Investigating the Impact of Na 2 WO 4 Doping in La 2 O 3 -Catalyzed OCM Reaction: A Structure–Activity Study via In Situ XRD-MS

: The La 2 O 3 catalyst exhibits good performance in OCM reactions for its promising C 2 selectivity and yield. Previous studies have affirmed that the formation of carbonates in La 2 O 3 impedes the catalyst’s activity as a result of poisoning from CO 2 exposure. In this study, a series of Na 2 WO 4 -impregnated La 2 O 3 catalysts were synthesized to investigate the poisoning-resistant effect. The bulk phase and kinetics of the catalysts were analyzed in reactors employed with in situ XRD-MS and online MS, focusing on the CO 2 adsorption on La 2 O 3 and the phase transition process to La 2 O 2 CO 3 in temperature zone correlated to OCM light-off. In situ XRD analysis revealed that, with Na 2 WO 4 doped, CO 2 exposure at elevated temperatures formed La 2 O 2 CO 3 in tetragonal crystal phases, exhibiting distinctive differences from the hexagonal phase carbonates in undoped commercial La 2 O 3 . The ability to develop tetragonal or monoclinic La 2 O 2 CO 3 was suggested as a descriptor to assess the sensitivity of La 2 O 3 catalysts to CO 2 adsorption, a tunable characteristic found in this study through varying Na 2 WO 4 doping levels. Coupled XRD-MS analysis of CO 2 adsorption uptake and phase change further confirmed a positive dependence between the resistivity of La 2 O 3 catalyst to CO 2 adsorption and its low-temperature C 2 selectivity. The results extended the previous CO 2 poisoning effect from multiple perspectives, offering a novel modification approach for enhancing the low-temperature performance of La 2 O 3 catalysts in OCM.


Introduction
The oxidative coupling of methane (OCM) reaction is deemed a promising industrial process, as it enables the direct conversion of CH 4 into C 2 H 4 and C 2 H 6 , which are essential precursors in the industrial realm [1][2][3][4].The obstacle preventing the large-scale industrialization of the OCM reaction is the restricted C 2 yield and selectivity at high reaction temperatures [5][6][7][8].Hence, catalysts exhibiting high catalytic activity at low temperatures in OCM reactions are indispensable to meet industrial requirements.Following the groundbreaking contributions by Keller and Bhasin, numerous catalytic materials have undergone investigation for the OCM reaction [8,9].Currently, La 2 O 3 has undergone extensive research and is acknowledged as one of the most effective catalysts for OCM reactions [2,3,8,10].In general, most OCM reaction systems, including La 2 O 3 , require very high temperatures around or above 800 • C [11], which is important to accomplish the cleavage of C-H bond and CH 3 radical desorption to form methyl radicals.On the other hand, the overall OCM reaction is highly exothermic [12].While OCM is generally recognized as a catalytic process, its efficiency is influenced by synergistic gas-phase chemistry, which is affected by factors, such as residence time, pressure, and the ratio of CH 4 to O 2 .
Experimental studies indicate that the rate of C 2 formation is heavily reliant on CH 3 concentrations [13].The formation rate of methyl radicals is only mildly affected by temperature Catalysts 2024, 14, 150 2 of 15 but varies significantly depending on the specific catalyst employed.Considering both the thermodynamics and kinetics, it is imperative to improve the performance of OCM at low temperatures.
As a single metal oxide species, La 2 O 3 demonstrates susceptibility to carbonation by CO 2 exposure, either from the atmosphere or the OCM reaction circumstance, resulting in the formation of La 2 O 2 CO 3 , which is exclusively generated at temperatures exceeding 500 • C [11,[14][15][16][17][18][19].Several relevant studies have focused on La 2 O 2 CO 3 , regarding whether CO 3  2− exerts an inhibitory or promotive effect on the OCM reaction activity of La 2 O 3 [15,17,[20][21][22][23].CO 2 is the primary by-product of La 2 O 3 -catalyzed OCM [24], of which the yield is around 10% accompanied by a 25% CH 4 conversion rate and nearly 50% selectivity of C 2 in the reaction [12].The in situ surface and bulk structure formation of La 2 O 2 CO 3 from pure La 2 O 3 under a variety of OCM-correlated conditions have been investigated by our group in detail using XPS and XRD [11,25].
Taylor et al. conducted a comparison of various synthesized lanthanum salts and suggested that the rank of C 2 activity is as follows: La 2 O 2 CO 3 > La 2 (CO 3 ) 3 > La(OH) 3 > La 2 O 3 [26].Other studies also agreed that synthesized lanthanum catalyst containing carbonate exhibits excellent performance [27].However, under high-temperature OCM reaction conditions, most La 2 O 2 CO 3 will undergo thermal decomposition and be reduced back into purified La 2 O 3 on the catalytic surface.On the other hand, most in situ structure or real-time activity-related studies, in which the CO 2 input is applied as a reaction parameter, concluded that carbonates have a negative impact in La 2 O 3 -catalyzed OCM reaction.Through 16 O and 18 O isotope and SSITKA analysis, Lacombe et al. concluded that, when exposed to CO 2 at 750 • C, La 2 O 3 will strongly inhibit surface oxygen dissociation [28,29].Yide Xu et al. observed a negative impact on the CH 4 conversion rate, C 2 selectivity, and C = 2 /C 0 2 ratio when CO 2 was added to the OCM reaction catalyzed by 10% La 2 O 3 /ZnO [30].In our previous publication, by employing in situ XRD-MS, La 2 O 2 CO 3 to La 2 O 3 ratio was quantitatively estimated and controlled by adjusting CO 2 exposure.Characterization afterward revealed that the higher the initial La 2 O 2 CO 3 level in the catalyst bulk phase, the higher the OCM light-off temperature of both CO x and C 2 products in OCM, which is a typical poisoning descriptor [11].XPS results after in situ CO 2 -correlated treatment also revealed that the sample pre-carbonated by CO 2 exhibits an activation temperature nearly 65 • C higher than the one with a pristine La 2 O 3 surface [25].Combined with in situ kinetic correlated structure studies and detailed DFT modeling, it revealed a thermodynamically favored peroxide structure as the active oxygen site.On La 2 O 3 , this peroxide is formed on the subsurface six-coordinated oxygen site, which is also the same site of carbonate formation.The model provided a reasonable preliminary explanation from the perspective of chemical pathways that the La 2 O 2 CO 3 acts as a poisoning species in La 2 O 3 -catalyzed OCM reaction; however, La 2 O 2 CO 3 is a very robust species.In an inert environment, such as in a vacuum or an Ar-filled reactor, the complete decomposition of La 2 O 2 CO 3 back to purified La 2 O 3 requires a high temperature of around 800 • C [11].In an atmosphere with CO 2 at the same high temperatures, the catalyst sample could still contain both components [11,12,15,23,[31][32][33][34].This explains why OCM requires a very high reaction temperature as it is a necessity to decompose the poisoned sites.
Further in situ studies correlated the higher resistance to carbonation of La 2 O 3 catalysts to its better low-temperature OCM activity [35].In situ XRD-MS was applied to investigate the CO 2 adsorption behavior and OCM reactivity of two La 2 O 3 catalysts, a lab-synthesized nanorod La 2 O 3 [23,33,34] and a commercial La 2 O 3 supplied by [11].In the identical temperature range, the apparent activation energy for the OCM reaction of nanorod La 2 O 3 is approximately 60 kJ mol −1 lower than that observed for commercial La 2 O 3 [35].The online MS CO 2 adsorption curve and the coupled in situ XRD indicate that, under the same exposure condition of CO 2 concentration and heating temperature, nanorod La 2 O 3 always absorbs less CO 2 than commercial La 2 O 3 .This case by itself strongly suggests that increasing the resistivity of CO 2 adsorption of La 2 O 3 is an effective method to enhance the low-temperature OCM performance of La 2 O 3 , which is also a natural consequence of the above theory that CO 2 acts as the poison in La 2 O 3 .
On the other hand, Na 2 WO 4 is widely acknowledged as a highly effective additive in catalyst formulations, particularly in the case of the Mn x O y -Na 2 WO 4 /SiO 2 catalyst, which demonstrates excellent performance in the OCM reaction [36][37][38][39][40][41][42]  -adsorption-correlated structure behavior and the low-temperature OCM reaction activity.In situ XRD allows real-time monitoring of the pure oxide to dioxymonocarbonate phase changes in La 2 O 3 during the linear heating process under CO 2 exposure, while the coupled online MS collects the real-time CO 2 adsorption from reaction outlet gas providing a quantitative estimation of the uptake.By combining the changes in bulk structure and gas composition through the time temperature correspondence, the resistance of the samples to CO 2 adsorption can be directly compared through the experiment results.It was found that, after doping with Na 2 WO 4 , a series of new behaviors were induced in the commercial La 2 O 3 catalysts.First, the online MS revealed that the CO 2 adsorption temperature on the doped La 2 O 3 sample is significantly increased.The in situ XRD also observed that the formed La 2 O 2 CO 3 crystal structure after CO 2 exposure changed from hexagonal to monoclinic/tetragonal with Na 2 WO 4 doped over La 2 O 3 .Both the bulk phase crystal structure of the formed dioxymonocarbonates and the CO 2 uptake temperature are characteristics of the nanorod La 2 O 3 [23].In the end, the OCM reaction activity results were acquired in a microreactor coupled with online MS, and the doped samples were found to have higher conversion and selectivity at 500-650 • C than the undoped La 2 O 3 .Combining the CO 2 adsorption and OCM activity results, it becomes evident that the CO 2 adsorption resistance of the La 2 O 3 catalyst is critical for better low-temperature OCM performance.The results of this article provide a new approach and possibility for improving La 2 O 3 .As generally reported, it is expected that Na 2 WO 4 exists in the bulk phase as a cubic crystal structure [44].However, the XRD pattern of all three M-La 2 O 3 _nW samples at room temperature did not reveal any signal correlated to Na 2 WO 4 .This absence could be attributed to the low content and effective dispersion of Na 2 WO 4 , so the grain size is not adequate to generate diffraction patterns.

Results and Discussion
To confirm that Na 2 WO 4 was indeed doped into the sample, XPS analysis was performed for M-La 2 O 3 _1W, 3W, and 5W samples (calcined at 800 • C) revealing the surface elements, their content, and electronic structure.As shown in Figure 1b (left), W 4f spectra exhibit the main W 4f 7/2 peak at 35.5 eV confirming the +6 oxidation state for all samples [45].While the only possible La oxidation state is +3 (metallic La is not expected at these conditions), different La 3+ compounds are known to demonstrate various La 3d 5/2 multiplet splitting.The La 3d 5/2 spectra presented in Figure 1b (right) show the splitting of 4.5 eV, which is clearly associated with La 2 O 3 (without hydroxide or carbonate) [46,47].With the increase in Na 2 WO 4 doping amount, the W 4f peaks intensity grows being in agreement with the samples preparation procedure.The atomic percentages of W and La (normalized to the total of W and La, other elements not considered) given in Table 1 demonstrate the increase in W content on the catalyst surface in agreement with the increase in Na 2 WO 4 loading.
Catalysts 2024, 14, x FOR PEER REVIEW 4 of 15 exhibit the main W 4f7/2 peak at 35.5 eV confirming the +6 oxidation state for all samples [45].While the only possible La oxidation state is +3 (metallic La is not expected at these conditions), different La 3+ compounds are known to demonstrate various La 3d5/2 multiplet splitting.The La 3d5/2 spectra presented in Figure 1b (right) show the splitting of 4.5 eV, which is clearly associated with La2O3 (without hydroxide or carbonate) [46,47].With the increase in Na2WO4 doping amount, the W 4f peaks intensity grows being in agreement with the samples preparation procedure.The atomic percentages of W and La (normalized to the total of W and La, other elements not considered) given in Table 1 demonstrate the increase in W content on the catalyst surface in agreement with the increase in Na2WO4 loading.As mentioned in previous studies, it was observed that the La2O3 catalysts with identical bulk structures have different OCM performance, whereas the nanorod La2O3 catalyst shows better low-temperature performance than the M-La2O3 catalyst.Distinctions of the bulk phases between these two catalysts were only found after carbonate formed under identical CO2 treatment conditions [11,35].In the subsequent experiments, the three M-La2O3_nW catalysts and M-La2O3 catalysts were exposed to identical CO2 flow.

Adsorption of CO2 and OCM Reaction Performance
As mentioned in the introduction, previous in situ XRD-MS studies [35] indicated that, under the same exposure condition of CO2 concentration and heating temperature, the same in situ XRD-MS was applied for simultaneously monitoring of CO2 adsorption  As mentioned in previous studies, it was observed that the La 2 O 3 catalysts with identical bulk structures have different OCM performance, whereas the nanorod La 2 O 3 catalyst shows better low-temperature performance than the M-La 2 O 3 catalyst.Distinctions of the bulk phases between these two catalysts were only found after carbonate formed under identical CO 2 treatment conditions [11,35].In the subsequent experiments, the three M-La 2 O 3 _nW catalysts and M-La 2 O 3 catalysts were exposed to identical CO 2 flow.

Adsorption of CO 2 and OCM Reaction Performance
As mentioned in the introduction, previous in situ XRD-MS studies [35] indicated that, under the same exposure condition of CO 2 concentration and heating temperature, the same in situ XRD-MS was applied for simultaneously monitoring of CO 2 adsorption processes over the four catalyst samples during linear heating.To compare with this previous result, constant CO 2 exposure treatments are performed on the M-La 2 O 3 _5W sample with 10% CO 2 in the first in situ measurement.The XRD signal intensity in this region is plotted in a topographical map style (Figure 2a), and the XRD patterns clearly show the peaks at around 29.1 • and 30.1 • in this range, representing hexagonal La 2 O 3 (002) and ( 101) are not reduced.The results showed that the doped M-La 2 O 3 _5W is always robust to CO 2 exposure, remaining oxidic under the same 10% CO 2 exposure when the temperature reaches 600 • C. Furthermore, it still shows no bulk phase change after additional heating to temperature region up to 740 • C (Figure 2a).The simultaneously obtained online MS signal also shows little change in CO 2 output, confirming that there is no CO 2 adsorption rate significant enough to change its partial pressure in the in situ XRD-MS cell during the whole process (Figure 2b).In previous studies [35], the optimized nanorods with better low-temperature activity also showed a stable La 2 O 3 phase while exposed to 10% CO 2 both throughout the 600 • C isotherm and after cooling down to room temperature.In this study, the M-La 2 O 3 sample only after a simple Na 2 WO 4 doping modification shows a similar behavior of resistance to carbonation.
processes over the four catalyst samples during linear heating.To compare with this previous result, constant CO2 exposure treatments are performed on the M-La2O3_5W sample with 10% CO2 in the first in situ measurement.The XRD signal intensity in this region is plotted in a topographical map style (Figure 2a), and the XRD patterns clearly show the peaks at around 29.1° and 30.1° in this range, representing hexagonal La2O3 (002) and ( 101) are not reduced.The results showed that the doped M-La2O3_5W is always robust to CO2 exposure, remaining oxidic under the same 10% CO2 exposure when the temperature reaches 600 °C.Furthermore, it still shows no bulk phase change after additional heating to temperature region up to 740 °C (Figure 2a).The simultaneously obtained online MS signal also shows little change in CO2 output, confirming that there is no CO2 adsorption rate significant enough to change its partial pressure in the in situ XRD-MS cell during the whole process (Figure 2b).In previous studies [35], the optimized nanorods with better low-temperature activity also showed a stable La2O3 phase while exposed to 10% CO2 both throughout the 600 °C isotherm and after cooling down to room temperature.In this study, the M-La2O3 sample only after a simple Na2WO4 doping modification shows a similar behavior of resistance to carbonation.Complete carbonation over the M-La2O3_nW samples with significant CO2 uptake (MS) and carbonate phase change can only be more easily detected under nearly 100% CO2 exposure.In this way, their CO2 adsorption behavior can be directly compared to the undoped M-La2O3 sample.The calibrated MS signal, presented as CO2 consumption percentage normalized to the input during the adsorption process, was plotted vs. real-time linear temperature profile, as illustrated in Figure 3a.The simultaneously obtained oxide to carbonate full phase change, represented by normalized intensities of La2O3 (011) at 30.1° and the La2O2CO3 (103) peak at 29.9°, is plotted in Figure 3b.In this series of measurements, as the temperature reaches 500 °C, the M-La2O3 sample initiates CO2 adsorption at approximately 510 °C, reaching its peak value at 527 °C.M-La2O3_1W sample initiates CO2 adsorption at 556 °C, reaching its peak at 574 °C.In comparison to M-La2O3, the temperature for CO2 adsorption is approximately 47 °C higher.M-La2O3_3W sample starts CO2 adsorption around 570 °C, with the maximum value achieved at 585 °C, which is about 14 °C higher than the 1% wt Na2WO4-doped La2O3 sample.M-La2O3_5W sample initiates CO2 adsorption at approximately 570 °C, reaching its maximum at 589 °C.For all four samples, after reaching 700 °C, CO2 adsorption saturated, reaching around zero rate.By integrating each online MS peak, the yielded total CO2 uptake molar amounts are 208, 265, 283, and 307 µmol for the M-La2O3, M-La2O3_1W, M-La2O3_3W, and M-La2O3_5W samples, respectively.As the sample loading is around 300 µmol (100 mg, with molar mass around 326 g/mol), this CO2 uptake represents almost full carbonation (CO2 uptake samples, respectively.As the sample loading is around 300 µmol (100 mg, with molar mass around 326 g/mol), this CO 2 uptake represents almost full carbonation (CO 2 uptake to La 2 O 3 loading ratios are 68%, 86%, 92%, and 100%, respectively), which can be represented by the following equation: sorbed CO2 quantities in sequential of the four samples are as follows: M-La2O3 > 1% > 3% > 5%, revealing a singular positive dependent correlation with the increment in Na2WO4 doping (Figure S1).As a solid base, it may be expected that Na2WO4 add-on species will decrease the uptake temperature.But instead, it actually increases the temperature.It could be because Na2WO4 is only doped at an overall low level, not as the main component.The preceding results have indicated that, in comparison to M-La2O3, the three doped La2O3 M-La2O3_nW catalysts display more resistance to CO2 adsorption.Based on this observation, OCM reactivity evaluation was conducted over these four samples in the microreactor.The exhaust gas from the reaction was collected for online MS analysis of C2 and COx products, and the reaction properties, including conversion, product yields, and selectivity, were extracted.All results are presented in Figure 4. Figure 4a reveals that the OCM reaction light-off temperature of M-La2O3 is higher than the other three samples and at low temperatures below 500 °C, conversion from this system is also the lowest.The CH4 conversion rate, C2 yield, C2 selectivity, and COx yield results at 600 °C and 650 °C are illustrated in the column charts of C2 yield and C2 activation temperature for the four samples, as depicted in Figure 4b,c.For all four samples, the C2 selectivity always shows a singular positive dependence on the Na2WO4 loading level.The CH4 conversion rates of The result shows that adding Na 2 WO 4 slightly raises the CO 2 uptake amount, approaching a 1:1 molar ratio.In addition to the CO 2 uptake amount increase, there is a noticeable increasing trend in the CO 2 uptake peaking temperature.In the case of the three doped M-La 2 O 3 _nW samples, the CO 2 adsorption temperature is 47 • C-62 • C higher than that of M-La 2 O 3 .Furthermore, at equal temperatures below 650 • C, the estimated adsorbed CO 2 quantities in sequential of the four samples are as follows: M-La 2 O 3 > 1% > 3% > 5%, revealing a singular positive dependent correlation with the increment in Na 2 WO 4 doping (Figure S1).As a solid base, it may be expected that Na 2 WO 4 add-on species will decrease the uptake temperature.But instead, it actually increases the temperature.It could be because Na 2 WO 4 is only doped at an overall low level, not as the main component.
The preceding results have indicated that, in comparison to M-La 2 O 3 , the three doped La 2 O 3 M-La 2 O 3 _nW catalysts display more resistance to CO 2 adsorption.Based on this observation, OCM reactivity evaluation was conducted over these four samples in the microreactor.The exhaust gas from the reaction was collected for online MS analysis of C 2 and CO x products, and the reaction properties, including conversion, product yields, and selectivity, were extracted.All results are presented in Figure 4. Figure 4a reveals that the OCM reaction light-off temperature of M-La 2 O 3 is higher than the other three samples and at low temperatures below 500 • C, conversion from this system is also the lowest.The CH 4 conversion rate, C 2 yield, C 2 selectivity, and CO x yield results at 600 • C and 650 • C are illustrated in the column charts of C 2 yield and C 2 activation temperature for the four samples, as depicted in Figure 4b,c.For all four samples, the C 2 selectivity always shows a singular positive dependence on the Na 2 WO 4 loading level.The CH 4 conversion rates of M-La 2 O 3 , after its later light-off, increased faster than all the other systems to 17.8% at 650 • C, while the M-La 2 O 3 _1W and M-La 2 O 3 _3W are notably lower at 14.0% and 13.0%, respectively.The overall C 2 product yields of M-La 2 O 3 _1W and M-La 2 O 3 _3W are also lower than that (7.8%) of M-La 2 O 3 , measuring 6.3% and 5.8%, respectively.However, the C 2 selectivity of the undoped M-La 2 O 3 remained below 50% and its excessive conversion mostly turned into unwanted CO x by-products.For the two most important OCM evaluation properties, C 2 yield and selectivity, the M-La 2 O 3 _5W surpasses the undoped M-La 2 O 3 system from light-off to 650 • C, reaching 9% and 58.2%, respectively.The total CH 4 conversion rate of M-La 2 O 3 _5W also rebounds to 17.3%.Consequently, tuning the Na 2 WO 4 level doped on La 2 O 3 shows a positive effect on the catalyst OCM performance, especially in the low-temperature region.
°C, while the M-La2O3_1W and M-La2O3_3W are notably lower at 14.0% and 13.0%, respectively.The overall C2 product yields of M-La2O3_1W and M-La2O3_3W are also lower than that (7.8%) of M-La2O3, measuring 6.3% and 5.8%, respectively.However, the C2 selectivity of the undoped M-La2O3 remained below 50% and its excessive conversion mostly turned into unwanted COx by-products.For the two most important OCM evaluation properties, C2 yield and selectivity, the M-La2O3_5W surpasses the undoped M-La2O3 system from light-off to 650 °C, reaching 9% and 58.2%, respectively.The total CH4 conversion rate of M-La2O3_5W also rebounds to 17.3%.Consequently, tuning the Na2WO4 level doped on La2O3 shows a positive effect on the catalyst OCM performance, especially in the low-temperature region.The C2 yield did directly increase as the loading of Na2WO4 on the catalyst increased.However, there is an obvious singular dependence between the light-off temperatures and the Na2WO4 doping level on M-La2O3, which is plotted in Figure 5.For C2 products, lightoff temperature of M-La2O3 is 481 °C, while those of the M-La2O3_1W, M-La2O3_3W, and  C lower.These two significant decreasing trends vs. the Na 2 WO 4 doping level clearly contrast the reversed trend of CO 2 adsorption uptake temperature.As mentioned in the introduction, CO 2 adsorption results in carbonate formation, which poisons the active oxygen sites.In OCM, CO 2 is the main by-product, which will cause sample carbonation as an autocatalysis process, especially with the high temperature of the reaction.The Na 2 WO 4 doping will increase the catalyst resistance to carbonation under CO 2 exposure, as already proved in Figure 3.For those doped samples, fewer catalyst surface sites will be poisoned by the CO 2 products, and a significant reaction rate becomes sustainable at lower temperatures.As a result, doping La 2 O 3 with Na 2 WO 4 is a worthy approach to reduce the activation temperature of C 2 products and achieve better low-temperature (<650 • C) OCM reaction activity as compared to traditional M-La 2 O 3 .
significant decreasing trends vs. the Na2WO4 doping level clearly contrast the reversed trend of CO2 adsorption uptake temperature.As mentioned in the introduction, CO2 adsorption results in carbonate formation, which poisons the active oxygen sites.In OCM, CO2 is the main by-product, which will cause sample carbonation as an autocatalysis process, especially with the high temperature of the reaction.The Na2WO4 doping will increase the catalyst resistance to carbonation under CO2 exposure, as already proved in Figure 3.For those doped samples, fewer catalyst surface sites will be poisoned by the CO2 products, and a significant reaction rate becomes sustainable at lower temperatures.As a result, doping La2O3 with Na2WO4 is a worthy approach to reduce the activation temperature of C2 products and achieve better low-temperature (<650 °C) OCM reaction activity as compared to traditional M-La2O3.

Differences in Carbonate Bulk Phase between Doped and Undoped Samples
During the CO2 uptake measurements by online MS, simultaneous in situ time-resolved XRD experiments were performed to validate phase changes under CO2 exposure during linear heating on the three doped M-La2O3_nW samples and undoped M-La2O3.The wide range full-scan in situ XRD patterns were collected immediately after the specific treatment, as illustrated in Figure 5.The diffraction peak angle at high temperature (Figure 5a) exhibits a 2θ decrease of 0.18° from the room temperature (Figure 1) due to lattice thermal expansion.After the sample cooled down, direct exposure to CO2 (99% balanced with 1% Ar) at room temperature did not change (Figure 5b) the bulk structure so there was no bulk CO2 uptake.Figure 5c demonstrates the M-La2O3_nW samples under CO2 exposure at 800 °C and the bulk phase completely converted to La2O2CO3.Interestingly, the dioxymonocarbonates formed in M-La2O3_nW samples evidently exhibit a tetragonal La2O2CO3 crystal phase, as can be directly compared with the PDF database plotted in the same figure.After cooling to room temperature, all three tetragonal La2O2CO3 phases transformed into monoclinic La2O2CO3 (Figure 5d), characterized by the single diffraction

Differences in Carbonate Bulk Phase between Doped and Undoped Samples
During the CO 2 uptake measurements by online MS, simultaneous in situ timeresolved XRD experiments were performed to validate phase changes under CO 2 exposure during linear heating on the three doped M-La 2 O 3 _nW samples and undoped M-La 2 O 3 .The wide range full-scan in situ XRD patterns were collected immediately after the specific treatment, as illustrated in Figure 5.The diffraction peak angle at high temperature (Figure 5a) exhibits a 2θ decrease of 0.18 • from the room temperature (Figure 1) due to lattice thermal expansion.After the sample cooled down, direct exposure to CO 2 (99% balanced with 1% Ar) at room temperature did not change (Figure 5b) the bulk structure so there was no bulk CO 2 uptake.Figure 5c demonstrates the M-La 2 O 3 _nW samples under CO 2 exposure at 800 • C and the bulk phase completely converted to La 2 O 2 CO 3 .Interestingly, the dioxymonocarbonates formed in M-La 2 O 3 _nW samples evidently exhibit a tetragonal La 2 O 2 CO 3 crystal phase, as can be directly compared with the PDF database plotted in the same figure.After cooling to room temperature, all three tetragonal La 2 O 2 CO 3 phases transformed into monoclinic La 2 O 2 CO 3 (Figure 5d), characterized by the single diffraction peak at 31.5 • splitting into two peaks at 2θ = 30.7 • and 31.3 • .On the other hand, the undoped M-La 2 O 3 after complete carbonate formation only shows a hexagonal phase, either at high or room temperature, which was identical to our previous published results [11,35].The results suggest that this phase change to tetragonal structure dioxymonocarbonate is strongly correlated to the high resistance to CO 2 adsorption.
The time-resolved XRD rapid scan collected during the heating of the M-La 2 O 3 _nW samples in the CO 2 flow is plotted in Figure 6a-c.These data are collected simultaneously with the online MS data collected in Figure 3.The 28.5~33.5 • window perfectly captured the phase change from hexagonal La 2 O 3 to tetragonal La 2 O 2 CO 3 as it covers the characteristic peaks of La 2 O 3 (011) at 30.1 • and the La 2 O 2 CO 3 (103) and (110) peaks at around 29.9 and 31.5 • , respectively.As explained in Section 2.2, the phase change temperature can only be more roughly estimated as the scan time resolution sets a low limit of the temperature range of each scan at 38 • C.However, it still indicates that the higher doped M-La 2 O 3 _3W and M-La 2 O 3 _5W samples have a phase change temperature (570-606 • C) higher than that of the lower doped M-La 2 O 3 _1W sample (530-570 • C).It can also be noted from this figure that the phase change temperature difference is within 38 • C.This phase change temperature result agrees with the MS analysis results on the CO 2 uptake temperature of the four samples.
Previous research also indicates that after complete carbonation, the M-La 2 O 3 sample only shows a uniform hexagonal phase.The tetragonal/monoclinic phase carbonates formation could be found on M-La 2 O 3 , but only after a partial phase change, which can be controlled by heating in low CO 2 concentration (<30%) to 600 • C for a limited time [35].On the other hand, the nanorod La 2 O 3 catalyst, with a higher low-temperature OCM reactivity compared to the M-La 2 O 3 catalyst, also formed tetragonal La 2 O 2 CO 3 at high temperatures after carbonation and subsequently transformed into monoclinic La 2 O 2 CO 3 crystal phase upon cooling to room temperature.In a word, after the M-La 2 O 3 was doped with Na 2 WO 4 , its carbonate formation behavior became similar to the nanorod La 2 O 3 .The previous studies, based on the comparison between the nanorod La 2 O 3 and M-La 2 O 3 catalysts behavior, already suggest that the special tetragonal carbonate phase change correlates with the higher CO 2 adsorption resistance.In this study, as the La 2 O 3 materials are all the same for the four catalysts, the analysis data certainly strengthen this conclusion as it can be correlated as follows: Tetragonal carbonate ↓ CO 2 adsorption/poisoning resistance ↓ Low-temperature OCM activity As a result, we strongly recommend that being able to form tetragonal carbonate is an experimental descriptor for a La 2 O 3 catalyst with better low-temperature activity in OCM.

Catalyst Preparation
Through impregnation in Na 2 WO 4 solution, La 2 O 3 can be modified to obtain samples with diverse weight loading percentages.Both Na 2 WO 4 (S859551, ≥98%) and commercial La 2 O 3 (L812319, ≥99.99%, denoted as M-La 2 O 3 ) were from Shanghai Macklin Biochemical Co., Ltd.(Shanghai, China).Previous BET measurement (Kr) of M-La 2 O 3 yields a relatively low specific surface area of 3.4 m 2 g −1 (56 µmol g −1 assuming lattice density of 10 15 cm −2 ) [11].Initially, 5, 10, and 15 mg of Na 2 WO 4 were precisely weighed and introduced into separate 20 mL portions of deionized water.Subsequently, 495, 490, and 485 mg of La 2 O 3 were gradually added to the solution under continuous magnetic stirring.The solution was fully dried at 70 • C for 24 h and then calcined at 800 • C in a muffle furnace for 4 h with a heating rate of 5 • C min −1 .The contents of Na 2 WO 4 were calculated to be 1%, 3%, and 5% wt, respectively (denoted as M-La 2 O 3 _nW, where n refers to the weight percentage of Na 2 WO 4 -loaded).The precipitation does not significantly change the BET surface area results and dispersion.All freshly prepared samples were characterized immediately.Prior to commencing the CO 2 absorption and OCM experiment, the sample was pretreated at 800 • C in an Ar atmosphere for 30 min in the in situ XRD or online MS microreactors, which aims to prevent sample hydroxylation or carbonation before the introduction of experimental gases.

In Situ XRD-MS
The XRD employs the Bruker D8 Advance powder X-ray diffractometer equipped with a Cu K α target with a wavelength of λ = 0.154 nm (8.05 KeV).The in situ heating reaction cell (Anton Paar XRK 900) is capable of conducting testing at 900 • C and 10 bar.The specific information on the relevant equipment is mentioned in the literature, including the design of the reaction gas pathway and coupled online mass spectrometry [11].Before further characterization, the loaded 0.1 g sample (~300 µmol) was heated in situ to 800 • C at a rate of 10 • C min −1 under a 20 sccm Ar airflow, followed by a 30 min dwelling at 800 • C. Previous studies [25,47] show that this treatment effectively depletes the bulk and surface impurities including hydroxyl and carbonates, leaving only purified La 2 O 3 .
For online CO 2 adsorption uptake studies, once cooling back to room temperature, the sample was exposed to CO 2 -contained gas at 20 sccm.After a 20 min gas flushing, the XRD pattern was collected at room temperature with a step size of 0.02 • and a sensor exposure time of 1.5 s, spanning from 15 to 60 • .Subsequently, the temperature was raised from room temperature to 800 • C at a rate of 10 • C min −1 , and the XRD pattern at 800 • C was collected using the same parameters.During the heating process, the continuous fast scanning XRD patterns were collected in the range of 28.5~33.5 • , with a step size of 0.05 • and a sensor exposure time of 1.0 s each for M-La 2 O 3 _nW samples, totally 100 s for each scan.The hardware still requires extra time after each scan to re-calibrate the initial alignment of the arms; so for the linear heating at the rate of 10 • C min −1 , each scan covers a temperature variation of about 38 • C. The custom-designed coupled online mass spectrometry system (SRS RGA 200) collects real-time data of the exhaust gas released from the XRD in situ reaction cell during the heating process using capillary tubes [48].In situ XRD-MS measurements were performed with two CO 2 input conditions, 10% and nearly 100% (99% CO 2 balanced with 1% Ar).The 10% condition is close to the practical CO 2 concentration under real OCM reaction conditions above 400 • C. The nearly 100% condition makes it easier to capture full phase change from La 2 O 3 to La 2 O 2 CO 3 as shown in the data results section.Due to the inability of mass spectrometry to detect the consumption of pure gas uptake signal, such as 100% CO 2 , 1% Ar gas was mixed with 99% CO 2 for labeling.In this way, accompanied by the CO 2 partial consumption in an adsorption process, mass spectrometry will detect the partial pressure increase in the Ar signal from the sampled mixture.The relative CO 2 uptake ratio thus can be re-calibrated from the Ar signal change.
The grain size on a specific face in the XRD result was calculated by Debye-Scherrer Formula: where K in Equation ( 2) is 0.89 for the (100) face, which is calculated using the FWHM of the diffraction peak.

Online MS Microreactor
OCM reaction kinetics data for all samples were obtained using a specialized online mass spectrometry (MS) microreactor designed to operate under high-temperature and high-pressure conditions.The sample was sealed in a ¼-inch quartz tube using sealing rings and sleeves to ensure airtightness.The gas manifold offers diverse reaction input options, enabling real-time online sampling and gas characterization via a mass spectrometer (Pfeiffer PrismaPlus).More detailed information about the experimental setup is described in our previous publications [49].
For online OCM reactivity evaluation, 0.1 g samples were loaded in the center of quartz tubes plugged with quartz wool on both ends.Before the OCM reaction, the sample undergoes calcination in Ar up to 800 • C to eliminate carbonates and hydroxides.After a 15 min OCM reaction gas (CH 4 :O 2 :Ar = 5:1:4, GHSV = 8000 mL•h −1 gcat −1 ) purging at room temperature, the reactor is linearly heated to 740 • C at a rate of 10 the selectivity of the product is determined using the following equation: the yield of the product is determined using the following equation: where n in Equations ( 4) and ( 5) is the carbon number of the product molecules.
The TOF of the product is estimated using the following equation: where a is the surface La atom density on the (001) surface of La 2 O 3 , which is calculated in average one La atom per 0.134 nm 2 (about 7.5 × 10 14 cm −2 ), applying the La 2 O 3 (001) surface model; S is the specific surface area acquired from BET measurement (3.4 m 2 /g).The product of a•S yields the catalyst surface La site density of 42 µmol/g cat ; m is the loading quality of the sample; yield is calculated by Equation (5).

XPS Measurement
The XPS surface analysis is performed utilizing a ThermoFischer ESCALAB 250Xi photoelectron spectrometer, applying monochromatic X-ray irradiation AlKα (hv = 1486.7 eV) and using a 180 • double-focusing hemispherical analyzer with a six-channel detector.For each sample, C 1s, O 1s, La 3d, La 4d, W 4d, W 4f, and Na 1s core level spectra are collected with 30 eV pass energy and 0.1 eV step.The survey XPS scans are monitored as well to demonstrate no impurities on the sample surfaces.The binding energy scale of the collected spectra is calibrated to the adventitious carbon C 1 s peak at 284.8 eV.The atomic percentages for W and La are calculated from the corresponding photoelectron peaks after background subtraction taking into account transmission function and atomic sensitivity factors.

Conclusions
The carbonation behavior of La 2 O 3 catalysts, as an inherent phenomenon during the OCM reaction process, is studied and correlated with OCM activity.This study carried out a comprehensive investigation of the bulk structure and kinetic analysis of CO 2 adsorption on M-La 2 O 3 _nW catalysts.The CO 2 kinetic adsorption curve indicated that an elevated Na 2 WO 4 doping level raised the temperature at which CO 2 absorption occurred, at least 46 • C higher than that of commercial La 2 O 3 .These results suggest that Na 2 WO 4 -doped La 2 O 3 is more difficult to form La 2 O 2 CO 3 .Furthermore, online MS results indicated that at low temperatures (650 • C), the M-La 2 O 3 _nW catalyst exhibited higher C 2 yield, and C 2 selectivity, and produced less CO 2 compared to commercial La 2 O 3 .In situ XRD-MS results also revealed that with Na 2 WO 4 doped on the M-La 2 O 3 _nW catalyst, CO 2 induced transform into tetragonal La 2 O 2 CO 3 after high-temperature CO 2 treatment.This behavior diverges from the hexagonal carbonation pattern observed in undoped M-La 2 O 3 at elevated temperatures.The results correlate the tetragonal La 2 O 2 CO 3 formation upon CO 2 exposure to higher resistance to CO 2 poisoning and better low-temperature OCM activity.Coupled the XRD measurement with simultaneous kinetic analysis of CO 2 adsorption, this

2 . 1 .
Differences in La 2 O 3 and La 2 O 2 CO 3 Phase Structure After in situ calcination at 800 • C, the results comparison of bulk phase analysis obtained by XRD from M-La 2 O 3 samples with different loading levels of Na 2 WO 4 are shown in Figure 1.Little different from the undoped M-La 2 O 3 sample, the XRD patterns of all three M-La 2 O 3 _nW catalysts exhibit similar crystal structures after high-temperature purification.The diffraction patterns all align well with the hexagonal phase of La 2 O 3 .

Figure 1 .
Figure 1.(a) In situ XRD patterns of M-La2O3 and M-La2O3_nW samples at RT compared with PDF database.The diffraction peak marked with a diamond is from the sample stage.(b) The W 4f (left) and La 3d5/2 (right) core level XPS spectra of M-La2O3_1W (black), 3W (green), and 5W (blue) samples calcined at 800 °C.

Figure 1 .
Figure 1.(a) In situ XRD patterns of M-La 2 O 3 and M-La 2 O 3 _nW samples at RT compared with PDF database.The diffraction peak marked with a diamond is from the sample stage.(b) The W 4f (left) and La 3d 5/2 (right) core level XPS spectra of M-La 2 O 3 _1W (black), 3W (green), and 5W (blue) samples calcined at 800 • C.

Figure 2 .
Figure 2. (a) In situ XRD and (b) online MS result of CO 2 uptake on M-La 2 O 3 _5W with 10% CO 2 -Ar (20 sccm); sample loading is 0.10 g.Complete carbonation over the M-La 2 O 3 _nW samples with significant CO 2 uptake (MS) and carbonate phase change can only be more easily detected under nearly 100% CO 2 exposure.In this way, their CO 2 adsorption behavior can be directly compared to the undoped M-La 2 O 3 sample.The calibrated MS signal, presented as CO 2 consumption percentage normalized to the input during the adsorption process, was plotted vs. realtime linear temperature profile, as illustrated in Figure 3a.The simultaneously obtained oxide to carbonate full phase change, represented by normalized intensities of La 2 O 3 (011) at 30.1 • and the La 2 O 2 CO 3 (103) peak at 29.9 • , is plotted in Figure 3b.In this series of measurements, as the temperature reaches 500 • C, the M-La 2 O 3 sample initiates CO 2 adsorption at approximately 510 • C, reaching its peak value at 527 • C. M-La 2 O 3 _1W sample initiates CO 2 adsorption at 556 • C, reaching its peak at 574 • C. In comparison to M-La 2 O 3 , the temperature for CO 2 adsorption is approximately 47 • C higher.M-La 2 O 3 _3W sample starts CO 2 adsorption around 570 • C, with the maximum value achieved at 585 • C, which is about 14 • C higher than the 1% wt Na 2 WO 4 -doped La 2 O 3 sample.M-La 2 O 3 _5W sample initiates CO 2 adsorption at approximately 570 • C, reaching its maximum at 589 • C. For all four samples, after reaching 700 • C, CO 2 adsorption saturated, reaching around zero rate.By integrating each online MS peak, the yielded total CO 2 uptake molar amounts are 208, 265, 283, and 307 µmol for the M-La 2 O 3 , M-La 2 O 3 _1W, M-La 2 O 3 _3W, and M-La 2 O 3 _5Wsamples, respectively.As the sample loading is around 300 µmol (100 mg, with molar mass around 326 g/mol), this CO 2 uptake represents almost full carbonation (CO 2 uptake to La 2 O 3 loading ratios are 68%, 86%, 92%, and 100%, respectively), which can be represented by the following equation:

Figure 3 .
Figure 3. (a) The calibrated MS signal presented as CO2 adsorption uptake rate results obtained from online MS and (b) the simultaneously obtained in situ XRD results of full phase change from normalized La2O3 (011) to La2O2CO3 (103) intensity over the four samples.The black and red lines represent the La2O3 and La2O2CO3 phase transitions.

Figure 3 .
Figure 3. (a) The calibrated MS signal presented as CO 2 adsorption uptake rate results obtained from online MS and (b) the simultaneously obtained in situ XRD results of full phase change from normalized La 2 O 3 (011) to La 2 O 2 CO 3 (103) intensity over the four samples.The black and red lines represent the La 2 O 3 and La 2 O 2 CO 3 phase transitions.

Figure 4 .
Figure 4. (a) The OCM reactivity evaluation results of four M-La2O3_nW samples and the analysis of CH4 conversion, C2 yield, COx yield (right scale) and C2, COx selectivity in OCM at (b) 600 °C and (c) 650 °C.The discrete points in (a) mean the data directly from online MS results; the smoothed lines are the fitting lines obtained through the locally weighted regression.

Figure 4 .
Figure 4. (a) The OCM reactivity evaluation results of four M-La 2 O 3 _nW samples and the analysis of CH 4 conversion, C 2 yield, CO x yield (right scale) and C 2 , CO x selectivity in OCM at (b) 600 • C and (c) 650 • C. The discrete points in (a) mean the data directly from online MS results; the smoothed lines are the fitting lines obtained through the locally weighted regression.The C 2 yield did directly increase as the loading of Na 2 WO 4 on the catalyst increased.However, there is an obvious singular dependence between the light-off temperatures and the Na 2 WO 4 doping level on M-La 2 O 3 , which is plotted in Figure 5.For C 2 products, light-off temperature of M-La 2 O 3 is 481 • C, while those of the M-La 2 O 3 _1W, M-La 2 O 3 _3W, and M-La 2 O 3 _5W samples are 475 • C, 460 • C, and 415 • C, respectively.The light-off temperatures of CO x also have the same dependence but are all around 70• C lower.These two significant decreasing trends vs. the Na 2 WO 4 doping level clearly contrast the reversed trend of CO 2 adsorption uptake temperature.As mentioned in the introduction, CO 2 adsorption results in carbonate formation, which poisons the active oxygen sites.In OCM, CO 2 is the main by-product, which will cause sample carbonation as an autocatalysis process, especially with the high temperature of the reaction.The Na 2 WO 4 doping will increase the catalyst resistance to carbonation under CO 2 exposure, as already proved in

Figure 5 .
Figure 5.In situ XRD patterns of four stages in CO2 treatment: (a) holding at 800 °C in Ar, (b) switching to CO2 at RT after cooling in Ar, (c) holding at 800 °C in CO2, and (d) cooling down to RT in CO2.The black, orange, purple and green curves represent M-La2O3, M-La2O3_1W, 3W, and 5W catalysts, respectively.The single diffraction peak at 34.8° represents impurities in the sample stage.

Figure 5 .
Figure 5.In situ XRD patterns of four stages in CO 2 treatment: (a) holding at 800 • C in Ar, (b) switching to CO 2 at RT after cooling in Ar, (c) holding at 800 • C in CO 2 , and (d) cooling down to RT in CO 2 .The black, orange, purple and green curves represent M-La 2 O 3 , M-La 2 O 3 _1W, 3W, and 5W catalysts, respectively.The single diffraction peak at 34.8 • represents impurities in the sample stage.

Figure 6 .
Figure 6.In situ XRD scanning results of M-La 2 O 3 : (a) 1W, (b) 3W, and (c) 5W, respectively, in 20 sccm CO 2 from room temperature to 800 • C. Diffraction peak intensity change is contrasted by color gradient change (red to purple).
. Pengwei Wang et al. synthesized a TiO 2 -doped Mn 2 O 3 -Na 2 WO 4 /SiO 2 catalyst and revealed that it has accepted CH 4 conversion of 22% and C 2-3 selectivity of ~62% at a lower temperature (650 • C) and established a reaction mechanism model supported by the active oxygen transforming on W 4+ -W 6+ [42].Recently, Shihui Zou et al. innovatively developed a novel two-part catalyst comprising Na 2 WO 4 /SiO 2 and La 2 O 3 ; this catalyst involves modifying the La 2 O 3 catalyst with additional Na 2 WO 4 , achieving a C 2 yield of 10.9% at 570 • C [43].Studies indicate the potential of incorporating Na 2 WO 4 as an active additive in modified catalysts.Motivated by the Na 2 WO 4 -modified catalyst, Na 2 WO 4 is introduced into the catalyst as an approach of optimization, especially on the low-temperature performance, in this study.Based on the above research foundation of Na 2 WO 4 and La 2 O 3 , we used in situ XRD-MS characterization to investigate the effect of doping Na 2 WO 4 into La 2 O 3 samples, focusing on CO 2

Table 1 .
W and La atomic percentages (other elements are not taken into consideration) for M-La2O3_1W, 3W, and 5W samples.

Table 1 .
W and La atomic percentages (other elements are not taken into consideration) for M-La 2 O 3 _1W, 3W, and 5W samples.
• C min −1 .The CH 4 conversion rate, C 2 selectivity, and yields of C 2 and CO x are calculated by the following equation, derived from carbon balance: