Effect of calcination temperature on CrOx-Y2O3 catalysts for fluorination of 2-chloro-1,1,1-trifluoroethane to 1,1,1,2-tetrafluoroethane

Article abstract of DOI:10.1016/j.jcat.2007.11.005

A series of CrO_x-Y₂O₃ catalysts were prepared by a deposition-precipitation method and tested for the fluorination of 2-chloro-1,1,1-trifluoroethane (CF₃CH₂Cl) to synthesize 1,1,1,2-tetrafluoroethane (CF₃CH₂F). The highest activity was obtained on a pre-fluorinated catalyst calcined at 400 °C, with 19% of CF₃CH₂Cl conversion at 320 °C. The effect of the calcination temperature on the CrO_x species was investigated. X-ray diffraction and Raman results indicated that the CrO_x species (Cr(VI)) were well dispersed on the catalyst surface when the catalyst was calcined at 400 °C. With increasing calcination temperature, most of the CrO_x species changed from high oxidation state Cr(VI) to low oxidation state Cr(V) or Cr(III) species, which resulted in difficulty in pre-fluorination of the catalyst. It was also found that the CrF_x, CrO_xF_y or Cr(OH)_xF_y phases originated from high oxidation state Cr(VI) species were the active sites for the fluorination reaction.

Full text of DOI:10.1016/j.jcat.2007.11.005

Journal of Catalysis 253 (2008) 1–10
www.elsevier.com/locate/jcat
Effect of calcination temperature on CrO –Y O catalysts for fluorination
x
2 3
of 2-chloro-1,1,1-trifluoroethane to 1,1,1,2-tetrafluoroethane
a
a
a
a
b
a
Jun He , Guan-Qun Xie , Ji-Qing Lu , Lin Qian , Xue-Liang Zhang , Ping Fang ,
a
a,∗
Zhi-Ying Pu , Meng-Fei Luo
a
Zhejiang Key Laboratory for Reactive Chemistry on Solid Surfaces, Institute of Physical Chemistry, Zhejiang Normal University, Jinhua 321004, China
b
Zhejiang Qu Hua Fluor-Chemistry Company Limited, Quzhou 324004, China
Received 14 August 2007; revised 28 September 2007; accepted 5 November 2007
Abstract
A series of CrOx–Y O catalysts were prepared by a deposition–precipitation method and tested for the fluorination of 2-chloro-1,1,1-
2
3
trifluoroethane (CF CH Cl) to synthesize 1,1,1,2-tetrafluoroethane (CF CH F). The highest activity was obtained on a pre-fluorinated catalyst
3
2
3
2
◦
◦
calcined at 400 C, with 19% of CF CH Cl conversion at 320 C. The effect of the calcination temperature on the CrOx species was investigated.
3
2
X-ray diffraction and Raman results indicated that the CrOx species (Cr(VI)) were well dispersed on the catalyst surface when the catalyst was
◦
calcined at 400 C. With increasing calcination temperature, most of the CrOx species changed from high oxidation state Cr(VI) to low oxidation
state Cr(V) or Cr(III) species, which resulted in difficulty in pre-fluorination of the catalyst. It was also found that the CrFx, CrOxFy or Cr(OH)xFy
phases originated from high oxidation state Cr(VI) species were the active sites for the fluorination reaction.
©
2007 Elsevier Inc. All rights reserved.
Keywords: Fluorination; 1,1,1,2-tetrafluoroethane; Cr catalyst; Y O ; Active sites
2
3
1
. Introduction
Chlorofluorocarbons (CFCs) are considered as stratosphere
was the best catalyst for the fluorination of CF3CH2Cl and the
catalytic activity was greatly affected by the pretreatment con-
ditions. Besides, they have found that the chromium fluoride
CrFx or oxyfluoride CrOxFy were the active sites for the reac-
tion while crystalline Cr2O3 was an inactive catalyst. In addi-
tion, it was found that the CrO F phase exhibited significantly
ozone depletion substances (ODS) [1]. According to the Mon-
treal Protocol, CFCs will be abandoned by 2010 in develop-
ing countries and has been forbidden since 2000 in developed
countries. As CFCs are widely used in the refrigeration field
and their importance in the industry, preparation of hydrofluo-
rocarbon (HFC) alternatives has attracted much attention. For
example, 1,1,1,2-tetrafluoroethane (CF3CH2F) has been con-
sidered as a promising alternative of dichlorodifluoromethane
x
y
higher catalytic activity in dismutation of CCl F and fluori-
2
2
nation of CF CH Cl than the pure oxide (Cr O ) and fluoride
3
2
2
3
(α-CrF and β-CrF ) [5,7,8]. Quan et al. [9] found that Cr
3
3
species with high oxidation state shows higher catalytic activ-
ity than Cr(III) in fluorination of CH Cl to CH F . However,
2
2
2 2
(CF2Cl2) for its zero ozone depletion potential (ODP) and low
more evidence is needed to further understand the active sites,
because at present evolution of the active sites such as CrFx
and CrOxFy remains unclear. Furthermore, as thermal pretreat-
ment would cause structural change of the Cr catalysts, effect
of calcination temperature on the structure of the catalysts and
catalytic activity is of great importance.
global warming potential (GWP) values.
One of the synthesis routes for CF3CH2F is gas-phase fluor-
ination of 2-chloro-1,1,1-trifluoroethane (CF3CH2Cl) with HF
over CrF3 and CrOx catalysts supported on MgF2, AlF3, MgO
or Al2O3 [2–6]. Cho et al. [4] found that Cr/MgO catalyst
In this work, yttrium oxide (Y O ) was used as a support
2
3
*
for chromium oxide catalysts. The catalytic performance of the
Corresponding author. Fax: +86 579 82282595.
E-mail address: mengfeiluo@zjnu.cn (M.-F. Luo).
CrOx–Y2O3 catalysts was tested for fluorination of CF3CH2Cl
0021-9517/$ – see front matter © 2007 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2007.11.005

2
J. He et al. / Journal of Catalysis 253 (2008) 1–10
to CF3CH2F. By using various characterization techniques, ef-
fect of calcination temperature on the structure and composition
of the catalysts was investigated and correlated to the catalytic
performance. Also, the evolution of the active sites for this re-
action was discussed.
X-ray diffraction (XRD) patterns were collected on a Philips
PW3040/60 powder diffractometer operating at 40 kV and
40 mA using CuKα radiation in the 2θ range from 10 to
◦
◦
−1
80 with a scan rate of 0.3 min . The in situ XRD experi-
ments were carried out using an Anton Paar TCU 750 cham-
ber linked to the X-ray diffraction diffractometer, with a scan
◦
−1
2
. Experimental
rate of 0.3 min and the XRD patterns were recorded in air
atmosphere. Phase composition and cell parameter were calcu-
lated by Rietveld [10] method using MID-JADE 6.5 software.
TG-DSC experiments of the catalysts were performed using
a Netzsch STA 449C instrument. The experiments were carried
2
.1. Catalyst preparation
The CrOx–Y2O3 catalysts were prepared by a deposition–
◦
out to a maximum temperature of about 800 C with a heating
precipitation method. A detailed process is as follows: an aque-
ous solution of Cr(NO3)3 was mixed with a Y(OH)3 powder,
then an aqueous solution of (NH4)2CO3 (1 M) was added to the
mixture under stirring until a precipitated slurry was obtained.
The resulting slurry was aged for 2 h and then separated from
◦
−1
−1
rate of 10 C min in air (50 ml min ).
Raman spectra were obtained on a Renishaw RM1000 con-
focal microscope with exciting wavelength of 514.5 nm un-
der ambient conditions, and the catalysts were hydrated. UV–
visible diffuse reflectance spectra were recorded on a Thermo
Evolution 500 spectrophotometer equipped with a labsphere
RSA-UC-40. X-ray photonelectron spectroscopy (XPS) exper-
iments were carried out on a VG ESCALAB MK2 system
(Perkin–Elmer) with AlKα radiation (hν = 1486.6 eV). The
pass energy was fixed at 20 eV to ensure sufficient sensitiv-
ity. Binding energies were calibrated by using the contaminant
carbon (C 1s = 284.6 eV).
the mother liquid, washed with deionized water and dried at
◦
1
8
20 C overnight. Finally, it was calcined at 400, 500, 600 and
◦
00 C for 4 h. The catalysts were denoted as CrYO-4, CrYO-5,
CrYO-6 and CrYO-8, and the Cr content was 20 wt% in the cat-
alyst.
2
.2. Catalyst activation
Prior to use, the prepared catalyst was subject to a pre-
Reducibility of the catalysts was measured by means of
H temperature-programmed reduction (H -TPR) technique.
fluorination process in order to activate the catalyst. The pre-
fluorination was carried out in a stainless steel tubular reactor
with a diameter of 1 cm and a length of 30 cm. 3 g of prepared
2
2
2
0 mg of the catalyst was placed in a quartz reactor, which was
connected to a home made TPR apparatus. 5% H2 in N2 was
◦
−
1
CrYO catalyst was loaded into the reactor and dried at 260 C
introduced with a flow rate of 25 ml min . The catalyst was
◦
−1
◦
◦
−1
for 2 h and at 350 C for 2.5 h in N2 at a flow of 30 ml min
.
heated from 150 to 800 C at a heating rate of 20 C min .
Then the N2 flow was stopped and a mixture of HF (80 ml
The amount of H2 uptake during the reduction was measured
using a thermal conductivity detector (TCD). Water produced
during the reduction was trapped with a 5A molecular sieve.
Contents of Cr in the catalysts were determined by the in-
ductively coupled plasma (ICP) technique.
−
1
−1
◦
min ) and N2 (20 ml min ) was introduced at 260 C for 2 h
◦
and subsequently at 350 C for 2.5 h. The CrYO-4, CrYO-5,
CrYO-6 and CrYO-8 catalysts after the activation process were
denoted as CrYF-4, CrYF-5, CrYF-6 and CrYF-8, respectively.
2
.3. Fluorination reaction
3
. Results
The fluorination reactions were carried out in the same re-
actor after the catalyst activation, under atmospheric pressure.
g of the pre-fluorinated catalyst was loaded, corresponding
3
.1. Structural characterization
3
Table 1 lists the surface areas of the CrYO and the CrYF cat-
to a catalyst volume of 3 ml. Flow rates of CF3CH2Cl and
alysts. For the CrYO catalyst, the surface area first increased
then decreased with further calcination temperature. For the
CrYF catalyst, its surface area was much larger than the cor-
responding CrYO catalyst. Besides, as calcination temperature
increased, the surface area declined dramatically. For exam-
◦
HF pre-heated at 30 C were carefully controlled using mass
flow controllers. The molar ratio of HF/CF3CH2Cl = 10 and
−1
the GHSV was 3000 h . The gaseous products were analyzed
by a gas chromatograph (Shimadzu GC-14C) equipped with a
flame ionization detector (FID) and a HP GS-GASPRO capil-
lary column.
◦
ple, the CrYF-4 catalyst calcined at 400 C had a surface area
Table 1
2
.4. Catalyst characterization
BET surface areas of CrYO and CrYF catalysts
2
−1
)
Calcination
temperature ( C)
BET surface area (m g
Surface areas of the catalysts were determined by the modi-
◦
CrYO catalysts
CrYF catalysts
fied BET method from the N2 sorption isotherms at 77 K on an
Auto-sorb-1 apparatus. Scanning electron microscopy (SEM)
images of the catalysts were obtained an a Hitachi S-4800 mi-
croscope equipped with an energy dispersion X-ray (EDX) at-
tachment.
400
7.9
19.1
16.6
12.3
60.7
46.1
25.3
11.6
500
600
800

J. He et al. / Journal of Catalysis 253 (2008) 1–10
3
Fig. 1. N sorption isotherms of CrYO-4 and CrYF-4 catalysts.
2
Table 2
Surface composition analysis of CrYO and CrYF catalysts by EDX
Calcination
Atom (%)
temperature
CrYO catalyst
CrYF catalyst
◦
(
C)
Cr
Y
O
F
Cr
Y
O
F
4
5
6
8
00
14.1
9.4
13.7
16.8
41.0
50.7
53.6
54.1
44.9
39.9
32.7
29.1
/
/
/
/
12.2
8.6
14.8
16.1
45.1
47.0
36.7
56.0
0
42.7
42.3
38.3
0
00
00
00
2.1
10.2
27.9
2
−1
2
−1
of 60.7 m g , while only 11.6 m g was obtained on the
◦
CrYF-8 catalyst calcined at 800 C.
Fig. 2. SEM images for (a) CrYO-4 and (b) CrYF-4 catalysts.
Fig. 1 shows the N2 sorption isotherms of the CrYO-4 and
the CrYF-4 catalysts. The N2 sorption isotherms of the CrYF-4
catalyst was of type IV, indicating that it is a porous material.
However, the isotherm of the CrYO-4 catalyst suggested that
the catalyst is non-porous. Pore volumes of the CrYO-4 cata-
−
1
−1
lyst and the CrYF-4 catalyst are 0.089 ml g and 0.076 ml g ,
respectively. It should be noticed that the pore volume of the
CrYO-4 catalyst is probably due to the particle packing ef-
fect. The microstructure of the representative catalyst calcined
◦
at 400 C is shown in Fig. 2. It can be seen that the surface of
CrYO-4 catalyst was very smooth while the CrYF-4 catalyst
had some pore structure. Meanwhile, the surface composition
analysis of the catalysts is listed in Table 2. For the CrYO
catalysts, it was found that the oxygen concentration on the cat-
alyst surface gradually decreased from CrYO-4 to CrYO-8. For
the CrYF catalysts, oxygen concentration gradually increased
while fluorine concentration gradually decreased with calci-
nation temperature. It was interesting to find that the CrYF-8
catalyst had same surface composition as the CrYO-8 catalyst,
indicating that the CrYO-8 catalyst could hardly be fluorinated
during the activation process.
Fig. 3 shows the XRD patterns of the CrYO and the CrYF
catalysts calcined at different temperatures. Phase compositions
of the catalysts were summarized in Table 3. From Fig. 3a
and Table 3, it can be seen that the CrYO-4 catalyst remained
amorphous and the CrOx species were well dispersed on the
support surface. However, the characteristic peaks for YCrO4
and Y2O3 were observed in the CrYO-5 and the CrYO-6 cata-
Fig. 3. XRD patterns of (a) CrYO and (b) CrYF catalysts.

4
J. He et al. / Journal of Catalysis 253 (2008) 1–10
Table 3
Phase and phase composition of CrYO and CrYF catalysts
Calcination
temperature ( C)
Phase composition (wt%)
CrYO catalyst
◦
CrYF catalyst
400
500
600
800
Amorphous
CrF (6.8%) + YF (93.2%)
3
3
YCrO (67.2%) + Y O (37.8%)
Cr O (3.8%) + YF (96.2%)
4
2
3
2
3
3
YCrO (55.2%) + Y O (44.8%)
Cr O (18.6%) + YF (81.4%)
4
2
3
2
3
3
YCrO (68.8%) + Y O (31.2%)
YCrO (77.9%) + Y O (22.1%)
3
2
3
3
2 3
Fig. 5. TG/DSC curves of CrYO-4 catalyst.
Fig. 4. In situ XRD patterns of CrYO-4 catalyst.
Table 5
Raman shifts of the CrYO and CrYF catalysts
Table 4
−
1
Calcination
temperature ( C)
Raman shift (cm
CrYO catalyst
365, 890, 1100
)
Phase and phase composition of CrYO-4 catalyst by in situ XRD results
◦
◦
CrYF catalyst
209, 282, 343, 545, 698, 848, 1378
Temperature ( C)
Phase composition (wt%)
Amorphous
400
500
600
800
400
500
600
700
800
365, 545, 825, 865 209, 282, 343, 545, 698, 848, 1378
365, 545, 825, 870 209, 282, 343, 545, 698, 848, 1378
220, 280, 340, 490, 209, 217, 280, 340, 487, 545, 698,
YCrO
4
YCrO (61.6%) + Y O (38.4%)
4
2 3
YCrO (35.1%) + YCrO (26.8%) + Y O (38.1%)
4
3
2 3
545, 698, 825, 1400 848, 1378
YCrO (66.0%) + Y O (34.0%)
3
2 3
The TG/DSC experiments were conducted to further illus-
lysts, and the intensity of Y2O3 peak increased with calcination
temperature. When the catalyst was calcined at 800 C, a new
YCrO3 phase appeared accompanied by the disappearance of
the YCrO4 phase.
As seen in Fig. 3b, for the CrYF-4 catalyst, characteristic
peaks of CrF3 and YF3 were observed. Besides, crystalline YF3
and Cr2O3 phases were detected in the CrYF-5 and the CrYF-6
catalysts, while the CrYF-8 catalyst had same structure as the
CrYO-8 catalyst.
Fig. 4 presents in situ XRD patterns of the CrYO-4 cata-
lyst, which clearly demonstrated the transformation of the CrOx
species during the heating process. The catalyst was amorphous
◦
trate the transformation of CrOx species for the CrYO-4 cat-
alyst during the calcination process and the results are shown
in Fig. 5. As seen in Fig. 5, an obvious weight loss was ob-
◦
served below 200 C, which could be attributed to the loss of
crystalline H2O. In addition, a main weight loss was observed
◦
between 500 and 650 C, accompanied by a strong endother-
◦
◦
mic peak at 610 C. Also, a weight loss at about 750 C was
observed with an endothermic peak.
The Raman spectra of CrYO and CrYF catalysts with differ-
ent calcination temperatures are shown in Fig. 6 and the Raman
shifts are summarized in Table 5. Note that no Raman band
−1
at 376 cm assigned to Y2O3 was observed for all the cata-
◦
◦
at 400 C, while the YCrO4 phase emerged at 500–600 C.
With further increasing temperature, the YCrO3 and Cr2O3
phases appeared. This result is in good agreement with the XRD
results obtained from the catalysts calcined at different tem-
peratures, indicating the CrOx species transformed from high
oxidation state (Cr(VI)) to low oxidation state (Cr(III)). Fur-
thermore, phase composition analysis summarized in Table 4
shows that the CrOx species composition is in good accordance
with that in Table 3.
lysts. For the CrYO-4 catalyst, two Raman peaks were observed
−1
at 365 and 890 cm , which were assigned to the symmetric
stretching and bending modes of Cr(VI), respectively, corre-
sponding to isolated tetrahedral surface-chromate species that
was distorted by its interaction with the oxide support surface
[4,11–13]. For the CrYO-5 and the CrYO-6 catalysts, they show
−
1
−1
a strong band at 865 cm and a shoulder band at 825 cm due
to oligomeric (dimer to tetramer) species of CrO [4,14]. Ad-
x
−
1
ditionally, a Raman band at 545 cm indicated the presence

J. He et al. / Journal of Catalysis 253 (2008) 1–10
5
Fig. 7. UV–vis diffuse reflectance spectra of (a) CrYO and (b) CrYF catalysts.
◦
in Cr(VI) oxidation state were predominant at 400 C. But the
Fig. 6. Raman spectra of (a) CrYO and (b) CrYF catalysts.
CrYO-5 and the CrYO-6 catalysts show two absorption peaks
5
+
−
of crystalline Cr2O3 [11,15,16], while the band at 365 cm 1
at 410 and 600–700 nm, which were assigned to the Cr and
3+
Cr , respectively [16,17]. When the calcination temperature
increased to 800 C, the peak intensity at 370 nm correspond-
almost disappeared. For the CrYO-8 catalyst, in addition to a
band at 825 cm due to oligomeric species of CrOx, a new
band at 220 cm due to the Cr–O–Cr band vibration of dichro-
mate was observed [11,17]. The bands at 489, 545, 698 and
380 cm were assigned to the crystalline Cr2O3. The weak
peak intensity for the CrYO-6 catalyst was due to the dark green
color of the catalyst. The shift of the main Raman band (800–
00 cm ) towards lower frequency indicated that the degree
◦
−
1
ing to Cr6 dramatically decreased, while peaks at 475 nm
and 600–730 nm appeared, suggesting that the oxidation state
changed from Cr(VI) to Cr(III) at high calcination temperature.
In Fig. 7b, all the CrYF catalysts show a band at 370 nm,
+
−
1
−1
1
indicating some Cr6 species still existed. However, for the
CrYF-4 catalyst, two new bands at 465 and 610 nm due to Cr
+
3+
−1
9
appeared. For the CrYF-5 and CrYF-6 catalysts, the intensity of
these two bands was slightly weaker. Like the XRD and Raman
results, the UV–visible diffuse spectra of CrYF-8 catalyst re-
mained the same as the CrYO-8 catalyst.
of polymerization of the CrOx species in the catalyst increased
with calcination temperature [4,14].
Fig. 6b shows the Raman spectra of the CrYF catalysts. It
should be noted that YF3 gave no Raman signal under the am-
bient conditions. For all the catalysts, the bands at about 343
XPS experiments were conducted to further investigate the
−1
CrO species on the catalysts surface. Fig. 8 shows the Cr 2p
and 848 cm due to the Cr–O stretching and bending modes
of the Cr(VI) oxides were observed. Besides, the bands as-
x
XPS spectra of the CrYO and the CrYF-4 catalysts. For the
CrYO-4 catalyst, the Cr 2p3/2 peak was resolved into two peaks
at 579.4 and 576.8 eV. The former peak was assigned to the
CrO3 and the latter one was assigned to the amorphous Cr2O3
phase [8,21–23]. For the CrYO-5 catalyst, the Cr 2p3/2 core
level shows two peaks at 578.5 and 576.3 eV, which were as-
signed to the YCrO4 phase and the Cr2O3 phase, respectively.
For the CrYO-6 catalyst, the peaks at 578.5 and 577.3 eV were
ascribed to the YCrO4 and YCrO3 phases, respectively. For
the CrYO-8 catalyst, the peak at 578.7 eV was assigned to the
YCrO3 phase and the peak at 575.8 eV indicated the presence
of Cr2O3. The shift of the main peak at 579.4 eV to lower bind-
ing energy suggested the transformation of high oxidation state
−
1
signed to Cr–O–Cr linkages (220 cm ) and crystalline Cr2O3
−1
(
548 cm ) were also observed. All the catalysts show a band at
378 cm assigned to the crystalline Cr2O3, which was prob-
−1
1
ably formed during the fluorination process. For the CrYF-8
catalyst, no obvious change was found compared to the CrYO-
8
catalyst.
In order to further illuminate the oxidation state of Cr
species in the CrYO and CrYF catalysts, the UV–visible dif-
fuse reflectance experiments were carried out and the results
are shown in Fig. 7. In Fig. 7a, a broad absorption peak at
about 370 nm assigned to Cr(VI) [18–20] was detected for
the CrYO-4 catalyst, indicating that the surface CrO3 species

6
J. He et al. / Journal of Catalysis 253 (2008) 1–10
Fig. 9. TPR profiles of CrYO catalysts.
Table 7
Total amounts of H consumption values obtained from TPR profiles
2
−1
)
cat
Catalyst
H consumption (mmolg
2
CrYO-4
CrYO-5
CrYO-6
CrYO-8
3.3
2.6
1.4
0.3
Cr(VI) species to low oxidation state Cr(III) species during the
calcination process.
The contents of Cr species with different valences in the
CrYO catalysts were listed in Table 6, derived from the peak
area of the corresponding Cr species from the XPS results. For
the CrYO-4 catalyst, the contents of Cr6 and Cr species
+
3+
were 53.5 and 46.5%, respectively. For the CrYO-5 and the
5+
CrYO-6 catalysts, the content of Cr species decreased while
the content of Cr3 species increased with calcination temper-
+
3+
ature. For the CrYO-8 catalyst, the content of Cr species
(
YCrO3 + Cr2O3) was 100%.
The Cr 2p XPS spectrum of the CrYF-4 catalyst was also
recorded in Fig. 8. For the CrYF-4 catalyst, the Cr 2p3/2 core
level shows three peaks at 579.4, 577.9 and 575.9 eV. The peak
at 579.4 eV was assigned to CrF3, and the peak at 577.9 eV may
be attributed to the formation of Cr(OH)xFy by fluorination of
CrOx [24], and peak at 575.9 eV was assigned to Cr2O3 phase.
It should be noticed that EDX results of the CrYF-4 catalyst
(Table 2) did not reveal the existence of oxygen; however, it is
probably due to the detection limit of the apparatus (0.5 at%).
The H2-TPR profiles of the CrYO catalysts are shown in
Fig. 9. The total amounts of H2 consumption are listed in Ta-
ble 7. The CrYO-4 catalyst shows one reduction peak centered
Fig. 8. XPS spectra of Cr 2p for the CrYO and CrYO-4 catalysts.
◦
◦
at 610 C and one weak reduction peak at 680 C. The former
was related to the more easily reducible oligomeric chromate
while the latter to the less reducible monochromate [14]. The
reduction peaks shifted to the lower temperature at 550 and
Table 6
Amount of Cr species with different valences on catalysts by XPS analysis
Catalyst
Cr content (%)
◦
◦
6
40 C when the calcination temperature increased to 500 C.
6
+
3+
CrYO-4
CrYO-5
CrYO-6
CrYO-8
Cr (53.5%) + Cr (46.5%)
Moreover, these two peaks incorporated into one when calci-
5
+
3+
Cr (68.6%) + Cr (31.4%)
◦
nation temperature increased to 600 and 800 C, and the peak
5+
3+
Cr (29.8%) + Cr (71.2%)
position shifted to a lower temperature. For the CrYO-8 cat-
3+
Cr (100.0%)
◦
alyst, only a reduction weak peak at 450 C was observed.

J. He et al. / Journal of Catalysis 253 (2008) 1–10
7
Fig. 12. XRD patterns of CrYF-4 catalyst after different reaction time.
Fig. 10. Fluorination of CF CH Cl over various CrYF catalysts. Reaction con-
3
2
−
1
ditions: HF/CF CH Cl = 10, S.V. = 3000 h
.
3
2
Fig. 13. Raman spectra of CrYF-4 catalyst after different reaction time.
Fig. 11. Fluorination of CF CH Cl over CrYF-4 catalyst with time on stream
3
2
◦
−1
.
at 320 C. Reaction conditions: HF/CF CH Cl = 10, S.V. = 3000 h
3
2
CrYF-4 catalyst after different reaction time. CrF3·H2O phase
was observed in the CrYF-4 catalyst after 12 h reaction, and the
peak intensity increased with further 72 h reaction. Also, Ra-
man spectra of the CrYF-4 catalyst after different reaction time
is shown in Fig. 13. With increasing reaction time, the peak in-
As seen in Table 7, H2 consumption amount varies from 3.3
−1
to 0.3 mmol H2 g as the calcination temperature increased
cat
◦
from 400 to 800 C.
−1
tensity of CrO became stronger, and a new band at 1602 cm
x
3
.2. Catalytic testing
due to carbon was observed [25].
Fig. 10 shows the effect of calcination temperature on the
4. Discussion
catalytic performance of the CrYF catalysts. Note that the selec-
tivity to CF CH F was not given for it was higher that 98%. It
3
2
4.1. Effect of calcination temperature on the catalyst structure
can be seen that calcination temperature has great influence on
the catalytic activity. The catalyst calcined at low temperature
had higher activity than the one calcined at high temperature.
For example, about 19% of CF CH Cl conversion was obtained
on the CrYF-4 catalyst at 320 C, while less than 5% of conver-
sion was obtained on the CrYF-8 catalyst.
In this work, a series of CrYO catalysts were prepared and
the effect of calcination temperature on the CrO species was
x
3
◦
2
investigated. For the CrYO catalysts, relatively low surface ar-
eas were obtained, while the pre-fluorinated catalysts (CrYF)
have much higher surface area than the CrYO catalysts ex-
cept the CrYF-8 catalyst (Table 1). The pre-fluorination process
probably caused a change in catalyst structure because of the re-
action between HF and the CrYO catalyst, as evidenced by the
Fig. 11 demonstrates stability of the pre-fluorinated CrYO-4
◦
catalyst (CrYF-4) at 320 C under reaction conditions. It could
be seen that the conversion slightly decreased in the initial pe-
riod and it reached steady state after 3 h.
N sorption isotherms (Fig. 1) and the SEM results (Fig. 2).
2
In order to investigate the change of the catalyst structure af-
ter different reaction time, Fig. 12 shows XRD patterns of the
The porous CrYF-4 catalyst results in a higher surface area
compared to the CrYO-4 catalyst. Additionally, by compar-

8
J. He et al. / Journal of Catalysis 253 (2008) 1–10
ing the images of the CrYO-4 and the CrYF-4 catalysts, the
more porous morphology of the CrYF-4 catalyst suggests that
it could be responsible for the increase in surface area after pre-
fluorination.
The XRD (Figs. 3 and 4) and the Raman (Fig. 6) results re-
veal the general phase transformation of the chromium species.
In the CrYO-4 catalyst, the chromium oxide are well dispersed
on the support and mainly exist in the form of high oxidation
state Cr(VI). With increasing calcination temperature, phases of
YCrO4 and YCrO3 with support Y2O3 were observed. The in-
teraction between CrO3 and Y2O3 at low and high temperature
may follow the equations [26]:
weaker because of the growth of the crystalline size of CrOx
species, therefore the reduction of the CrO species becomes
x
easier. In other hand, as reported previously, the reduction of
polymeric chromate is easier than the monochromate [14], high
calcination temperature makes a higher degree of polymeriza-
tion of CrO species, as evidenced by Raman results (Fig. 6a),
x
which results in a lower reduction temperature.
H consumption amounts of the CrYO catalysts decrease
2
with increasing calcination temperature (Table 7). Note that the
reduction of CrO may follow two steps:
3
CrO3 + (3/2)H2 → (1/2)Cr2O3 + (3/2)H2O,
Cr2O3 + (3/2)H2 → (1/2)Cr + (3/2)H2O.
(3)
(4)
2
CrO3 + Y2O3 → 2YCrO4 + (1/2)O2,
(1)
(2)
YCrO4 → YCrO3 + (1/2)O2.
However, the reduction of Cr2O3 to Cr (Eq. (2)) is not likely
◦
to occur below 800 C [27,28]. Also, the consumption values
As seen from the Raman spectra (Fig. 6a), CrOx species in
the CrYO-4 catalyst was probably in form of CrO3. The isoelec-
tric point (IEP) of Y2O3 is 8.1, which is close to that of Al2O3
are much lower than the stoichiometric value of H2 consump-
−
1
tion (5.8 mmolg ), if one assumes that all the Cr(VI) species
cat
could be reduced to Cr(III), especially for the catalysts calcined
at high temperature. Note that the XRD, Raman and UV–vis re-
sults (Figs. 3a, 6a and 7a) clearly show that the high oxidation
state Cr content decreases with calcination temperature. There-
fore, the reduction peaks found in the present work is likely
due to the reduction of fractional high oxidation state Cr(VI)
or Cr(V) on the catalyst surface to Cr(III), as the XRD results
(
8.9). Previous research pointed out that the chromium species
on the Al2O3 contains chromate and dichromate [12], which
is consistent with the current study, as evidenced by the H2-
TPR results (Fig. 9). As the calcination temperature increased
◦
from 500 to 800 C, the crystalline Cr2O3 was observed and the
isolated CrOx species transformed to the oligomeric (dimer to
tetramer) species. For the CrYO-8 catalyst, the intensity of the
crystalline Cr2O3 Raman band further increases, which indi-
cates a larger proportion of Cr2O3 in the catalyst. Additionally,
(Fig. 3a) and the Raman results (Fig. 6a) clearly show the exis-
tence of Cr(VI) and Cr(V) species.
−
1
in the Raman spectra, the band due to Y2O3 (376 cm ) was
not observed probably due to the fact that the CrOx species
aggregates on the catalyst surface and cover the Y2O3 sup-
port, particularly at high Cr loading (20%). UV–vis (Fig. 7)
and XPS (Fig. 8) results clearly identified the different Cr
species in the CrYO and the CrYF catalysts, which again con-
firmed the different Cr species in the catalysts. The existence
of the bands at 465 and 610 nm in the CrYF-4 catalysts sug-
gested the emergence of Cr3 species during the fluorination
process.
4
.2. Active sites for the fluorination reaction
Since the pre-fluorination process is necessary in order to ac-
tivate the catalyst, it is more important to investigate the CrYF
catalysts because it may provide relevant information of the ac-
tive sites for the fluorination reaction.
The phase composition of the CrYF catalysts was also con-
firmed by the XRD (Fig. 3b) and the Raman (Fig. 6b) results.
It was found that the high oxidation state Cr species (Cr(VI))
transformed to low oxidation state Cr species (Cr(III)). Also, in
the CrYF-4, CrYF-5 and CrYF-6 catalysts, the support Y2O3
transformed to YF3. Moreover, comparing to the CrYF-8 cata-
lyst with the CrYO-8 catalyst, XRD results (Fig. 3) found that
the features merely changed, indicating that the CrYO-8 cata-
lyst is very stable during the fluorination process. Raman results
(Fig. 6b) indicated that the CrOx species become polymeric
during the fluorination process, and the polymeric chromates
are difficult to be fluorinated [4]. EDX composition analysis
(Table 2) shows that the F concentration in the CrYF catalysts
decreased with calcination temperature. Note that the CrYO-4
catalyst contains isolated Cr(VI) species while the CrYO-8 cat-
alyst contains oligomeric Cr(VI) and dominant Cr(III) species.
XPS results (Fig. 8) on the CrYF-4 catalyst reveal the existence
of CrF3, Cr(OH)xFy and Cr2O3 species, so it could be con-
cluded that the evolution of these species was originated from
the well isolated high oxidation state Cr(VI) on the catalyst sur-
face, whereas the low oxidation state chromium compounds
YCrO3 and Cr2O3 or the Cr(VI) species in oligomeric form
+
TG/DSC result (Fig. 5) indicated that two endothermic peaks
◦
◦
between 600 and 800 C were observed. The peak at 610 C is
probably assigned to the transformation of Cr species from the
high oxidation state Cr(VI) to the low oxidation state Cr(V), and
◦
the peak below 800 C is due to the transformation from Cr(V)
to the lower oxidation state Cr(III) [26]. This is consistent with
the in situ XRD result.
The binding energy values of Cr 2p3/2 in the CrYO cata-
lysts (Fig. 8) shift to lower values with increasing calcination
temperature, indicating the average oxidation state of CrOx be-
comes lower. These results are in line with the XRD, Raman
and UV–vis results.
H2-TPR results (Fig. 9) shows a clear shift of the reduction
peaks. The shift is probably related to interaction between the
CrOx species and the support. For the catalysts calcined at low
temperature, the interaction of CrOx species with the support is
very strong, which makes the reduction of the CrOx species dif-
ficult. While for the catalysts calcined at high temperature, the
interaction between the CrOx species and the support becomes

J. He et al. / Journal of Catalysis 253 (2008) 1–10
9
5
. Conclusions
A series of CrOx–Y2O3 catalysts with different calcination
temperatures have been prepared for vapor-phase fluorination
of CF3CH2Cl to synthesize CF3CH2F. The highest activity was
◦
obtained on a pre-fluorinated catalyst calcined at 400 C, with
1
◦
9% of CF3CH2Cl conversion at 320 C. The calcination tem-
perature of the catalyst has great influence on the CrOx species.
With increasing calcination temperature, the structure of the
CrOx species transformed from the highly dispersed mono-
chromate (CrO3) to low oxidation state CrOx species (Cr(V)
and Cr(III)) and oligomeric chromate. During the activation
process, partial well dispersed monochromate (CrO3) trans-
formed to the catalytically active species CrFx, CrOxFy or
Cr(OH)xFy. As the content of the well dispersed monochro-
mate decreased with increasing calcination temperature, it led
to a decline in activity.
Fig. 14. Turnover frequency (TOF) for fluorination of CF CH Cl over CrYO
3
2
catalysts calcined at different temperatures (calculated based on the total Cr
◦
content in the catalyst, reaction temperature 320 C).
Acknowledgments
is difficult to be fluorinated by HF. Furthermore, UV–vis re-
sults (Fig. 7b) confirmed the transformation of Cr species. In
the CrYF-4, the appearance of Cr(III) species indicates that par-
tial Cr(VI) was reduced during the pre-fluorination process, and
the higher intensity of the Cr(III) in the CrYF-4 catalyst than in
other catalysts suggests that higher proportion of Cr(VI) species
in the CrYO-4 catalyst was transformed.
The prepared CrYO-4 catalyst has higher reactivity com-
pared to the best catalyst that had been reported. Cho et al. [4]
prepared a Cr/MgO catalyst and found that the catalytic activity
was greatly affected by the support and activating conditions.
The highest conversion obtained on the Cr/MgO catalyst was
This work is financially supported by the Zhejiang Provin-
cial Nature Science Foundation of China (Grant No. Y407179).
We also thank Prof. Wei-xin Huang (University of Science and
Technology of China) for XPS peak fitting.
References
[
[
1] M.L. Molina, F.S. Rowland, Nature 249 (1974) 810.
2] S. Brunet, B. Requieme, E. Colnay, J. Barrault, M. Blanchard, Appl. Catal.
B 5 (1995) 305.
[
[
[
3] S. Brunet, B. Requieme, E. Matouba, J. Barrault, M. Blanchard, J. Ca-
tal. 152 (1995) 70.
4] D.H. Cho, Y.G. Kim, M.J. Chung, J.S. Chung, Appl. Catal. B 18 (1998)
251.
5] H. Lee, H.S. Kim, H. Kim, W.S. Jeong, I. Seo, J. Mol. Catal. A 136 (1998)
◦
−1
about 10% at 320 C (HF/CF3CH2Cl = 8, GHSV = 1800 h ).
In another group, Lee et al. [6] prepared a Cr/MgF2 catalyst
and the best initial catalytic activity was 0.68 mmolh
−
1
−1
g
cat
◦
at 320 C (HF/CF3CH2Cl = 8). In our study, the highest con-
85.
◦
version obtained on the CrYO-4 was about 19% at 320 C
[6] H. Lee, H.D. Jeong, Y.S. Chung, H.G. Lee, M.J. Chung, S. Kim, H.S. Kim,
J. Catal. 169 (1997) 307.
[7] B. Adamczyk, O. Boese, N. Weiher, S.L.M. Schroeder, E. Kemnitz, J. Flu-
orine Chem. 101 (2000) 239.
−
1
(
HF/CF3CH2Cl = 10, GHSV = 3000 h ), corresponding to
−
1
−1
a formation rate of 2.03 mmolh
g
. Besides, the catalyst is
cat
stable even after 72 h reaction time. This indicates that the Cr
catalysts supported on Y2O3 could be a promising candidate for
the fluorination reaction.
[
8] Y.S. Chung, H. Lee, H.D. Jeong, Y.K. Kim, H.G. Lee, H.S. Kim, S. Kim,
J. Catal. 175 (1998) 220.
[9] H.D. Quan, M. Tamura, Y. Matsukawa, J. Mizukado, T. Abe, A. Sekiya,
J. Mol. Catal. A 219 (2004) 79.
Turnover frequency (TOF) of the reaction calculated based
on the total Cr content in the catalyst is shown in Fig. 14.
It was found that the TOF decreased with increasing calcina-
tion temperature. As the total amount of Cr in the catalysts
is nearly the same, however, the amount of the CrOx species
with high oxidation state (>Cr(III)) decreased from CrYO-4 to
CrYO-8. Cho et al. [4] and Adamczyk et al. [7] concluded that
chromium fluoride CrFx or oxyfluoride CrOxFy species are the
active sites for the fluorination of CF3CH2Cl to CF3CH2F. Our
finding supports their conclusion because the activity closely
correlates with the pre-fluorinated Cr species. Furthermore, the
new finding in this study is that the origin of the fluorinated Cr
species was identified, that is, the well isolated Cr(VI) species
on the catalyst surface could be transformed to CrFx, CrFxOy
or Cr(OH)xFy species.
[
10] R.A. Young, The Rietveld Method, Oxford Univ. Press, Oxford, 1993,
p. 22.
[11] B.M. Weckhuysen, I.E. Wachs, J. Chem. Soc. Faraday Trans. 92 (1996)
1
969.
[12] B.M. Weckhuysen, I.E. Wachs, R.A. Schoonheydt, Chem. Rev. 96 (1996)
327.
3
[13] B.M. Weckhuysen, R.A. Schoonheydt, J.M. Jehng, I.E. Wachs, S.J. Cho,
R. Ryoo, S. Kijlstra, E. Poels, J. Chem. Soc. Faraday Trans. 91 (1995)
3245.
[14] S.D. Yim, I.S. Nam, J. Catal. 221 (2004) 601.
[15] M. Cherian, M.S. Rao, A.M. Hirt, I.E. Wachs, G. Deo, J. Catal. 211 (2002)
82.
4
[16] O.F. Gorriz, L.E. Cadús, Appl. Catal. A 180 (1999) 247.
[17] F.D. Hardcastle, I.E. Wachs, J. Mol. Catal. 46 (1988) 173.
[18] B. Sun, E.P. Reddy, P.G. Smirniotis, Appl. Catal. B 57 (2005) 139.
[19] R. Galindo, J.A. Badenes, M. Llusar, M.A. Tena, G. Monrós, Mater. Res.
Bull. 42 (2007) 437.

10
J. He et al. / Journal of Catalysis 253 (2008) 1–10
[
[
[
[
20] M. Cherian, M.S. Rao, W.T. Yang, J.M. Jehng, A.M. Hirt, G. Deo, Appl.
[24] J.M. Rao, A. Sivaprasad, P.S. Rao, B. Narsaiah, S.N. Reddy, V. Vijayaku-
mar, S.V. Manorama, K.L. Krishna, K. Srinivas, K.R. Krishman, S. Badri-
narayanan, J. Catal. 184 (1999) 105.
Catal. A 233 (2002) 21.
21] D. Chidambaram, G.P. Halada, C.R. Clayton, Appl. Surf. Sci. 181 (2001)
[
25] R. Escribano, J.J. Sloan, N. Siddique, N. Sze, T. Dudev, Vib. Spectrosc. 26
2001) 179.
283.
(
22] E. Kemnitz, A. Kohne, I. Grohmann, A. Lippitz, W.E.S. Unger, J. Ca-
tal. 159 (1996) 270.
[26] T. Tachiwaki, Y. Kunifusa, M. Yoshinaka, K. Hirota, O. Yamaguchi, Int. J.
Inorg. Mater. 3 (2001) 107.
23] P.G. Harrison, N.C. Lloyd, W. Daniell, J. Phys. Chem. B 102 (1998)
[27] X. Wang, Y.C. Xie, Appl. Catal. B 35 (2001) 85.
[28] X. Wang, Y.C. Xie, Catal. Lett. 75 (2001) 73.
10672.