Effect of preparation method on performance of Cu-Fe/SiO2 catalysts for higher alcohols synthesis from syngas

Article abstract of DOI:10.1039/c6ra07366b

Five Cu-Fe/SiO₂ catalysts were prepared by deposition-precipitation, solid state impregnation, solid-state chemical reaction, citric acid combustion, and ultrasound-assisted wet impregnation, and their physicochemical and catalytic properties for higher alcohols synthesis were investigated. The results showed that the Cu-Fe/SiO₂ catalyst prepared by ultrasound-assisted impregnation had a higher dispersion and reducibility of active metal oxides, and the largest S_Cu, which increased the amount of adsorbed CO and the ability for CO dissociation. As a result, the CO conversion and space time yield of total alcohols of the catalyst reached 14.04% and 104.3 g kg^-1 h^-1, respectively, which were better than the other catalysts.

Full text of DOI:10.1039/c6ra07366b

RSC Advances
View Article Online
View Journal | View Issue
PAPER
Effect of preparation method on performance of
Cu–Fe/SiO catalysts for higher alcohols synthesis
2
Cite this: RSC Adv., 2016, 6, 55233
from syngas
Chao Sun, Dongsen Mao,* Lupeng Han and Jun Yu
2
Five Cu–Fe/SiO catalysts were prepared by deposition–precipitation, solid state impregnation, solid-state
chemical reaction, citric acid combustion, and ultrasound-assisted wet impregnation, and their
physicochemical and catalytic properties for higher alcohols synthesis were investigated. The results
showed that the Cu–Fe/SiO
2
catalyst prepared by ultrasound-assisted impregnation had a higher
Received 21st March 2016
Accepted 2nd June 2016
dispersion and reducibility of active metal oxides, and the largest SCu, which increased the amount of
adsorbed CO and the ability for CO dissociation. As a result, the CO conversion and space time yield of
DOI: 10.1039/c6ra07366b
ꢀ1 ꢀ1
total alcohols of the catalyst reached 14.04% and 104.3 g kg
the other catalysts.
h , respectively, which were better than
www.rsc.org/advances
SiO
compared with that of catalyst supported on larger and smaller
Higher alcohols synthesis (HAS) from coal, natural gas or pore SiO , which was contributed to the well dispersion of
lignocellulosic biomass via synthesis gas (syngas, CO + H ) has active metals and high diffusion of products in the bimodal
2
exhibited higher selectivity of C2+OH in total alcohols
1
. Introduction
2
2
18
attracted increasing attention due to their potential application structures. Although some promising results have been ob-
as liquid fuel, additives for gasoline, hydrogen carriers for fuel tained on the Fe modied Cu-based catalyst, there are still
1
–4
cells, and intermediates for chemicals.
systems have been studied extensively so far for HAS such as hols, high hydrocarbons selectivity and products with plenty of
Several catalytic several problems to be solved, such as low selectivity to alco-
1
8,20,21
noble Rh-based catalysts, modied methanol synthesis cata- water.
Thus, how to further improve the activity and
lysts, modied Fischer–Tropsch (F–T) synthesis catalysts and selectivity of Cu–Fe based catalyst for HAS remains the key of
2,4–6
14,18
Mo-based sulphur tolerance catalysts.
Noble Rh-based research.
catalysts have little attraction for commercial utilization due
In the present study, for the purpose of improving the cata-
to the high cost and limited availability in large scale produc- lytic performance of Cu–Fe/SiO
tion.
2
catalysts for HAS, they were
Among the non-noble metal-based catalysts, cobalt prepared by ve methods including deposition–precipitation,
modied Cu-based catalyst has been considered as one of the solid-state impregnation, solid-state chemical reaction, citric
7
–10
3
,11,12
most promising catalysts,
selectivity and poor stability limit its large scale industrial order to clearly elucidate the inuence of preparation methods,
however, the low total alcohol acid combustion, and ultrasound-assisted wet impregnation. In
13,14
application.
In recent years, Fe modied Cu-based catalysts were also ized by X-ray diffraction (XRD), N
the physicochemical properties of the catalysts were character-
adsorption–desorption,
For example, Lin et al. Fourier transform infrared (FT-IR) spectroscopy,
reported that the catalytic performance of zinc and manganese temperature-programmed reduction (H -TPR), X-ray photo-
doubly promoted Cu–Fe catalyst was superior to that of the electron spectra (XPS), temperature-programmed desorption of
2
11,13–27
investigated by many researchers.
H
2
2
zinc or manganese singly promoted Cu–Fe catalyst, which was adsorbed CO (CO-TPD), and
attributed to the synergistic effect between zinc and manga- chromatography.
N
O
reactive frontal
2
1
4
nese on the Cu–Fe catalyst. Zhang et al. found that plasma-
promoted Cu–Fe/SiO2 catalyst possessed higher CO conver-
sion, total alcohol selectivity and selectivity of C OH in alco-
2
. Experimental
2
+
1
6
2.1. Materials and catalyst preparation
hols than those of conventional sample for HAS. Ding et al.
reported that Cu–Fe catalyst supported on the bimodal pore A commercial silica gel (Qingdao Haiyang Chemicals Company,
ꢁ
China) was rst dried at 110 C overnight in an electric oven,
and then used as the support in this study. The Cu–Fe/SiO2
Research Institute of Applied Catalysis, School of Chemical and Environmental
catalysts were prepared by ve different methods, viz.,
ultrasound-assisted wet impregnation (WI), deposition–
Engineering, Shanghai Institute of Technology, Shanghai 201418, P.R. China.
E-mail: dsmao@sit.edu.cn; Fax: +86 21 6087 3625; Tel: +86 21 6087 3625
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 55233–55239 | 55233

View Article Online
RSC Advances
Paper
precipitation (DP), solid-state impregnation (SI), solid-state 2.2. Catalyst testing
chemical reaction (SR), and citric acid combustion (CC). Both
the Cu and Fe loadings of the catalysts were 10% mole fraction
CO hydrogenation reaction was carried out in a xed-bed micro-
reactor with length of 350 mm and internal diameter of 5 mm.
26
relative to support SiO . The Cu(NO ) $3H O, Fe(NO ) $9H O,
2
3 2
2
3 3
2
The fresh catalyst (0.3 g) was loaded between quartz wool and
axially centred in the reactor tube, with the temperature
monitored by a thermocouple close to the catalyst bed. Prior to
NaOH, oxalic acid, and citric acid were of analytical purity and
purchased from Chinese Sinopharm Chemical Reagent Co.,
Ltd., China.
ꢁ
ꢁ
reaction, the catalyst was heated to 300 C (heating rate ¼ 3 C
Preparation of Cu–Fe/SiO
impregnation method. For the preparation of Cu–Fe/SiO
appropriate amounts of Cu(NO ) $3H O and Fe(NO ) $9H O
2
by ultrasound-assisted wet
ꢀ1
ꢀ1
min ) and reduced with a H /N mixture (50 mL min , V /V
N
2
2
2
H
2
2
(WI),
¼
1 : 9) for 3 h at atmospheric pressure. The catalyst was then
3
2
2
3 3
2
ꢁ
cooled down to 250 C and the reaction started as gas ow was
were dissolved in 4 mL of distilled water. To this solution, the
ꢀ1
switched to a H
2
/CO mixture (30 mL min , VH2/VCO ¼ 2 : 1) at 3
SiO was soaked in the solution under ultrasonic irradiation.
ꢁ
2
MPa. All post-reactor lines and valves were heated to 150 C for
preventing the possible condensation of products. The products
Sonication was carried out on a KQ-250E instrument (Kunshang
Ultrasonic Instrument Co., Ltd., Jiangsu, China) with power of
were analyzed for CO, CO
by Agilent GC 6820 equipped with a ame ionization detector
FID) and a thermal conductivity detector (TCD). Preliminary
2
, alcohols and hydrocarbons on-line
250 W at 40 kHz for 30 min. Then this solution was slowly
evaporated under ambient atmosphere, aer that the impreg-
(
ꢁ
nated sample was dried at 110 C for 10 h and then calcined in
experiments with respect to possible inuence caused by mass
transfer limitation conrmed that such limitation could be
ruled out under the reaction conditions used in this work.
The conversion of CO was calculated based on the fraction of
ꢁ
static air at 350 C for 4 h.
Preparation of Cu–Fe/SiO2 by deposition–precipitation
method. For the preparation of Cu–Fe/SiO (DP), appropriate
2
amounts of Cu(NO
solved in 10 mL distilled water. Required amount of SiO
3
)
2
$3H
2
O and Fe(NO
3
)
3
$9H
2
O were dis-
was
CO that formed carbon-containing products according to: %
P
2
conversion ¼ (
n
i
M
i
/MCO) ꢂ 100, where n
i
is the number of
is the percentage of product i
dispersed in this solution and mixed thoroughly with
a magnetic stirrer. To this, an aqueous NaOH solution (0.1 M)
was added drop wise with continuous stirring at ambient
temperature, till complete precipitation (pH ¼ 7–9) was ach-
carbon atoms in product i; M
i
detected, and MCO is the percentage of CO in the syngas feed.
The selectivity of a certain product was calculated based on
P
carbon efficiency using the formula %S ¼ (n C / n C ) ꢂ 100,
i
i
i
i i
ieved. Then, the suspension was aged for 3 h in water bath at
where n
i
and C
i
are the carbon number and molar concentration
ꢁ
7
0 C. Subsequently, the precipitate was ltered and washed
of the product i, respectively. The carbon balance was 100 ꢃ 5%.
several times with distilled water to remove the sodium ions
ꢁ
and then dried in air at 110 C overnight. The dried sample was
ꢁ
2.3. Sample characterization
calcined at 350 C in static air for 4 h to obtain the nal
catalyst.
XRD patterns were recorded on a PANalytical X'Pert instrument
Preparation of Cu–Fe/SiO
2
by solid-state impregnation. For using Ni b-ltered Cu K
a
radiation (l ¼ 0.15418 nm) at 40 kV
ꢁ
ꢁ
the preparation of Cu–Fe/SiO (SI), appropriate amounts of and 40 mA. Two theta angles (2q) ranged from 10 to 70 with
2
ꢁ
ꢀ1
Cu(NO ) $3H O, Fe(NO ) $9H O and SiO were mixed and a scanning rate of 4 min
.
3
2
2
3 3
2
2
ground for 30 min at room temperature. Aer that the mixture
BET specic surface areas (SBET) and pore volumes of the
ꢁ
was dried in air at 110 C overnight, and then the dried sample catalysts were measured by N2 adsorption–desorption
ꢁ
ꢁ
was calcined at 350 C for 4 h in static air to obtain the nal isotherms at ꢀ196 C using a Micromeritics ASAP 2020M + C
catalyst.
Preparation of Cu–Fe/SiO
For the preparation of Cu–Fe/SiO
adsorption apparatus aer degassing the samples under
ꢁ
2
by solid-state chemical reaction. vacuum at 200 C for 6 h. SBET was calculated using a value of
2
2
(SR), appropriate amounts of 0.162 nm for the cross-sectional area of the nitrogen molecule;
Cu(NO ) $3H O, Fe(NO ) $9H O, SiO , and oxalic acid (molar pore volume was determined by BJH adsorption cumulative
3
2
2
3 3
2
2
ratio relative to active metals was 1.5 : 1) were mixed and volume of pores.
ground for 30 min at room temperature. Aer that the sample
The active copper surface areas (SCu) in the reduced catalysts
ꢁ
was dried at 110 C overnight in an electric oven, and then the were determined by the technique of N O reactive frontal
2
ꢁ
ꢁ
dried sample was calcined at 350 C for 4 h in static air.
chromatography at 60 C assuming a Cu : N O ¼ 2 titration
2
1
9
Preparation of Cu–Fe/SiO
preparation of Cu–Fe/SiO
Cu(NO $3H O and Fe(NO
2
by citric acid combustion. For the stoichiometry and a surface atomic density of 1.46 ꢂ 10
2
2
(CC), appropriate amounts of copper atoms per m , respectively.
3
)
2
2
3
)
3
$9H
2
O were dissolved in 10 mL
H
2
2
temperature-programmed reduction (H -TPR) was
distilled water, and the SiO was dispersed in this solution. The carried out in a quartz micro-reactor. Firstly, 0.05 g of the
2
ꢁ
prepared citric acid aqueous solution was added drop wise with prepared catalyst was pre-treated at 300 C in N for 1 h prior to
2
continuous stirring at ambient temperature, till the molar ratio a TPR measurement. During the TPR experiment, H /N
2
2
ꢀ1
of citric acid to active metals of Cu and Fe was 1.2. Then the mixture gas with VH2/VN2 ¼ 1 : 9 was used at 50 mL min and
ꢁ
ꢁ
sample was treated under ultrasonic irradiation to obtain the temperature was ramped from 50 to 500 C at a rate of 10 C
ꢁ
ꢀ1
orange gel. It was evaporated at 350 C in a muffle roaster till it min while the effluent gas was analyzed with a TCD.
ꢁ
combusted; aer that the sample was further calcined at 350 C
for 4 h to yield the nal catalyst.
CO adsorption properties of the catalysts were studied using
a Nicolet 6700 IR spectrometer equipped with a diffuse
55234 | RSC Adv., 2016, 6, 55233–55239
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
reectance infrared Fourier transform (DRIFT) cell with CaF
2
found that the orders of the catalysts as to different reactivity
windows. The sample in the cell was pre-treated in H /N (50 mL parameters are the following:
2
2
ꢀ
ꢀ
1
ꢁ
min , VH2/VN2 ¼ 1 : 9) at 300 C for 1 h, followed by N (50 mL
CO conversion: WI > SR > SI > CC > DP;
ROH selectivity: SI > CC > DP > WI > SR;
2
1
ꢁ
min , Ultrahigh-purity) ushing at 300 C for 0.5 h. Aer the
ꢁ
temperature was dropped to 30 C, the background was scan-
x
CH selectivity: SR > CC > WI z SI > DP;
ned in N
2
ow. Followed by introducing 0.5% CO/N
2
(50 mL
C2+OH selectivity: DP > CC > SI > SR > WI (calculated from
ꢀ
1
min ) into the IR cell, the IR spectrum of CO adsorbed on the ROH selectivity and C2+OH distribution in ROH);
catalyst was recorded at 30 C, when adsorption state remained
steady. The spectral resolution was 4 cm and the number of
scans was 64.
ꢁ
STY of ROH: WI > SI > SR > CC > DP.
Evidently, the performance of the Cu–Fe/SiO catalyst was
2
inuenced greatly by its preparation method and the one
ꢀ
1
Temperature-programmed desorption of adsorbed CO (CO- prepared by ultrasound-assisted wet impregnation showed the
TPD) was carried out in a quartz micro-reactor. The catalyst highest STYROH due to the highest CO conversion, which
ꢁ
ꢀ1 ꢀ1
(
0.1 g) was rstly reduced for 1.5 h at 350 C in H
2
(50 mL reached 104.3 g kgcat
h
and 14.04%, respectively.
ꢀ
1
ꢁ
min ), and then cooled down to 50 C in He ow. The next step
was CO adsorption at 50 C for 30 min until the surface was
saturated. Then the catalyst was swept with He stream for 1 h.
ꢁ
3
.2. Structural and textural properties
Subsequently, the sample was heated in a owing He stream (50 Fig. 2 shows the XRD patterns of the different Cu–Fe/SiO
2
ꢀ1
ꢁ
ꢁ
ꢀ1
ꢁ
mL min ) up to 600 C at a rate of 10 C min . The desorbed catalysts aer calcination at 350 C for 4 h. As shown, the peak
ꢁ
species were detected with a quadrupole mass spectrometer at 2q of 22.5 is attributed to the characteristic diffraction peak
ꢁ
ꢁ
ꢁ
(
4
QMS, Balzers Omnistar 200). MS signals at m/z ¼ 28 (CO) and of amorphous SiO ; the peaks at 2q of 35.6 , 38.8 and 48.7 are
2
4 (CO
2
) were continuously recorded.
attributed to characteristic diffraction peaks of crystal CuO,
X-ray photoelectron spectra (XPS) experiments were tested while that at 2q of 33.4 can be ascribed to characteristic
ꢁ
20,25
on the Thermo Scientic ESCALAB 250Xi spectrometer with an diffraction peak of a-Fe O .
In addition, the peaks at 2q of
spinel
2
3
ꢁ
ꢁ
ꢁ
2 4
Al anode for Ka (1486.6 eV) radiation. Charging effects were 30.1 , 57.0 , and 62.8 are attributed to CuFe O
11,28
corrected by adjusting the binding energy of C1s peak from phase.
carbon contamination to 284.6 eV.
As shown in Fig. 2, the Cu–Fe/SiO (WI) sample only exhibits
2
characteristic diffraction peaks of CuO phase with no Fe
phase observable, indicating amorphous nature or the small
crystallite size of Fe (less than 5 nm). Neither CuO nor
Fe O peaks can be clearly observed on Cu–Fe/SiO (DP) cata-
2 3
O
3
. Result and discussion
2 3
O
3
.1. CO hydrogenation performance of the catalysts
Typical time dependent changes of CO conversion and ROH
selectivity on the representative Cu–Fe/SiO (WI) catalyst are
2
3
2
lyst, suggesting the crystallite sizes of CuO and Fe O are less
2
3
than 5 nm or the active components are amorphous. The
diffraction peaks of CuO over Cu–Fe/SiO (SI) sample are
sharper than those of the other catalysts, indicating the largest
CuO crystallites on Cu–Fe/SiO (SI); the crystallite size of CuO
calculated by Scherrer equation reaches 20.9 nm (Table 2). In
addition, the peak of Fe phase can be clearly observed,
indicating the poor dispersion of Fe O on Cu–Fe/SiO (SI).
2
shown in Fig. 1. It can be seen that both the CO conversion and
ROH selectivity changed greatly during the rst 10 h on stream,
and remained relatively constant aer 15 h on stream. There-
fore, the data taken at 15–24 h on stream were used as indexes
for performance of the catalysts and listed in Table 1. It was
2
2
2 3
O
2
3
2
Compared to Cu–Fe/SiO (SI) catalyst, the characteristic
2
diffraction peak of a-Fe
are wider on Cu–Fe/SiO
CuO and Fe on it. As for Cu–Fe/SiO
peaks of CuO and CuFe are clearly observed. The formation
of CuFe phase can be due to the high combustion
temperature.
2
O
3
disappears and the peaks of CuO
(SR), indicating the smaller sizes of
(CC), the diffraction
2
2
O
3
2
2 4
O
2 4
O
1
7
Specic surface areas and pore volumes of the SiO support
2
and Cu–Fe/SiO catalysts are listed in Table 2. Compared with
2
the support SiO
2 2
, all the Cu–Fe/SiO catalysts show smaller
surface areas and pore volumes, which can be attributed to the
1
9
2
deposition of oxides of Cu and Fe in pores of SiO . Thus, the
better dispersion of oxides of Cu and Fe will result in the larger
surface area and pore volume of the catalysts. Among the
catalysts, Cu–Fe/SiO (DP) catalyst shows the largest surface
2
area and pore volume, due to the smallest sizes of oxide
particles as evidenced by the result of XRD. Similarly, the
smallest surface area of Cu–Fe/SiO (SI) is due to the biggest
2
Fig. 1 CO conversion and alcohol selectivity vs. time-on-stream on
ꢁ
the Cu–Fe/SiO
2
(WI) catalyst. Reaction conditions: 250 C, 3.0 MPa,
ꢀ1
V(H
2
)/V(CO) ¼ 2, GHSV ¼ 6000 mL (g h)
.
sizes of CuO and Fe
O
particles. Combining the textural
2
3
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 55233–55239 | 55235

View Article Online
RSC Advances
Paper
d
2
Table 1 Catalytic performance of Cu–Fe/SiO catalysts prepared by different methods in CO hydrogenation
Selectivity (%)
CO
Alcohol distribution (%)
CO conv.
(%)
a
x
b
c
catꢀ1 ꢀ1
Catalyst code
2
CH
ROH
STYROH (g kg
h
)
CH
3
OH
C2+OH
WI
DP
SI
SR
CC
14.0
5.6
10.2
12.4
7.4
18.2
18.1
13.5
16.1
12.2
55.0
51.7
54.7
61.9
56.9
26.8
30.2
31.8
22.0
31.0
104.3
47.0
89.7
75.5
63.3
68.4
56.3
65.9
60.6
62.4
31.6
43.7
34.1
39.4
37.6
a
b
c
d
ꢁ
CH
x
represents hydrocarbons. ROH represents alcohols. STYROH represents space time yield of alcohols. Reaction conditions: 250 C, 3.0
ꢀ1
MPa, VH2/VCO ¼ 2, SV ¼ 6000 mL (gcat h)
.
In addition, the active copper surface areas (SCu) of the Cu–
Fe/SiO catalysts aer in situ reduction are also shown in Table
2
2
. Generally, smaller size of CuO particles contributes to larger
active copper surface area. However, the CuO particles may
2
accumulate in the pores of SiO , difficult to be reduced by H ,
2
26
which leads to the smaller Cu surface area. The combined
effects of these two factors account for the difference of SCu of
these Cu–Fe/SiO catalysts. As well known, the catalyst with
2
larger SCu possesses more active copper species, which is
benecial to the improvement of the activity for hydrogenation.
Thus, the order of SCu (WI > SR > SI ¼ CC > DP) of these Cu–Fe/
2
SiO catalysts is fairly consistent with that of CO conversion (WI
>
SR > SI > CC > DP). Similar result was also reported by Mahdavi
et al. who found that the activity of Cu–Co /ZnO catalyst
2 3
O
30
increased linearly with SCu
.
3.3. Reducibility of the catalysts
2
Fig. 2 XRD patterns of Cu–Fe/SiO catalyst prepared by different
methods.
2 2
Fig. 3 displays the H -TPR proles of the Cu–Fe/SiO catalysts.
For the Cu–Fe/SiO
observed at 213 C and 230 C, corresponding to the reduction
2
(WI) catalyst, two peaks (a and b) could be
ꢁ
ꢁ
16,26
parameters (Table 2) and catalytic performance (Table 1) of the of highly dispersed and bulk CuO respectively.
catalysts, it can be inferred that the surface area of catalyst is broad peak (g) detected at 300–500 C is attributed to the
Moreover, the
ꢁ
26,31
not the main factor affecting the CO conversion and ROH stepwise reduction of Fe
O
3
(Fe
(WI), the temperature of Fe
higher distribution of C OH alcohols of Cu–Fe/SiO (DP) and reduction peak of Cu–Fe/SiO (SI) and Cu–Fe/SiO (CC) shis
(SR) can be attributed to the larger pore volume, toward higher temperatures, indicating that the reduction of
since the larger pore volume is benecial to diffusion of reac- Fe on these two catalysts is more difficult; the reason can
tant molecules in the surface of active site, thereby promoting respectively be attributed to the larger Fe particles and the
(ref. 32) as evidenced by the above XRD
O / Fe O / FeO / Fe).
2 3 3 4
2
selectivity of the Cu–Fe/SiO catalyst. On the other hand, the Compared with Cu–Fe/SiO
O
2 3
2
2
2
+
2
2
2
Cu–Fe/SiO
2
2 3
O
2 3
O
29
the formation of long-chain alcohols.
formation of CuFe
result.
2 4
O
Table 2 Physicochemical properties of the SiO
2
2
support and Cu–Fe/SiO catalysts
a
2
ꢀ1
b
3
ꢀ1
c
d
2
ꢀ1
Sample
SiO
Cu–Fe/SiO
Cu–Fe/SiO
Cu–Fe/SiO
Cu–Fe/SiO
Cu–Fe/SiO
S
BET (m g
)
V
p
(cm g
)
d
CuO (nm)
S
Cu (m g
)
2
261.4
203.5
245.9
192.4
227.9
199.2
0.89
0.53
0.58
0.54
0.58
0.51
—
13.1
—
20.9
9.9
14.1
—
2
2
2
2
2
(WI)
(DP)
(SI)
(SR)
(CC)
4.4
2.6
3.0
3.2
3.0
a
b
c
The error of SBET is ꢃ1%. The error of V
p
is ꢃ2%. Calculated from the CuO(111) peak of the XRD spectra according to the Scherrer equation.
d
Determined by the technique of N
2
O reactive frontal chromatography.
55236 | RSC Adv., 2016, 6, 55233–55239
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
Fig. 3
H
2
-TPR curves of the different Cu–Fe/SiO
2
catalysts.
Fig. 4 CO
2
-TPD profiles of Cu–Fe/SiO
2
catalyst prepared by different
methods.
The peak areas of the reduction peaks of the active constit-
3.5. Infrared spectra of CO adsorption
uents are shown in Table 3. The Cu–Fe/SiO (DP) and Cu–Fe/
2
SiO (CC) catalysts have smaller areas of the reduction peaks of IR spectra of CO adsorption on the different Cu–Fe/SiO cata-
2
2
ꢀ1
active species compared with the other catalysts, indicating the lysts are shown in Fig. 5. Apparently, one peak at ꢄ2120 cm is
lower contents of active metals on the surface of catalysts. This observed on Cu–Fe/SiO (WI), corresponding to the CO linearly
2
36
is responsible for the lower CO conversion of Cu–Fe/SiO
and Cu–Fe/SiO (CC).
2
2
(DP) absorbed on copper species. The intensity of the peak
increased in the order of DP < SI < CC < SR < WI, which is the
same order of CO conversion on them except the catalyst Cu–
33
Fe/SiO
2
(CC). The unexpected low activity of the catalyst Cu–Fe/
3
.4. CO
2
-TPD aer CO adsorption
SiO (CC) can be attributed to the weak bond of Cu–CO, which
2
Fig. 4 shows the CO -TPD proles aer CO adsorption on the
Cu–Fe/SiO
peaks of CO
2
can be evidenced by its position at higher wavenumber of 2127
2
catalysts prepared by different methods. Several
desorption are found on all the catalysts, indi-
ꢀ1
3
7,38
cm
.
2
cating that there are several active sites for CO dissociation
adsorption. It perhaps took place the CO dissociation or the
disproportionation reaction (2CO / C + CO
2
) on the catalyst
34
surface. The peak of CO
2
desorption at low temperature
represents the strong active site for CO dissociation adsorption.
Therefore, the larger areas of the peaks at low temperature
suggest more amounts of strong active sites, thus resulting in
3
5
higher activity of the catalyst. In this work, the area of CO
2
ꢁ
desorption peak at low temperature (<175 C) follows the order
of WI > SR > SI > CC > DP, which is in good agreement with that
of CO conversion as shown in Table 1.
Table 3 The temperatures and areas of reduction peaks for different
Cu–Fe/SiO catalysts
2
ꢁ
Temperature ( C)
Area (a.u.)
Catalyst
a
b
g
a + b
g
Cu–Fe/SiO
Cu–Fe/SiO
Cu–Fe/SiO
Cu–Fe/SiO
Cu–Fe/SiO
2
2
2
2
2
(WI)
(DP)
(SI)
(SR)
(CC)
213
210
215
185
208
230
227
235
230
235
384
482
432
381
432
829.4
727.5
917.3
870.2
615.4
683.1
484.6
771.5
711.9 Fig. 5 IR spectra of CO chemisorbed on different Cu–Fe/SiO
627.6 catalysts.
2
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 55233–55239 | 55237

View Article Online
RSC Advances
Paper
Table 4 Binding energies of core electrons of the different Cu–Fe/
SiO catalysts
2
Binding energy (eV)
Cu2p3/2
Catalyst
Fe2p3/2
Cu–Fe/SiO
Cu–Fe/SiO
Cu–Fe/SiO
Cu–Fe/SiO
Cu–Fe/SiO
2
2
2
2
2
(WI)
(DP)
(SI)
(SR)
(CC)
933.3
934.0
933.5
933.4
933.5
710.4
710.1
711.3
711.1
711.0
16,21
other catalysts.
It is well known that the copper has the
function of methanol synthesis and the iron element plays an
important role in promoting the chain growth. Therefore, the
synergism between copper and iron benets the formation of
27,28
higher alcohols.
Accordingly, the strong interaction between
copper and iron on the Cu–Fe/SiO (DP) catalyst can account for
2
39,40
its highest distribution of C OH alcohols (Table 1).
2
+
4. Conclusions
2
Five Cu–Fe/SiO catalysts were prepared by ultrasound-assisted
wet impregnation, deposition–precipitation, solid state
impregnation, solid-state chemical reaction, citric acid
combustion, and the inuence of preparation methods on the
physicochemical and catalytic properties for higher alcohols
synthesis by CO hydrogenation were investigated. The results
indicate that the catalyst prepared by ultrasound-assisted wet
impregnation has higher dispersion of active components (Cu,
Fe), the maximum of SCu, higher reducibility, and the strongest
CO adsorption capacity and dissociation capability, leading to
the highest CO conversion and STY of alcohols.
Acknowledgements
Financial supports provided by the Science and Technology
Commission of Shanghai Municipality (13ZR1461900,
2
Fig. 6 XPS spectra of the different Cu–Fe/SiO catalysts: (a) Cu2p; (b)
Fe2p.
08520513600) are gratefully acknowledged.
Notes and references
3
.6. XPS study
The XPS spectra for Cu2p and Fe2p of CuO, Fe
the decomposition of Cu(NO ) $3H O and Fe(NO ) $9H O,
2
O
3
(derived from
1 Z. Wang and J. J. Spivey, Appl. Catal., A, 2015, 507, 75.
2 K. G. Fang, D. B. Li, M. G. Lin, M. L. Xiang, W. Wei and
Y. H. Sun, Catal. Today, 2009, 147, 133.
3 M. Gupta, M. L. Smith and J. J. Spivey, ACS Catal., 2011, 1,
641.
3
2
2
3 3
2
respectively) and the different Cu–Fe/SiO2 catalysts are dis-
played in Fig. 6. For the CuO sample, the Cu2p3/2 and Cu2p1/2
binding energies at 933.5 and 953.4 eV, could be ascribed to the
presence of Cu species. For the Fe
Fe2p1/2 binding energies at 711.0 and 723.6 eV, could be
2
+
2
O
3
sample, the Fe2p3/2 and
4 V. R. Surisetty, A. K. Dalai and J. Kozinski, Appl. Catal., A,
2011, 404, 1.
3+
21
ascribed to the presence of Fe .
5 J. J. Spivey and A. Egbebi, Chem. Soc. Rev., 2007, 36, 1514.
6 V. Subramani and S. K. Gangwal, Energy Fuels, 2008, 22, 814.
7 M. Gupta and J. J. Spivey, Catal. Today, 2009, 147, 126.
8 X. H. Mo, Y. T. Tsai, J. Gao, D. S. Mao and J. G. Goodwin Jr, J.
Catal., 2012, 285, 208.
9 G. Prieto, S. Beijer, M. L. Smith, M. He, Y. Au, Z. Wang,
D. A. Bruce, K. P. de Jong, J. J. Spivey and P. E. de Jongh,
Angew. Chem., Int. Ed., 2014, 53, 6397.
The binding energies of Cu2p3/2 and Fe2p3/2 for the different
Cu–Fe/SiO catalysts are summarized in Table 4. It can be seen
that the Cu2p3/2 binding energy of the Cu–Fe/SiO (DP) catalyst
is the largest while its Fe2p3/2 binding energy is the smallest.
This result indicates that the interaction between copper and
2
iron on the Cu–Fe/SiO (DP) catalyst is stronger than that on the
2
2
55238 | RSC Adv., 2016, 6, 55233–55239
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
RSC Advances
10 W. Gao, Y. Zhao, H. Chen, H. Chen, Y. Li, S. He, Y. Zhang, 25 M. Y. Ding, J. L. Tu, J. G. Liu, N. Tsubaki, T. J. Wang and
M. Wei, D. G. Evans and X. Duan, Green Chem., 2015, 17,
525.
1 K. Xiao, X. Qi, Z. Bao, X. Wang, L. Zhong, K. Fang, M. Lin and
L. L. Ma, Catal. Today, 2014, 234, 278.
26 R. L. Lu, D. S. Mao, J. Yu and Q. S. Guo, J. Ind. Eng. Chem.,
2014, 25, 338.
1
1
1
1
1
1
Y. Sun, Catal. Sci. Technol., 2013, 3, 1591.
2 Y. Z. Fang, Y. Liu and L. H. Zhang, Appl. Catal., A, 2011, 397,
27 Y. Lu, B. Cao, F. Yu, J. Liu, Z. Bao and J. Gao, ChemCatChem,
2014, 6, 473.
28 H. Guo, H. Zhang, F. Peng, H. Yang, L. Xiong, C. Wang,
C. Huang, X. Chen and L. Ma, Appl. Catal., A, 2015, 503, 51.
29 H. Ran, K. Fang, M. Lin and Y. Sun, Nat. Gas Chem. Ind.,
2010, 35, 1.
30 V. Mahdavi, M. H. Peyrovi, M. Islami and J. Y. Mehr, Appl.
Catal., A, 2005, 281, 259.
31 C. H. Zhang, Y. Yang, B. T. Teng, T. Z. Li, H. Y. Zheng,
H. W. Xiang and Y. W. Li, J. Catal., 2006, 237, 405.
1
83.
3 Y. W. Lu, F. Yu, J. Hu and J. Liu, Appl. Catal., A, 2012, 429–
30, 48.
4 M. Lin, K. Fang, D. Li and Y. Sun, Catal. Commun., 2008, 9,
869.
4
1
5 X. M. Yang, Y. Wei, Y. L. Su and L. P. Zhou, Fuel Process.
Technol., 2010, 91, 1168.
16 H. Zhang, W. Chu, H. Y. Xu and J. Zhou, Fuel, 2010, 89, 3127.
17 D. S. Mao, Q. S. Guo, J. Yu, L. P. Han and G. Z. Lu, Acta Phys.– 32 J. Xu and W. Wang, Chin. J. Catal., 1992, 13, 420.
Chim. Sin., 2011, 27, 2639.
33 H. Xu, W. Chu, L. Shi, H. Zhang and S. Deng, J. Fuel Chem.
Technol., 2009, 37, 212.
34 C. Mazzocchia, P. Gronchi, A. Kaddouri, E. Tempesti,
L. Zanderighi and A. Kiennemann, J. Mol. Catal. A: Chem.,
2001, 165, 219.
1
1
2
2
2
8 M. Y. Ding, J. G. Liu, Q. Zhang, N. Tsubaki, T. J. Wang and
L. L. Ma, Catal. Commun., 2012, 28, 138.
9 Q. S. Guo, D. S. Mao, J. Yu and L. P. Han, J. Fuel Chem.
Technol., 2012, 40, 1103.
0 J. G. Liu, M. Y. Ding, T. J. Wang and L. L. Ma, Acta Phys.- 35 A. Kiennemann, A. Barama, S. Boujana and M. M. Bettahar,
Chim. Sin., 2012, 28, 1964. Appl. Catal., A, 1993, 99, 175.
1 W. Gao, Y. F. Zhao, J. M. Liu, Q. W. Huang, S. He, C. M. Li, 36 R. Xu, Z. Y. Ma, C. Yang, W. Wei, W. H. Li and Y. H. Sun, J.
J. W. Zhao and M. Wei, Catal. Sci. Technol., 2013, 3, 1324. Mol. Catal. A: Chem., 2004, 218, 133.
2 Z. H. Bao, K. Xiao, X. Z. Qi, X. X. Wang, L. S. Zhong, 37 R. Xu, Z. Y. Ma, C. Yang, W. Wei and Y. H. Sun, React. Kinet.
K. G. Fang, M. G. Lin and Y. H. Sun, J. Energy Chem., 2013,
2, 107.
Catal. Lett., 2004, 81, 91.
38 G. Blyholder and M. Lawless, Surf. Sci., 1993, 290, 155.
2
2
3 J. G. Liu, M. Y. Ding, T. J. Wang and L. L. Ma, Pet. Process. 39 R. Xu, C. Yang, W. Wei, W. Li, Y. Sun and T. Hu, J. Mol. Catal.
Petrochem., 2013, 44, 22. A: Chem., 2004, 221, 51.
4 S. Kiatphuengporn, M. Chareonpanich and J. Limtrakul, 40 R. Xu, W. Wei, W. H. Li, T. D. Hu and Y. H. Sun, J. Mol. Catal.
Chem. Eng. J., 2014, 240, 527. A: Chem., 2005, 234, 75.
2
This journal is © The Royal Society of Chemistry 2016
RSC Adv., 2016, 6, 55233–55239 | 55239