Enhanced Long-term Stability and Carbon Resistance of Ni/MnxOy-Al2O3 Catalyst in Near-equilibrium CO2 Reforming of Methane for Syngas Production

Herein we study the catalytic activity/stability of a new generation of cheap and readily available Ni and Al-based catalysts using two Mn precursors, namely Mn(NO3)2 and Mn(EDTA)2complex in the reaction of CO2 reforming of methane. In this respect, Ni/Al2O3 and two types of Ni/MnxOy-Al2O3 catalysts were successfully synthesized and characterized using various analytical techniques: TGA, ICP, XRD, BET, FTIR, TPR-H2, SEM-EDX, TEM, XPS, and TPO-O2. Utilization of Mn(EDTA)2as synthetic precursor successfully furnished Ni/Al2O3-MnxOyY (Y = EDTA) catalyst which was more active during CO2 reforming of methane when compared to Ni/MnxOy-Al2O3 catalyst, synthesized using Mn(NO3)2 precursor. Compared to Ni/MnxOy-Al2O3, Ni/Al2O3-MnxOyY catalyst afforded near-equilibrium conversion values at 700 °C (ca. 95% conversion for CH4 and CO2, and H2/CO = 0.99 over 50 h reaction time). Also, Ni/Al2O3-MnxOyY showed more resistance to carbon formation and sintering; interestingly, after 50 h reaction time, the size of Ni0 particles in Ni/MnxOy-Al2O3 almost doubled while that of Ni/Al2O3-MnxOyY remained unchanged. The elevated conversion of CO2 and CH4 in conjunction with the observed low carbon deposition on the surface of our best catalyst (Ni/Al2O3-MnxOyY) indicated the presence of MnxOy oxide positioning mediated simultaneous in-situ carbon elimination with subsequent generation of oxygen vacant sites on the surface for more CO2 adsorption. Copyright © 2020 BCREC Group. All rights reserved


Introduction
Global warming is one of the major environmental challenges of our time. A major cause of use of greenhouse gas (CO2 and CH4) and/or by reducing energy consumption. Catalytic dry reforming of methane (DRM) (also known as CO2 reforming of methane) converts both CO2 and CH4 to syngas (CH4 + CO2 ↔ 2CO + 2H2); the reaction has been proposed by many researchers as one of the possible solutions targeted at reducing the amount of CO2 and CH4 in our environment. The resulting intermediate is widely applied in the Fischer-Tropsch synthesis [1,2], to make other value-added hydrocarbons. However, the dry reforming of methane requires a high temperature (typically above 640 °C) [3]. DRM has been widely investigated over Ni-based catalysts; rapid deactivation of the utilized catalysts is primarily due to carbon formation and sintering of the active phase. Carbon deposition on the surface of catalysts mainly occurs from the following side reactions (Equations (1), (2), (3) and (4)). The reverse water-gas shift reaction can also occur as a sidereaction in which some of the H2 from the CH4 decomposition reacts with the CO2 to produce H2O and CO (Equation (5)) [4][5][6][7][8][9].
CH4 decomposition: CH4 → C + 2H2 (1) Boudouard reaction: 2CO → C + CO2 (2) CO2 reduction: CO2 + 2H2 → C + 2H2O (3) CO reduction: H2 + CO → H2O +C (4) Reverse water-gas shift reaction: CO2 + H2 → CO + H2O (5) In this regard, many efforts were devoted to improving the activity and coke-resistance of nickel-based catalysts. Among the various strategies proposed is the incorporation of other metal promoters such as Mn, Zr, La, Zn, Fe, Mg, Ce and Cu [10][11][12][13][14], alongside increasing the catalysts' surface basicity [15,16], and additional control of the active element particle size [17,18]. The addition of metal promoters had also attracted much attention from the scientific community [19]. More recently, the effect of Mn promoter in Ni-based catalysts used for DRM reactions were investigated by many researchers [20][21][22][23][24]. Yao et al. [24] studied the effect of Zr and Mn promotion in Ni/SiO2 catalysts. The authors found that the addition of Mn lead to the formation of smaller particle size of Ni, which enhanced the interaction with the support. Thus, Ni-Mn/SiO2 catalyst showed higher activity and better ability to restraining carbon deposition. The introduction of Zr increased the reducibility of Ni/SiO2 catalyst and enhanced the initial catalytic activity, but suffered from carbon deposition had which lead to obvious deactivation.
The promotion effect of Mn on Ni-based catalysts was also studied by Mousavi et al. [22] The authors concluded that the incorporation of Mn into the cerium oxide 10%Ni/Ce1-xMnxO2 (0, 0.05, 0.25, 0.50, 0.75 and 1) increased the BET surface area and improved the catalytic activity and resistance against carbon formation. They also found that the catalytic activity decreased in the presence of excess Mn. Recently, our group studied a new catalyst system based on manganese incorporated within NiAl hydrotalcite derived structures in CO2 reforming of methane reaction [10]. NiAl and MnAl hydrotalcites were synthesized following co-precipitation method and NiAl-MnY (Y = ethylenediaminetetraacetic acid (EDTA)) catalyst was synthesized via intercalation of MnY 2-in the interlayer space of NiAl hydrotalcite type structure following an ionic exchange protocol. It was found that during DRM reaction, NiAl-MnY catalyst achieved the best result compared to NiAl catalyst in terms of minimizing carbon deposition and enhancing gas conversions at 700 °C. The elimination of carbon from the surface of the catalysts was explained by the possibility of in-situ redox loopi n g m e c h a n i s m i n t h e fo r m o f Mn3O4/C/MnO/CO2 [10].
The present paper aims at studying the effect of MnxOy positioning within the structure of Ni/Al2O3 hydrotalcite-derived catalysts systems to generate newly derived catalysts for application in DRM reaction. For comparison, we apply the co-precipitation method to synthesize Ni/Al2O3, Ni/MnxOy-Al2O3 and Ni/Al2O3-MnxOyY hydrotalcite-derived catalysts, where Mn 2+ is incorporated in the brucite-like layer of Ni/Al2O3 catalyst to form Ni/MnxOy-Al2O3 catalyst while in the case of Ni/Al2O3-MnxOyY, MnY 2-complex is intercalated in the interlayer space of Ni/Al2O3 hydrotalcite-derived catalysts.

Catalysts Preparation
Ni/Al2O3, Ni/MnxOy-Al2O3 and Ni/Al2O3-MnxOyY hydrotalcite-derived catalysts were prepared via co-precipitation method following procedure previously described [10]. Ni/Al2O3 catalyst was synthesized with molar ratio (nNi 2+ /nAl 3+ ) = 2, in order to induce the formation of hydrotalcite structures. The aqueous solutions of nitrate metals were added dropwise to a vigorously stirring solution of NaOH (2 M) at room temperature while pH was maintained at 12. The obtained slurry was heated at 80 °C, kept under stirring for 15 hours for maturation, and then filtered, washed with water and finally dried at 100 °C in oven overnight. In order to study the effect of Mn positioning within the Ni/Al2O3 hydrotalcite-derived catalysts structure (i.e., in the brucite layer or in the interlayer space of hydrotalcite), we embarked on the synthesis of Ni/MnxOy-Al2O3 sample using the co-precipitation method reported above with molar ratio (nNi 2+ /nAl 3+ ) = 2.4.
The synthesis of Ni/Al2O3-MnxOyY hydrotalcite-derived catalysts layered double hydroxide (LDH) included two steps. Firstly, an aqueous solution of MnY 2− complex was prepared at room temperature following a procedure previously described [10]. Mn 2+ nitrate solution (20 mmol) was added dropwise to tetrasodium salt of (EDTA=Y) dissolved in 50 mL of water. NaOH solution (1 M) was added up to pH = 8. After stirring for 2 h, the resulting solution was filtered to remove excess (NaNO3) solids. The second step of synthesis included dropwise addition of aqueous solution of Ni 2+ and Al 3+ nitrates to a solution of MnY 2-. PH was adjusted to 12 by addition of aqueous solution of NaOH (2 M). The obtained slurry was stirred at 80 °C for 15 h under Nitrogen. The resulting MnY 2-intercalated in-between brucite layers of Ni/Al2O3 hydrotalcite-derived was filtered, washed with ultra-pure water and finally dried at 100 °C in oven overnight. Finally, the furnished materials, was subjected to calcination at 800 °C (increments of 5 °C/min) for 6 h, to afford the desired hydrotalcite-derived material.

Catalysts Characterization
Thermogravimetric analysis (TGA) was performed using Thermal Analyzer Setaram Set Sys 16/18 from room temperature to 900 °C with a heating rate of 10 °C/min in the presence of air. The chemical composition was established using inductively coupled plasma atomic emission spectroscopy (ICP-AES) in the presence of a multichannel Thermo Jarrel Ash ICAP 957 spectrometer. X-ray diffraction patterns of powdered samples were obtained using Siemens D-5000 diffractometer with Cu-K radiation (= 1.5418 Å). The average particle size (dhkl), following reduction, were estimated using the Debey-Scherer formula (Equation (6)) [25]. (6) where  is the radiation's wavelength, hkl is the half width of the peak, and  is the Bragg diffraction angle. The specific surface area measured by nitrogen adsorption at -196 °C using Micrometrics Tristar 3000 and evaluated using the BET equation. Temperatureprogrammed reduction (TPR) and temperature programmed oxidation analysis (TPO-O2) profiles were determined using a TriStar3000 V6.01A apparatus equipped with a TCD detector. Prior to reduction, the sample (20 mg) was flushed with argon in a quartz reactor tube, heated at 120 °C for 2 h under argon flow with a rate of 5 °C/min and cooled afterward to room temperature. The sample was submitted to a 5% H2/Ar up flow to 900 °C with a 10 °C/min rate. The same method was also used for TPO-O2 of the used catalysts. An oxidizing gas stream (3% O2/He) was employed in this analysis and other steps were similar to those mentioned for the TPR-H2 analysis. Scanning electron microscopy (SEM) images were obtained using Jeol 320 instrument. The samples were affixed to the sampling plate with carbon black tape and coated with Pt to prevent any charging effect during analysis. Transmission Electron Microscopy (TEM) images of the catalysts before and after reaction were obtained with high resolution field emission transmission electron microscope at 100 kV using JEOL-JEM-1200EX. FTIR spectra were recorded with Alpha Bruker (single reflection diamond ATR) spectrometer in the region 400-4000 cm -1 . X-ray photoelectron spectroscopy (XPS) surface analysis of NiAl, NiMnAl and NiAl-MnY was performed in Escalab 200R spectrometer equipped with a hemispherical electron analyzer and an Al K X-ray source (1486.6 eV).

Catalytic Activity
Catalytic testing experiments were carried out in a continuous flow system at atmospheric pressure using a fixed-bed tubular quartz reactor. Prior to catalytic reaction, 100 mg of each sample was reduced in-situ under constant hydrogen flow at 750 °C for one hour. After that, the temperature was cooled down to the initial reaction's temperature and a feed gas mixture containing CH4:CO2: Ar in a ratio of 20:20:60 was passed through. The total gas flow rate was set to 20 mL/min. The reaction products were analyzed using gas chromatograph (Delsi), equipped with a thermal conductivity detector (TCD). The TCD uses two-meters-long stainless-steel Carboseive column, in the presence of carrier gas argon. The chromatography (GC) operating conditions: Oven temperature = 100 °C; Injector and detector temperature = 100 °C; and Temperature for the accessory = 70 °C.

Thermogravimetric Analysis (TG/DTG)
Both TG and DTG curves of the investigated samples are given in Figure S1 in the Supplementary Materials. In every case, three losses of mass were observed; the first weight loss is attributed to the removal of interlayer water molecules. At this stage, a higher weight loss for Ni/MnxOy-Al2O3 and Ni/Al2O3-MnxOyY was observed compared to that of Ni/Al2O3 ( Table  1). The cause of this is the increase in the number of interlayer water molecules resulting from the addition of manganese. The second weight loss was due to the dehydroxylation of the layered structure and the removal of interlayer anions (NO3 -). In the case of Ni/Al2O3-MnxOyY sample, a large weight loss was observed, this could be due to the decomposition of EDTA. In the third step, the weight loss could be assigned to the decomposition of all residual salt within the samples (Table 1).

Catalyst Composition Using ICP
By looking at the data obtained from elemental the chemical analysis and taking into account the amount of water loss in the samples using thermogravimetric analysis, we generate the formulae summarized in Table 2.
The experimental values of the molar ratios nNi 2+ /nAl 3+ of the synthesized samples are very close to the theoretical values nNi 2+ /nAl 3+ = 2, with the exception of the solid Ni/MnxOy-Al2O3 which showed a molar ratio (nNi 2+ + nMn 2+ )/nAl 3+ = 2.84 instead of 2.40. This difference could be attributed to the partial incorporation of aluminum cations inside the brucite layer.

Measurement of Specific Surface Areas
The specific surface areas for all samples are shown in Table 3. The surface areas of the hydrotalcite intermediate prior to calcination showed relatively high SBET values, except for Ni/Al2O3-MnxOyY sample (13 m 2 /g). This result can be explained by the penetration of EDTA species into the pores of the materials. However, following calcination of all samples, a change in SBET was observed. The increase in SBET in the case of Ni/Al2O3 and Ni/MnxOy-Al2O3 samples can be explained by the removal of water from the interlayer space of the hydrotalcite structures [26]. On the other hand, the large increase in SBET in the case of Ni/Al2O3-MnxOyY sample was mainly attributed to the elimination of inter-lamellar water and the pyrolysis of Y 4-ligand after heat treatment [10,27].

Samples
First weight loss (%)   Figure 1A shows the X-ray diffractograms of the un-calcined samples. The corresponding Xray diffractograms for Ni/Al2O3, and Ni/MnxOy-Al2O3 are consistent with those expected for hydrotalcite-like structures with sharp and symmetric reflections for (003), (006), (110) and (113) planes and broad asymmetric reflexions for (012), (015) and (018) [JCPDS file . The position of the first two diffraction peaks (003) and (006), are related to the spacing between the sheets "c" (with c = 3 × d003 = 6 × d006) [18]. This parameter represents the distance between two adjacent sheets. The insertion of MnY 2-complex causes a separation of the sheets of the LDH, which is manifested by the appearance of two additional peaks located at 2 = 6 and 18°.
It can be readily seen that the diffractogram of Ni/Al2O3-MnxOyY has a specific set of reflection peaks that significantly differ from the counterpart materials. Indeed, the intercala-tion of MnY 2-in Ni/Al2O3 sample which lead to the obtention of Ni/Al2O3-MnxOyY prompted an important shift towards lower 2 for reflection plane (003), indicating a significant increase in the interlayer separation from 8.946 to 14.496 Å ( Figure 1A). The results of XRD experiments for the calcined samples (at 800 °C) are presented in Figure 1B. Following thermal treatment, the hydrotalcite structure of the samples were destroyed. XRD patterns of Ni/Al2O3 sample showed the formation of NiO species [JCPDS file 47-1049] and NiAl2O4 spinel [JCPDS file 10-0339]; an amorphous aluminum oxide phase (Al2O3) should also be formed (not detected in XRD) [28]. The Mn-containing samples exhibited peaks corresponding to NiO, NiAl2O4 and Mn3O4 [JCPDS file 24-0734].

FTIR Analysis
FTIR spectroscopy can provide valuable information on the identification of the anion bonds in the inter lamellar space. FTIR spectra  of Ni/Al2O3 and Ni/MnxOy-Al2O3 samples are shown in Figure 2A, while that of Ni/Al2O3-MnxOyY solid is shown in Figure 2B. The bands present are almost similar; the characteristic vibrational bands of the existing groups within the LDH structure are similar to those previously listed in literature [12,26]. In addition, we also note those bands forming part of the EDTA structure inserted into the inter lamellar space of Ni/Al2O3-MnxOyY. The very strong peak at 1378 cm -1 can be assigned to the stretching vibration of NO3 -group. The apparent band at 3443 cm -1 is attributed to the O-H stretching vibration of physically adsorbed water. The band around 1620 cm -1 is attributed to the deformation vibrations of water δ(H2O). The FTIR spectra also confirmed the successful intercalation of MnY 2-in-between the layers of the brucite. Two strong bands located at 1627 cm -1 and 1386 cm -1 are attributed to the asymmetric and symmetric vibrations of carboxylate group (-COO -) in EDTA [10,27,29]. In the low frequency region (1000 and 400 cm -1 ), the adsorption peaks correspond to the lattice vibration modes (Al-O, Ni-O and Mn-O).

Temperature Programmed Reduction (H2-TPR)
H2-TPR profiles for all samples are plotted in Figure 3. H2-TPR of Ni/Al2O3 sample showed a profile with a main peak at 530 °C, which corresponds to the reduction of free NiO. A broad peak was observed at around 677 °C, this reduction peak corresponds to the reduction of nickel species in NiAl2O4 phase, which was previously observed during XRD analysis.
H2-TPR of Ni/MnxOy-Al2O3 and Ni/Al2O3-MnxOyY showed three reduction peaks. The first peak was observed at 450 °C and 470 °C for Ni/Al2O3-MnxOyY and Ni/MnxOy-Al2O3 respectively, corresponding to the reduction of free NiO on the surface of the catalyst, the second peak was observed around 650 °C which can be attributed to reduction of nickel strongly bound to Al2O3 phase. The last peak centered at 780 °C could be assigned to the reduction of Ni 2+ and/or MnxOy further situated inside the amorphous structure of Al2O3 phase. Although H2 reduction of MnxOy species was found to generally occur at lower temperature range, [24,30] previous findings by our group had shown elevated reduction temperatures for MnxOy in materials derived from MnAl hydrotalcite systems (H2-TPR not shown here) [10]. Based on previous studies [10], and in agreement with results from other literature material [31], we can conclude that the presence of Mn in Al solids favored a displacement of the reduction peak of Ni towards lower temperatures, this could be explained by the presence of strong interactions between Mn and Al2O3, which consequently weakens the interaction between Ni and Al2O3. H2-TPR also helps us assessing the amount of reducible species of a material. The hydrogen consumptions, expressed in mmol/g of catalyst, are summarized in Table 4. In all catalysts, the H2 consumption between 300 and 600 °C roughly corresponds to that expected for the complete reduction of the NiO phase to metallic Ni 0 . In the second H2 consumption (600-750 °C), the amount of hydrogen is equal to that needed for reducing nickel in NiAl2O4 phase to Ni 0 . However, H2 consumption between 750 and 850 °C was observed in the case of Ni/Al2O3-MnxOyY and Ni/MnxOy-Al2O3 catalyst, which could correspond to the surface reduction of Mn3O4 to    MnO, as shown in the XRD results after reduction ( Figure 4).

X-ray diffraction after reduction
For all samples, XRD patterns (Figure 4) obtained after reduction showed the presence of Ni 0 [JCPDS file 04-0850], identified by the diffraction lines located at 2 = 44.50°, 51.84° and 76.01°. In addition, XRD profile of the catalyst after reduction demonstrated the partial presence of NiO. Mn3O4 were also observed in the case of Mn-containing samples. XRD pattern of Ni/MnxOy-Al2O3 also shows a small peak at 2 = 33.7° which corresponds to the presence of MnO. Interestingly, the Ni particle size in Ni/Al2O3 (19 nm) was found to be greater than that of Ni/MnxOy-Al2O3 (7.5 nm) and Ni/Al2O3-MnxOyY (7.2 nm) catalysts (Table 3).

BET surface of reduced samples
Evaluation of the specific surface area of all catalysts after reduction was also studied. Indeed, data shown in Table 3 demonstrated that the reduction step lead to a small decrease in the specific surfaces areas in the case of Ni/Al2O3 and Ni/MnxOy-Al2O3 catalysts. On the other hand, in the case of Ni/Al2O3-MnxOyY, the specific surface area remained constant after reduction (126-124 m 2 /g). Albeit the decrease in surface areas in Ni/Al2O3 and Ni/MnxOy-Al2O3 compounds was in accordance with the general sintering of nanoparticles forming part of the catalysts, we clearly notice a different trend with regards to the larger surface area observed in Ni/Al2O3-MnxOyY, which indicates a much more stable surface morphology that prompted resistance in surface particle agglom-eration. These results can be explained by the introduction of Mn shall minimize Ni particle agglomeration via the prevention of van der Waals forces between Ni particles. Therefore, small crystallite size of Ni particles is formed, thus, leading to the obtention of high dispersion in Ni/Al2O3-MnxOyY.
Following reduction, all catalysts demonstrated the presence of smaller and evenly distributed Ni particles on the surface of the catalysts.
The average particle size of Ni 0 in the reduced samples, calculated from the SEM and TEM images, are in good agreement with the average size calculated by the XRD method (Table 3). It is interesting to point out that the size of Ni 0 nickel particles are smaller in the case of Mn-containing catalysts, which can suggest that the addition of Mn in the solid matrix has a considerable effect in decreasing the particles sizes.

XPS after reduction
XPS spectra of Ni/Al2O3, Ni/MnxOy-Al2O3 and Ni/Al2O3-MnxOyY catalysts after reduction are displayed in Figure 7. Figure 7A shows the Ni 2p3/2 XPS spectrum of three catalysts after reduction at 750 °C (before reaction). The figure shows two peaks having binding energies of 852.4 and 856.3 eV for all catalysts, which are assigned, respectively, to metallic Ni and NiO, these results are in agreement with previously reported data in the literature [32,33]. The presence of NiO on the surface of the reduce catalysts (prior reaction) might have been owing to the re-oxidation of Ni 0 when the reduced samples were exposed to air. Figure 7B shows the XPS spectra of Mn 2p on Ni/MnxOy-Al2O3 and Ni/Al2O3-MnxOyY catalysts. The   [34,35]. In addition, all three samples showed Al 2p peak ( Figure 7C) located at a binding energy of 75.5 eV, which can be ascribed to Al 3+ ions [36,37]. In all samples, the positions of this peak did not change, while the intensity level was stronger in Ni/Al2O3 catalyst. The corresponding XPS spectra of O 1s in all catalysts had two distinct peaks shown in Figure 7d. One located at 529.2 eV, attributed to the lattice oxygen, while the peak placed at 532.2 eV can be ascribed to oxygen species adsorbed on the surface. The ratio of OAdsorption/OLattice is listed in The catalytic performance of our synthesized catalysts were evaluated at 700 °C in the reaction of CO2 reforming reaction following an ''in situ'' H2 reduction at 750 °C for 1 h. The catalytic performance and selectivity results are presented in Figure 8. Mn-based catalysts showed higher catalytic activity than Ni/Al2O3. The conversion of CH4 increased from 82% (Ni/Al2O3) to 86% (Ni/MnxOy-Al2O3) and 94% (Ni/Al2O3-MnxOyY), whereas CO2 conversion increased from 86% (Ni/Al2O3) to 88% and 94% in the case of Ni/MnxOy-Al2O3 and Ni/Al2O3-MnxOyY, respectively. For Ni/Al2O3 and Ni/MnxOy-Al2O3 catalysts, it was found that the conversion of CO2 was higher than that of CH4 while H2/CO molar ratio ( Figure 8C) in the resulting syngas was slightly lower than the desired stoichiometric ratio of 1 after 50 h of reaction time; this was owing to the presence of concurrent hydrogenation of carbon dioxide (CO2 + 2H2 → C + 2H2O) and/or reverse watergas shift reaction (CO2 + H2 → CO + H2O).
Ni/Al2O3 showed relatively lower CH4 and CO2 conversions when compared to Ni/MnxOy-Al2O3 and Ni/Al2O3-MnxOyY catalysts. In addition, it is important to note that in Ni/Al2O3 catalyst, the activity started to decrease after 30 hours of reaction time, such phenomenon was not observed in the case of Ni/MnxOy-Al2O3 and Ni/Al2O3-MnxOyY according to Ray et al. [38] this result is due to the incorporation of Mn in Ni-based catalyst which plays a good role in promoting a synergistic effect that further stabilizes the catalyst. The addition of Mn leads to the formation of well dispersed Ni 0 nanoparticles on the surface of the catalyst, and thus induces large specific surface areas after reduction (Table 3). In addition, previous work by our group also elucidated the possibility of deposited carbon being oxidized in the presence of manganese oxide (Mn3O4) to release carbon monoxide CO: (Mn3O4 + C ↔ 3MnO + CO) [10]. This was confirmed by the presence of Mn3O4 and MnO species prior to and after reaction, respectively. This hypothesis was further exploited by Gan et al. [39].
Guo et al. [30] also showed that the addition of manganese to the Ni/Al2O3 catalyst treated by N2 glow discharge plasma, improved the catalytic performance by about 10% at 700 °C for the methane dry reforming reaction. The authors found that the percentage of Mn plays a very important role in catalytic activity; but that there is an optimal manganese content as for any promoter element of the catalytic activity. They reported that the catalyst (5% Mn-Ni/Al2O3) gave better catalytic activity compared to other catalysts (10% Mn-Ni/Al2O3) and (15% Mn-Ni /Al2O3). Mousavi et al. [22] also investigated the effects of the incorporation of Mn to cerium oxide 10% Ni/Ce1-xMnxO2 for the methane dry reforming. The catalytic results indicated that the incorporation of Mn into the catalyst carrier slightly decreased the catalytic activity. However, the catalytic stability was improved upon the addition of Mn. 10% Ni/Ce0.95Mn0.05O2 catalyst exhibited the highest activity and stability compared to nickel catalyst supported on pure ceria, this was due to the low content of manganese, interaction be- tween manganese and ceria, and formation of the solid state solution, which improves the number of oxygen vacancies. The increase in oxygen vacancies was also found to improve the activity and coke resistance. Liu et al. [20] previously added La, Al and Mn to Fe-clay based Ni. The introduction of La, Al and Mn was found to affect the specific surface area and catalyst basicity. Also, the presence of La, Al and Mn resulted in smaller and further promoted Ni dispersion. Based on DRM results, La/Alpromoted catalysts led to improvement of the catalytic performance whereas Mn-promoted catalysts inhibited it.

Influence of Mn Position on Catalyst Stability and Carbon Formation
In order to gain more understanding about the high performance of Ni/Al2O3-MnxOyY catalyst, the spent catalyst was characterized using XRD, SEM, TEM, TPO-O2 and XPS. The surface areas of all spent catalysts are incorporated in Table 3. It is clear that following CO2 reforming of methane reaction, we observe a decrease in the surface area. In the case of Ni/Al2O3 catalyst, the surface area decreased from (105-75 m 2 /g). Incorporation of manganese in the brucite layer also resulted in a decrease in surface area, (121-101 m 2 /g). The surface area of Ni/Al2O3-MnxOyY catalyst, where manganese is located in the interlayer space, remained unchanged (126-123 m 2 /g), suggesting a better resistance to sintering and/or carbon deposition. These results also suggests that the positioning of Mn within the HDL structure plays an important role in the thermal stability of the solid. Figure 9 shows XRD profiles of all spent catalysts. Phases corresponding to NiO and Ni 0 were observed at around 2 = 37°, 44°, 62°, 76° and 2 = 44°, 51°, 76°, respectively. Also, the diffractograms of the manganese-containing catalysts have shown the presence of MnO and Mn3O4. The presence of NiO phase suggested that the re-oxidation of a small part of Ni 0 species following contact with air, given the fact that the analysis was carried out ex-situ. XRD analysis of Ni/Al2O3 and Ni/MnxOy-Al2O3 catalysts also demonstrated the presence of carbon deposition at 2 = 27 [JCPDS file , however, this peak is much smaller in Ni/MnxOy-Al2O3 catalyst. It is also important to note that this peak is totally absent in the diffractograms of Ni/Al2O3-MnxOyY catalyst. Probably, this result is due to the presence of manganese which lead to the total or partial elimination of carbon according to the reaction (Mn3O4 + C ↔ 3MnO + CO). Manganese oxides Mn3O4 and MnO are clearly observed in XRD of both catalysts. Figure 9 shows further examination of these results revealed that, depending on the position of manganese in the hydrotalcite structure, surface carbon deposition is totally or partially eliminated. In our study, we found a clear distinction between catalysts derived from Mn being in the interlayer space of the Ni/Al2O3 hydrotalcite structure (i.e., Ni/Al2O3-MnxOyY), where the elimination of carbon was much more efficient, and when Mn is partially substituted for nickel in the hydrotalcite matrix, (i.e., Ni/MnxOy-Al2O3), where we find the presence of coke deposition but at a lower extent to that of Ni/Al2O3 catalyst.
SEM and TEM study of post-reaction samples are presented in Figure 10 and 11 respectively. It is clear from these images that after CO2 reforming reaction, the carbon detected on Ni/Al2O3 and Ni/MnxOy-Al2O3 catalysts is in the form of nano-fibers. However, the absence of carbon traces was clearly noted when Ni/Al2O3-MnxOyY was used as catalyst; such result is in agreement with the obtained XRD results. In Table 3, we have incorporated the size values of Ni 0 particles obtained from SEM, TEM and XRD for all used catalysts after 50 hours of reaction time at 700 °C; this is performed in order to examine the evolution of the size of particles on the surface of the catalyst. Ni 0 particle size, estimated by SEM and TEM, are in agreement with those obtained by the XRD analyses.
The size of nickel particles increased after reaction for Ni/Al2O3 and Ni/MnxOy-Al2O3 (from 19 nm to 27 nm for the former and from 7.5 nm to 13.5 nm for the latter). The increase in particle size is owed to the sintering phenomenon of Ni particles after reaction. However, a very Figure 9. XRD of the used catalysts. •Mn3O4, ❖NiO, Ni 0 , ▲MnO and carbon.  small increase in the size of Ni 0 is observed in the case of Ni/Al2O3-MnxOyY catalyst (7.20 nm to 8.30 nm). These results show the beneficial effect of intercalation of Mn 2+ cations in the inter-layer space of the catalyst; it seems that in this case, the particles would be more resistant to agglomeration during the reaction.
TPO-O2 curves of different catalysts after reaction are shown in Figure 12. The TPO-O2 profile of the spent catalyst shows the presence of a broad peak between 450-700 °C in the case of Ni/MnxOy-Al2O3 catalyst. On the other hand, TPO of Ni/Al2O3 sample showed a broad peak with high intensity around 600 °C. The only oxidation peak appearing in the TPO-O2 profile of the spent catalysts (Ni/Al2O3 and Ni/MnxOy-Al2O3) was attributed to the oxidation of one type of carbonaceous species (filamentous carbon) on the surface of both samples [40,41]. In the case of Ni/Al2O3-MnxOyY catalyst, no oxidation peak was detected. As can be seen, the addition of Mn within the brucite-like layer of Ni/Al2O3 catalyst, prompted less carbon deposition during the reaction of CO2 reforming of methane. Therefore, we concluded that the addition of Mn in the interlayer space plays a much important role in the removal of surface carbon during the reaction of CO2 reforming of methane.
In order to study the effect of Mn positioning within the structure of our catalysts on the carbon deposition process, the valence state of Mn on the surface of these catalysts, after being tested under DRM for 50 h, is studied using XPS. Gan et al. [39] previously mentioned that the split of the binding energy (ΔE) in Mn-3s spectrum is dependent on the oxidation state of Mn.
XPS spectra of the catalysts after reaction are displayed in Figure 13. ΔE value of Mncontaining catalysts was 6.1 eV. This result indicated the presence of Mn 2+ species. The deposited carbon could be oxidized by Mn3O4 as previously mentioned (Mn3O4 + C ↔ 3MnO + CO). The improvement of catalytic performance of Ni/Al2O3-MnxOyY catalyst is mainly attributed to the enhanced resistance to carbon deposition, as demonstrated by TEM, SEM, XRD and TPO-O2.
In addition, the presence of MnxOy in Ni/Al2O3 hydrotalcite derived catalyst can also enhance the basicity of the support which can lead to a better CO2 adsorption on the surface. In our case, this clearly explains the elevated CO2 conversion observed during the reaction of CO2 reforming of methane. In an interesting work by Guo et al. [30] the authors successfully demonstrated how the addition of MnxOy to Ni/Al2O3 catalysts, synthesized by glow discharge plasma, influenced the amount of CO2 adsorption on the surface of their synthesized catalysts. Using CO2-TPD measurements, the authors showed that higher Mn content (10%) prompted better CO2 adsorption. In addition to the superior promotion of both CO2 and CH4 on the surface of our synthesized materials, we envisage that the presence of an in-situ elimination of carbon via Mn3O4/C/MnO/CO2 redox looping cycle could play a synergetic role which can lead to lowering the amount of surface carbon during the reaction of CO2 reforming of methane [40][41][42]. Where carbon residues are formed on the surface of reduced Ni particles, migration of carbon to neighboring Mn3O4 sites could be anticipated to enable the in-situ elimination of carbon.

Conclusion
Different types of catalysts of the type containing Ni/Al2O3-MnxOy have been prepared by thermal decomposition of Ni/Al2O3, Ni/MnxOy-Al2O3 and Ni/Al2O3-MnxOyY (Y = EDTA) hydrotalcite precursors. The synthesized catalysts were successfully characterized using TGA, ICP, XRD, BET, FTIR, TPR-H2, SEM-EDX, TEM, XPS, and TPO-O2 analysis, and applied for syngas production from CO2 reforming of methane at 700 °C. The hydrotalcite structures were confirmed using XRD analysis, while the intercalation of Mn in the interlayer space was verified using both XRD and FTIR. The XPS spectra indicated that the Mn species were reduced from 3+ to 2+ oxidation state, whereas Ni species were reduced from 2+ to 0; this was further confirmed by X-ray diffraction after reduction. The addition of Mn to Ni/Al2O3 increased the number of oxygen vacant sites on the support, which prompted an increase in CO2 adsorption/conversion. The results also showed that the activity and stability of Mnbased catalysts were related to the Mn positioning; the best catalytic activity was observed with Ni/Al2O3-MnxOyY. The characterization of Mn-based catalysts after reaction was also effected using XRD, SEM, TEM and TPO-O2; results showed that the lack of formation of carbon deposits responsible for catalyst deactivation corroborated the stability of the reaction tests for the studied material. However, in order to gain more insight about the mechanism responsible for the low carbon formation on the surface of our Mn containing catalysts, more work using state-of-the-art on-stream techniques will need to be performed to enable better understanding of the overall situation.