Studies on molybdenum carbide supported HZSM-5 (Si/Al = 23, 30, 50 and 80) catalysts for aromatization of methane

2.1 Catalyst synthesis

For nominal loadings of 6 wt% of molybdenum, about 1.104 g of ammonium heptamolybdate tetrahydrate was dissolved in 100 mL of deionized water. To this solution 9.4 g of HZSM-5 (Si/Al = 23, 30, 50 and 80) was added and the resultant mixture was aged for 2 h. The excess water was evaporated by using hot plate at 120 °C. The obtained solid particles were dried in a preheated oven at 110 °C. The dried samples were calcined at 500 °C in static air for 4 h. The resultant samples were denoted as MoO₃/HZSM-5 = 23 for 6 wt% of molybdenum deposited HZSM-5 with Si to Al ratio = 23 and so on so forth. Further, Mo/HZSM-5 catalyst denotation represents both MoO₃/HZSM-5 and spent Mo₂C/HZSM-5 samples respectively with Si to Al ratio = 23, 30, 50 and 80.

2.2 Carburization

For example, 0.5 g of calcined MoO₃/HZSM-5 = 23 was placed in a quartz reactor and CH₄/H₂ gas mixture in a volume ratio of 1/4 with a flow rate of 50 mL.min⁻ was passed through the loaded sample. The reactor temperature was maintained at 700 or 750 °C for 30 min. Subsequently, the reactor was purged with N₂ gas at a flow rate of 20 mL min⁻ for 30 min followed by non-oxidative MDA reaction was carried out on these catalysts. Identical procedure was followed for the carburization of MoO₃/HZSM-5 catalysts with different Si to Al ratio.

2.3 Characterization of samples

All the studied HZSM-5 (with Si to Al ratio = 23, 30, 50 and 80), MoO₃/HZSM-5 and spent Mo₂C/HZSM-5 at 700 or 750 °C samples were characterized by using the following techniques.

The BET surface area-pore size analysis was done by using Quantachrome Nova Station (USA). The samples were pretreated under vacuum for 2 h at 200 °C. XRD analysis was done by using EQUINOX 1000 Inel XRD machine at Co Kα = 1.7902 Å with acquisition of 2θ from 10° to 110°. XPS analysis was done by using SPECS GmbH analysis system containing Mg Kα 1253.6 eV X-ray source. The samples binding energy (BE) was attained by using C1s 284.8 eV correction.

NH₃-TPD and TPO analysis results were obtained by using micromeritics AutoChem HP 2950 V3.02 instrument. The mass spectral data was obtained by using ThermoStarTM GSD 320 quad core mass spectrometer. The m/z values followed were: m/z = 17 (NH₃), m/z = 18 (H₂O), m/z = 28 (CO) and m/z = 44(CO₂). In a typical experiment, spent samples of 100 mg each was placed in a quartz tube and pretreated at 200 °C in He flow (10 cm³ min⁻¹) for 2 h. Subsequently, the temperature of the sample was brought to 40 °C and saturated with 5% NH₃-N₂ for 1 h. Then, the sample was purged with He flow (10 cm³ min⁻¹) for 1 h. Desorption of ammonia was performed over the temperature range of 40–800 °C at a ramping rate of 10 °C min⁻¹. Desorption stream was analyzed simultaneously using a TCD and mass detector by means of an automated split valve.

For TPO experiments about 100 mg of each spent catalyst was loaded in quartz tube and 1%O₂-He gas mixture was used as a probe gas. Desorption of CO₂ was recorded over the temperature range of 40–800 °C at a ramping rate of 10 °C min⁻¹. Desorption stream was analyzed by TCD and mass detector.

SEM-EDX was done by using field emission scanning electron microscopy (FE-SEM, Quanta FEG450, FEI) equipped with an Everhart Thornley detector (ETD, HV mode) and a solid-state back scattering electron detector (VCD). High resolution transmission electron microscopy (HR-TEM) and fast fourier transform (FFT) results of spent Mo₂C/HZSM-5 = 23 and 80 was collected on Tecnai 200 kV D1234 SuperTwin microscope with camera length of 97 cm. EPR spectra were recorded by using JEOL JES-FA 100 EPR spectrometer operating in the X-band with standard TE₀₁₁ cylindrical resonator at 123 K. TG/DTG analysis was carried out on STA-449 F3, NETZSCH instrument by using 20 mg of spent sample loading.

2.4 Activity tests

Catalytic activity tests were carried out in a continuous flow fixed bed quartz reactor at ambient pressure with methane GHSV of 1800 mL.g_cat⁻.h⁻. Pre-carburized catalyst at 700 or 750 °C was introduced with CH₄/N₂ gas mixture at a volume ratio of 9/1. The product stream was analyzed by a gas chromatograph equipped with a HaySep D 80/100 column, to analyze gases, such as H₂, N₂, CO, CO₂, CH₄, C₂H₄, and C₂H₆, by TCD and a HP-1 capillary column was used to analyze condensable aromatics, such as benzene, toluene, xylenes, naphthalene and methyl naphthalene by FID. The resultant catalysts were labeled as spent Mo₂C/HZSM-5 = 23, 700 °C, for 6 wt% of molybdenum deposited HZSM-5 with Si to Al ratio = 23 at 700 °C reaction temperature and spent Mo₂C/HZSM-5 = 23, 750 °C, for 6 wt% of molybdenum deposited HZSM-5 with Si to Al ratio = 23 at 750 °C reaction temperature and so on so forth. $$\%\text{Conversion}\text{C}{\text{H}}_{\text{4}}=\frac{\text{moles}\text{of}\text{C}{\text{H}}_{\text{4}\text{in}}-\text{moles}\text{of}\text{C}{\text{H}}_{\text{4}\text{out}}}{\text{moles}\text{of}\text{C}{\text{H}}_{4\text{in}}}\times 100$$ $$\%\text{Selectivity}=\frac{\text{moles}\text{of}\text{produc}{\text{t}}_{\text{out}}}{\text{moles}\text{of}\text{C}{\text{H}}_{4\text{in}}-\text{moles}\text{of}\text{C}{\text{H}}_{\text{4}\text{out}}}\times 100$$ $$\%\text{Yield}=\frac{\text{moles}\text{of}\text{produc}{\text{t}}_{\text{out}}}{\text{moles}\text{of}\text{C}{\text{H}}_{4\text{in}}}\times 100$$

3.1 XRD studies

Powder X-ray diffraction analysis (XRD) results of HZSM-5 and Mo/HZSM-5 samples are presented in Fig. 1a–d. Essentially, support HZSM-5 demonstrated X-ray diffraction signals at 2θ = 9.8, 10.2, 28.0, 29.0 and 29.5° ascribed to zeolite framework (MFI). A part from zeolite signals, other X-ray signals related to MoO₃ (2θ = 30.5, 32.5, 39.8 and 46.0°) and Mo₂C (2θ = 40.0, 44.0 and 45.5°) phases were observed in Mo/HZSM-5 samples. Primarily, MoO₃ phase was observed in calcined MoO₃/HZSM-5 = 23 and 30 samples. Whereas, MoO₃ phase was not detected in MoO₃/HZSM-5 = 50 and 80 samples. It might be associated with the presence of small sized MoO₃ (<4nm) crystals on high surface area HZSM-5 = 50 and 80. Evaluation of support, calcined and spent samples XRD results in Fig. 1a–d suggests Mo₂C phase formation was taken place by MoO₃ reduction under studied reaction conditions. However, presence of MoO₃ species in 700 °C spent Mo₂C/HZSM-5 = 50 and 80 samples emphasize partial reduction of MoO₃ into molybdenum carbide at this temperature. Further, at 750 °C, Mo₂C phase formation was facile over 700 °C spent Mo₂C/HZSM-5 samples. Interestingly, spent Mo₂C/HZSM-5 = 23 and 30 samples exhibited new X-ray diffraction signals at 2θ = 12.0, 22.0, 30.0, 34.0 and 42.5° are associated with part of HZSM-5 structure as reported in PDF-00-049-0657. On the other hand, X-ray diffraction signals related to part of HZSM-5 structure were minimized in spent Mo₂C/HZSM-5 = 50 and 80 samples. According to Gao et al. (2014) part of zeolite structure consists of external framework Al and Si sites which cut out from ZSM-5 zeolite cluster. Hence, in our case, part of HZSM-5 structure was observed after reaction suggests fraction of Mo species anchoring with extra-framework Al and Si sites in Mo₂C/HZSM-5 samples.

Close

Fig. 1

XRD analysis of HZSM-5 and Mo/HZSM-5 samples.

Export to PPT

3.2 TEM/HR-TEM analysis

TEM and HR-TEM analysis of spent Mo₂C/HZSM-5 = 23 and 80 was presented as Fig. 2a–e. The TEM image in Fig. 2a shows finely distributed molybdenum carbide particles on the surface of HZSM-5. The distributed Mo₂C particles were sized in the range of 6–14 nm with distorted spherical shape. High resolution TEM image in Fig. 2b depicts the micro crystalline structure of molybdenum carbide with d spacing of ≈0.229 nm for crystalline plane (1 0 1) in Mo₂C. Further, FFT analysis (Fig. 2c and e) emphasizes Mo₂C (1 0 1) and HZSM-5 crystal plane existence in spent Mo₂C/HZSM-5 = 23 sample. Essentially, TEM analysis of spent Mo₂C/HZSM-5 = 80 (Fig. 2d) also showed fine distribution of Mo₂C particles with particle size between 6 and 22 nm. The increase in the Mo₂C particle size might be associated with its aggregation due to weak interactions with fewer acidity HZSM-5 = 80. It is obvious from these images that fine dispersion of Mo₂C particles on HZSM-5 in Mo₂C/HZSM-5 samples.

Close

Fig. 2

(a) TEM analysis of Mo₂C/HZSM-5 = 23 (b) HR-TEM analysis of Mo₂C/HZSM-5 = 23 (c) FFT analysis for Mo₂C (d) TEM analysis of Mo₂C/HZSM-5 = 80 and (e) FFT analysis for HZSM-5.

Export to PPT

3.3 XPS analysis

XPS analysis of spent Mo₂C/HZSM-5 samples at 700 °C are presented as Fig. 3a–d. After deconvolution of XPS signal, Mo⁶⁺ (≈232.2 eV), Mo⁴⁺ (≈230.5 eV), Mo²⁺ (≈228.5 eV) and Mo⁰ (≈227.2 eV) molybdenum species were observed in the studied samples. In general, existence of Mo⁶⁺ oxidation state for molybdenum suggests the presence of MoO₃. Whereas, Mo⁴⁺ oxidation state reflects the presence of MoO₂, a partially reduced species of MoO₃. Further, lower oxidation state molybdenum (Mo²⁺ and or Mo³⁺) suggests the presence of Mo₂C in the studied samples (Oshikawa et al., 2001). Existence of Mo⁰ suggests the metallic molybdenum species in the spent samples. However, surface metallic molybdenum formation was found high in higher Si to Al ratio samples over lower Si to Al ratio samples (Supplementary Table 1). It could be due to the ease reduction of Mo₂C species into metallic molybdenum for higher Si to Al ratio samples (Si/Al = 50 and 80) and associated with weak interactions between Mo₂C and HZSM-5 = 50 and 80 with fewer acidity. Moreover, slightly enhanced metallic molybdenum formation was observed in 750 °C spent Mo₂C/HZSM-5 = 50 and 80 samples suggests increase in the reaction temperature can also reduce Mo₂C into metallic Mo in these samples.

Close

Fig. 3

XPS analysis of spent samples at 700 °C (a) Mo₂C/HZSM-5 = 23 (b) Mo₂C/HZSM-5 = 30 (c) Mo₂C/HZSM-5 = 50 and (d) Mo₂C/HZSM-5 = 80.

Export to PPT

3.4 TPO studies

The nature of carbon deposits on spent Mo₂C/HZSM-5 samples were studied by TPO-mass technique (m/z = 44, CO₂) and the results are presented as Fig. 4. In general, carbon deposits formation in non-oxidative MDA was due to dehydrogenated methane (“CH_x”) in parallel with desired C—C bond formation step on Mo₂C and or oligomerization of lower molecular weight intermediates (“C₂H_y”) to desired product (benzene, naphthalene) on the Bronsted acid sites of zeolite (Spivey et al., 2014). Four types of CO₂ desorption signals were observed for spent Mo₂C/HZSM-5 = 23 and 30 samples respectively at T_max = 345, 440, 475 and 600 °C. On the other hand, only three types of CO₂ desorption signals were observed for Mo₂C/HZSM-5 = 50 & 80 samples at T_max = 340, 440 and 475 °C respectively. Among the CO₂ desorption signals, the maximum desorption was observed in the temperature range of 350–550 °C for all the studied samples. The CO₂, desorption upto 550 °C might be associated with more reactive coke that forms directly from dehydrogenated methane and or degradation of linear hydrocarbons on Mo₂C, metallic molybdenum and extra-framework Mo species. Desorption of CO₂ at about 600 °C might be associated with inert coke deposited on the Bronsted acid sites inside the zeolite channels (Tessonnier et al., 2008). Among the Studied catalysts the carbon deposits found high on spent Mo₂C/HZSM-5 = 80 (571 µmol g⁻¹) followed by Mo₂C/HZSM-5 = 50 (466 µmol g⁻¹), Mo₂C/HZSM-5 = 30 (428 µmol g⁻¹) and Mo₂C/HZSM-5 = 23 (390 µmol g⁻¹). It is obvious from the TPO results that non-aromatic carbon deposits (<550 °C) formation was high in higher Si to Al ratio samples over lower Si to Al ratio ones.

Close

Fig. 4

TPO analysis of (a) Spent Mo₂C/HZSM5 = 23 (b) Spent Mo₂C/HZSM5 = 30 (c) Spent Mo₂C/HZSM5 = 50 (d) Spent Mo₂C/HZSM5 = 80 samples.

Export to PPT

3.5 NH₃-TPD-mass analysis

NH₃-TPD-mass analysis results are presented as Fig. 5. Two distinct ammonia desorption peaks were observed for all the spent Mo₂C/HZSM-5 samples. The first desorption was appeared in the temperature range of 75–275 °C with temperature maxima (T_max) about 140 °C. The second desorption was appeared in the temperature range of 325–575 °C with T_max about 350 °C. After deconvolution, six ammonia desorption peaks were observed for spent Mo₂C/HZSM-5 = 23 and 30 samples respectively at T_max about 135, 185, 270, 350, 390 and 490 °C. Whereas, five deconvolution signals were observed for spent Mo₂C/HZSM-5 = 50 and 80 samples at T_max about 120, 155, 230, 350 and 430 °C respectively. It is obvious from the results that ammonia desorption peaks were shifted to lower temperature with increase in the Si to Al ratio of HZSM-5. Acid sites of weak strength was reported in the temperature range of 150–190 °C and acid sites of strong strength was reported in the temperature range of 360–390 °C for HZSM-5 support through ammonia TPD analysis (Rodríguez-González et al., 2007). Acid sites of weak strength were attributed to silanol groups and the acid sites of strong strength were ascribed to bridging hydroxyl groups (Si-OH-Al) of HZSM-5 (Yu et al., 2019). On the other hand, calcined 4 wt% Mo/HZSM-5 = 15, 25 and 40 samples ammonia TPD analysis results yielded three desorption peaks at 200, 280 and 450 °C respectively for acid sites of weak, moderate (extra-framework aluminum) and very strong strength (located inside the channels of zeolite) (Tessonnier et al., 2008). Furthermore, bulk molybdenum carbide exhibited two distinct ammonia desorption peaks in the temperature range of 50–250 °C and 360–500 °C respectively (Bej et al., 2003). In the present study, the ammonia desorption below 150 °C might be associated with Mo₂C acid sites exhibited due to the charge transfer between molybdenum and C atoms (Bej et al., 2003). The remaining ammonia desorption peaks might be associated with HZSM-5 and or part of HZSM-5 structure with acid sites of weak, moderate, strong and very strong strength. The decreasing order of acidity for spent Mo₂C/HZSM-5 samples at 700 °C as follows: Mo₂C/HZSM-5 = 23 (856 µmol g⁻¹) > Mo₂C/HZSM-5 = 30 (646 µmol g⁻¹) > Mo₂C/HZSM-5 = 50 (393 µmol g⁻¹) > Mo₂C/HZSM-5 = 80 (277 µmol g⁻¹).

Close

Fig. 5

NH₃-TPD-mass analysis of spent Mo₂C/HZSM-5 catalysts.

Export to PPT

3.6 EPR studies

EPR spectra of spent Mo₂C/HZSM-5 samples are presented as Fig. 6. The EPR spectra was recorded in the magnetic field range of 300 to 360 mT. The intensive narrow EPR signal appeared at g factor equals to 2.00 was the characteristic band for carbon deposits in the studied samples. Apart from this signal, three more signals were observed for spent Mo₂C/HZSM-5 samples at g factor of 1.97, g_|| =1.89 and g_⊥= 1.93. EPR signal at 1.97 was observed for Mo/γ-Al₂O₃ system by Kostova et al. (2005) and it was attributed to the interaction between Mo and the support. Further, Kucherov et al. (2014) reported EPR signal at 1.96 for Mo/HZSM-30 system and attributed to the transition of molybdenum cations in the zeolite to paramagnetic Mo⁵⁺ ions under the reductive reaction atmosphere. Hence, in the present study, the signal observed at g factor of 1.97 is attributed to the presence of Mo⁵⁺ ions at the exchange sites of zeolite. On the other hand, the axial signal exhibited at g_|| =1.8966 and g_⊥= 1.9326 for Mo⁵⁺ paramagnetic ions was due to the aluminum molybdate (Al₂(MoO₄)₃) extra phase formed upon dealumination of the zeolite in the presence of molybdenum (Ha et al., 2013). However, all these signals were found weak in higher Si to Al ratio samples studied under identical EPR conditions. These results are in agreement with XRD data wherein, part of HZSM-5 structure was observed in spent Mo₂C/HZSM-5 samples.

Close

Fig. 6

EPR analysis of spent Mo₂C/HZSM-5 samples at 700 °C.

Export to PPT