Ni-exchanged Y-zeolite: An efficient heterogeneous catalyst for acetylene hydrocarboxylation

Article abstract of DOI:10.1016/j.apcata.2014.07.036

A series of Ni-modified Y-zeolites with varying Ni loading in the presence of cupric salt as promoter were studied for acetylene hydrocarboxylation performed in a batch reactor. The catalysts were characterized by elemental analysis, H₂-TPR, XRD, NH₃-TPD, pyridine-FTIR, SEM, TG-DTG and Raman. It was found that the catalytic activity showed a pronounced dependence on the supports, metal introduction method, promoters and reaction conditions. The nickel species present as charge compensation cations in the zeolite framework constitute the active sites, and the acid sites help to promote the performance of carbonylation. Moreover, two types of coke were observed, and the remarkable reusability of NiY is attributable to the location of the coke outside the zeolite crystals. High catalytic performance was obtained over a NiY(7.0) catalyst with 62 g_{acrylic acid}/(g _cat. · h) of yield at 235 °C, 3.6 MPa of initial total pressure and 0.8 mM/l of cupric bromide within 40 min of reaction time. This is the most effective heterogeneous system for synthesizing acrylic acid by carbonylation of acetylene to date.

Full text of DOI:10.1016/j.apcata.2014.07.036

Accepted Manuscript
Title: Ni-exchanged Y-zeolite—An efficient heterogeneous
catalyst for acetylene hydrocarboxylation
Author: Tie Jun Lin Xuan Meng Li Shi
PII:
S0926-860X(14)00466-9
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2014.07.036
APCATA 14929
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
10-5-2014
14-7-2014
28-7-2014
Please cite this article as: T.J. Lin, X. Meng, L. Shi, Ni-exchanged Y-zeolitemdashAn
efficient heterogeneous catalyst for acetylene hydrocarboxylation, Applied Catalysis A,
General (2014), http://dx.doi.org/10.1016/j.apcata.2014.07.036
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Ni-exchanged Y-zeolite— An efficient heterogeneous catalyst for
acetylene hydrocarboxylation
Tie Jun Lin, Xuan Meng, and Li Shi∗
The State Key Laboratory of Chemical Engineering, East China University of Science and
Technology, Shanghai 200237, People's Republic of China
Highlights
ꢀ Production of acrylic acid by the first highly effective heterogeneous catalysts
ꢀ The role of metal and acidic sites were investigated
ꢀ Promoters such as copper salt can reduce the activation period of the reaction
ꢀ The catalytic activity depends on the Ni loading and the state of nickel in zeolite
ꢀ Coke formed during the reaction was mainly located outside the channel of
zeolite
ABSTRACT:
A series of Ni-modified Y-zeolites with varying Ni loading in the presence of
cupric salt as promoter were studied for acetylene hydrocarboxylation performed in a
batch reactor. The catalysts were characterized by elemental analysis, H2-TPR, XRD,
∗
To whom correspondence should be addressed. E-mail: yyshi@ecust.edu.cn Tel.:
021-64252274.
1
Page 1 of 43

NH3-TPD, pyridine-FTIR, SEM, TG-DTG and Raman. It was found that the catalytic
activity showed a pronounced dependence on the supports, metal introduction method,
promoters and reaction conditions. The nickel species present as charge compensation
cations in the zeolite framework constitute the active sites, and the acid sites help to
promote the performance of carbonylation. Moreover, two types of coke were
observed, and the remarkable reusability of NiY is attributable to the location of the
coke outside the zeolite crystals. High catalytic performance was obtained over a
NiY(7.0) catalyst with 62 gacrylic acid/(gcat.•h) of yield at 235℃,3.6MPa of initial total
pressure and 0.8mM/l of cupric bromide within 40 min of reaction time. This is the
most effective heterogeneous system for synthesizing acrylic acid by carbonylation of
acetylene to date.
Keywords: Carbonylation, Acrylic acid, Heterogeneous catalysts, Acetylene, Zeolite;
Coke characterization
1. Introduction
In order to reduce the dependence of bulk chemical commodity on natural mineral
oil and develop a diversified chemical industry, the coal and alkynes are now again
being seen as important alternate raw materials for chemical feedstock.[1,2] Alkynes
and, more specifically, acetylene that derived from coal, nature gas or shale gas could
be transformed into various base chemicals. [3-5] Among the approaches for these
chemicals, one of the most interesting routes is the Reppe carbonylation (Scheme1),
known as reaction of acetylene with carbon monoxide and compounds with
displaceable hydrogen atoms such as water, alcohols, thiols or aimnes et al, for
2
Page 2 of 43

yielding carboxylic monomers [5,7]
Acrylic acid is the primary building blocks of acrylate, polymers, plastics and
numerous other industrial and consumer products. The industrial process for
producing acrylic acid is based on acetylene carbonylation before the replacement of
this route by propylene oxidation.[8] Metal-containing compounds, such as
Ni(CO)4[7], NiX2-CuX2 (X=halogen)[9,10], Ni(OAc)2-CuBr2-CH3SO3H-PPh3[11], are
commonly used as active and selective homogeneous catalysts for the acetylene
hydrocarboxylation. All of these reported catalysts are dissolved in the solvent media
and thus cause the difficulties in the separation of catalyst and product. Recently, as
the significant improvement and progress in the design of catalyst and application of
new material and reaction technology, the carbonylation of acetylene has been seen as
an interesting topic and more recent works were focused on the development of
heterogeneous catalysts[12], which would provide the benefit of facile separation of
products.[5]
O
CuBr2/NiY
OH
H2O
CH CO
CH
T,P
Scheme 1. Hydrocarboxylation of acetylene with CO and water to yield acrylic acid.
The supported nickel catalysts are expected to be effective heterogeneous catalysts
for this reaction, and a series of experiments had been done by Bhattacharyya and
coworkers[13-16] using nickel salts supported on silica gel as catalyst. It should be
noted that, however, the maximum conversion of acetylene by nickel iodide-silica
gel catalyst ((Ni/SiO2= 70/30) is just 8.56%, namely that all of these heterogeneous
3
Page 3 of 43

catalyst investigated showed little active for the hydrocarboxylation of acetylene to
acrylic acid, and much room have been left for us to develop a highly effective solid
catalyst systems. It is reported that the both transition metal complex catalyst in
homogeneous system and transition metal cations embed in zeolites display catalytic
activities in various reactions such as carbonylation[17] and polymerization[18] of
acetylene, which suggests some analogies of them in catalytic behavior. Thus, it is
supposed that the nickel-modified zeolite would favor the carbonylation of acetylene
to acrylic acid.
In this context, it appeared very appealing to investigate the activity of nickel
exchanged Y zeolite in the Reppe carbonylation to acrylic acid. In order to understand
the role of nickel ions in the catalytic performance, a series of Ni ion-exchanged Y
zeolites with varying Ni weight loadings were synthesized and characterized. For
comparison, nickel impregnated on SiO2 and NaY were prepared. Our attention has
also focused on the influence of acid sites and reaction conditions on the
hydrocarboxylation of acetylene in a batch reactor mode. At last, the solid deposition
was also characterized by different techniques.
2. Experimental
2.1 Catalyst preparation.
The starting materials Na–Y zeolite (Si/Al = 3.26, Nankai University catalyst Co.,
Ltd.), was used as received and in powder form. The nickel-zeolite catalysts were
synthesized by two different methods for metal introduction. In the ion-exchange
procedure, 10g of NaY was exchanged with 200 mL of 5-200 mM Ni(NO3)2 solution.
4
Page 4 of 43

The exchange was performed at 363 k for 24 h in a round-bottomed flask. After ion
exchange, the exchanged zeolite was filtered, washed thoroughly using deionized
water, dried at 120℃ for 12h, and then calcined in a muffle furnace in static air
atmosphere. The resulting zeolite denoted as NiY(x), where x represents nickel mass
concentrations. In the incipient wetness impregnation (IWI) procedure, the parent
zeolite were outgassed at room temperature under vacuum for 1 h; and then a solution
with proper concentration of the Ni(NO3)2 was added followed by introducing the
impregnated samples into a rotary evaporator under vacuum at 60℃ for 30
min．Subsequently, the dried samples were calcined in air at 550℃ for 5 h. The
resulting zeolite was denoted as Ni/NaY(x), where x represents nickel mass
concentrations. The Ni/SiO2 with 7.0 wt% of Ni loading was prepared in exactly the
same way as the Ni/NaY except the replacement of NaY with silica gel, previously
calcined at 550℃. It should be noted that all of the calcined catalysts were not
reduced before the activity test.
2.2 Catalyst characterization
2.2.1 Chemical element analysis
The chemical compositions (Ni, Na, Si, Al) in prepared catalyst were determined
by inductively coupled plasma atomic emission spectroscopy (ICP-AES, Varian
710ES). The samples were previously dissolved by acid digestion. The content of
nickel leached into the liquid was also measured by elemental analysis.
2.2.2 X-ray diffraction
The crystal structure of catalyst was characterized by X-ray diffractometer (XRD,
5
Page 5 of 43

Bruker D8, Germany) equipped with an atmosphere and temperature control stage and
using Cu Kα radiation (λ=1.5406 Å) operated at 40 kV and 100 mA. The scanning
range from 10 to 80° (2θ), 0.02 step, and 1 s/step.
2.2.3 Temperature-programmed Reduction (TPR)
The TPR experiments were performed on a Micromeritics AutoChem II 2920.
About 50 mg of catalyst was loaded into a quartz U-tube and dried in Ar atmosphere
at 373 K for 1 h before the reduction. The catalyst was then heated in 5% H2 diluted
in argon at flow rate of 30 ml/min with a heating rate of 10 K/min from
room-temperature to 1073 K, the consumption of H2 was monitored using a thermal
conductivity detector.
2.2.4 Acidity characterization
The acid properties of samples were measured by ammonia temperature-
programmed desorption (NH3-TPD, Micrometrics 2910) and Fourier transform
infrared( FT-IR) spectroscopy (Nicolet company, Model Magna-IR 550). For the
NH3-TPD, The samples was pre-treated in the stream of helium (50ml/min) at 450 for
2h and afterwards cooled to 180 ℃ and then saturated under an ammonia stream for
30 min. The physically adsorbed ammonia was removed by flowing He with 50ml/
min at 180℃ for 90min. Subsequently, the sample was heated to 550℃ at 10℃/min
in flowing He to remove the chemically adsorbed ammonia. The amount of desorbed
ammonia was continuously monitored with a thermal conductivity detector (TCD).
For the pyridine-FTIR, the characteristic absorption peak of Brønsted acid is located
at1540 cm−1, and that of Lewis acid is located at1450 cm−1. The detailed information
6
Page 6 of 43

about the identification and determination of acid amount, acid density and acid
variety can be found elsewhere. [19]
2.2.5Thermal analysis
The fresh and recovered catalysts were measured by thermo-gravimetric analysis
(TGA). The sample were loaded on a platinum microcrucible and were heated up to
800℃ at 10 ℃/min with flowing air at100 ml/min, recording continuously the
weight of each sample.
2.2.6 SEM and Raman spectra
The morphologies of recovered catalysts were obtained using a Nova Nano SEM450.
A Micro-Raman System (HORIBA Jobin Yvon) using He-Ne laser beam with a
wavelength of 632nm was used to analyzing the nature of coke in the spent catalyst.
The quality was identified by Raman spectroscopy using the 514nm line of an argon
laser operated at a laser power of 50 MW.
2.3 Hydrocarboxylation procedure
The catalytic hydrocarboxylation of acetylene was performed in a 0.5L Parr
autoclave made of stainless steel-316 having facilities for gas inlet and outlet, a
rupture disk as a safety measure in case of excessive pressure buildup, intermediate
sampling, temperature-controlled heating, and variable agitation speed.
In a typical experiment, known quantities of catalyst powders (0.5g), promoters
(0.08mM/l), H2O(15ml) , and acetone (150ml) were charged into a stirred pressure
reactor. The reactor was firstly flushed with nitrogen for several times and
subsequently pressurized with acetylene to 0.5MPa and then to 3.6 MPa of initial total
7
Page 7 of 43

pressure with CO (both controlled by mass flowmeter) at room temperature. The
reactor was then heated to235℃ and the reaction proceeded until no pressure drop
was observed under constant agitation (800 rpm). After the reaction, the autoclave
was quickly cooled to room temperature and the tail gas and liquid were collected,
calculated and analyzed. For recycling experiments, the catalyst used in the previous
run was separated by filtration and washed with acetone for several times and reused
after drying under vacuum. The detailed schematic drawing of the reactor is shown in
Fig. S1 (supplemental materials).
The conversion of acetylene，selectivity and yield of acrylic acid and the space time
yield( STY) are defined as follows:
n
n
p
n - n
p
0
1
Yield =
×100% ，
×100%
，
Conversion=
×100%
，Selectivity=
n
n
n - n
0
0
0
1
g
g
AA
AA
STY =
×100% or STY =
×100%
g
• h
molNi • h
cat.
Where, n0, amount of acetylene for feedstock before reaction; n1, amount of acetylene
for residue after reaction; np, amount of acetylene for forming acrylic acid; gAA, the
amount of acrylic acid produced during reaction; gcat., the amount of catalysts for
input before reaction; molNi, the molar content of nickel contained in the catalysts for
input before reaction.
3. Results and discussion
3.1 Characterization of Catalyst
The chemical elements of Na-Y and Ni-exchanged Y zeolite were determined by
element analysis. The Ni loading, degree of site exchange, and the ratio of total
8
Page 8 of 43

silicon to aluminum are present in Table 1. Fig. S2(supplemental materials) shows the
effect of Ni(NO3)2 concentration in the exchange aqueous solution on the Ni loading
in NiY. As can be seen, the Ni loading nearly increases linearly with an increase in
Ni(NO3)2 concentration from 5mM to 60mM, and almost remains unchanged with a
further increase in the Ni(NO3)2 concentration, which suggests that the Ni loading can
be controlled by varying the Ni(NO3)2 concentration in the aqueous solution. The
maximum exchange degree was determined to approximately 62% and the
corresponding Ni loading was 7.0 wt% in this study. It is reported that the metal
cations introduced to the zeolite by ion-exchange process are in the form of a skeleton
balance-charge, namely that one Ni2+ cation exchanged into the zeolite charges
compensate two change sites, and thus ratio of n(2Ni2++ Na+)/n(Al) should be a
constant value (≈1.0)[19], which is consistent with the results showed in Table 1.
Moreover, the observation that the increase in n(Si)/n(Al) ratio from 3.26 to 3.28
indicates the ion-exchange process caused slightly collapse of the molecular sieve
structure and dealumination of the framework.
X-ray diffraction technology was used to characterize the physicochemical
properties of the NaY, Ni/NaY and Ni-exchanged Y zeolite with various Ni loadings,
the result is shown in Fig. 1. There is no significant structural modification after
nickel ion exchange compared to the parent NaY. However, the X-ray diffraction
peaks of the NiY samples are less intense, indicating the slightly decrease in the
crystallinities of zeolite during the exchange process and subsequent calcination. The
XRD patterns only show characteristic diffraction peaks of NaY, and none of nickel
9
Page 9 of 43

oxide or other nickel species is detected. All of these facts suggest that the nickel ions
exchanged into the zeolite samples resided as charge-compensating and the nickel
ions were highly dispersed in exchanged zeolite. While for the Ni/NaY prepared by
incipient wetness impregnation (IWI), the characteristic peaks of NiO at 2θ of 37.25°,
43.28°and 31.44° (JCPDS47-1049) were clearly observed, indicating that the nickel
ions mainly present as in the form of bulk NiO on the impregnated catalysts.
The TPR spectrum of Ni/NaY prepared by IWI and Ni-exchanged Y zeolite are
shown in Fig.2. For NiY catalyst after calcination, the spectrum exhibits only one
high temperature strong reduction at approximately 1043K, which is attributable to
the reduction of charge-compensating Ni2+ cations. The similar result was observed
for Ni-exchanged other zeolites.[21,22]While for the Ni/NaY sample, one weak
reduction peak at high temperature zone (756 K ) and three reduction peaks
composing of one strong reduction peak at about 450K and two weak shoulder weak
at appropriately 823K and 1043K, respectively, are observed. According to the
literatures[23], the former peak is attributable to the reduction of bulk NiO particles
located mainly outside the pores of zeolite support[22], the latter two reduction peaks
may correspond to the presence of smaller NiO particles located inside the channels
of zeolite, which are less reducible caused by a stronger metal-support interaction
than the bulk NiO particles.[24]The reduction peak at 756 K is analogous to that
observed in ion-exchanged NiY samples and is attributed to the reduction of nickel
located inside the framework of zeolites. These facts clearly indicates that all of the
Ni cations in NiY are distributed exclusively as charge-compensation cations in the
10
Page 10 of 43

zeolite framework, as shown in Fig. 4, while most of Ni ions are dispersed on the
outer surface and presented as the form of NiO in the impregnated catalyst after
calcination, and fewer nickel ions were introduced into the zeolite framework as
compensation cations during the IWI process[25]．These nickel species distributed as
compensation cations are difficult to reduce and require higher reduce temperature
than that of nickel oxide particles.[26]
Ammonia TPD profiles of parent NaY, NiY, Ni-NH4Y and Ni-HY are plotted in
Fig.4. Fig. S3 (Supplementary Material) shows the Pyridine-FTIR spectra of the NaY
and NiY at 200℃ and 450℃, and the quantification results of their acid properties
characterized by Pyridine-FTIR are listed in Table 2.It is evident that the acidic
properties of zeolite changed during the ion exchange of Na+ with Ni2+. For the parent
NaY, only one large ammonia desorption band between 50~200℃ with a maximum
intensity at about 125℃ is observed, which is assigned to the weak acid sites and
further confirmed to be weak Lewis sites by pyridine-FTIR. The ammonia desorption
at about 300 and 425℃,respectively, indicates the occurrence of medium and strong
acid sites in sample after Na ions are substituted by nickel ions. From the results of
pyridine-FTIR, the total number of acid sites and, specifically, Lewis acid increases
significantly, which is agreement with the point proposed by Minchev et al[27] that
the nickel cations play the role of Lewis acid centers. Moreover, the additional
Brønsted acid sites would be formed and each Ni2+ produces two acid sites after Ni2+
ion exchange of Na-zeolite. [28-30] The acidity of Ni-NH4Y and Ni-HY is measured
by NH3-TPD to explain the role of acid sites on the hydrocarboxylation of acetylene,
11
Page 11 of 43

and this aspect will be discussed in the next section.
3.2 Catalytic activity
3.2.1 Role of nickel sites
It had been disclosed that[9]the production of acrylic acid by bring acetylene and
carbon monoxide into contact with water in the presence of compounds of metals of
VIII Group, i.e. iron, nickel and cobalt can be carried out at elevated pressures and
temperatures. To stand out the role of metals in the acetylene hydrocarboxylation,
various catalysts including FeY, CoY, NaY, HY, Ni/SiO2, Ni/NaY(imp.,3.1),
Ni/NaY(imp.,7.0) , NiY(7.0) ,Ni-NH4Y and Ni-HY were tested. As can be seen from
the Table 3, all of Ni free samples were completely inactive for acetylene
hydrocarboxylation within 2 hours, whereas the Ni-containing catalysts showed
varied degree of activity under the identical conditions, suggesting that the nickel sites
play an indispensable role in the activity of acetylene hydrocarboxylation. As concern
the nickel-base catalysts, it was found that the supports and metal introduction
methodology have
a great influence on activity. The reaction rate of
NiY(ion-exchange, 7.0) was four times of that of Ni/NaY(imp.7.0). These differences
may be attributable to the different state of nickel species in the catalysts. For the
catalysts prepared by impregnation, the bulk NiO species existed on the zeolite are
considered to be low activity for the carbonylation of acetylene. Interestingly, Huang
et al[31-33] hold that the NiO are the active species in hydroesterification of acetylene
with CO and methyl formate to yield methyl acrylate. Further comparison between
the Ni/NaY and Ni/SiO2 (both prepared by incipient wetness impregnation), the higher
12
Page 12 of 43

activity of the former is because of some nickel ions distributed as charge-compensate
in the framework of zeolite (verified by TRR result in Fig. 2) during the preparation
process using NaY as support, and those parts of Ni2+ work more in the reaction,
while only NiO species were detected in Ni/SiO2 sample. What is noteworthy is that
the Ni/NaY with 7.0 wt% Ni loading shows lower activity than that with 3.0 wt% Ni
loading after calcination, which indicates that the formation of nickel oxide become
favored when raising the nickel amount via impregnation, the similar phenomena
were observed and confirmed by Maia et al.[23]
The high activity of NiY prepared by ion-exchange among the investigated
catalysts further demonstrates the dominant role of Ni species distributed as
compensation cations in the reaction. In order to further investigate the effect of Ni
loading on the activity, the NiY with different degrees of exchange were compared.
The tested reaction is terminated when no pressure drop was observed, and the
reaction time for each run was 120 min, 90 min, 64 min, 44 min and 40 min,
respectively. As shown in Fig. 4, the yield and space time yield (STY) increase with
the increase in Ni loading. For the NiY with Ni loading exceeds 3.1wt%
(corresponding to 28% of exchange degree), no obvious changes were observed in
yield, but significant increase in the STY, indicating that the samples with lower Ni
loading require more time to acquire the same targeted product yield. On one hand,
there are no enough active molecules that can coordinate with reactants for these
catalysts with lower nickel loadings; on the other hand, the possible migration of
nickel ions from inaccessible hexagonal prism and sodalite cages to zeolite supercage
13
Page 13 of 43

within the zeolite during the reaction would also account for this. In regard to these
NiY catalysts with the degree of cation exchange less than 40% (NiY(0.8), NiY(3.1)),
the Ni cation resides mainly inside the hexagonal prisms and sodalite cages, or SI and
SII sites of the faujasite zeolite, respectively, these Ni2+ cations are not active for the
reaction due to their inaccessiblility to the reactant.[22,34] Under the reaction
conditions, the Ni cations will migrate into the supercage from SI sites and work with
reactants to form active sites. For these samples with higher exchange degree, the Ni
cations would locate directly in the zeolite supercage, and thus require less time to
activate the reaction. The pressure-time plot (Fig S4, supporting materials) for these
catalysts with different nickel loadings agreed well with our hypothesis.
3.2.2 Effect of reaction temperature and initial total pressure
Fig.5(a) presents the effect of reaction temperature on the hydrocarboxylation of
acetylene using NiY(7.0) as catalyst. The run was carried out in acetone, at 3.6MPa of
initial total pressure with reaction temperature selected at 200℃，220℃，235℃ and
250℃, respectively. No acrylic acid was detected at a temperature of about 200℃
after 2h of reaction time . As the temperature increased to 220℃, the yield of acrylic
acid enhanced drastically and the yield goes through a maximum at about 235℃(40
min of reaction time). With further increase of temperature to 250℃,however, the
pronounced fall in space time yield(STY) was observed. After separation by filtration,
obvious coke deposition and polymer were observed, which suggests that the
occurrence of polymerization and decomposition of acetylene and/or acrylic acid. We
note that the similar phenomena were observed by Bhattacharyya,[14] and these
14
Page 14 of 43

sediments would cause serious problems, thus a proper temperature of about 235℃
was selected in this reaction.
The effect of initial total pressure on the activity in the acetylene
hydrocarboxylation was investigated under the same moral ration of CO/C2H2, and
the result is shown in Fig. 5(b). As expected, the activity increased drastically by
increasing the initial total pressure from 2.5 to 3.6 MPa. The higher pressure will
increase the solubility of gas in the liquid, and favor its contact and coordination with
metal ions in the zeolite catalyst and to form active species. But on further increase in
the initial pressure will cause the decomposition of acetylene and bring the
operational risk and cost, so the further experiments will be performed at 3.6 MPa of
initial total pressure.
3.2.3 Role of copper salt
The catalytic properties of NiY(7.0) with a variety of copper salt as promoter were
evaluated for the hydrocarboxylation of acetylene under the same conditions at 235
℃，3.6MPa of initial total pressure and acetone as solvent. As can be seen from the
Fig. 6(a), different copper salts displayed varied catalytic activity. Among these
promoters investigated, the cupric bromide gave the best activity, while the
substitution of bromide with acetate or nitrate proved counterproductive and led to
decreased production of acrylic acid and increased production of unwanted polymer.
Cupric halide(CuX2) and cuprous halides(CuX) were all found to be effective in
improving the carbonylation, but gave the different activities depend on the halogen
ions. It has been reported that the cupric ion plays an important role in activating the
15
Page 15 of 43

acetylene[11,20]. Thus, the differences between these tested copper salts could be
attributed to their different coordinative behaviors with acetylene. Some copper salts
tend to cause the formation of polymer and that would be trapped near the entrance of
the channels or cages and therefor result in the problems of mass transfer. Fig. 6(a)
also indicates that bromide ion plays a special role in the forming intermediates. Exact
nature of the influence of bromide ion is unknown. It may be speculated that the
bromide ion serves as a bridge between cupric ion and acetylene and thus increasing
the kinetics of electron transfer and stabilizing the copper-acetylide species and
favoring its attack and insertion to Ni2+-(CO)n intermediates species. Moreover, the
cupric bromide is deemed to improve the solubility of carbon monoxide[11] and thus
increasing the diffusion rate of gas into to liquid-phase. Note that the improvement of
reaction rate by introducing the copper salts into the catalyst system was also
observed in other types of reaction.[35,36]
In order to further explain the role of cupric bromide, the effect of its
concentration was studied and the result was shown in Fig. 6(b). As can be seen, a
marked increase in the conversion of acetylene is observed with increase in cupric
bromide concentration from 0 to 0.8 mmol/l(mM/l), and the conversion increases
slightly with further increase in the concentration. The increase of selectivity to
acrylic acid form 58% to 81% with the increase of concentration of CuBr2 indicates
that the copper plays a positive role in the carbonylation of acetylene. However, the
selectivity decreases drastically when the concentration of CuBr2 exceeds 0.8mM/l. In
consideration of the change trend of conversion and additional byproducts detected in
16
Page 16 of 43

gas chromatography and precipitate in the products, it can be inferred that the excess
of cupric salt resulted in the decomposition of acetylene and the formation of polymer.
That is, the cupric salt plays a double role in the carbonylation process, and the
suitable concentration should be controlled.
3.2.4 Influence of acid sites
It is of importance to clarify the role of acid sites in the hydrocarboxylation of
acetylene. Most of the previous studies of acetylene reaction are carried out on acidic
zeolites[37,38], and it is believed that the acetylene is easy to interact with acid sites
forming hydrogen-bound species[39]. Thus, it is necessary to answer the question
about whether the presence of acid sites is required in this reaction. For this purpose, a
special experiment run was carried out in the presence of Ni-NH4-Y catalyst, which is
the non-calcined Ni-HYsample. The NH3-TPD profiles of Ni-NH4Y and Ni-HY are
shown in the Fig. 3. Compared to NH4-form catalyst, the H-form sample shows a
higher density of weak, medium and strong acid sites. Table 3(entry 8 and entry 9)
gives the comparative results of catalytic performance of Ni-NH4Y and Ni-HY. It is
evident that the Ni-NH4Y shows relative lower catalytic activity, while the Ni-HY
gives a better result. It is noteworthy that the coke deposition in the recovered Ni-HY
is a little more, indicating that the higher density of acid sites are more likely to cause
the decomposition of acetylene. In addition, a Ni-SiO2 (the Ni is present solely as
isolated Ni2+ after impregnation and calcination rather than bulk NiO in the catalyst)
without acid sites were prepared according to the literature[22], and it is found that
the catalytic activity towards acrylic acid is very low. These findings suggest that the
17
Page 17 of 43

acid sites have a certain effect on the catalytic performances, and the moderate density
of acid sites helps to promote the carbonylation. A possible explanation is that the acid
sites would interact with acetylene and promote the protonation of acetylene to form
active intermediates.
3.2.5 Stability of catalyst
Reusability is crucial for prospective uses in a continuous tubular reactor and
further applications in industrial scale. No information is available regarding the
stability and recyclability of zeolite catalysts in liquid-phase hydrocarboxylation of
acetylene and herein we firstly report the results for the most promising NiY(7.0)
catalyst.
Table 4 gives the reusability test of NiY catalyst for several repeated run. It is
surprising that the catalytic activity remained stable over four consecutive runs. No
obvious loss in the conversion and selectivity towards acrylic acid was observed,
notwithstanding the catalyst powder, which was originally while, turned black
because of coke formation and accumulation after the fourth catalytic run.
Interestingly, upon removal of the deposits throng calcination at 823 K for 5h, the
catalytic performance of heat-treated NiY(7.0) catalysts was identical to that of
pristine catalysts, suggesting that a great part of the coke is located on the external
surface of the zeolite catalyst. Moreover, a very small amount of Ni ions (less than
0.3ppm) in the reaction liquid detected by ICP-AES indicates the stability of nickel
ions exchanged into the framework of zeolite.(seeing Fig S5 in the supplementary
materials)
18
Page 18 of 43

In order to study the state of catalysts after reaction, the recovered catalysts were
characterized by XRD and TPR. Fig. 7 gives the X-ray diffraction patterns of the
fresh and recovered NiY catalyst. No obvious peaks changing in XRD and TPR (not
shown here) profile were observed between them, indicating that the state of nickel
remains nearly unchanged during the reaction. Although the types of peak in XRD
patterns remain unchanged, the slightly decrease in intensity of peak for the recovered
catalyst may be caused by the slightly decrease in the crystallinities of zeolite and
collapse in zeolite structure during the reaction or by the coke or some polymer
deposited on external surface of zeolites.
3.3 Characterization of solid deposition
Coke deposition and polymer are the most common and serious problem in the
hydrocarboxylation of acetylene under elevated temperature and pressure. In the
preceding study[20,40], the formation of solid deposition in the homogeneous system
was detailed studied, and the amount of sediments strongly depend on the catalysts
used and reaction conditions. The deposition, mainly referred to coke, in zeolites is
formed mainly due to the decomposition and/or polymerization of acetylene. The
coke may be located both on the external surface and inside the channels of the
zeolites. In this section, the fresh and recovered catalysts were characterized by a
number of techniques, such as TG-DTG, SEM and Raman, in order to gain
information about the nature of the coke.
The TG and DTG results from the temperature-programmed oxidation of the tested
samples are shown in Fig.8. As can be seen from Fig.8(B), a large peak ranges of
19
Page 19 of 43

50~200℃ in all of the tested samples is attribute to the moisture vaporization peak,
and none of other peaks can be found in the fresh NiY and parent NaY sample. The
amount of coke can be roughly calculated according to the weight loss of sample
between 200℃ to 700℃.It is apparent that the amount of coke in the recovered
Ni/NaY is larger than that of NiY sample calculated based on the TG analysis shown
in Fig.8(A). In the case of NiY, the amount of coke is about 12% referred to the used
zeolite sample, which is higher than the maximum amount of that could be
accumulated inside the NiY zeolite micropores. Considering the observation of a
small amount of black particles in the recovered catalyst and remarkable reusability of
the catalyst, it can be inferred to that most of this coke is deposited over the external
surface of zeolite, and the active site remain accessible for the reactant molecules.
Moreover, based on the previous study[20], the decomposition of acetylene is mainly
caused by the copper salt. For those activated acetylene outside the zeolite channel, it
is easy to decompose, but if they transfer to the zeolite micropores and function with
the metal site, the acetylene becomes more stable. Namely, the reaction towards the
carbonylation is superior to decomposition for the acetylene coordinated with the Ni
site inside the zeolite supercage.
The profile of derivative of the TG weight loss as a function of temperature shown
in Fig.8(B) also provides information about the location and nature of the coke
deposits. From the DTG profile, two exothermic peaks were observed at the
temperature ranges from 200 to 500℃ for the catalyst prepared by incipient wetness
impregnation and ion-exchange. No obvious oxidation peak was found at temperature
20
Page 20 of 43

exceeds 550℃. All of these facts suggest that there are two types of deposition exist
in the spent catalyst. The first type of coke, soft coke or polymeric coke, can be
removed at a relative lower temperature, while the remove of the second coke
(refractory coke) need higher temperature. [41-42] In terms of the location of coke,
the oxide peak at around 300℃may correspond to the burn out of coke deposited on
the external surface of the zeolite, and the relatively broader peak indicates that there
is little coke deposited within the zeolite micropores.
SEM results of fresh and recovered catalyst are shown in Fig. 9. It is evident that
the zeolite retains its microstructure as octahedral or hexagonal prism sheet structure
before and after reaction, but the location, morphology and nature of coke show some
differences. In the case of the recovered NiY samples, the coke deposited on the
surface of the zeolite is assigned to filamentous coke, which is in the form of carbon
nanotubes with filamentous lines structures. Most of those nanotubes are separated
from the zeolite crystals and tend to aggregate to form clusters (Fig. 9(b)). The feature
of the coke location in the zeolite explains why the NiY catalysts keep a remarkable
catalytic activity after several times of consecutive catalytic run. But for the recovered
Ni/NaY(imp.) samples( Fig. 9(c,d)), most of the zeolite grain are covered by a layer of
tiny particles. Moreover, the filamentous coke is also found in this sample. These facts
imply that the different formation mechanism of coke between NiY and Ni/NaY(imp.)
catalyst. The results of SEM analysis of these samples are line with the findings in the
TG-DTG analysis, and are agreement with the above results of reusability test of
catalyst.
21
Page 21 of 43

Additional information about the nature of coke in the recovered NiY catalyst can
be obtained from the Raman characterization. As shown in the Fig. 10, there are two
large bands can be observed at 1360 and 1596 cm-1 in the spectra of the recovered
catalysts, which are known as D-band and G-band, respectively. The feature of the
coke characterized by Raman is the same as that observed in the homogeneous system.
[20]The G-band provides information about the electronic properties of the
filamentous carbon and is used to characterize the ordered carbon, while the D-band is
usually used to characterize the polycrystalline and imperfect graphite and other types
of carbon. The Raman spectra also indicates that the nanotubes observed by SEM
present multiwalled morphology.[43] In a word, the Raman profiles clearly indicates
that the coke formed on the NiY present as quite different nature, as above concluded
from TG-DTG results.
4. Conclusion
The Ni-modified Y-style zeolites were examined as a case of highly efficient
heterogeneous catalyst for the hydrocarboxylation of acetylene with carbon monoxide
and water to yield acrylic acid.
A careful comparison of catalysts prepared by different supports and nickel
introduction methodologies (incipient wetness impregnationand aqueous ion
exchange) reveals that the Ni-exchanged Y zeolites with the presence of cupric salt as
promoter give the best catalytic activity(62gacrylic acid/(gcat.•h)). The multiple reuse of
NiY with stable activity in a batch reactor was demonstrated at 235℃ and 3.6MPa of
initial total pressure in a short period of reaction time.
22
Page 22 of 43

The catalytic activity shows a pronounced dependence of the reaction conditions.
The reaction velocity increases with the initial total pressure increases from 2.5 to 3.6
MPa. The temperature exceeds 250℃ would cause the obvious decomposition of
acetylene and thus the decrease in the selectivity to acrylic acid, and that under 220℃
would bring about the lower reaction rate. The types of copper salt influence the yield
of target production, and the activity goes through a maximum using cupric bromide
as promoter.
Elemental analysis, XRD and TPR characterization indicates the nickel introduced
into the zeolite framework presents as charge-compensation, and remains the state
throughout the reaction. The loading of Ni in the zeolite can be controlled by
controlling the concentration of nickel nitrate in the aqueous exchange solution. It is
found that the bulk NiO is detrimental to the reaction, while the Ni2+ ions in a high
coordinative unsaturation in the zeolite are proposed to be catalytic sites in acetylene
hydrocarboxylation. Pyridine-FTIR and NH3-TPD suggest that the acidity of NiY
increases remarkable after ion-exchange compared with parent NaY, and the moderate
acidity favors the protonation of acetylene.
Characterization of recovered catalysts by TG-DTG, SEM and Raman indicate
that the different coke amount and location between NiY and Ni/NaY(imp.) catalysts.
The coke deposition on NiY presents a relative ordered filamentous nature, and much
of them located mainly outside the zeolite crystals and thus keeps a remarkable
reusability in spite of the relative higher amount of coke deposited after many times of
recovery.
23
Page 23 of 43

Acknowledgement
We are grateful for the financial support from China Postdoctoral Science
Foundation (No: 2013M541488) and Opening Project of State Key Laboratory of
Chemical Engineering of East China University of Science and Technology (No.
SKL-ChE-11C04).
Appendix A. Supplementary data
Supplementary data associated with this article can be found in the online version.
References
[1] A.Corma, Chem. Rev. 114(2014) 1545−1546.
[2] Nexant Incorporation, Multiclient Prospectus-Chemicals from acetylene, back to
the future? http://www.nexant.com/, 2007.
[3] R.Chinchilla, C.Nájera, Chem. Rev. 114(2014) 1783−1826.
[4] H. Schobert, Chem.Rev. 114(2014) 1743−1760
̧ ̈th, Chem. Rev. 114(2014) 1761− 1782.
[5] I.T. Trotus, T. Zimmermann, F. Schu
[6] G. Kiss, Chem. Rev. 101(2001) 3435− 3456.
[7] W. Reppe, Acetylene Chemistry, Meyer, New York, 1949.
[8] R.J. Tedeschi, Acetylene-Based Chemicals from Coal and Other Natural
Resources, Marcel Dekker, New York, 1982.
[9] W. Repee, R. Stadler, US Patent 3023237(1962).
[10] K. Weissermel, H.J. Arpe, Industrial Organic Chemistry, Wiley-VCH,
Weinheim, Germany,1993, pp. 272.
[11] C.M. Tang, Y. Zeng, P. Cao, X.G. Yang, G.Y. Wang, Catal.Lett. 129(2009)
24
Page 24 of 43

189-193.
[12] C.M. Tang, P. Cao, Y.Zeng, X.G. Yang, G.Y. Wang, Petrochemical Technology,
37(2008) 1089-1094.
[13] S.K. Bhattacharyya, A.K. Sen, Journal of Chemical Technology and
Biotechnology 13(1963) 498–505.
[14] S.K. Bhattacharyya, A.K. Sen, Ind. Eng. Chem. Proc. Des. Dev 3(1964)
169–176.
[15] S.K. Bhattacharyya, D.P. Bhattacharyya, J. Appl. Chem. 16(1966) 18–21.
[16] S.K. Bhattacharyya, D.P. Bhattacharyya, J. Appl. Chem. 16(1966) 202–205.
[17] S.L. Phillip, N.J. Woodbury, US Patent 3334132 (1967).
[18] P. Pierre, C.V. Jacques, G. Pierre, I. Boris, J. Catal. 32(1974) 190–203.
[19] D.Z. Yi, H. Huang, X. Meng, L. Shi, Appl. Catal. B: Environm. 148-149(2014)
377–386.
[20] T.J. Lin, X. Meng, L. Shi, Ind. Eng. Chem. Res. 52(2013) 14125−14132.
[21] V. Hulea, F. Fajula, J. Catal. 225(2004) 213–222.
[22] A.N. Mlinar, G.B. Baur, G.G. Bong, A. Getsoian, A.T. Bell, J. Catal. 296(2012)
156–164.
[23] A.J. Maia, B. Louis, Y.L. Lam, M.M. Pereira, J. Catal. 269(2010) 103–109.
[24] C. Wu, L. Wang, P.T. Williamsa, J. Shi, J. Huang, Appl. Catal. B: Environm.
108–109(2011) 6–13.
[25] A. Carrero, J.A. Calles, A.J. Vizcaíno, Appl. Catal. A: Gen. 327(2007) 82–94.
[26] B. Pawelec, R. Mariscal, R.M. Navarro, J.M. Campos-Martin, J.L.G. Fierro,
25
Page 25 of 43

Appl. Catal. A: Gen. 262(2004) 155–166.
[27] C. Minchev, S.A. Zubkov, V. Valtchev, V. Minkov, N. Micheva, V. Kanazirev,
Appl. Catal. A: Gen. 119(1994) 195–204.
[28] B.I. Mosqueda-Jiménez, A. Jentys, K. Seshan, J.A. Lercher, Appl. Catal. B:
Environm. 46(2003) 189–202.
[29] B.I. Mosqueda-Jiménez, A. Jentys, K. Seshan, J.A. Lercher, J. Catal. 218(2003)
375–386.
[30] M.O. Lallemand, A. Rusu, E. Dumitriu, A. Finiels, F. Fajula, V. Hulea, Appl.
Catal. A: Gen. 338(2008) 37–43.
[31] X.H. Huang, X.G.Yang, J.Q. Zhang, Z.T. Liu, Chinese Chem. Lett. 11(2000)
521–524.
[32] X.H. Huang, X.G. Yang, J.Q. Zhang, Z.T. Liu, J. Nat. Gas. Chem. 10(2001)
42–50.
[33] X.H. Huang, Z.T. Liu, J.Q. Zhang, X.G. Yang, J.W. Wang, J. Nat. Gas. Chem.
9(2000) 197–204.
[34] P. Gallezot. B. Imelik, J. Phys. Chem. 77(1973) 652–656.
[35] B.C. Dunn, C. Guenneau, S.A. Hilton, J. Pahnke, E.M. Eyring, Energy& Fuels
16(2002) 177-181.
[36] J.L. Hu, Y.L. Gu, Z.H. Guan, J.J. Li, W.L. Mo,; T. Li, G.X. Li, ChemSusChem
4(2011) 1767-1772.
[37] C. Pereira, G.T. Kokotailo, R. J. Gorte, J. Phys. Chem. 95(1991) 705–709.
[38] P. Tsai, J.R. Anderson, J. Catal. 80(1983) 207–214.
26
Page 26 of 43

[39] G. Onyestyák, L.V.C. Rees, Appl. Catal. A: Gen. 223(2002) 57–64.
[40] J.M. Qiu, Y.Z. An, D.H. Yang, L.R. Feng, Nat. Gas Chem. Ind. 19(1994) 10–14.
[41] J.J. Guo, H. Lou, X.M. Zheng, Carbon 45(2007) 1314–1321.
[42] X.J. Zhang, Y. Wang, F. Xin, Appl. Catal. A: Gen. 307 (2006) 222–230.
[43] A.M. Rao, E. Richter, Shunji Bandow, Bruce Chase, P.C. Eklund, K.A. Williams,
S. Fang, K.R. Subbaswamy, M. Menon, A. Thess, R.E. Smalley, G. Dresselhaus,
M. S. Dresselhaus, Science 275(1997) 187–191.
27
Page 27 of 43

Table 1 Elemental analysis of Na-Y and Ni-Na-Y exchanged with different Ni(NO3)2
concentration
Elemental analysis
(wt%)
Sample
Si/Al
Cations/Ala
% Exchangeb
Ni
Na
NaY
0.0
0.8
3.1
6.8
7.0
8.97
8.3
3.26
3.26
3.27
3.28
3.28
0.99
0.99
0.98
0.99
0.99
0
NiY（5mM)c
NiY（20mM)
NiY（60mM)
NiY（200mM)
7
6.44
3.60
3.45
28
60
62
a Calculated assuming that every positive charge from the exchanged cations can
charge compensate an exchange site ((Na+2Ni2+)/Al);
b Determined as the amount of Na+ replaced by the exchanged cation Ni.
c NiY(x), x represents the concentration of Ni(NO3)2 in the exchange aqueous solution
28
Page 28 of 43

Table 2 Acidic properties of NaY and NiY（7.0）calculated according to the result of Pyridine-FTIR spectra ×10−4
mol g−1).
Sample
Na‐Y
T
TL
TB
0
SB
0
SL
0
WB
0
WL
1.95
20.5
1.95
17.46
1.95
16.02
3.04
0
1.44
3.04
Ni‐Y（7.0）
T—Total acid; TL—total Lewis acid; TB—total Brønsted acid; SL—strong Lewis acid; SB—strong Brønsted acid;
WL—weak Lewis acid; WB—weak Brønsted acid.
29
Page 29 of 43