Improvement of catalytic stability and carbon

i n t e r n a t i o n a l j o u r n a l o f h y d r o g e n e n e r g y x x x ( 2 0 1 8 ) 1 e1 1

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.elsevier.com/locate/he

Improvement of catalytic stability and carbon resistance in the process of CO2 reforming of methane by CoAl and CoFe hydrotalcite-derived catalysts N. Aider a,b, F. Touahra b,c,*, F. Bali b, B. Djebarri b, D. Lerari c, K. Bachari c, D. Halliche b Departement de Chimie, Faculte des Sciences (UMMTO), Tizi-ouzou, Algeria Laboratory of Natural Gas Chemistry, Faculty of Chemistry (USTHB), BP, 32, 16111, Algiers, Algeria c Centre de Recherche Scientifique et Technique en Analyses Physico-chimiques (CRAPC), BP 384-Bou-Ismail, RP42004, Tipaza, Algeria a

b

article info

abstract

Article history:

The CoAl-LDH and CoFe-LDH hydrotalcite-like compounds (LDH ¼ Layered Double Hy-

Received 14 January 2018

droxides named also hydrotalcite), were successfully synthesized following co-

Received in revised form

precipitation method at pH ¼ 12. A several characterization techniques including TGA,

24 February 2018

ICP, XRD, N2 adsorption and desorption, H2-TPR, O2-TPO, SEM-EDX and TEM, were utilized

Accepted 17 March 2018

to determine the structure function relationship for the obtained catalysts. These catalysts

Available online xxx

were evaluated in CO2 reforming of methane under continuous flow with CH4/CO2 ration equal to 1, at atmospheric pressure and a temperature range between 400 and 700 C.

Keywords:

The iron addition to the cobalt showed improved resistance to coke deposition while a

Hydrotalcite

slight decrease in methane conversion was observed compared to CoAlcal-R catalyst

g-CoFe alloy

(cal ¼ after calcination and R ¼ after reduction) derived from CoAl-LDH precursor.

Carbone deposition

© 2018 Published by Elsevier Ltd on behalf of Hydrogen Energy Publications LLC.

Sintering Methane dry reforming

Introduction Production of hydrogen and synthesis gas is conventionally produced via the process of steam reforming reaction (Eq. (1)). This reaction has been established and already widely commercialized. The target several recent studies on natural gas reforming is shifted to the catalyst development for carbon dioxide reforming also named dry reforming of methane

(DRM; Eq. (2)). So far there is no established commercially industrial process due to problem associated with this reaction comes in the form of sintering of the active phase and carbon formation [1]. This latter could be formed by direct decomposition of methane (Eq. (3)); inverse boudouard reaction (Eq. (4)) or following the direct reduction of CO with H2 (Eq. (5)) [2]. The reverse water-gas shift reaction (RWGS; Eq. (6)) can also occurring during DRM.

* Corresponding author. Laboratory of Natural Gas Chemistry, Faculty of Chemistry (USTHB), BP, 32, 16111, Algiers, Algeria. E-mail address: [email protected] (F. Touahra). https://doi.org/10.1016/j.ijhydene.2018.03.118 0360-3199/© 2018 Published by Elsevier Ltd on behalf of Hydrogen Energy Publications LLC. Please cite this article in press as: Aider N, et al., Improvement of catalytic stability and carbon resistance in the process of CO2 reforming of methane by CoAl and CoFe hydrotalcite-derived catalysts, International Journal of Hydrogen Energy (2018), https://doi.org/ 10.1016/j.ijhydene.2018.03.118

2


CH4 þ H2 O4CO þ 3H2

(1)

CH4 þ CO2 42CO þ 2H2

(2)

CH4 4C þ 2H2

(3)

2CO4C þ CO2

(4)

H2 þ CO 4C þ H2 O

(5)

CO2 þ H2 4CO þ H2 O

(6)

Wang et al. [3], inferred from the values of standard free energies to calculate the minimum operating temperatures for CO2 reforming of CH4, decomposition of methane, inverse boudouard reaction and reaction reverse water-gas shift (RWGS). DRM proceeds in the forward direction above 640 C. Similarly, carbon formation can occur due to decomposition of methane above 557 C and due to inverse boudouard reaction below 700 C. Thus, they suggested that carbon formation can occur due to both decomposition of methane and inverse boudouard reaction in 557e700 C range. Ni or Co-based catalysts have attracted much attention in recent years for this reaction, but excessive deposition of carbon deactivates Ni or Co catalysts [4]. Many attempts have been made to solve these problems; among those of prime focus were noble metal systems which were found to produce most promising results [5e7]. On the other hand, the stability of nickel catalyst may be enhanced by introduction of support or promoters with basic and/or redox properties [8] also derived from layered double hydroxide (LDHs) have shown an enhanced catalytic performance and excellent resistance to sintering and coking compared to catalysts obtained by conventional techniques [9e12]. Layered double hydroxides (LDHs), also called anionic clays, are one of the most studied layered materials [13]. LDHs materials are quite rare in nature but relatively very inexpensive and simple to synthesize in the laboratory. Zhang et al. [9] investigated dry reforming of methane over Ni/MgO/g-Al2O3 catalyst prepared by impregnated method and NieMgeAl-LDHs/g-Al2O3 catalyst derived from layered double hydroxides (LDHs). The catalyst derived from LDHs showed excellent catalytic performance compared with a reference catalyst of Ni/MgO/g-Al2O3 prepared by impregnation. In recent years, the application of bimetallic catalysts in DMR, have been attracted a great deal of attention. San-Jose Alonso et al. [14] Zhang et al. [15] and Wang et al. [16] found that bimetallic CoeNi catalysts with the highest cobalt content are the most active and stable for CH4 reforming with CO2, but they produce a large amount of carbon. Tsoukalou et al. [17] also investigated dry reforming of methane with Nibased bi-metallic catalysts derived via the reduction of perovskite precursors, i.e. LaNi0.8M0.2O3 (M ¼ Ni, Co and Fe). NieCo catalyst showed excellent catalytic properties in dry reforming of methane, whereas NieFe showed no activity. The authors observed that the addition of Co increases the rate of La2O2CO3 formation, which in turn enhances the removal of carbonaceous deposits from neighbouring active sites, thus, leading to a catalyst with an increased stability and activity. The poor activity of NieFe was explained by the encapsulation of active Ni particle by LaFeO3.

Thus in this present work, the hydrotalcite-derived mixed oxides based Co, Al and Fe were tested in DRM reaction an aim to minimize carbon deposition and sintering.

Experimental Catalyst synthesis CoAl-LDH and CoFe-LDH solids were synthesized easily, according to the co-precipitation method described elsewhere [18]. The mixture of metals nitrate solution were added dropwise to a vigorously stirring solution of NaOH (2 M) and Na2CO3 (0.4 M) at room temperature while the pH was maintained constant at 12. The precipitate is brought to reflux at T ¼ 70 C, with stirring for 15 h. The recovered solid is finally dried in an oven at 100 C overnight and finally crushed until a homogeneous powder is obtained. These compounds are then calcined at 800 C in an oven for 6 h with a temperature rise of 5 C/min, after which the compounds are named M2þ M3 þcal.

Material characterization Thermogravimetric analysis TG-DTA, was carried out with Thermal Analyst Setaram Set Sys 16/18 in the presence of air, from ambient temperatures to 900 C using a slower heating rate of 10 C per minute. Inductively coupled plasma emission spectrometer (ICP-ES) was employed to the determination of elemental chemical analysis of each solids, using Thermo Jarrel Ash ICAP 957. Before to each analysis, 100 mg of the solids were dissolved in 20 mL of aqua regia. The solution was diluted to 100 ml with distilled water. A small volume of ample was subjected to ICP-AES analyses. The specific surface areas (BET) were determined by nitrogen adsorptionedesorption isotherms using Micromeritics Tristar 3000 at 196 C. XRD patterns of all samples were performed with BRUKER D8 Advance diffractometer equipped with Cu-Ka radiation (l ¼ 1.54056A); the data were collected in the 2q range between 2 and 80 . The average particle size dhkl of Co0 following reduction estimated from Scherrer's formula (Eq. (7)) [19,20]: dhkl ¼

k:l bhkl :cos q

(7)

where l is the X-ray wavelength, qhkl is the Bragg's diffraction angle and bhkl is the full-width at half-maximum in radian. Scanning electron microscopy (SEM) and Transmission Electron Microscopy (TEM) were using to investigate the metal particle size distribution and morphology of the carbon deposition on catalysts after reaction. SEM 3D images were effected using Quanta 250 and TEM images were taken by JEOL-JEM-1200EX at 100 kV. Hydrogen temperature programmed reduction H2-TPR and O2-TPO were carried out in a quartz U-tube (ID ¼ 4 mm) in the presence of 50 mg of the catalyst. In first, the catalyst flushed with helium for 1 h at 200 C followed by cooling to room temperature in helium. After flushed, the He was replaced by 5%H2/Ar in case of H2TPR and was replaced by 3% O2/He in case of O2-TPO and the temperature was then raised to 900 C.

Please cite this article in press as: Aider N, et al., Improvement of catalytic stability and carbon resistance in the process of CO2 reforming of methane by CoAl and CoFe hydrotalcite-derived catalysts, International Journal of Hydrogen Energy (2018), https://doi.org/ 10.1016/j.ijhydene.2018.03.118

3


Catalytic activity Catalytic tests were performed under atmospheric pressure in a fixed-bed tubular quartz reactor (ID ¼ 6 mm, L ¼ 16 cm) and place in furnace. Prior to reaction, 100 mg of the catalyst was reduced in-situ under constant hydrogen flow at 750 C for 1 h. After reduction, the temperature was lowered under argon to the initial reaction's temperature and a feed gas mixture containing CH4:CO2:Ar in a ratio of 20:20:60. The total flow rate was 20 mL min1. The reaction products were analyzed using gas chromatograph (Delsi), equipped with a thermal conductivity detector (TCD) and carbosieve B column, in the presence of carrier gas, argon.

Results and discussion

The results obtained for the chemical composition and the structural formula of the samples are summarized in Table 2. The results of the elemental analysis by ICP (Inductively coupled plasma) show the presence of minor anomalies is mainly associated with deficiencies in the total incorporation of the Fe cation within the brucite-like layer.

BET surface area analyses Table 2 shows the specific surface area values (SBET) of the CoAl-LDH and CoFe-LDH solids before calcination and after calcination at 800 C. This results show that the specific surface area of CoFe-LDH solid is lower than CoAl-LDH. After calcination, the sample exhibited relatively high BET surface area compared to uncalcined solids (Table 2) associated with the removal of water, nitrate and carbonate from the interlayer space of the hydrotalcite structures.

Thermal analysis (TG/DTG) X-ray diffraction The thermal decomposition of the hydrotalcite samples was studied by conventional thermal methods (TGA/DTA). The TGA/DTA curves of synthesized samples are reported in Fig. S1. According to the literature [19,21e23], the thermal evolution of hydrotalcite solids in the carbonate form present three well-defined weight losses. The first peak is associated with the elimination of water, physically adsorbed on the external surface of the crystallite [24,25]. The second loss at around 250 C accounting for 25%, due to dehydroxylation and nitrate removal in agreement with literature [25]. Finally, the last loss of mass is characteristic of the departure of the carbonates in CO2 form [26]. TGA values of analyzed samples are shown in Table 1. In addition, it observed shift in last peak position towards high temperatures, when Al3þ is completely substituted by Fe3þ. This shows that the thermal stability of the hydrotalcites changes as a function of the nature of the cations existing in the layers. The strong interactions between the metal and the carbonates, explain this result. As a result, CoFe-LDH solid are more thermally stable than CoAl-LDH solid due to the higher af2 finity between Fe and CO2 3 than Al and CO3 .

Chemical analyses Hydrotalcite-type clay can be expressed by the following formula: mþ 3þ x M2þ n Mm ðOHÞ2ðnþmÞ ; Am=x :y H2 O

where: M2þand M3þare di- and trivalent cations respectively, Ax is the interlayer anion, x ¼ charge of the anion, n > m, and y ¼ number of water molecules of the interlayer layer [27,28].

Fig. 1 (A) show the XRD patterns of precursors catalyst. X-ray diffraction of CoAl-LDH and CoFe-LDH showed the presence of crystalline phases of hydrotalcite-like materials; sharp and symmetric reflections for (003), (006), (110) and (113) planes and broad and asymmetric peaks for (012), (015) and (018) respectively, which correspond respectively to one another in a hexagonal mesh at the diffraction angles 2q ¼ 11.6 ; 23.5 ; 35.3 ; 39.7 ; 47.2 ; 61.4 ; 62.8 and 66.4 . After calcination at 800 C, the layered structures were completely destroyed and only the presence of mixed oxides phase was detected (Fig. 1 (B)). The XRD patterns of CoAlcal solid show peaks of the spinel phases CoAl2O4 [JCPDS file No. 44-0160], Co3O4 [JCPDS file No. 43-1003] and Co2AlO4 [JCPDS file No. 38-0814]. Consequently, an amorphous aluminum oxide Al2O3 phase should be also formed (not detected in XRD). The spectrum of CoFecal sample, depicted two series of broad peaks corresponding to: CoFe2O4 [JCPDS file No. 22-1086] and Co3O4. These results is good agreement with the results obtained by several researchers [29e32].

Reducibility of catalysts by H2-TPR and H2-XRD The reduction behavior of CoAlcal and CoFecal catalysts was studied by H2-TPR and XRD after reduction under H2 atmosphere at variable temperature. H2-XRD was undertaken to interpret the results of the analysis by reducibility of CoAlcal and CoFecal by programmed temperature reduction (RTP). The H2-XRD patterns obtained are represented in Fig. 2. The reduced solids under hydrogen flow are noted below CoAlcal-R and CoFecal-R. It can be noted that the reduction of CoAlcal-R and CoFecal-R solids leads to the formation of following phases:

Table 1 e TG results of two synthesized materials. Samples

First weight loss (%)

Temperature ( C)

Second weight loss (%)

Temperature ( C)

Third weight loss (%)

Temperature ( C)

CoAl-HDL CoFe-HDL

7 13

90 129

25 15

220 207

60 5

310 609


4


Table 2 e Chemical composition, specific surfaces and particle size (nm) of the reduced samples. Samples

Proposed formula

Molar ratio (M2þ/M3þ)

SBET (m2/g)

particle size (nm)

Precursor Calcined reduced CoAl-HDL [Co0.68 Al0.32(OH)2] [(NO3)0$014(CO3)0.187, 0.40 H2O] CoFe-HDL [Co0.69 Fe0.31(OH)2] [(NO3)0$01(CO3)0.182, 0.85 H2O] a b

2.19 2.22

87 43

116 82

87 50

a

(Co0) 18 17

b

(Co0) 17 19

Calculated from the XRD (Fig. 5). Calculated from SEM (Fig. 6).

In the case of CoAlcal-R solid, at 400 C and 500 C, the X-ray diffractograms (Fig. 2 (A)) show the appearance of the CoO peaks [JCPDS file No 43-1004] located at 2q ¼ 36.5 ; 42.4 ; 61.5 ; 73.7 and 77.5 which is explained by the reduction of Co3O4 to CoO. At a temperature of 600 C, the onset of the appearance of the Co0 metal phases is observed [JCPDS file No15-0806]. For the CoFecal-R sample, at 400 C, Fig. 2 (B) show peak corresponding to CoFe2O4 and Co3O4 and other peaks consistent with CoO phase formation due to partial reduction of Co3O4 to CoO. The beginning of the

Fig. 1 e X-ray diffractograms of (A) CoAl-LDH and CoFe-LDH and (B) calcined solids at 800 C. hydrotalcite, CoAl2O4 or Co3O4 or Co2AlO4 and CoFe2O4.

destruction of the CoFe2O4 spinel structure under hydrogen flow occurs around 500 C, it leads to the formation of the Fe2O3 phase. At this same temperature, we also notice the formation of Co0. At 700 C, the simultaneous appearance of the Co0 and Fe0 (2q ¼ 44.8 and 65.2 ) lines attributed to the g-CoFe alloy [JCPDS file No 44-1433]. Fig. 3 depicts the temperature-programmed reduction (H2TPR) analysis of CoAlcal-R and CoFecal-R. The H2-TPR profile of the CoAlcal-R samples shows two reduction peaks. The first peak observed at 380 C can be assigned to the reduction of Co3O4 spinel in CoO (Eq. (8)), while the second observed at 710 C is attributed reduction of CoO species to Co0 (Eq. (9)), and/or reduction of Co2þ in CoAl2O4 spinel or Co2AlO4,

Fig. 2 e X-ray diffractograms of the solids: (A) CoAlcal-R and (B) CoFecal-R reduced at different temperatures.



5

Fig. 4 e RX diffractograms of reduced solids at 750 C; CoO, Co0, ↑Fe0 and g-CoFe.

Fig. 3 e Reduction temperature profiles (5% H2/Ar) of the solids calcined at 800 C: CoAlcal-R and CoFecal-R.

according to some studies [33e36]. In our case, the results of DRX after reduction do not show any characteristic spinel peak after reduction between 400 C and 800 C (see Fig. 3) and other results of the literature show that the reduction of Co2þ in CoAl2O4 can not be total at 700 C [33,35,36]. According to our results, the phase obtained after calcination at 800 C of the CoAl-LDH solid is the Co3O4 spinel. 2 2þ 2 Co2þ Co3þ þ H2 O 2 O4 þ H2 /3Co O

(8)

Co2þ O þ H2 /Co0 þ H2 O

(9)

SEM & TEM analyses after reduction Fig. 5 shows SEM images of CoAlcal-R and CoFecal-R. TEM image of CoFecal-R is also presented in the same figure. SEM images of CoAlcal-R revealed the homogeneous distribution of Co0 particles on the surface of the catalyst. The Co0 particles

For the CoFecal-R sample, the H2-RTP profile is distinguished by three reduction peaks located at about 380 C, 640 C and 730 C. The first peak may be associated with the reduction of Co3O4 spinel to CoO oxide, the second and third peaks can be attributed to the reduction of Fe3þ to Fe0, accompanied by reduction of Co2þ to Co0, These results are consistent with those of DRX (Fig. 2) and are in perfect agreement with those already reported in the literature [37,38].

X-ray diffraction after reduction Fig. 4 presents XRD patterns of the two catalysts following H2 reduction at 750 C over the course of 1 h. In case of CoAlcal-R catalyst, the spectra showed the presence of species Co0 and CoO. The presence of Co0, Fe0 and g-CoFe alloy were observed in case of CoFecal-R catalyst. After reduction, the average particle size of Co0 are estimated from X-ray diffractograms (Fig. 5), using the DebyeScherrer relation. The average particle size Co0 in the CoAlcal and CoFecal catalysts is 18 and 17 nm, respectively (Table 2).

BET surface of reduced samples The surface areas of the reduced catalysts are summarized in Table 2. The BET surface area of two catalysts was decreased in approximately 20% after reduction in the case of the CoAlcal-R and decrease by 30% in the case of CoFecal-R catalyst.

Fig. 5 e SEM for (A): CoAlcal-R and (B): CoFecal-R. (Reduced to 750 C for 1 h).


6


size measurement is 17 nm, as compared to that of 18 nm for that calculated by X-ray (Table 2). The CoFecal-R catalyst SEM image shows a less uniform appearance than the previous one. It is reveal the presence of small particles of spherical shape corresponding to the Co0 and Fe0 nanoparticles and large agglomerates corresponding to the g-CoFe alloy of different shapes. The MET (Fig. 6) images and EDX (Fig. S2) confirm the formation of g-CoFe alloy. The elemental distribution of g-CoFe alloy formation after H2-TPR process was investigated using energy-dispersive Xray spectroscopy (EDX)-STEM mapping (Fig. 7). Both Co (pink) and Fe (red) elements were distributed uniformly in the sample after reduction, implying the alloy formation.

Catalytic activity tests Activity as a function of reaction temperature The catalytic performances of CoAlcal-R and CoFecal-R has been tested in CO2 reforming of methane reaction under identical experimental conditions. The results obtained for variations of CH4 and CO2 conversions and H2/CO ratio in the range 400e700 C are presented in Fig. 8. Conversions of CH4, CO2, H2/CO selectivity and carbon balance (%) at 400 and 700 C are summarized in Table 3. Initially, CoAlcal-R and CoFecal-R catalysts were active towards the reforming of methane reaction, the conversion of CH4 and CO2 increased rapidly with increasing reaction temperature is in agreement with the thermodynamicity of the process (endothermic reaction) [39,40]. CH4 conversion increased from 10.5% and 5.1% at 400 C to 66.4% and 54.5% at 700 C for CoAlcal-R and CoFecal-R respectively (Table 3). Theoretically, conversions of CH4 and CO2 must therefore be equimolar, thus the H2/CO ratio equal to 1 should be the preferred condition for dry reforming.

Fig. 6 e MET for CoFecal-R.

Fig. 8 e (A) CH4 conversion, (B) CO2 conversion and (C) H2/ CO ratio obtained during DRM.

Fig. 7 e Maps showing the distribution of Cobalt, Iron and g-CoFe alloy.

In the case of two catalysts, in the temperature range [400e500 C], the CO2 conversion rate is generally higher than that of CH4 (Fig. 8), while H2/CO ratio was lower to 1 (Fig. 8 (C)). This suggests the participation of the reverse water gas shift reaction (RWGS), which tends to increase CO2



7

Table 3 e Conversions of CH4 and CO2, H2/CO selectivity and carbon balance (%), in the presence of the CoAlcal-R and CoFecal-R catalysts. catalysts

CH4 conversion (%) CO2 conversion (%) H2/CO Carbon balance (%)

CoAlcal-R

CoFecal-R

400 C

700 C

400 C

700 C

10.5 14.4 0.87 100

66.4 70.3 1.07 87.5

5.1 7.2 0.92 100

54.5 60.2 0.94 98.5

conversion and the production of CO, which becomes higher than that of H2. At high temperatures [500e700 C], the H2/CO syngas ratio is greater than the theoretical value 1 in the presence of CoAlcal-R. This deviation from stoichiometry of the reaction, may involve the participation of the reactions such as the conversion of CO via inverse boudouard reaction and/or the methane decomposition reaction. In addition, the CO2 conversion is higher than the CH4 conversion and the carbon balance obtained was lower than 100% (Table 3). This suggests that the carbon deposition occurs via the inverse boudouard reaction. At the same temperatures [500e700 C], in the case of CoFecal-R catalyst, CO2 conversion was always higher than that of CH4 while the ratio H2/CO is close to 1, indicates the occurrence of the reverse water gas shift reaction (RWGS) favored by the presence of iron [34]. The stability of our catalyst was also studied over a prolonged period of time.

Selectivity as a function of time of the reaction According to study of the effect of the reaction temperature, a quasi-total conversion of methane can be achieved at 700 C. We have chosen this temperature for facilitate the comparison of the stability of the various catalysts over time. The results are summarized in Fig. 9. We also present a comparative study of the catalytic performance of the both catalysts in Table 4. Under the experimental conditions used, the aluminum catalyst is more active than the iron-substituted catalyst. It should also be noted that in the case of an iron-based catalyst, a stationary regime is rapidly reached within the first minute of contact with the reaction mixture and remains constant throughout the test period (Fig. 9). Furthermore, we have found that the catalytic activity of aluminum catalysts evolves over time to reach a plateau after 2 h of reaction and this stage is maintained beyond 10 h of catalytic test. At this level, we have tried to interpret the behavior of the catalysts during the reaction as a function of certain physical characteristics such as the specific surface area and the dispersion of the cobalt particles. In the presence of CoAlcal-R, a higher activity was observed in terms of particles formed upon reduction and also associate with higher SBET value obtained for CoAlcal-R (87 m2/g) compared to that CoFecal-R (50 m2/g) (Table 4). Indeed, the large surface area permits a better access of the reagents and thus constitutes a necessaryalbeit not sufficient-condition for high performance. Previously, San Jose and all [41] studied two series of catalysts (Co and Ni alumina supported catalysts) with 1, 2.5 and 4 wt %

Fig. 9 e (A) CH4 conversion, (B) CO2 conversion as a function of time on stream in the DRM at 700 C. prepared by impregnation method. Remarkably, the Co(1) catalyst is deactivated during the first minutes of reaction due to the formation of CoAl2O4, but Ni(1) and Co(2.5) catalysts present high specific activity high stability and a very low amount of deposited carbon. Equally, Ewbank and co-workers [35]. We envisaged Co/Al2O3 catalysts synthesis by two method, dry impregnation (DI) and controlled adsorption (CA) for methane dry reforming reactions. Under similar conditions to those utilized in our work (Treaction ¼ 700 C, CH4: CO2 ratio 1:1), they found that the initial conversion of methane for 2CoCA catalyst is 68% and 71% of the carbon dioxide fed while 2CoDI catalyst initially converted 52% of each of the reactants fed. These catalysts deactivate over the 8 h reaction time studied.

Influence of Co, Al and Fe on carbon formation For CoAlcal-R and CoFecal-R, the conversion of CO2 was higher than that of CH4 in all tests and ratio of H2/CO was slightly higher than one in case of CoAlcal-R (1.07), can be attributed to the carbon deposition by the inverse Boudouard reaction. While in the case of CoFecal-R, the ratio H2/CO was always less than one (0.89) and also noticed a lower amount of carbon deposition (2.06) compared to CoAlcal-R (14.52) (Table 5). These results can be explained by the participation of the reaction


8


Table 4 e Conversions of CH4 and CO2, H2/CO ratio, carbon balance (%), specific surface area and size of Co0 particles of reduced catalysts. catalysts

Conversion CH4 (%)

CO2 (%)

66.4 54.5

70.3 60.2

CoAlcal-R CoFecal-R a

H2/CO

Xc (%)

Reduced catalysts 2

1.07 0.94

87.5 98.5

S.BET (m /g)

Particle size aCo0 (nm)

87 50

17 19

Calculated from SEM (Fig. 6).

reverse water-gas shift reaction (RWGS) in the case of CoFecalR. We can assume that the carbon formed on the surface of the CoFecal-R catalyst is removed by the oxygen species formed from FeOx (Eq. (10)). The latter is obtained after the dissociation of g-CoFe alloy in the presence of CO2. ðFeOx þ CNi 4xCO þ Ni þ FeÞ

(10)

Our results correlate well with several previous works by Theofandis et al. [35e43] report that the carbon species formed on the metallic sites nickel, are eliminated by the oxygen species formed from iron oxide. In our case, therefore, the decrease in the catalytic performance of iron-based catalyst with respect to the aluminum-based catalyst is not attributable to carbon deposition, it is due to the re-oxidation of the active sites Co0 by H2O (Eq. (11)) formed to the inverse reaction of the gas with the water favored in the presence of iron (Eq. (12)) ðCo þ H2 O4CoO þ H2 Þ

(11)

ðFeOx þ H2 4Fe þ H2 OÞ

(12)

This observation is in line with that of Ando et al. [35,37e44] and Duvenhage et al. [35,38e45], where they are noticed that the iron has good catalyst for the water gas shift WGS reaction. This is favored at high temperature and could lead to rapid re-oxidation of metallic species and cause a drop in catalytic activity.

Physicochemical features of CoAlcal-R and CoFe after the DRM reaction

cal-R

catalysts

The specific surfaces of all the catalysts, measured by the BET method after 12 h of reaction, are shown in Table 5. This table also shows the average particle size Co0 (nm) calculated from the SEM images, before and after reaction. From the results shown in Table 5, the specific surface after reaction was decreased for the aluminum catalyst and remained substantially stable in the case of iron-based catalyst. In the case of the CoAlcal-R, the reduction of surface area due to the sintering of the Co0 particles during the reaction. While the reduction of surface area after reaction in the case of CoFecal-R catalyst may be due to due to the re-oxidation of the active sites Co0.

The deactivation of the catalysts in the dry reforming reaction of methane has been extensively studied in the literature [35e48]. This deactivation is due to the formation of carbon or to the sintering of the metal particles. In order to demonstrate the causes of decreased activity of the CoFecal-R catalyst with respect to CoAlcal-R, we followed by scanning electron microscopy the evolution of the texture after 12 h of test of stability at 700 C. The images obtained are presented in Fig. 10. The SEM micrographs showed the presence of filament carbon formed on the surface of the various catalysts during the methane dry reforming reaction. Carbon is much more observed on the surface of CoAlcal-R than on CoFecal-R, which shows that the observed filament carbon is detrimental to the activity of the aluminum catalyst. In order to demonstrate the presence or absence of the second deactivation phenomenon: “sintering”, the sizes of nickel and cobalt particles were measured from SEM for two catalysts after 12 h reaction. In the case of CoFecal-R, the presence of small particles always well dispersed on the surface of the catalyst is noted. Thus the decrease in the catalytic activity of the CoFecal-R is due to the re-oxidation of the active sites with H2O formed via the reaction reverse water-gas shift reaction (RWGS). The reoxidation of the active sites was confirmed by MEB-EDX analysis which shows the formation of CoO oxide after reaction (Fig. 10 and Fig. S3). XRD patterns of used catalysts after DRM reaction at 700 C for 12 h are showed in Fig. 11. The phases corresponding to CoO and Co0 were clearly indicated in two catalysts and Fe0 [JCPDS file No 06-0696] have also been observed in NiFe. However, a new weak diffraction peak corresponding to carbon graphite was observed at 2 q ¼ 26.38 [JCPDS file 41-1487], this was mainly detected in case of CoAlcal. However, no carbon was deposited on the CoFe catalyst, such result conform well with those obtained from SEM (Fig. 10) and TPO (Fig. 12). To confirm the interpretations made previously, the two catalysts were investigated by the technique of Analyzes by Temperature Oxidation Programmed (O2-TPO). Coke deposited on catalysts has been investigated by continuously monitoring evolved CO2 during the temperature programmed oxidation (O2-TPO) according to the following equation:

Table 5 e Specific surfaces and mean particle size Co0 (nm) before and after DRM reaction and Estimated carbon deposited. S.BET (m2/g)

Samples

CoAlcal-R CoFecal-R

Particle size Co0 (nm)

After reduction

After reaction

duction After re

87 50

75 45

16 19

Estimated carbon deposited

action After re

100- carbon balance

(%) of carbon by EDX

22 20

12.5 1.5

14.52 2.06



9

Fig. 12 e TPO profiles of CoAlcal-R and CoFecal-R catalysts after DRM reaction.

On the other hand, the O2-TPO profile of the solid CoFecal-R shows peak extremely low. This is related to the presence of iron.

Conclusions

Fig. 10 e SEM of used catalysts after reaction at 700 C for 12 h.

C þ O2 4CO2

(13)

The TPO profiles shows in Fig. 12. The O2-TPO profiles of CoAlcal-R show one-pic significant CO2 peaks at 635 C, which suggests the presence of mainly one type of carbon deposit.

Catalysts CoAlcal-R and CoFecal-R, obtained by calcination and reduction of hydrotalcite-type materials characterized by difference techniques including TGA, ICP, XRD, N2 adsorption and desorption, FTIR, H2-TPR, SEM-EDX and TEM. The calcination of the precursor at 800 C for 15 h and reduced at 750 C led to the formation of Co0, Fe0 and g-CoFe alloy as confirmed by XRD. The CoAlcal-R catalyst exhibited rather high catalytic activity and stability during the reaction of CO2 reforming of methane compared to CoFecal-R catalyst. The high performance of the catalytic process can be explained by: i) the presence of a very homogeneous Co0 particle, ii) a very high specific surface area and iii) lower size of cobalt. The low reactivity of the iron-based catalysts is correlated with the re-oxidation of the active phase by the water formed via the reverse water-gas shift reaction (RWGS) favored by the presence of iron. The carbon formed on the surface of the iron catalysts during the catalytic test was removed by the oxygen of the iron oxide (FeOx).

Acknowledgements The authors of this paper would like to thank the Ministry of Higher Education and Scientific Research (MESRS), Algeria and the General Directorate for Scientific Research and Technological Development (DGRSDT), Algeria, for their financial help that prompted the accomplishment of this scientific material.

Appendix A. Supplementary data Fig. 11 e RX diffractograms of used catalysts (Treaction ¼ 700 C); CoO, Co0, ↑ Fe0 and Carbon.

Supplementary data related to this article can be found at https://doi.org/10.1016/j.ijhydene.2018.03.118.


10


references

[1] Cao Y, Lu M, Fang J, Shi L, Zhang D. Hexagonal boron nitride supported meso SiO2-confined Ni catalysts for dry reforming of methane. Chem Commun 2017;53:7549e52. € ck E-M, Klo € tzer B, Schachinger T, Wallisch W, [2] Kogler M, Ko Henn R, et al. High-temperature carbon deposition on oxide surfaces by CO disproportionation. J Phys Chem C Nanomater Interfaces 2016;120:1795e807. [3] Wang SB, Lu GQM, Millar GJ. Carbon dioxide reforming of methane to produce synthesis gas over metal-supported catalysts: state of the art. Energy Fuels 1996;10:896e904. [4] Zhao X, Lu M, Li H, Fang J, Shi L, Zhang D. In situ preparation of Ni nanoparticles in ceriumemodified silica aerogels for coking- and sintering-resistant dry reforming of methane. New J Chem 2017;41:4869e78. [5] Oemar U, Hidajat K, Kawi S. High catalytic stability of Pd-Ni/ Y2O3 formed by interfacial Cl for oxy-CO2 reforming of CH4. Catal Today 2017;281:276e94. [6] Ghelamallah M. http://www.sciencedirect.com.www.sndl1. arn.dz/science/article/pii/S0926860X14004517-!, Granger P. http://www.sciencedirect.com.www.sndl1.arn.dz/science/ article/pii/S0926860X14004517-!. Supported-induced effect on the catalytic properties of Rh and Pt-Rh particles deposited on La2O3 and mixed a-Al2O3-La2O3 in the dry reforming of methane. Appl Catal A 2014; 485:172e180. [7] Xie T, Shi L, Zhang J, Zhang D. Immobilizing Ni nanoparticles to mesoporous silica with size and location control via a polyol-assisted route for coking and sintering-resistant dry reforming of methane. Chem Commun 2014;50:7250e3. [8] Djebarri B, Gonzalez-Delacruz VM, Halliche D, Bachari K, Saadi A, Caballero A, et al. Promoting effect of Ce and Mg cations in Ni/Al catalysts prepared from hydrotalcites for the dry reforming of methane. React Kinet Mech Catal 2014;111:259e75. [9] Zhang X, Wang N, Xu Y, Yin Y, Shang S. A novel NieMgeAlLDHs/g-Al2O3 Catalyst Prepared by in-situ synthesis method for CO2 reforming of CH4. Catal Commun 2014;45:11e5. [10] Zhang H, Dong YY, Fang WP, Lian YX. Effects of composite oxide supports on catalytic performance of Ni-based catalysts for CO methanation. Chinese J catal 2013;34:330e5. [11] Zhu Y, Zhang S, Chen B, Zhang Z, Shi C. Effect of Mg/Al ratio of NiMgAl mixed oxide catalyst derived from hydrotalcite for carbon dioxide reforming of methane. Catal Today 2016;264:163e70. [12] Shiratori Y, Sakamoto M, Uchida T, Le H, Quang-Tuyen T, Sasaki K. Hydrotalcite-dispersed paper-structured catalyst for the dry reforming of methane. Int J Hydrogen Energy 2015;40:10807e15. [13] Yu J, Wang Q, O'Hare D, Sun L. Preparation of two dimensional layered double hydroxide nanosheets and their applications. Chem Soc Rev 2017;46:5950e74. -Alonso D, Juan-Juan J, Illan-Gomez MJ, Roman[14] San-Jose Martinez MC. Ni, Co and bimetallic Ni-Co catalysts for the dry reforming of methane. Appl Catal A 2009;371:54e9. [15] Zhang X, Yang C, Zhang Y, Xu Y, Shang S, Yin Y. Ni-Co catalyst derived from layered double hydroxides for dry reforming of methane. Int J Hydrogen Energy 2015;40:16115e26. [16] Wang C, Sun N, Zhao N, Wei W, Zhao Y. Template-free preparation of bimetallic mesoporous Ni-Co-CaO-ZrO2 catalysts and their synergetic effect in dry reforming of methane. Catal Today 2016;281:268e75. [17] Tsoukalou A, Imtiaz Q, Kim SM, Abdala PM, Yoon S, Mu¨ller CR. Dry-reforming of methane over bimetallic NieM/ La2O3 (M¼Co, Fe): the effect of the rate of La2O2CO3 formation and phase stability on the catalytic activity and stability. J Catal 2016;343:208e14.

[18] Touahra F, Ketir W, Halliche D, Trari M. Photocatalytic hydrogen evolution over the hetero-junction CoAl2O4/ZnO. React Kinet Mech Catal 2014;111:805e16. s N. Study of the composition of ternary [19] Melo F, Morlane mixed oxides: use of these materials on a hydrogen production process. Catal Today 2008;133:374e82. [20] Makri MM, Vasiliades MA, Petallidou KC, Efstathiou AM. Effect of support composition on the origin and reactivity of carbon formed during dry reforming of methane over 5 wt% Ni/Ce1-x MxO2-d (M¼Zr4þ, Pr3þ) catalysts. Catal Today 2016;259:150e64. s V. [21] Caldararu H, Caragheorgheopol, Corma A, Rey F, Forne One-electron donor sites and their strength distribution on some hydrotalcite and MgO surfaces as studied by EPR spectroscopy. J Chem Soc Faraday Trans 1994;90:213e8. [22] Touahra F, Sehailia M, Ketir W, Bachari K, Chebout R, Trari M, et al. Effect of the Ni/Al ratio of hydrotalcite-type catalysts on their performance in the methane dry reforming process. Appl Petrochem Res March 2016;6:1e13. € hnert J, [23] Mette K, Ku¨hl S, Tarasov A, Willinger MG, Kro Wrabetz S, et al. High-temperature stable Ni nanoparticles for the dry reforming of methane. ACS Catal 2016;6:7238e48. [24] Sobhana SSL, Mehedi R, Malmivirta M, Paturi P, Lastusaari M, Dıˆrtu MM, et al. Heteronuclear nanoparticles supported hydrotalcites containing Ni(II) and Fe(III) stable photocatalysts for Orange II degradation. Appl Clay Sci 2016;132e133:641e9. [25] Qi C, Amphlett JC, Peppley BA. H2 production via the steam reforming of methanol over NiAl-layered double hydroxide derived catalysts. Catal Surv Asia 2009;13:16e21. [26] Braterman PS, Xu ZP, Yarberry F. Layered double hydroxides (LDHs). Handb of Layered Mater 2004:373e474. [27] Cavani F, Trifiro F, Vaccari A. Hydrotalcite-type anionic clays: preparation, properties and applications. Catal Today 1991;11:173e301. [28] Lebedeva O, Tichit D, Coq B. Influence of the compensating anions of Ni/Al and Ni/Mg/Al layered double hydroxides on their activation under oxidising and reducing atmospheres. Appl Catal A 1999;183:61e71. [29] Zhao X, Xu S, Wang L, Duan X, Zhang F. Exchange-biased NiFe2O4/NiO nanocomposites derived from NiFe-layered double hydroxides as a single precursor. Nano Research 2010;3:200e10. [30] Lv M, Guo X, Wang Z, Wang L, Li Q, Zhang Z. Synthesis and characterization of CoeAleFe nonstoichiometric spineltype catalysts for catalytic CO oxidation. RSC Adv 2016;6:27052e9. Filho M, Ferreira OP, [31] Carvalho DC, Ferreira NA, Josue Soares JM, Oliveira AC. NieFe and CoeFe binary oxides derived from layered double hydroxides and their catalytic evaluation for hydrogen production. Catal Today 2015;250:155e65. [32] Zhang X, Wang Zhang DS, Ma Y, Yang W, Wang Y, Awaji S, et al. Effect of high magnetic field annealing on the microstructure and magnetic properties of CoeFe layered double hydroxide. J Magn Magn Mater 2010;322:3023e7. [33] Rudolf C, Dragoi B, Ungureanu A, Chirieac A, Royer S, Nastro A, et al. NiAl and CoAl materials derived from takovite-like LDHs and related structures as efficient chemoselective hydrogenation catalysts. Catal Sci Technol 2014;4:179e89. [34] James OO, Maity S. Temperature programme reduction (TPR) studies of cobalt phases in-alumina supported cobalt catalysts. J Pet Technol and Altern Fuels 2016;7:1e12. [35] Ewbank JL, Kovarik L, Kenvin CC, Sievers C. Effect of preparation methods on the performance of Co/Al2O3 catalysts for dry reforming of methane. Green Chem 2014;16:885e96.



[36] Budiman AW, Song SH, Chang TS, Choi MJ. Preparation of a high performance cobalt catalyst for CO2 reforming of methane. Adv Powder Technol 2016;27:584e90. € fberg A, Rubbens A, Bordes-Richard E, [37] Benrabaa R, Lo Vannier R, Barama A. Structure, reactivity and catalytic properties of nanoparticles of nickel ferrite in the dry reforming of methane. Catal Today 2013;203:188e95. [38] Ryl’tsova I, Nestroinaya O, Lebedeva O, Vorontsova O, Kosova N, Kurzina IA. New magnesium cobalt iron double hydroxides with hydrotalcite structure: synthesis and characterization. Russ J Inorg Chem 2014;59:1403e10. [39] Yang E, Noh Y, Ramesh S, Lim SS, Moon DJ. The effect of promoters in La0.9M0.1Ni0.5Fe0.5O3 (M ¼ Sr, Ca) perovskite catalysts on dry reforming of methane. Fuel Process Technol 2015;134:404e13. [40] Naeem MA, Al-Fatesh AS, Khan WU, Abasaeed AE, Fakeeha AH. Syngas production from dry reforming of methane over nano Ni polyol catalysts. Int J Chem Eng Appl 2013;4:315e20. -Alonso D, Illan-Gomez MJ, Roman-Martinez MC. [41] San-Jose Low metal content Co and Ni alumina supported catalysts for the CO2 reforming of methane. Int J Hydrogen Energy 2013;38:2230e9.

11

[42] Theofanidis SA, Batchu R, Galvita VV, Poelman H, Marin GB. Carbon gasification from FeeNi catalysts after methane dry reforming. Appl Catal B 2016;185:42e55. [43] Theofanidis SA, Galvita VV, Poelman H, Marin GB. Enhanced carbon-resistant dry reforming Fe-Ni catalyst: role of Fe. ACS Catal 2015;5:3028e39. [44] Ando H, Matsumura Y, Souma Y. A comparative study on hydrogenation of carbon dioxide and carbon monoxide over iron catalyst. J Mol Catal A Chem 2000;154:23e9. [45] Duvenhage DJ, Coville NJ. Fe:Co/TiO2 bimetallic catalysts for the FischereTropsch reaction: Part 2. The effect of calcination and reduction temperature. Appl Catal A 2002;233:63e75. [46] Muraza O, Galadima A. A review on coke management during dry reforming of methane. Int J Energy Res 2015;39:1196e216. [47] Rahbar Shamskar F, Rezaei M, Meshkani F. The influence of Ni loading on the activity and coke formation of ultrasoundassisted co-precipitated Ni-Al2O3 nanocatalyst in dry reforming of methane. Int J Hydrogen Energy 2017;42:4155e64. [48] Zhang Q, Zhang T, Shi Y, Zhao B, Wang M, Liu Q, et al. A sintering and carbon-resistant Ni-SBA-15 catalyst prepared by solid-state grinding method for dry reforming of methane. J CO2 Util 2017;17:10e9.
