*2.2. Catalyst Preparation*

In the case of zirconium-based support (for 5Ni-10La-ZrO2), the required amounts of zirconia (2.62 g) in the form of ZrOCl2·8H2O was ground completely and poured into the empty crucible. Then, the desired weights of La(NO3)3·6H2O (0.265 g) and that of Ni(NO3)2·6H2O (0.25 g) were added to the crucible containing the support to ge<sup>t</sup> a powder mixture. The mixture was ground well in the crucible to obtain a homogenous mixture. Purified water was poured slowly to the mixture to produce a paste while mixing. The paste was set to evaporate under room temperature condition until it dried. Thereafter, the dried sample was calcined at 700 ◦C for 3 h. The obtained catalysts were denoted as 5Ni-ZrO2 for 5 wt% Ni supported over ZrO2, and 5Ni-xLa-ZrO2, where x = 10, 15, 20 wt%.

In the case of alumina-based support, the above-mentioned procedure was adopted by replacing zirconia with the required amounts of mesoporous γ-Al2O3. The obtained catalysts were denoted as 5Ni-Al2O3 for 5 wt% Ni supported over Al2O3 and 5Ni-xLa-Al2O3, where x = 10, 15, 20 wt%.

#### *2.3. Catalyst Characterization and Activity*

The catalyst activity test and characterization are described in detail in the supplementary information.

#### **3. Results and Discussion**

The surface texture of the catalysts was assessed via the nitrogen adsorption–desorption isotherms. Figure 1A shows the nitrogen adsorption isotherms of the fresh catalysts (5NixLa2O3-ZrO2, x = 0, 10, 15, and 20 wt%) and according to IUPAC labelling, catalysts are showing type IV isotherm with capillary condensation appearing at a relative pressure below the saturation pressure, and H1 hysteresis loop. These features are associated with mesoporous materials that have cylindrical pore geometry with narrow size distribution as well as relatively high uniformity [39]. Figure 1B exhibits the nitrogen adsorption isotherms of the fresh catalysts (5Ni-x%La-Al2O3**)**. All samples indicate typical type IV adsorption/desorption isotherms with H1 hysteresis loop. Point of inflection at a relative pressure in the range of 0.6–0.75 corresponds to the capillary condensation which indicates the uniformity of the pores in mesoporous material [40,41].The textural properties of the catalyst are given in Table S1 of the supplementary.

**Figure 1.** N2 adsorption–desorption isotherms for fresh (**A**) 5Ni-xLa2O3+ZrO2 and (**B**) 5Ni-x La2O3+Al2O3 catalyst calcined at 700 ◦C, (x = 0, 10, 15, and 20 wt%).

The reducibility of 5%Ni-x% La2O3-ZrO2 and 5%Ni-x% La2O3-Al2O3 (x = 0, 10, 15, and 20) catalysts were examined by TPR and the patterns are shown in Figure 2. For the 5%Ni-x% La2O3-ZrO2 catalysts (Figure 2A), two prominent reduction maxima are detected over the entire temperature range, which may be attributed to the reduction of different NiO species (NiO→Ni0). The first reduction peak situated in region I at Tmax = 298 ◦C could be ascribed to the reduction of free NiO that is not attached to the support, and hence reduces easily at low temperature. Further, the peak in the region II at Tmax = 468 ◦C is allotted to the NiO reduction, attached to ZrO2 by a moderately strong link and its reduction requires higher thermal energy. After support modification by means of La2O3, substantial variations in reduction kinetics were detected. Reduction maxima illustrating NiO reduction for the higher temperature peaks has shifted towards lower temperatures [42]. In the case of Figure 2B, the 5% Ni-x% La2O3-Al2O3 catalysts (x = 0, 10, 15, and 20) display a single peak in region III at higher temperatures. The un-promoted 5%Ni-Al2O3 catalyst shows a broad peak at 750 ◦C indicating that the reduced NiO is strongly attached to the support. When the support is modified with the addition of different loadings of La2O3, peaks of relatively lower areas appear at high temperatures. The La2O3 modified support catalyst is shifted to higher temperatures [43]. The highest La2O3 loading catalyst gives the highest peak shift. This means that the addition of La2O3 increases further the interaction between the NiO and the modified support. The

quantitative analysis of H2 consumption during H2-TPR is displayed in Table S2 of the supplementary.

**Figure 2.** H2-TPR profiles of (**A**) for 5Ni-xLa2O3-ZrO2 and (**B**) 5Ni-xLa2O3-Al2O3 (x = 0, 10, 15, and 20 wt%) catalysts.

Figure 3 presents the powder X-ray diffraction patterns of the calcined 5Ni-xLa2O3- ZrO2 and 5Ni-xLa2O3-Al2O3 catalysts (x = 0, 10, 15, and 20 wt%). Figure 3A displays the XRD patterns for 5Ni-xLa2O3-ZrO2. There are no peaks attributable to La2O3 in these patterns since similar patterns are obtained for La2O3 modified and un-modified catalysts denoting the homogenous distribution of La2O3. The peaks at 28.3◦ and 31.6◦ are attributed to NiO phase (JCPDS No. 47–1049). The ZrO2 support in Figure 3A has two crystalline phases, tetragonal zirconia (t-ZrO2) and monoclinic zirconia (m-ZrO2). The peaks at 25.5◦ and 34.3◦ are ascribed to m-ZrO2 [44,45], while the peaks at 40.9◦, 50.2◦, and 55.5◦ are credited to t-ZrO2 [46]. In Figure 3B, there is no peak ascribable to La2O3 in the patterns and therefore La2O3 is well dispersed in the alumina matrix. The characteristics peaks at 37.3◦ (311), 45.6◦ (400), and 67.0◦ (440), all are correspondingly allocated to the Al2O3 structure (JCPDS 10–0425) [47,48].

**Figure 3.** XRD patterns of the (**A**) 5Ni-xLa2O3+ZrO2 (**B**) 5Ni-xLa2O3+Al2O3 catalysts (x = 0, 10, 15, and 20 wt%) calcined at 700 ◦C).

Figure 4 displays the hydrogen yield versus time on stream for the dry reforming reaction at 700 ◦C. The impact of La2O3 addition on the DRM catalytic performance of Ni-ZrO2 is discussed in this section. The initial hydrogen yield of the 5Ni-ZrO2 catalyst in Figure 4 is lower than the La2O3 modified catalysts. An evident trend is noted when La2O3 is added causing a significant improvement in the DRM performance. The improvement profile escalates as the following: 5Ni-ZrO2 < 5Ni-20La2O3-ZrO2 < 5Ni-15La2O3-ZrO2 < 5Ni-10La2O3-ZrO2. The highest hydrogen yield of about 80% is recorded using 10% La2O3. The improvement due to the La2O3 addition is attributed to the fact that La2O3 increases the dispersion of Ni particles on the supports and reduces the agglomeration of Ni particles during the reforming reaction as depicted in Figure 2A. Moreover, La2O3 increases the basicity as shown in the TPD profiles and therefore adsorb and react with CO2 to form La2O2CO3 species on the surface of catalyst which can speed up the conversion [49].

**Figure 4.** H2 yield vs. time on stream over the 5Ni-xLa2O3-ZrO2 (x = 0, 10, 15, and 20 wt%) catalysts at 700 ◦C for 420 min.

In Figure 5, the hydrogen yield obtained is close for La2O3 doped and non-doped catalysts. The La2O3 addition improved marginally the hydrogen yield in the following manner 5Ni-Al2O3 < 5Ni-20La2O3-Al2O3 < 5Ni-10La2O3-Al2O3 < 5Ni-15La2O3-Al2O3. The 15% La2O3 gave the highest hydrogen yield of 84%. It can be inferred that the effect of La2O3 loading affects differently the hydrogen yield productivity depending on the type of the support. Table 1 describes the efficiency of the present work and some of the literature.

**Figure 5.** H2 yield vs. time on stream over the 5Ni-xLa2O3-Al2O3 (x = 0, 10, 15, and 20 wt%) catalysts at 700 ◦C for 420 min.


**Table 1.** Hydrogen yield performances obtained in CO2 reforming of methane of present and past work.

Figure 6 exhibits the CO2-TPD profiles of the (A) 5Ni-xLa2O3-ZrO2 and (B) 5NixLa2O3-Al2O3 (x = 0, 10, 15, and 20 wt%) spent catalysts obtained at 700 ◦C reaction temperature. This was performed to scan the surface basicity of the catalysts, which plays a vital role in the catalytic DRM reaction [54]. It is commonly understood that a greater desorption temperature of CO2 reveals a stronger basicity and a bigger amount of CO2 desorption although signifying that more basic sites are presented on the surface of catalyst [55]. In Figure 6A, three chief CO2 desorption peaks were identified in the experimented temperature range from 50 to 700 ◦C for higher loadings of La2O3 (5Ni-15La2O3-ZrO2 and 5Ni-20La2O3-ZrO2) modified catalysts and only two peaks for the un-modified (5Ni-ZrO2) and lower loading La2O3 (5Ni-10La2O3-ZrO2) modified catalysts, which denoted that three types of basic sites existed in the 5Ni-xLa2O3 + ZrO2 (x = 0, 10, 15, and 20 wt%) catalysts. The CO2 desorption peaks appeared at 80 ◦C, 260 ◦C and 550 ◦C. The peaks correspond to the weak adsorption of CO2 on OH groups, moderate adsorption of CO2 and strong CO2 adsorption on the metal–oxygen pairs and O2− anions, respectively [56,57]. It is clear that CO2 was favorably absorbed on the strong basic sites as the support was modified with La2O3, rather than bare zirconia support. This result showed that the adsorption of CO2 had altered from physical adsorption to chemical adsorption because of the addition of La2O3. Figure 6B shows two main CO2 desorption peaks at 80 ◦C and 250 ◦C corresponding to weak and moderate basic sites. Hence, the addition of La2O3 to the support did give significant variation in basicity of the 5NixLa2O3 + Al2O3 catalysts. The 5Ni-15La2O3-Al2O3 displays larger peak intensity than the remaining catalysts, which is in accordance with the better activity observed. Table 2 shows a summary of a quantitative assessment of CO2 adsorption of the spent catalysts via CO2-TPD for ZrO2 and Al2O3 supported catalysts.

**Figure 6.** CO2-TPD profiles of the spent ( **A**) 5Ni-xLa2O3-ZrO2 and (**B**) 5Ni-xLa2O3-Al2O3 (x = 0, 10, 15, and 20 wt%) catalysts obtained at 700 ◦C reaction temperature.

a


**Table 2.** Quantitative assessment of CO2adsorption of the spent catalysts via CO2-TPD.

For comparison and basicity assessment of catalysts, the sum of all basic sites of 5Ni-ZrO2 is set to 1 and 5Ni-Al2O3 is set to 1.

> Figure 7A,B contain images obtained from Energy-dispersive X-ray spectroscopy (EDX) analysis of the fresh samples of both 5Ni-10La2O3-ZrO2 and 5Ni-15La2O3-Al2O3. These results show the elemental composition of the as-prepared catalyst samples. First, the analysis confirmed the presence of all the elemental constituents that were mixed together during the catalyst synthesis. Moreover, the percentage loadings revealed by the EDX analysis are virtually the same as intended in the calculation and catalyst preparation with error values of 10% for Ni and 6% for La.

**Figure 7.** EDX analysis of fresh (**A**) 5Ni-15La2O3-Al2O3 and (**B**) 5Ni-10La2O3- ZrO2 showing the elemental composition of prepared catalysts.

Broad examination of the morphology of catalysts was performed via TEM. Typical TEM overviews of the fresh catalysts (Figure 8A,C) and spent catalyst samples (Figure 8B,D) were acquired after DRM at 700 ◦C for 420 min time on stream. There is a difference in the morphology of the fresh and used catalysts. The fresh catalyst seems to have particles clumped on the surface, while the spent catalysts depict rough and dispersed particles on the catalyst surface. In addition to the recognized metal particles, carbon in the form of nanotubes can be seen in the images of the used catalysts. The image of the fresh 5Ni-10La2O3-ZrO2 catalyst is shown in Figure 8A. It is evident from the TEM images that the Ni is homogeneously dispersed over the surface of the support. In contrast to its spent sample as shown in Figure 8B, big quantities of coke are formed commonly in the form of nanotubes. The TEM images showed well dispersed Ni species ((average particle diameter of 4 nm) over La2O3-Al2O3 support (Figure 8C). However, after the reaction (Figure 8D) size of Ni species has grown (average particle diameter of 7–8 nm).

**Figure 8.** TEM image of (**A**) Fresh, (**B**) spent of 5Ni/10La2O3+ZrO2, (**C**) Fresh, (**D**) spent of 5Ni/15La2O3 + Al2O3 catalysts.

The Fourier Transform Infrared Spectroscopy (FTIR) analysis of fresh 5Ni-ZrO2, 5Ni-10La2O3-ZrO2, 5Ni-Al2O3 and 5Ni-15La2O3-Al2O3 catalysts were done to investigate the existing bonds within the catalyst system. The infrared spectra of absorption for these samples are shown in Figure 9A,B.

**Figure 9.** The FTIR spectra of fresh (**A**) Al2O3, 5Ni-Al2O3, 5Ni-15La2O3-Al2O3 and (**B**) ZrO2, 5Ni-ZrO2, 5Ni-10La2O3-ZrO2 showing the existing stretching vibration.

In Figure 9A, Al2O3 is well known for its tendency to adsorb moisture from the atmosphere onto itself. Thus, the distinct band representing the stretching vibration of O-H, within the wavelengths 3430–3460 cm<sup>−</sup><sup>1</sup> for all the samples can be ascribed to [OH]−<sup>1</sup> groups interaction and/or physisorbed moisture interaction that is adsorbed onto the Al2O3 support [58]. It is noticeable from the figure that the extent of hydration of the catalyst surface changed with metal addition as the OH band's intensity decreased on adding active metal. Moreover, the vibration bands centered at around 1400, 1520, 1649, and 2366 cm<sup>−</sup><sup>1</sup> can be seen for the support and the other samples. This implies that these bands can be associated with the support, the Al–O bond stretching to be specific [59]. The small less noticeable peaks appearing at wavelengths 1237 and 1719 cm<sup>−</sup><sup>1</sup> for the 5Ni-Al2O3 sample can be said to be the stretching vibration of NiO and NiAl2O4 species, respectively. The latter is thought to have stronger interaction with the support, in light of that it appeared at a higher wavelength.

As for the ZrO2 support and ZrO2 supported catalysts (Figure 9B); the band within 3357–3440 cm<sup>−</sup><sup>1</sup> can be said to be hydrogen-bonded bending and stretching of the OH groups as a result of adsorbed moisture while the peaks at 1632–1640 cm<sup>−</sup><sup>1</sup> can be assigned to the vibration of the water molecules [60]. These peaks are seen to decrease in intensity with the addition of nickel and lanthanum oxide to the support. The peaks that are centered at around 457 and 752 cm<sup>−</sup><sup>1</sup> represent the asymmetric stretching of the Zr–O–Zr bond [61]. The vibration owing to the presence of La2O3 was not discovered. This further supports the results obtained from the XRD analysis of samples with La2O3.

Figure 10A displays TGA profiles of spent 5Ni-xLa2O3 + ZrO2 (x = 0, 10, 15, and 20 wt%) catalysts operated at 700 ◦C. The weight loss above 500 ◦C was due to the removal of deposited carbon. The extents of carbon deposition on the spent catalysts exhibit the following sequence: 5Ni-10La2O3-ZrO2 < 5Ni-15La2O3-ZrO2 < 5Ni-20La2O3-ZrO2 < 5Ni-ZrO2. The un-promoted ZrO2 catalyst gives the highest weight loss of 47.4%. The results are well established with the catalytic performance. Similarly, Figure 9B shows the TGA profiles of spent 5Ni-xLa2O3 + Al2O3 (x = 0, 10, 15, and 20 wt%) catalysts at 700 ◦C. The amounts of weight loss are close to each between the La2O3 promoted and non-promoted catalysts. Since values range between 9.0% for 5Ni-10La2O3-Al2O3 and 17.2% for 5Ni-ZrO2, this indicates the Al2O3 supported catalysts are more resistant to carbon deposition than ZrO2 supported catalysts.

**Figure 10.** TGA profiles of spent (**A**) 5Ni-xLa2O3-ZrO2, (**B**) 5Ni-xLa2O3-Al2O3 (x = 0, 10, 15, and 20 wt%) catalysts at 700 ◦C.

Figure 11 shows the Raman analysis of the used catalysts (5Ni-ZrO2, 5Ni-10La2O3- ZrO2, 5Ni-Al2O3, and 5Ni-15La2O3-Al2O3). The D (deformation) and G (graphitic) bands appear at nearly 1342 and 1580 cm<sup>−</sup><sup>1</sup> respectively except for spent 5Ni-Al2O3 with D and G bands appearing at about 1468 and 1532 cm<sup>−</sup>1, respectively. The spent catalysts are characterized by carbon deposits of different degree of graphitization. It is established that carbon deposits having high *IG* to *ID* ratio show better extent of graphitization [62]. From the figure, it can be seen that 5Ni-ZrO2 had the highest degree of graphitization followed by 5Ni-Al2O3. Moreover, it can be inferred that the graphitization decreases with the addition of La2O3. Thus, La2O3 promotes the formation carbons that are defective.

**Figure 11.** Raman spectra for the spent catalysts (5Ni-ZrO2, 5Ni-10La2O3-ZrO2, 5Ni-Al2O3, and 5Ni-15La2O3-Al2O3) showing the extent of graphitization of the carbon deposits.
