Effect of Citric Acid on MoO 3 / Al 2 O 3 Catalysts for Sulfur-Resistant Methanation

A series of MoO3/Al2O3 catalysts with different amounts (molar ratio of CA/Mo = 0, 1, 1.5, and 2) of citric acid (CA) prepared by simultaneous impregnation were evaluated for sulfur-resistant methanation. Based on the evaluation results, catalytic activity increased accompanied with the rise of citric acid amount. Combine with the analysis of N2-physisorption, XRD, H2-TPR, XPS, and TEM, the catalyst saturated loading capacity improved, resulting in increasing dispersion of Mo species on Al2O3 surface clearly. According to H2-TPR result, the Mo oxide precursors can be more easily sulfureted when citric acid is added. Moreover, based on the Raman analysis, increasingly tetrahedrally coordinated Mo6+ species with high methanation performance are generated after citric acid treatment. These factors probably together accelerate MoO3/Al2O3 catalytic activity growth for methanation.


Introduction
With the energy shortage and contamination crisis on the environment [1], the efficient and clean use of coal resources has become increasingly significant.The production of synthetic natural gas (SNG) from coal is an important method utilizing coal effectively [2].Methanation is an essential process for SNG production.Traditionally, Ni-based catalyst can be used for a methanation reaction [3,4], but it is extremely sensitive to sulfur and only occurs as the molar ratio of H 2 /CO is no less than 3:1.
Considering its good performance in water-gas shift (WGS) reactions [5], Mo-based catalysts can exhibit good methanation activity at relatively low H 2 /CO.As a consequence, the chemical equation for sulfur-resistant methanation on Mo-based catalyst can be expressed in the following equation [6].Moreover, the Mo-based catalysts resistance to sulfur poisoning has also been recognized [7][8][9][10][11].
Citric acid has been widely used as a chelating agent in hydrogenation reactions on Mo-based catalyst, and has been proven to be effective in increasing catalytic activity.Bergwerff et al. [12] reported that the agglomeration of Mo species can be removed by adding citric acid when the MoO 3 loading was beyond the saturated coverage of the catalyst.This suggested that citric acid could improve the dispersion of Mo species on the support surface efficiently.Furthermore, Wang et al. [10] found that citric acid could change the textural properties of catalysts and further increase the dispersion of active components.Similarly, Valencia et al. [13] and Peña et al. [14] also proved that citric acid was helpful in improving the Mo dispersion on support.Moreover, as one of the most important chelating agents, citric acid could increase the sulfidation of Mo species for sulfided Mo/Al 2 O 3 catalysts.Rinaldi et al. [15,16] reported that the addition of citric acid weaken the interaction of Mo species and Al 2 O 3 support obviously, resulting in a significantly increasing sulfidation of the catalyst and improved catalytic performance for the hydrodesulfurization (HDS) of thiophene.
Obviously, citric acid performs a great function in Mo-based catalysts for hydrogenation reactions.Our previous study [17] has reported the structure of MoO 3 /Al 2 O 3 catalysts and their catalytic performance for sulfur-resistant methanation reactions.In this work, the role of citric acid in MoO 3 /Al 2 O 3 catalyst was studied.A series of MoO 3 /Al 2 O 3 catalysts with different amounts of citric acid were prepared by simultaneous impregnation, and the prepared catalysts were tested for sulfur-resistant methanation and characterized by N 2 -physisorption, X-ray diffraction (XRD), H 2 -temperature programmed reduction (H 2 -TPR), transmission electron microscope (TEM), and Raman.

N 2 -Physisorption
The N 2 adsorption-desorption isotherms of all CAMo/Al 2 O 3 catalysts and Al 2 O 3 support are shown in Figure 1.All samples showed typical and similar IV curves, with hysteresis loops according to the IUPAC classification system.Therefore, it can be concluded that the mesoporous structure presents all catalysts and support.After adding citric acid into catalysts, the Brunauer-Emmett-Teller (BET) surface area, the average pore volume, and the pore size of all catalyst samples are smaller than the Al 2 O 3 support.Obviously, the impregnation of Mo may gather on the support and block the micropores of the catalysts.Therefore, the S BET , the average pore volume, and the pore size of all catalyst samples become smaller than the Al 2 O 3 support.The N 2 physisorption results of all CAMo/Al catalysts after the sulfidation are also listed in Table 1.Compared with the oxide catalysts, the sulfide catalysts possess a lower BET surface area and a higher pore size.With the increase in the CA/Mo ratio, the BET surface area and average pore volume of oxide catalysts and sulfide catalysts both increased gradually.Citric acid played an important role in modifying the texture properties of the catalysts.The change is attributed to the fact that the gas (CO 2 and water vapor), which is produced by the thermal decomposition of citric acid, can get through some holes inside the catalysts and block the agglomeration of MoO 3 .Moreover, the pore size diameters are not influenced by the addition of citric acid.Figure 2 shows the pore size distribution of all catalysts and Al 2 O 3 support, indicating that an average pore diameter of 3-5 nm is found among them.The impregnation of Mo does not change the pore size distribution of the catalyst.
Based on the textural properties of the catalysts, the molybdenum could be dispersed better on the support surface modified by citric acid.This viewpoint will be also confirmed by the results of XRD, TEM, and TPR in later sections.

XRD Analysis
The XRD patterns of Al Obviously, the lack of a crystalline MoO 3 peak indicates that MoO 3 is dispersed better as an amorphous form on the Al 2 O 3 supports modified by citric acid.Otherwise, according to our previous study [17], the Al 2 (MoO 4 ) 3 phase can be formed, as the MoO 3 loading exceeded the saturated monolayer coverage with calcination temperatures of 600 • C. Therefore, the disappearance of Al 2 (MoO 4 ) 3 should be attributed to the change in the catalyst saturated monolayer coverage.Combined with the increase in BET surface area and the average pore volume from the N 2 -physisorption results, it can be concluded that the saturated loading capacity of the catalyst can be improved when citric acid is added.

H 2 -TPR Analysis
The TPR profiles of the series of CAMo/Al catalysts, as well as Mo/Al catalyst, are shown in Figure 4.It can be seen that two main hydrogen consumption peaks around 430 and 840 • C were found in the profiles among them.According to Chary et al. [20], both tetrahedral and octahedral Mo 6+ species were present when the Mo loading was not low.The interaction between tetrahedral Mo 6+ and Al 2 O 3 is stronger than the interaction between octahedral Mo 6+ and Al 2 O 3 .The tetrahedral Mo 6+ species are difficult to reduce due to the strong interaction with support.Therefore, the first peak at the low temperature region is attributed to the partial reduction of octahedral Mo 6+ species from Mo 6+ to Mo 4+ and the reduction of small crystalline MoO 3 [21,22].Meanwhile, the second peak at the high temperature region is assigned to the reduction of the tetrahedral Mo 6+ species [23,24], and the further reduction of MoO 2 from Mo 4+ to Mo 0 [17].
As shown in Figure 4, the peaks in the low temperature region and the high temperature region both shifted to the left, significantly accompanied with the rise of citric acid amounts.On the one hand, the change in the peaks' positions implies that the reduction of different Mo species becomes easier when citric acid is added.It also means that the addition of citric acid can effectively weaken the interaction of Al 2 O 3 support and Mo species.As a result, it is easier to achieve sulfidation of Mo oxide precursors to generate sulfide catalyst with high CO methanation activity.
On the other hand, the peak at the low temperature zone consists of the superposition of two hydrogen consumption peaks.Specifically, the first hydrogen consumption peak is assigned to the octahedral Mo species (300-500 • C), and the second hydrogen consumption peak is attributed to the reduction of small crystalline MoO 3 to MoO 2 (500-600 • C).Therefore, the crystalline MoO 3 species decrease and the amorphous Mo species increase when the reduction peak in the low-temperature region shifts to the lower temperature side.As certified by the previous XRD analysis, the crystalline and agglomerate MoO 3 can be removed when citric acid is added.

TEM Analysis
Figure 5 shows the representative TEM images for Mo/Al and CAMo/Al catalysts with different molar ratios of CA/Mo.As shown in Figure 5a, the bulk MoO 3 species gathered on the surface of Al 2 O 3 .After the addition of citric acid into the precursors, the aggregation of MoO 3 is disappeared (see Figure 5b-d).Meanwhile, the needle-like Al 2 O 3 can be clearly observed, suggesting that the Mo species dispersed better on the Al 2 O 3 supports with the increase of citric acid content.
In order to explore the dispersion of Mo element in each catalyst, the mapping image of the catalysts has been observed.As shown in Figure 6, the yellow zone in the pictures stands for Mo; the black zone stands for other elements in the catalysts.In Figure 6a, the brightness of different yellow zones are not the same, and the distribution is unbalanced.However, in Figure 6b, the brightness and distribution of these yellow zones are totally different from those in Figure 6a.Therefore, the dispersion of Mo in the 2CAMo/Al catalyst is superior to that in the Mo/Al catalyst.To be summarized, based on the different characterization analysis above, citric acid performs a function in increasing the dispersion of Mo species on the support surface.As shown in Figure 3, the diffraction peaks of MoO 3 and Al 2 (MoO 4 ) 3 appear in the curve of the Mo/Al catalyst sample.However, no diffraction peaks corresponding to the crystalline MoO 3 and Al 2 (MoO 4 ) 3 were detected on any other CAMo/Al catalysts.Similarly, the H 2 -TPR profiles in Figure 4 also demonstrated that the crystalline MoO 3 decreases when citric acid content gradually increases.In fact, the results of N 2 -physisorption indicate that the disappearance of MoO 3 and Al 2 (MoO 4 ) 3 is due to the change of catalyst textural properties.Based on the increase of the catalyst BET surface area, the saturated loading capacity of the catalyst is improved.Therefore, the MoO 3 can disperse better on the CAMo/Al catalyst than on the Mo/Al catalyst.Moreover, this can be directly observed in the mapping images of TEM.

Raman Analysis
To obtain a deeper understanding about the function of citric acid on modifying catalysts, the Mo/Al and CAMo/Al catalysts were characterized by Raman analysis.The characterization results of all catalyst samples are presented in Figure 7.
According to Figure 7, the detection range of the Raman band for all samples is from 200 to 1000 cm −1 .The sharpest Raman peak, observed around 954 cm −1 , can be assigned to the symmetric stretching vibration modes of the terminal Mo=O bond, which is supposed to be an amorphous Mo 6+ species in an octahedral environment [25].Moreover, a relatively intense peak at 878 cm −1 and two weak peaks at 852 and 320 cm −1 are attributed to the Mo-O-Mo asymmetric stretching and to the bending modes of the tetrahedrally coordinated Mo 6+ species, respectively [26].Clearly, the peak intensity of tetrahedral and octahedral Mo 6+ species significantly changed when citric acid was added.The area ratio (S t /S o ) of the tetrahedral Mo 6+ specie peak with respect to the octahedral Mo 6+ specie peak can be used as a measure to detect the content of different Mo 6+ species among samples.As listed in Table 1, there is an apparent increase about the value of S t /S o from 0.08 to 0.13 when the ratio of CA to Mo is from 0 to 2. It is indicated that the amount of tetrahedral Mo 6+ species increases as citric acid content increased.
The addition of citric acid has an obvious effect on the change of the active catalytic precursor.In our previous study [17], the tetrahedrally coordinated Mo 6+ was proven to be an active catalytic precursor for sulfur-resistant methanation.Based on the variation trend of S t /S o for different catalysts from the Raman result in Table 1, citric acid helps to increase the relative content of tetrahedral Mo 6+ species in catalysts.Therefore, with the increase of tetrahedrally coordinated Mo 6+ , the methanation activity is supposed to gradually improve.This trend is consistent with the results of the catalytic activity evaluation described in later section.

XPS Analysis
Figure 8 shows the distribution of S in different catalysts after sulfidation.As shown in Figure 8, the ratio of S species to Al species on the catalyst surface is steadily increased with citric acid amounts.Especially when the value of n(CA)/n(Ce) is beyond 1, the growth rate of S/Al atomic ratio increases.Based on the XPS result, the addition of citric acid can increase the degree of sulfidation for the catalyst.Obviously, this conclusion is in agreement with the H 2 -TPR result.

Catalytic Activity Evaluation
The Mo/Al and a series of CAMo/Al catalysts were evaluated for CO methanation of syngas with a H 2 /CO ratio of 1:1 and 0.27% H 2 S. As shown in Figure 9, the CO conversion increased from 42% to 49% accompanied with the gradual rise of citric acid amounts.In addition, the hydrocarbon distribution of reaction products is listed in Table 2. Based on the data in Table 2, the selectivity of CH 4 , which is the destination product for all catalysts, maintains a high value beyond 96% during reaction.The selectivity of C 2 H 6 and C 3 H 8 are only around 3% and 0.2%, respectively.It is clear that citric acid cannot affect the selectivity of products.
According to the experimental results, the addition of citric acid can improve the activity of Mo/Al catalysts for methanation.On the one hand, the results of XRD (Figure 3) and H 2 -TPR (Figure 4) demonstrate that the inactive crystalline MoO 3 and Al 2 (MoO 4 ) 3 decreased or even disappeared after using citric acid.On the other hand, Raman analysis reveals that the amount of tetrahedral Mo 6+ species with high CO methanation activity increased.In addition, the H 2 -TPR and XPS also show that citric acid can weaken the Mo-support interaction and increase the degree of sulfidation for the catalyst.These factors are all in favor of increasing the CO methanation activity of the Mo/Al catalyst.
Meanwhile, as for both 1.5CAMo/Al and 2CAMo/Al catalysts, their catalytic performance are enhanced slowly in the first 15 h of reaction, and then hold steady.As the amount of active Mo oxide precursor increased, a growing number of Mo-S active sites were formed from reactions with syngas (including H 2 S) during this period.

Catalyst Preparation
The commercial γ-Al 2 O 3 powderwaspurchased from Yixing Qianye Non-metallic material Co. Ltd., Yixing, China, and the other reagents were purchased from Kemiou Chemical Reagent Co. Ltd., Tianjin, China.
All catalysts were prepared by simultaneous impregnating γ-Al 2 O 3 powder was impregnated with the aqueous solutions of (NH 4 ) 6 Mo 7 O 24 •4H 2 O and citric acid (CA).Then, the mixtures were continuously stirred at room temperature for 24 h, and subsequently dried at 120 • C for 12 h.Finally, the samples were calcined at 600 • C for 4 h at a heat rate of 5 • C/min.The MoO 3 content was fixed at 20 wt % with respect to Al 2 O 3 for all catalysts.The catalysts are denoted as xCAMo/Al, where x represents the molar ratio of citric acid to Mo and ranges from 0 to 2. When the molar ratio of citric acid to Mo is 0, the sample is denoted as Mo/Al.

N 2 -Physisorption
N 2 -physisorption of the catalysts was performed at −196 • C with a Tristar 3000 apparatus (Micromeritics, Norcross, GA, USA).As a pretreatment of the N 2 physisorption analysis, each fresh sample should be degassed at 300 • C for 4 h in vacuum.

XRD Analysis
The XRD was detected on a D/max-2500 X-ray diffractometer (Rigaku, Tokyo, Japan) with Ni-filtered Cu-Kα radiation (λ = 1.54056Å).The angles of scanning were from 10 • to 90 • with a rate of 5 • /min.The XRD phases were identified by comparison with powder diffraction data from the Joint Committee on Powder Diffraction Standards (JCPDS).

H 2 -TPR Analysis
TPR profiles of the prepared samples were obtained by 2910 Automatic chemical adsorption instrument (Micromeritics).Before each TPR measurement, approximately 200 mg of catalyst was flushed with 99.999% argon at 200 • C for 40 min to remove traces of water.A gas mixture of 10% (v/v) of hydrogen in argon was used at a flow rate of 30 mL/min, and the temperature was increased at a heating rate of 10 • C/min from 60 to 1000 • C.

TEM Analysis
The morphology and structure of the catalyst samples were observed by a Tecnai G2 F20 (200 kV) transmission electron microscope (TEM) (FEI, Eindhoven, The Netherland), with a high resolution of 0.15 nm/200 kV.The catalyst samples were prepared via ultrasonic dispersion in ethanol, and a drop of the suspension was then put on a lacey carbon film supported on a Cu grid.

Raman Analysis
The Raman spectra of Mo/Al and samples with a citric acid addition were obtained using an InVia-Reflex (Renishaw, Wotton-under-Edge, Gloucestershire, UK) laser Raman spectrometer with a highly sensitive system of integrated research grade microscopes.The 532 nm line of an Ar + laser was used.
3.2.6.X-ray Photoelectron Spectroscopy Analysis X-ray photoelectron spectroscopy was analyzed using Perkin Elmer PHI-1600 XPS spectrometer (PerkinElmer, Waltham, MA, USA) with Mg Kα X-ray radiation.The binding energy of C 1s at 284.6 eV was a reference to the binding energies.

Catalytic Activity Evaluation
The evaluation of sulfur-resistant methanation activity was carried out in a continuous-flow fixed-bed reactor (inner diameter: 12 mm; length: 700 mm) with a gas chromatograph equipped with a thermal conductivity detector and a hydrogen flame ionization detector.The loading quantity was 3 mL of catalyst for each experiment.The catalyst (3 mL) was sulfurized under a gas mixture of 3.0 vol % H 2 S/H 2 at 400 • C for 4 h before the catalytic activity test.The activity test conditions were as follows: syngas (H 2 /CO = 1.0) containing 0.27 vol % H 2 S and 10 vol % N 2 , a gas hourly space velocity of 5000 h −1 , at 550 • C and at 3 MPa.
The CO conversion and selectivity to hydrocarbons were calculated by using the following expressions:

Conclusions
In the present study, a series of CAMo/Al catalysts prepared by simultaneous impregnation were tested for sulfur-resistant methanation.The activity of the CAMo/Al catalysts enhanced with the increase in citric acid content.Based on the characterization analysis, the following conclusions can be drawn.First, the addition of citric acid can change the textural properties of a catalyst and improves the dispersion for all kinds of Mo species on catalyst surfaces.Second, it can translate crystalline MoO 3 and Al 2 (MoO 4 ) 3 into unformed Mo species, with relatively high catalytic activity.Moreover, the interaction of Al 2 O 3 support with Mo species was gradually weakened when different amounts of citric acid were added.Therefore, MoO 3 is more easily sulfureted to MoS 2 , with highly catalytic activity in H 2 /H 2 S. In addition, the more tetrahedrally coordinated Mo 6+ with high methanation activity can be formed using citric acid in this way.
2 O 3 support and a series of CAMo/Al catalysts with different molar ratio of CA/Mo are shown in Figure 3.For all catalysts, the diffraction peaks 2θ = 45.8 • and 67.1 • are assigned to γ-Al 2 O 3 (PDF-10-0425) [18].The Mo/Al catalyst exhibits diffraction peaks at 2θ = 23.4• and 25.7 • which are assigned to the crystalline MoO 3 phase [19], and other peaks at 2θ = 20.7 • , 22.0 • , 23.4 • and 25.7 • are attributed to the Al 2 (MoO 4 ) 3 phase.After the citric acid addition, the characteristic diffraction peaks of MoO 3 and Al 2 (MoO 4 ) 3 disappeared.All CAMo/Al catalysts and Al 2 O 3 support only exhibited typical diffraction peaks of γ-Al 2 O 3 .

Figure 8 .
Figure 8. Elemental distribution of S in different catalysts after sulfidation.

Figure 9 .
Figure 9. CO Conversion with time on stream on catalysts.

Table 1 .
Textural properties and Raman results of catalysts and Al 2 O 3 support.

Samples BET Surface Area (m 2 /g) Pore Size (nm) Pore Volume (cm 3 /g) S t /S o 1
/ Figure 2. Pore size distribution of catalysts and support.

Table 2 .
Catalytic behavior of catalysts.