Journal of Catalysis 42 (2021) 334–346

Table 2

EXAFS fitting parameters at the Au L3‐edge for Au‐N(CN)₂/AC and Au‐Cl/AC catalysts.

Sample

Au foil

KAuCl₄

Scattering path

Au‐Au

Au‐Cl

CN ^a

12

R (Å) ^b

2.86

2.27

2.28

2.05

2.27

2.86

2.27

3.22

σ²(Å²) ^c

0.0079

0.0024

ΔE₀(eV) ^d

3.9

R factor

0.004

0.018

3.9

1.7

1.0

3.0

0.5

1.4

5.3

9.2

Au‐P

Au‐C

Au‐Cl

Au‐Au

Au‐Cl

(PPh₃)AuCH₃

Au‐Cl/AC

0.0032

0.0012

0.0013

2.4

8.3

2.0

0.008

0.014

0.019

Au‐N(CN)₂/AC

Au‐N

^aCN: coordination number; ^bR: bond distance; ^cσ²: Debye‐Waller factor; ^dΔE₀: the inner potential correction. R factor: goodness of fit. Ѕ₀²was set to

0.829 by fixing CN as the known crystallographic value according to the experimental EXAFS fit of Au foil reference. The error bounds that character‐

ize the structural parameters obtained by EXAFS spectroscopy were estimated as CN ± 20%; R ± 1%; σ²± 20%; ΔE₀± 20%.

a

b

(PPh₃)AuCH₃

Au-C+Au-P

Au-Cl

0.81

1.14

(PPh₃)AuCH₃

KAuCl₄

Au foil

Au-Au

0.76

0.85

Au foil

Fresh Au-Cl/AC

Fresh Au-N(CN)₂/AC

Au-Cl

Fresh Au-Cl/AC

Au-Cl

Au-N

1.04

Fresh Au-N(CN)₂/AC

11900

11910

11920

11930 11940 11950

0

1

2

3

4

5

6

Energy (eV)

R(Å)

Fig. 2. The edge‐normalized XANES spectra at the Au L3‐edge for the Au‐N(CN)₂/AC and Au‐Cl/AC catalysts and reference Au samples. (a) White line

intensity and (b) R‐space.

of catalysts in the presence of Au‐N entities for Au‐N(CN)₂/AC.

3.2. Catalytic performance

[Bmim][N(CN)₂] IL can provide a suitable microenvironment

for Au species during the reaction.

The catalytic ability of the Au‐N(CN)₂/AC catalyst was fur‐

ther explored by analyzing the relationship among the GHSV,

Au loading, and VCM productivity (kg_VCMkg_Au^‒1min^‒1) (Figure

3b). In the three‐dimensional graph, it is shown that even at a

The promotion effect of ILs on the Au‐based catalysts over

acetylene

hydrochlorination

was

evaluated

using

[Bmim][N(CN)₂] and [Bmim][Cl], and Figure 3 and Figure S8a

show the evaluated results of the Au‐N(CN)₂/AC and Au‐Cl/AC

catalysts. Specifically, the catalytic performance of the two cat‐

alysts was studied first by evaluating their optimal reaction

temperature in the range of 140–220 °C (Figure S8a), with each

point determined by an isolated test to eliminate the interfer‐

ence of catalyst deactivation. The reaction temperature of 180

°C was defined as the optimal reaction temperature because

the highest VCM productivity for both Au‐N(CN)₂/AC and

Au‐Cl/AC was achieved at this temperature (Figures S8a and

S8b). After temperature evaluation, the catalytic ability of the

two Au‐SILP catalysts was investigated by stability tests on a

laboratory scale. In Figure 3a, the Au‐N(CN)₂/AC catalyst is

shown to have a higher VCM productivity and durability within

100 h on stream than the Au‐Cl/AC catalyst, indicating that the

relatively high GHSV and a low metal loading, the

Au‐N(CN)₂/AC catalyst still exhibited superior catalytic activity.

A turning point in productivity at an Au loading of 0.25 wt% is

observed under a variety of GHSV values. Meanwhile, the

Au‐N(CN)₂/AC catalyst is not sensitive to the change in the

GHSV, as shown by a productivity increase of only ~20 kg_VCM

kg_Au^‒1min^‒1with the increase of GHSV from 500 h^‒1to 2000 h^‒1

at the Au loading of 0.1 wt%; in addition, a relatively small

changing trend of the VCM productivity is observed with a fur‐

ther increase of the Au loading. As is shown in Figure S9b, the

catalytic activity of the Au‐N(CN)₂/AC catalyst increased with

the increase in the Au loading, and no noticeable decrease in

activity for all the catalysts was observed within 10 h at 180 °C

and 1000 h^‒1, indicating that the stabilization method in this

study is also effective for the ultra‐low Au content of 0.01 wt%.

Jia Zhao et al. / Chinese Journal of Catalysis 42 (2021) 334–346

339

Fig. 3. VCM productivity as a function of (a) time on stream, h, and at GHSV(C₂H₂) = 1000 h^‒1; (b) VCM productivity versus the GHSV and Au loading

for the investigated Au‐N(CN)₂/AC; (c) apparent activation energies of the Au‐Cl/AC and Au‐N(CN)₂/AC catalysts; (d) VCM productivity of

Au‐N(CN)₂/AC and other Au‐based catalysts reported in literature. Numbers 11–13 indicate the productivity of the Au‐N(CN)₂/AC catalyst at different

GHSV(C₂H₂) values; (e) VCM productivity as a function of the deactivation rate for various samples.

As the catalytic activity originates from the promotion of

[Bmim][N(CN)₂], the activation energies of the Au‐N(CN)₂/AC

and Au‐Cl/AC catalysts were calculated for further analysis of

the properties of this reaction. [Bmim][N(CN)₂] in the Au‐based

catalyst has a lower activation energy (E_a= 32.76 kJ mol^‒1) than

[Bmim][Cl] (E_a= 48.07 kJ mol^‒1), indicating that the acetylene

hydrochlorination reaction catalyzed by the Au‐N(CN)₂/AC

catalyst tends to be more easily triggered under the same reac‐

tion conditions. The relatively low activation energy of

Au‐N(CN)₂/AC may be responsible for the relatively high cata‐

lytic activity of the Au‐N(CN)₂/AC catalyst (Figure 3c). Addi‐

tionally, a comparison of Au catalysts reported in the literature

[1,17,18,24,33,41,62,70,73,74] revealed that the Au‐N(CN)₂/AC

catalyst shows considerable VCM productivity under different

hydrochlorination conditions (Figure 3d), which can rival the

performance of state‐of‐the‐art Au‐based catalysts at similar

3.3. Induction phenomenon and catalytic activity

Figure 3 shows the distinct catalytic behaviors (induction

phenomenon and catalytic activity) of the two investigated

catalysts. The reaction with the Au‐Cl/AC catalyst presented a

long induction period of 10 h (Figure 3a), with the VCM

productivity being maintained at about 50 kg_VCMkg_Au^‒1min^‒1

for 12 h, followed by a decline. In contrast, Au‐N(CN)₂/AC ex‐

hibited good catalytic stability within 100 h, with no evident

induction period being observed. This distinction indicated

their potential difference in the active species and/or inter‐

phase mass transfer. Hutchings mentioned that the induction

period for Au‐based catalysts was typically 3 h, which was al‐

lotted to the redox coupling of the Au(III)/Au(I) sites in the

Au‐catalyzed hydrochlorination of acetylene, including the ox‐

idation of Au(I) sites to predominant Au(III) sites during the

first 20 min under the reaction conditions, followed by the re‐

duction to Au(I) sites during the subsequent 160 min [36]. This

result indicates that the induction period necessary for this

reaction can be shortened by increasing the Au(III) content in

the Au(III)/Au(I) couple, which was verified by the higher

number of Au(III) species in Au‐N(CN)₂/AC than in Au‐Cl/AC in

the XAS results (Figure 2a). However, it should be noted that

the induction period of 10 h for the Au‐Cl/AC catalyst in this

work is overly high and evidently different from the induction

period of 2–5 h reported for the classical Au‐catalyzed acety‐

lene hydrochlorination (Table S4), suggesting the redox cou‐

pling process of Au(III)/Au(I) is unlikely to be the only deter‐

temperatures, although

a relatively high GHSV for the

Au‐N(CN)₂/AC catalyst was obtained in our reaction. Moreover,

the deactivation rates calculated in this work and other refer‐

ences (Table S3) suggest the excellent stability of the

Au‐N(CN)₂/AC catalyst. Furthermore, the correlation of VCM

production versus the deactivation rate of the catalysts was

established and is shown in Figure 3e, which reveals that the

Au‐N(CN)₂/AC catalyst has the highest VCM productivity but a

negligible deactivation rate. These results demonstrate that the

introduced [Bmim][N(CN)₂] contributes to the outstanding

activity and durability of the Au‐(CN)₂/AC catalyst for acetylene

hydrochlorination.

340

Jia Zhao et al. / Chinese Journal of Catalysis 42 (2021) 334–346

minant of the induction period. As the thin film dissolved

Au‐complex supported on AC is established on the

Au‐N(CN)₂/AC and Au‐Cl/AC catalysts, the dissolution and dif‐

fusion of reactants in this IL film should not be neglected, par‐

ticularly in the initial stage of this reaction.

COSMO‐RS method. As shown in Figures 4g–h, the solubilities

of C₂H₂and HCl decreased with increasing temperature. The

solubility values (herein expressed as the mole fraction) of C₂H₂

and HCl at the reaction temperature of 180 °C were 0.007 and

0.068 for Au‐N(CN)₂/AC, and 0.004 and 0.029 for Au‐Cl/AC,

respectively. The solubilities of C₂H₂and HCl were slightly

higher in [Bmim][Cl] than in [Bmim][N(CN)₂]. We have proven

above that C₂H₂and HCl undergoes predominantly chemical

dissolution in ILs, which can be affected by hydrogen‐bond

interaction between the substrates (C₂H₂and HCl) and the

basic atoms of the ILs. Therefore, the higher solubilities of C₂H₂

and HCl in [Bmim][Cl] than in [Bmim][N(CN)₂] can be ascribed

to the stronger hydrogen‐bond basicity in [Cl^–] (0.95) than in

The interfacial structure and property between an IL and

the reactants were studied first due to their intrinsic im‐

portance to understand the dissolution and diffusion behavior

of C₂H₂and HCl in ILs. C₂H₂is known to weakly bind to polar

organic molecules containing a heteroatom through interac‐

tions between X···H‒C(acetylene) (hydrogen bond) and

X‒H···C(acetylene) (where X can be a proton acceptor, such as

O, N, or halogen atoms) [46,60,75]. This means that C₂H₂can be

hydrogen‐bonded to the heteroatom or heteroatoms of the

anion of ILs. For the selected ILs in this work, the occurrence of

the hydrogen bond interaction between the H atom of C₂H₂and

the proton acceptor of the IL anion (Figures 4b, 4e) suggests

that the dissolution and diffusion behaviors of C₂H₂in the ILs

are affected by this hydrogen‐bond interaction. The distinct

activation mechanism of HCl was determined for Cl^–and

–

[N(CN)₂] (0.60) [75]. Meanwhile, the higher solubility of both

the substrate molecules in [Bmim][Cl] than in [Bmim][N(CN)₂]

suggests that solubility is probably not the main factor respon‐

sible for the long induction period and low catalytic activity of

Au‐Cl/AC relative to Au‐N(CN)₂/AC.

As a catalytic reaction is a process of continuous collision of

reaction molecules, the Wilke‐Chang correlation was used to

estimate the diffusion coefficients of C₂H₂and HCl, which were

9.07 × 10^–9m²s^‒1and 5.02 × 10^‒9m²s^‒1in [Bmim][Cl], and

1.14 × 10^‒8m²s^‒1and 9.57 × 10^‒9m²s^‒1in [Bmim][N(CN)₂]

(Figure 4i), respectively. This difference in the diffusion coeffi‐

cient can be attributed to the difference in the IL anions in‐

duced by identical cations. The analysis of the calculated results

revealed an apparently higher diffusion rate of substrates in

–

N(CN)₂‐based ILs, with HCl being more inclined to form hy‐

drogen bonds with the cation in the [Bmim][Cl] IL (Figure 4c)

and anion in [Bmim][N(CN)₂] (Figure 4f). Evidently, the reac‐

tants tend to dissolve in the IL film in the form of hydrogen

bonds and are activated at the same time.

Additionally, the solubilities of C₂H₂and HCl over

[Bmim][N(CN)₂] and [Bmim][Cl] were evaluated using the

c

a

b

2.205

HCl molecule

C₂H₂molecule

2.737

[Bmim][Cl]

2.999

d

g

f

e

C₂H₂molecule

2.592

2.104

HCl molecule

2.724

[Bmim][N(CN)₂]

h

i

C₂H₂-[Bmim][Cl]

HCl-[Bmim][Cl]

1.2

0.0042

0.0035

0.0028

0.0021

0.0014

0.0007

0.0000

0.08

0.06

0.04

0.02

0.00

HCl-[Bmim][N(CN)₂]

1.0

C₂H₂-[Bmim][N(CN)₂]

0.8

0.6

0.4

0.2

0.0

Au-Cl/AC

Au-N(CN)₂/AC

80 100 120 140 160 180 200 220

Temperature (^oC)

Fig. 4. Optimized structures of (a) [Bmim][Cl], (b) C₂H₂‐[Bmim][Cl], (c) HCl‐[Bmim][Cl], (d) [Bmim][N(CN)₂], (e) C₂H₂‐[Bmim][N(CN)₂], and (f)

HCl‐[Bmim][N(CN)₂]. The dotted lines represent the possible modes of interaction, with interatomic distances in angstroms. The solubilities of (g)

C₂H₂and (h) HCl in the investigated [Bmim][Cl] and [Bmim][N(CN)₂] ILs at different temperatures; (i) diffusion coefficients for C₂H₂and HCl in

[Bmim][Cl] and [Bmim][N(CN)₂]. Color code: H, white; C, gray; N, blue; Cl, light green.

Jia Zhao et al. / Chinese Journal of Catalysis 42 (2021) 334–346

341

[Bmim][N(CN)₂] than in [Bmim][Cl]. Similar results had also

a

b

been reported by Dai et al [76]. Apart from the initial content of

Au(III) in the Au(III)/Au(I) sites, the lower diffusion rate of

substrates in [Bmim][Cl] also contributed to the long induction

period of the Au‐Cl/AC catalyst. Therefore, the induction period

of the catalyst in the reaction process can be eliminated

through the strategy of ensuring a high Au(III) content in the

Au(III)/Au(I) sites, in addition to a sufficiently high substrate

diffusion rate in the IL layer.

3.4. Proposed mechanism

d

c

Au(III)

Au(0)

Au(I)

Au(III)

Au(I)

In this study, the stabilization mechanism of the cationic Au

species in the steady‐state process was also studied. As the

deactivation of the Au‐based catalysts in acetylene hydrochlo‐

rination is considered as the main result of cationic Au reduc‐

tion by C₂H₂, the remarkably higher catalytic stability of

Au‐N(CN)₂/AC than that of Au‐Cl/AC suggests that the

[Bmim][N(CN)₂]‐stabilized cationic Au species are stable

enough to resist the deactivation of the metallic Au reduced by

the strongly reducing C₂H₂. The characterization results of the

spent catalysts indicated that carbon deposition and Au leach‐

ing were not the main causes of catalyst deactivation (Table 3).

XPS analysis confirmed the presence of metallic Au(0) species

in the spent Au‐Cl/AC sample (Figure 5c), but not in the spent

Au‐N(CN)₂/AC catalyst (Figure 5d), demonstrating that the

cationic Au species are highly stable when Au‐Cl is substituted

by Au‐N. The agglomeration of atomically dispersed Au into

large clusters was also determined from the HAADF‐STEM im‐

ages of the spent Au‐Cl/AC catalyst (Figure 5a). For

Au‐N(CN)₂/AC, the atomically dispersed Au was still main‐

tained over the surface of the spent Au‐N(CN)₂/AC catalyst

(Figure 5b), further verifying that the Au(III) is stabilized by

[Bmim][N(CN)₂]. To the best of our knowledge, Au‐Cl sites or

AuCl₃, are labile and easy to dissociate/disproportionate in the

process of catalyst synthesis and evaluation [41]. The stability

constant of the Au‐N coordination is several orders of magni‐

tude higher than that of the Au‐Cl coordination [41], which

inhibits the migration of Au atoms during the preparation of

the catalyst. Accordingly, the strong interaction between the

Au‐N contributions stabilized and inhibited the reduction of the

cationic Au species in the process of catalytic preparation and

reaction.

Au(I)

4f_5/2

Au(III)

4f_7/2

4f_5/2

Au(I)

92 90 88 86 84 82 80

Binding Energy (eV)

98 96 94 92 90 88 86 84 82

Binding Energy (eV)

Fig. 5. Characterizations of the spent (a,c) Au‐Cl/AC and (b,d)

Au‐N(CN)₂/AC catalysts. (a,b) TEM images; (c,d) XPS results.

Au source for the catalyst synthesis and induction period of the

reaction process. Additionally, the core role of Au(III) in the

catalytic behavior is also shown by its dominance in the XPS

and EXAFS spectra of Au‐N(CN)₂/AC and Au‐Cl/AC. The

Hirshfeld charge of the Au atom in [Bmim][N(CN)₂] was lower

(0.294 e) than that (0.334 e) of [Bmim][Cl], which were close to

the values of Au(I) and Au(III) with charges of 0.204 e and

0.356 e when modeled as AuCl and AuCl₃(Table S6), respec‐

tively. In the Au/AC catalyst, Au(I) can be generated via the

thermal decomposition of Au(III) species [36,37]. Therefore,

the improved stability of the Au species in [Bmim][N(CN)₂] of

the Au‐N(CN)₂/AC catalyst can be attributed to the Au(I) char‐

–

acteristic of the [N(CN)₂]‐stabilized cationic Au(III). As is

shown in Figure S9, the average Hirshfeld charge on N atoms in

the [N(CN)₂] anion of AuCl₃‐[Bmim][N(CN)₂] complexes is

–

–0.175 e, which is much lower than that in the pure

[Bmim][N(CN)₂] IL (–0.274 e), suggesting that some of the

electrons of the N atoms in [N(CN)₂] transferred to Au atoms,

since the Hirshfeld charge on Au was decreased from 0.356 to

0.294 e. Meanwhile, when the catalyst adsorbs C₂H₂and HCl

molecules in the reaction process, the Hirshfeld charges on the

N and Au atoms do not change significantly (Au<0.034 e,

N<0.006 e), indicating the stabilization effect of [N(CN)₂] ani‐

ons on the electron structure of the Au species, which cannot be

achieved by [Cl^–] anions, as shown in Figure 6b (Au < 0.099 e).

The regulatory effect of the electronic structure of the Au

atoms on the activation of substrates was studied by thermal

programmed desorption mass spectroscopy (TPD‐MS). Previ‐

ous studies assigned the atomic masses of 26 and 36 to C₂H₂

and HCl [73,77], respectively. The peaks at 155 °C and 174 °C

(Figure S10a) represent the desorption profiles of C₂H₂from

the Au‐N(CN)₂/AC and Au‐Cl/AC catalysts. In general, the de‐

sorption temperatures in the TPD profiles reflect the binding

The electronic properties of the cationic Au species stabi‐

lized by [Bmim][N(CN)₂] and [Bmim][Cl] ILs were further in‐

vestigated by Hirshfeld charge analysis. In accordance with the

literature [41], we focused on AuCl₃in the following mechanis‐

tic study due to the role of cationic Au(III) in the selection of an

Table 3

Porous structure and metal content of the spent Au‐based catalysts.

S_BET

(m²g^‒1)^a

695

Volume

(cm³g^‒1) ^b

0.38

Au content

(wt%) ^c

0.09

Sample

Spent Au‐N(CN)₂/AC

Spent Au‐Cl/AC

787

0.37

0.11

b

a

Measured using the Brunauer‐Emmett‐Teller (BET) method; calcu‐

lated based on the adsorbed N₂volume;^cdetermined by ICP‐MS analy‐

sis.

342

Jia Zhao et al. / Chinese Journal of Catalysis 42 (2021) 334–346

a

1

6

2

7

3

4

9

5

8

10

b

0.389

0.4

0.3

0.2

0.1

0.0

0.371

0.356

0.334

0.328

0.32

0.295

0.294

0.29

0.204

0

1

2

3

4

5

6

7

8

9

10

Adsorbed state

Fig. 6. (a) Electron density and (b) the Hirshfeld charge of the Au atoms in various Au states.

strength of the adsorbed sites on the catalyst surface, and the

peak area correlates with the number of adsorbed sites

[46–48]. Accordingly, the adsorption of C₂H₂on Au‐N(CN)₂/AC

was inhibited due to the smaller peak area and lower desorp‐

tion temperature of C₂H₂in Au‐N(CN)₂/AC than in Au‐Cl/AC.

Furthermore, the selected catalysts were studied by HCl‐TPD

analysis. As shown in Figure S10b, the desorption content of

HCl on Au‐N(CN)₂/AC was lower than that of Au‐Cl/AC, due to

the small difference in the desorption temperature, which is in

contrast to the similar adsorption of HCl on both catalysts. The

C₂H₂and HCl desorption contents from TPD analysis were con‐

sistent with the solubility values from the theoretical calcula‐

tion.

Strong HCl adsorption was observed on the [Cl^–]‐stabilized Au

atoms, in contrast to the slight interaction detected on the

[N(CN)₂]‐stabilized Au atoms.

–

To gain more insights into the adsorption properties of the

substrates, we calculated the adsorption energies of C₂H₂and

HCl on Au‐N(CN)₂/AC and Au‐Cl/AC as –75.54 and –85.48 kJ

–

mol^‒1, suggesting that the [N(CN)₂]‐stabilized Au atoms

weakened the strong adsorption of C₂H₂, which was consistent

with the results obtained by Hirshfeld charge and TPD anal‐

yses. For the HCl adsorbed states, the adsorption energies were

–23.23 kJ mol^‒1on Au‐N(CN)₂/AC and –80.02 kJ mol^‒1on

Au‐Cl/AC, which were also consistent with the results of

Hirshfeld charge analysis. It should be noted that almost no

–

The above TPD results indicated that the involvement of the

electron is transferred between [N(CN)₂]‐stabilized Au atoms

–

ligand, [N(CN)₂], in the formation of the Au‐N contribution

and HCl for Au‐N(CN)₂/AC, and the adsorption energy is prob‐

ably due to the adsorption of the acidic HCl on the basic

through Au coordination enables the resulting structure to

notably inhibit the adsorption of C₂H₂, thereby stabilizing the

cationic Au species. The activation of C₂H₂on the Au(III) sites of

Au‐N(CN)₂/AC and Au‐Cl/AC catalysts was further elucidated

by the Hirshfeld charge analysis. Table S6 shows the calculated

charge distributions before and after C₂H₂adsorption on

Au‐N(CN)₂/AC and Au‐Cl/AC. A more positive charge was de‐

tected in the Au centers of Au‐N(CN)₂/AC and Au‐Cl/AC after

C₂H₂adsorption, indicating the flow of electrons from the Au

center to the C₂H₂substrate and suggesting the strong adsorp‐

tion of C₂H₂. It is worth noting that the electron flow was more

significant in Au‐Cl/AC than in Au‐N(CN)₂/AC due to the gen‐

eration of 0.037 e and 0.026 e Hirshfeld charges in the adsorp‐

tion process, indicating the higher probability of C₂H₂to be

adsorbed at the Au center stabilized by [Cl^–] ligands than that

–

[N(CN)₂] anion in Au‐N(CN)₂/AC.

The different adsorption properties of the C₂H₂and HCl

molecules on Au‐N(CN)₂/AC and Au‐Cl/AC induced the varia‐

tion of the reaction paths in acetylene hydrochlorination. The

electron cloud density of the basic structure is shown in Figures

S11–S13. Due to the lower adsorption energy of C₂H₂than that

of HCl, the reactions of Au‐N(CN)₂/AC and Au‐Cl/AC were likely

to be triggered by the adsorption and activation of C₂H₂. The

reaction pathway analysis did show that the coordination of

C₂H₂in the absence of HCl results in site blocking through the

formation of this stable metal cycling structure (Figure 7),

which was consistent with the inactivation observed under

C₂H₂exposure [1–7]. Based on DFT calculations, the proposed

catalytic cycles of Au‐N(CN)₂/AC and Au‐Cl/AC‐catalyzed acet‐

ylene hydrochlorinations consisted of eight elementary steps:

the coordination of AuCl₃on selected ILs (step 1), the adsorp‐

tion of C₂H₂(step 2) and HCl (step 3), the formation of the tran‐

sition state (step 4), the surface reaction (steps 5 and 6), and

the desorption of vinyl chloride (steps 7 and 8). For the

Au‐N(CN)₂/AC catalyst, C₂H₂was first adsorbed to the Au at‐

oms of AuCl₃(step 2), after which it interacted with HCl to form

the more stable HCl‐C₂H₂sites (step 3). The adsorption energy

–

by [N(CN)₂] ligands. In other words, the adsorption of C₂H₂

–

was weakened on [N(CN)₂]‐stabilized Au atoms, which was

consistent with the solubility and TPD measurements. When

HCl was adsorbed, the Hirshfeld charge values of the Au atoms

were changed from the initial 0.294 e and 0.334 e to 0.295 e

and 0.290 e for Au‐N(CN)₂/AC and Au‐Cl/AC catalysts, respec‐

tively, with –0.001 and 0.044 electrons being transferred from

the HCl atom to the Au atom of Au‐N(CN)₂/AC and Au‐Cl/AC.

Jia Zhao et al. / Chinese Journal of Catalysis 42 (2021) 334–346

343

was desorbed, and the active catalyst was released to complete

the catalytic cycle.

In the proposed reaction pathway of the Au‐N(CN)₂/AC and

Au‐Cl/AC catalysts, the dissociation of HCl and the generation

of vinyl chloride were the rate‐limiting steps for the catalytic

cycle with the requirements of 17 and 44 kJ mol^‒1, respectively.

This may explain why the catalytic activity of Au‐N(CN)₂/AC is

higher than that of Au‐Cl/AC (Figure 3a). Additionally, the

source of HCl from the gas phase for Au‐Cl/AC and its dissocia‐

tion on Au‐N(CN)₂/AC should not be ignored. Due to its disso‐

–

ciation by the [N(CN)₂] anion, the HCl is activated, leading to a

greater reduction in the adsorption of the chlorinated acetylene

intermediate, [Cl‐C₂H₂], on the Au site of Au‐N(CN)₂/AC rather

than on that of Au‐Cl/AC. The adsorption energies of the chlo‐

rinated acetylene over Au‐N(CN)₂/AC and Au‐Cl/AC were –100

and –184 kJ mol^‒1, respectively, indicating that the dissociation

of HCl can significantly inhibit the deactivation pathway of cat‐

ionic Au reduction caused by the strong adsorption of acety‐

lene. HCl dissociation by N ligands had already been reported

in the literature [73]. Owing to the low reaction energy re‐

quired in the rate‐limiting steps and the weak adsorption of

chlorinated acetylene, Au‐N(CN)₂/AC exhibited high catalytic

activity and stability toward acetylene hydrochlorination. Ad‐

ditionally, the adsorption energy of HCl on the individual

[Bmim][N(CN)₂] IL was calculated to be –60 kJ mol^‒1, and HCl

Fig. 7. Reaction pathways of the Au‐Cl/AC and Au‐N(CN)₂/AC catalysts.

was calculated as –102.51 kJ mol^‒1for Au‐N(CN)₂/AC during

this adsorption process. Subsequently, the adsorbed HCl was

dissociated to form N–H bonds between the H atom of HCl and

–

the N atom of the [N(CN)₂] anion with a N–H bond length of

1.03 Å in the intermediate (step 4); the Cl atom of HCl was

slightly closer to the Au atom, which was consistent with the

above Hirshfeld charge analysis. Subsequently, the latter in‐

termediate underwent one Cl atom transfer from the Au atom

to the C atom of C₂H₂, leading to the formation of chlorinated

acetylene (step 5). The N–H bond was activated from 1.03 Å to

1.58 Å, and the H atom was close to the C atom of C₂H₂(step 6).

Finally, the product was generated (step 7) and simultaneously

released (step 8) to complete the catalytic cycle.

–

could not dissociate at the [N(CN)₂] anion in the case of no

–

coordination between [N(CN)₂] and Au (Figure S14). When

HCl is adsorbed, the Hirshfeld charges of the N atoms with and

without Au show a transfer of 0.040 e (Figure 8a) and 0.023 e

For Au‐Cl/AC, C₂H₂was also firstly adsorbed, after which it

interacted with gaseous HCl via step 2 and step 3 (Figure 7).

–

(Figure 8b), respectively, indicating that the [N(CN)₂] anion is



Next, the Cl atoms from AuCl₄attacked C₂H₂to form a more

activated by Au to promote the dissociation of HCl.

stable adsorption structure (step 5) via an intermediate (step

4). Subsequently, the H atom was transferred from HCl to the C

atom of C₂H₂to form vinyl chloride (step 7) via a transition

state (step 6). Meanwhile, the Cl atom of HCl was added to the

consumed Cl atom of the AuCl₄^structure. Finally, the product

4. Conclusions

In this study, we systematically investigated the induction

and controversial reaction mechanism of the Au‐catalyzed hy‐

a

b

Fig. 8. Hirshfeld charge of the N atoms before and after HCl adsorption. (a) AuCl₃‐[Bmim][N(CN)₂]; (b) [Bmim][N(CN)₂].

344

Jia Zhao et al. / Chinese Journal of Catalysis 42 (2021) 334–346

[14] H. Y. Zhang, B. Dai, X. G. Wang, W. Li, Y. Han, J. J. Gu, J. L. Zhang,

Green Chem., 2013, 15, 829–836.

[15] J. H. Xu, J. Zhao, J. T. Xu, T. T. Zhang, X. N. Li, X. X. Di, J. Ni, J. G. Wang,

J. Cen, Ind. Eng. Chem. Res., 2014, 53, 14272–14281.

[16] H. Y. Zhang, W. Li, X. Q. Li, W. Zhao, J. J. Gu, X. Y. Qi, Y. Z. Dong, B.

Dai, J. L. Zhang, Catal. Sci. Technol., 2015, 5, 1870–1877.

[17] J. Zhao, B. L. Wang, X. L. Xu, Y. Yu, S. X. Di, H. Xu, Y. Y. Zhai, H. H. He,

L. L. Guo, Z. Y. Pan, X. N. Li, J. Catal., 2017, 350, 149–158.

[18] J. Zhao, Y. Yu, X. L. Xu, S. X. Di, B. L. Wang, H. Xu, J. Ni, L. L. Guo, Z. Y.

Pan, X. N. Li, Appl. Catal. B, 2017, 206, 175–183.

drochlorination of acetylene in a SILP system and successfully

prepared Au single‐atom catalysts. The reaction induction pe‐

riod of the Au‐N(CN)₂/AC catalyst could be effectively elimi‐

nated due to the relatively high diffusion rate of the substrates

in the [Bmim][N(CN)₂] IL layer and the high content of Au(III)

in the Au(III)/Au(I) mixed sites. Owing to the low reaction en‐

ergy of the rate‐limiting steps, the Au‐N(CN)₂/AC catalyst ex‐

hibits high catalytic activity in acetylene hydrochlorination. The

–

activation of the [N(CN)₂] anion for stabilizing Au atoms pro‐

motes the dissociation of HCl and weakens the deactivation of

cationic Au due to the strong adsorption of C₂H₂. This work not

only provides a new optimization strategy for the mechanism

of Au‐catalyzed acetylene hydrochlorination but also facilitates

the design of Au catalysts and the development of non‐Hg cata‐

lysts for the manufacture of PVC.

[19] X. Y. Li, M. Y. Zhu, B. Dai, Appl. Catal. B, 2013, 142–143, 234–240.

[20] M. Conte, C. J. Davies, D. J. Morgan, T. E. Davies, D. J. Elias, A. F.

Carley, P. Johnston, G. J. Hutchings, J. Catal., 2013, 297, 128–136.

[21] L. Wang, F. Wang, J. D. Wang, X. L. Tang, Y. L. Zhao, D. Yang, F. M.

Jia, T. Hao, React. Kinet. Mech. Catal., 2013, 110, 187–194.

[22] P. Li, H. B. Li, X. L. Pan, K. Tie, T. T. Cui, M. Z. Ding, X. H. Bao, ACS

Catal., 2017, 7, 8572–8577.

[23] J. M. Ma, S. J. Wang, B. X. Shen, React. Kinet. Mech. Catal., 2013, 110,

177–186.

References

[24] K. Zhou, J. C. Jia, C. H. Li, H. Xu, J. Zhou, G. H. Luo, F. Wei, Green

Chem., 2015, 17, 356–364.

[25] C. F. Huang, M. Y. Zhu, L. H. Kang, B. Dai, Catal. Commun., 2014, 54,

61–65.

[26] X. Y. Li, Y. Wang, L. H. Kang, M. Y. Zhu, B. Dai, J. Catal., 2014, 311,

288–294.

[27] T. V. Krasnyakova, I. V. Zhikharev, R. S. Mitchenko, V. I.

Burkhovetski, А. M. Korduban, T. V. Kryshchuk, S. A. Mitchenko, J.

Catal., 2012, 288, 33–43.

[28] S. A. Mitchenko, T. V. Krasnyakova, Kinet. Catal., 2014, 55, 722–728.

[29] Y. F. Pu, J. L. Zhang, X. Wang, H. Y. Zhang, L. Yu, Y. Z. Dong, W. Li,

Catal. Sci. Technol., 2014, 4, 4426–4432.

[30] F. Wang, L. Wang, J. D. Wang, Y. L. Zhao, Y. L. Wang, D. Yang, React.

Kinet. Mech. Catal., 2014, 114, 725–734.

[31] M. Conte, C. J. Davies, D. J. Morgan, A. F. Carley, P. Johnston, G. J.

Hutchings, Catal. Lett., 2014, 144, 1–8.

[32] J. G. Zhao, X. G. Cheng, L. Wang, R. H. Ren, J. J. Zeng, H. H. Yang, B. X.

Shen, Catal. Lett., 2014, 144, 2191–2197.

[1] B. Nkosi, N. J. Coville, G. J. Hutchings, M. D. Adams, J. Friedl, F. E.

Wagner, J. Catal., 1991, 128, 366–377.

[2] G. J. Hutchings, Catal. Today, 2005, 100, 55–61.

[3] M. Conte, A. F. Carley, C. H. Heirene, D. J. Willock, P. Johnston, A. A.

Herzing, C. J. Kiely, G. J. Hutchings, J. Catal., 2007, 250, 231–239.

[4] M. Conte, A. F. Carley, G. A. Attard, A. A. Herzing, C. J. Kiely, G. J.

Hutchings, J. Catal., 2008, 257, 190–198.

[5] G. J. Hutchings, Top. Catal., 2008, 48, 55–59.

[6] M. Conte, A. F. Carley, G. J. Hutchings, Catal. Lett., 2008, 124,

165–167.

[7] J. Li, J. T. Fan, S. Ali, G. J. Lan, H. D. Tang, W. F. Han, H. Z. Liu, B. Li, Y.

Li, Chin. J. Catal., 2019, 40, 141–146.

[8] Y. Yang, G. J. Lan, X. L. Wang, Y. Li, Chin. J. Catal., 2016, 37,

1242–1248.

[9] Y. Y. Qiu, S. Ali, G. J. Lan, H. Q. Tong, J. T. Fan, H. Y. Liu, B. Li, W. F.

Han, H. D. Tang, H. Z. Liu, Y. Li, Carbon, 2019, 146, 406–412.

[10] S. J. Wang, B. X. Shen, Q. L. Song, Catal. Lett., 2010, 134, 102–109.

[11] W. R. Liu, M. Y. Zhu, B. Dai, New J. Chem., 2018, 42, 20131–20136.

[12] J. L. Zhang, N. Liu, W. Li, B. Dai, Front. Chem. Sci. Eng., 2011, 5,

514–520.

[33] J. Zhao, B. L. Wang, Y. X. Yue, S. X. Di, Y. Y. Zhai, H. H. He, G. F. Sheng,

H. X. Lai, Y. H. Zhu, L. L. Guo, X. N. Li, J. Catal., 2018, 365, 153–162.

[34] J. Zhao, B. L. Wang, Y. X. Yue, G. F. Sheng, H. X. Lai, S. S. Wang, L. Yu,

Q. F. Zhang, F. Feng, Z. T. Hu, X. N. Li, J. Catal., 2019, 373, 240–249.

[35] Y. Y. Zhai, J. Zhao, X. X. Di, S. X. Di, B. L. Wang, Y. X. Yue, G. F. Sheng,

[13] G. Qin, Y. H. Song, R. Jin, J. Shi, Z. Y. Yu, S. K. Cao, Green Chem., 2011,

13, 1495–1498.

Graphical Abstract

Chin. J. Catal., 2021, 42: 334–346 doi: 10.1016/S1872‐2067(20)63617‐8

Acetylene hydrochlorination over supported ionic liquid

phase (SILP) gold‐based catalyst: Stabilization of cationic Au

species via chemical activation of hydrogen chloride and

corresponding mechanisms

Au-IL complexes

Jia Zhao*, Saisai Wang, Bolin Wang, Yuxue Yue, Chunxiao Jin,

Jinyue Lu, Zheng Fang, Xiangxue Pang, Feng Feng, Lingling Guo,

Zhiyan Pan, Xiaonian Li *

Au

Zhejiang University of Technology

SILP

Cationic Au species coordinate with [Bmim][N(CN)₂] to form

strong Au‐IL complexes, which greatly increase the thermal and

reaction stability of Au‐based catalysts.

Jia Zhao et al. / Chinese Journal of Catalysis 42 (2021) 334–346

345

H. X. Lai, L. L. Guo, H. Wang, X. N. Li, Catal. Sci. Technol., 2018, 8,

2901–2908.

Limbeck, K. Bica, H. Hoffmann, K. Kirchner, ACS Catal., 2018, 8,

1048–1051.

[36] G. Malta, S. A. Kondrat, S. J. Freakley, C. J. Davies, L. Lu, S. Dawson,

A. Thetford, E. K. Gibson, D. J. Morgan, W. Jones, P. P. Wells, P.

Johnston, C. R. A. Catlow, C. J. Kiely, G. J. Hutchings, Science, 2017,

355, 1399–1403.

[37] G. Malta, S. A. Kondrat, S. J. Freakley, C. J. Davies, S. Dawson, X. Liu,

L. Lu, K. Dymkowski, F. Fernandez‐Alonso, S. Mukhopadhyay, E. K.

Gibson, P. P. Wells, S. F. Parker, C. J. Kiely, G. J. Hutchings, ACS

Catal., 2018, 8, 8493–8505.

[58] T. Yasuda, E. Uchiage, T. Fujitani, K. Tominaga, M. Nishida, Appl.

Catal. B, 2018, 232, 299–305.

[59] P. Z. Zhang, Z. B. Jiang, Y. H. Cui, G. Q. Xie, Y. Z. Jin, L. L. Guo, Y. Q. Xu,

Q. F. Zhang, X. N. Li, Appl. Catal. B, 2019, 255,

117757/1–117757/10.

[60] R. Kukawka, A. Pawlowska‐Zygarowicz, J. Dzialkowska, M. Pie‐

trowski, H. Maciejewski, K. Bica, M. Smiglak, ACS Sustain. Chem.

Eng., 2019, 7, 4699–4706.

[38] K. Zhou, W. Wang, Z. Zhao, G. H. Luo, J. T. Miller, M. S. Wong, F. Wei,

ACS Catal., 2014, 4, 3112–3116.

[61] J. J. Peng, J. Y. Li, Y. Bai, H. Y. Qiu, K. Z. Jiang, J. X. Jiang, G. Q. Lai,

Catal. Commun., 2008, 9, 2236–2238.

[39] H. Xu, K. Zhou, J. K. Si, C. H. Li, G. H. Luo, Catal. Sci. Technol., 2016,

6, 1357–1366.

[62] J. Zhao, S. C. Gu, X. L. Xu, T. T. Zhang, Y. Yu, X. X. Di, J. Ni, Z. Y. Pan, X.

N. Li, Catal. Sci. Technol., 2016, 6, 3263–3270.

[40] M. Y. Zhu, Q. W. Wang, K. Chen, Y. W. Wang, C. H. Huang, H. Dai, F.

Yu, L. H. Kang, B. Dai, ACS Catal., 2015, 5, 5306–5316.

[41] P. Johnston, N. Carthey, G. J. Hutchings, J. Am. Chem. Soc., 2015,

137, 14548–14557.

[42] X. H. Tian, G. H. Hong, B. B. Jiang, F. P. Lu, Z. W. Liao, J. D. Wang, Y.

R. Yang, RSC Adv., 2015, 5, 46366–46371.

[63] J. Oliver‐Meseguer, A. Domenech‐Carbo, M. Boronat, A. Ley‐

va‐Perez, Angew. Chem. Int. Ed., 2017, 56, 6435–6439.

[64] S. X. Di, Y. Q. Xu, Q. F. Zhang, X. L. Xu, Y. Y. Zhai, B. L. Wang, H. H. He,

Q. T. Wang, H. Xu, Y. S. Jiang, J. Zhao, X. N. Li, RSC Adv., 2018, 8,

24094–24100.

[65] B. L. Wang, J. Zhao, Y. X. Yue, G. F. Sheng, H. X. Lai, J. Y. Rui, H. H. He,

Z. T. Hu, F. Feng, Q. F. Zhang, L. L. Guo, X. N. Li, ChemCatChem,

2019, 11, 1002–1009.

[43] A. Goguet, C. Hardacre, I. Harvey, K. Narasimharao, Y. Saih, J. Sa, J.

Am. Chem. Soc., 2009, 131, 6973–6975.

[44] X. Duan, X. Tian, J. Ke, Y. Yin, J. Zheng, J. Chen, Z. Cao, Z. Xie, Y. Yuan,

Chem. Sci., 2016, 7, 3181–3187.

[66] K. C. O’Connell, J. R. Monnier, J. R. Regalbuto, Appl. Catal. B, 2018,

225, 264–272.

[45] X. Duan, Y. Yin, X. Tian, J. Ke, Z. Wen, J. Zheng, M. Hu, L. Ye, Y. Yuan,

Chin. J. Catal., 2016, 37, 1794–1803.

[67] L. Wang, B. X. Shen, J. J. Zhao, X. T. Bi, Can. J. Chem. Eng., 2017, 95,

1069–1075.

[46] S. S. Shang, W. Zhao, Y. Wang, X. Y. Li, J. L. Zhang, Y. Han, W. Li, ACS

Catal., 2017, 7, 3510–3520.

[68] J. Liu, G. J. Lan, Y. Y. Qiu, X. L. Wang, Y. Li, Chin. J. Catal., 2018, 39,

1664–1671.

[47] Y. Li, Y. Z. Dong, W. Li, Y. Han, J. L. Zhang, Mol. Catal., 2017, 443,

220–227.

[69] J. Zhao, J. T. Xu, J. H. Xu, T. T. Zhang, X. X. Di, J. Ni, X. N. Li, Chem. Eng.

J., 2015, 262, 1152–1160.

[48] B. L. Wang, H. X. Lai, Y. X. Yue, G. F. Sheng, Y. Q. Deng, H. H. He, L. L.

Guo, J. Zhao, X. N. Li, Catalysts, 2018, 8, 351/1–351/13.

[49] S. Werner, N. Szesni, A. Bittermann, M. J. Schneider, P. Härter, M.

Haumann, P. Wasserscheid, Appl. Catal. A, 2010, 377, 70–75.

[50] A. Weiβ, M. Giese, M. Lijewski, R. Franke, P. Wasserscheid, M.

Haumann, Catal. Sci. Technol., 2017, 7, 5562–5571.

[51] J. Zhao, Y. X. Yue, G. F. Sheng, B. L. Wang, H. X. Lai, S. X. Di, Y. Y. Zhai,

L. L. Guo, X. N. Li, Chem. Eng. J., 2019, 360, 38–46.

[52] Y. X. Yue, B. L. Wang, G. F. Sheng, H. X. Lai, S. S. Wang, Z. Chen, Z.‐T.

Hu, J. Zhao, X. N. Li, New J. Chem., 2019, 43, 12767–12775.

[53] Y. Q. Cen, Y. X. Yue, S. S. Wang, J. Y. Lu, B. L. Wang, C. X. Jin, L. L. Guo,

Z.‐T. Hu, J. Zhao, Catalysts, 2020, 10, 24/1–24/14.

[54] M. Haumann, K. Dentler, J. Joni, A. Riisager, P. Wasserscheid, Adv.

Synth. Catal., 2007, 349, 425–431.

[55] M. Marchetti, C. Botteghi, S. Paganelli, M. Taddei, Adv. Synth. Catal.,

2003, 345, 1229–1236.

[56] L. Naicker, H. B. Friedrich, A. Govender, P. Mohlala, Appl. Catal. A,

2018, 562, 37–48.

[57] J. Brünig, Z. Csendes, S. Weber, N. Gorgas, R. W. Bittner, A.

[70] L. Ye, X. P. Duan, S. Wu, T. S. Wu, Y. X. Zhao, A. W. Robertson, H. L.

Chou, J. W. Zheng, T. Ayvali, S. Day, C. Tang, Y. L. Soo, Y. Z. Yuan, S.

C. E. Tsang, Nat. Commun., 2019, 10, 914/1–914/10.

[71] X. P. Duan, L. C. Ning, Y. Yin, Y. T. Huang, J. Gao, H. Q. Lin, K. Tan, H.

H. Fang, L. M. Ye, X. Lu, Y. Z. Yuan, ACS Appl. Mater. Interfaces,

2019, 11, 11317–11326.

[72] H. X. Lai, B. L. Wang, Y. X. Yue, G. F. Sheng, S. S. Wang, F. Feng, Q. F.

Zhang, J. Zhao, X. N. Li, ChemCatChem, 2019, 11, 3318–3326.

[73] S. K. Kaiser, R. Lin, S. Mitchell, E. Fako, F. Krumeich, R. Hauert, O. V.

Safonova, V. A. Kondratenko, E. V. Kondratenko, S. M. Collins, P. A.

Midgley, N. Lopez, J. Pérez‐Ramírez, Chem. Sci., 2019, 10, 359–369.

[74] J. Zhao, J. T. Xu, J. H. Xu, J. Ni, T. T. Zhang, X. L Xu, X. N. Li,

ChemPlusChem, 2015, 80, 196–201.

[75] A. F. Claudio, A. M. Ferreira, S. Shahriari, M. G. Freire, J. A. Coutinho,

J. Phys. Chem. B, 2011, 115, 11145–11153.

[76] R. Santiago, J. Bedia, D. Moreno, C. Moya, J. de Riva, M. Larriba, J.

Palomar, Sep. Purif. Technol., 2018, 204, 38–48.

[77] R. Lin, S. K. Kaiser, R. Hauert, J. Pérez‐Ramírez, ACS Catal., 2018, 8,

1114–1121.

乙炔氢氯化反应中的负载金-离子液体催化剂: 离子态金物种的稳定机制

赵佳^a,*, 王赛赛^a, 王柏林^a,b, 岳玉学^a, 金春晓^a, 陆金跃^a, 方正^a, 庞祥雪^a,

丰枫^a, 郭伶伶^a, 潘志彦^b, 李小年^a,#

^a浙江工业大学工业催化研究所, 浙江杭州310014

^b浙江工业大学环境工程学院, 浙江杭州310014

摘要: 聚氯乙烯(PVC)作为世界通用工程塑料之一, 具有优异的物理、化学和机械性能, 在工业、农业、建筑、包装、电力

等行业中应用广泛. 氯乙烯是生产聚氯乙烯的重要单体. 氯乙烯的生产主要有电石法和乙烯法两种工艺路线, 由于我国

346

Jia Zhao et al. / Chinese Journal of Catalysis 42 (2021) 334–346

“贫油、富煤、少气”的资源现状, 电石法产能占全部产能的83%以上. 电石法生产氯乙烯的原理是在氯化汞催化剂存在下,

将电石水解精制后的乙炔气与氯化氢加成直接合成氯乙烯. 随着节能减排及环保要求的逐渐提高和国际涉汞公约的实施,

开发新一代绿色无汞催化剂具有重要的战略意义. 近年来, 金基催化剂是无汞催化剂基础研究和技术开发中最重要的方

向. 在之前的工作中, 我们课题组首先报道了负载离子液体-金催化剂体系(Au-SILP)在电石法生产氯乙烯工艺中的应用,

并发现离子液体的存在可以显著提高金活性物种在载体表面的分散度和稳定其化学价态. 在后续研究中, 我们在负载离

子液体-金催化体系中引入金属铜离子(Cu²⁺), 利用反应过程中Au-Cu之间的氧化还原循环, 设计并制备了金属铜基配位离

子液体, 构建了负载离子液体-金-铜催化剂体系. 铜离子的引入形成了一个催化剂自身维持氧化态的微环境, 实现了被还

原金物种的原位氧化再生.

–

本文在上述研究基础上, 利用配位离子液体[Bmim][N(CN)₂]中[N(CN)₂]阴离子和阳离子金之间的强配位作用, 构建出

比Au-Cl键更稳定的Au–N键, 并通过X射线光电子能谱(XPS)、球差校正-扫描透射电镜(AC-STEM)和扩展X射线吸收精细

结构(EXAFS)表征证明了Au以单原子状态存在于载体表面. 制备的Au-N(CN)₂/AC催化剂在乙炔氢氯化反应中表现出比

–

Au-Cl/AC和Au/AC催化剂更高的稳定性和催化活性以及更短的诱导期. 进一步表征分析发现, [N(CN)₂]配体促进了阳离

子金和配体之间的电子转移, 提高了阳离子金的电子云密度, 削弱了乙炔在阳离子金上的吸附强度, 抑制了其还原, 提高了

–

催化剂的稳定性. 更重要的是, 与阳离子金配位的[N(CN)₂]配体使得反应过程中的氯化氢在氮位点发生化学解离, 促进了

氯化氢活化, 同时降低了反应能垒. 对负载配位离子液体-金催化体系反应诱导期的分析结果表明, 反应诱导期与反应物

(乙炔、氯化氢)分子在离子液体层中的溶解度无关, 而主要取决于催化剂中Au(III)物种的含量和反应物分子在离子液体中

的扩散速率. 上述研究结果进一步深化了离子液体和活性金物种之间电子的作用机理, 建立了负载离子液体-金催化剂体

系对反应物的活化机制和反应机理, 为进一步开发具有工业应用价值的乙炔氢氯化反应无汞催化剂提供了科学基础和参

考.

关键词: 乙炔氢氯化; 电子密度; 氯化氢活化; 稳定机理; 金基负载离子液体相催化剂

收稿日期: 2020-03-21. 接受日期: 2020-05-10. 出版日期: 2021-02-05.

*通讯联系人. 电话: (0571)88871565; 电子信箱: jiazhao@zjut.edu.cn

^#通讯联系人. 电话: (0571)88320002; 电子信箱: xnli@zjut.edu.cn

基金来源: 国家自然科学基金(21606199); 浙江省科技厅(LGG20B060004).

本文的电子版全文由Elsevier出版社在ScienceDirect上出版(http://www.sciencedirect.com/science/journal/18722067).

Article Doi

Acetylene hydrochlorination over supported ionic liquid phase (SILP) gold-based catalyst: Stabilization of cationic Au species via chemical activation of hydrogen chloride and corresponding mechanisms

DOI: 10.1016/S1872-2067(20)63617-8

Source and publish data:

Authors:

Article abstract of DOI:10.1016/S1872-2067(20)63617-8

Full text of DOI:10.1016/S1872-2067(20)63617-8

Products guided by the article

R&D Labs maybe for 75-01-4

Relevant to this article

Hot Product