Highly Active Nickel (II) Oxide-Supported Cerium Oxide Catalysts for Valorization of Glycerol into Oxygenated Fuel Additives

Acetylation of glycerol to yield monoacetin (MAT), diacetin (DAT), and triacetin (TAT) over NiO-supported CeO2 (xNiO/CeO2) catalysts is reported. The catalysts were synthesized utilizing a sol-gel technique, whereby different quantities of NiO (x = 9, 27, and 45 wt%) were supported onto the CeO2 substrate, and hexadecyltrimethylammonium bromide (CTABr) served as a porogen. The utilization of EDX elemental mapping analysis confirmed the existence of evenly distributed Ni2+ ion and octahedral NiO nanoparticles on the CeO2 surface through the DRS UV-Vis spectroscopy. The most active catalyst is 27NiO/CeO2 based on TAT selectivity in the glycerol acetylation with ethanoic acid, attaining 97.6% glycerol conversion with 70.5% selectivity to TAT at 170 °C with a 1:10 glycerol/ethanoic acid molar ratio for 30 min using a non-microwave instant heating reactor. The 27NiO/CeO2 is reusable without significant decline in catalytic performance after ten consecutive reaction cycles, indicating high structure stability with accessible active acidity.


Introduction
One of the most significant contributions in reducing greenhouse gas emissions is the production of biodiesel via the transesterification of vegetable oil or animal fat [1][2][3][4]. Meanwhile, glycerol is a by-product of biodiesel production that accounts for about 10% of the total production [5,6]. Due to the surplus of glycerol and the need to keep the chemical industry competitive, current research is mainly focused on the catalytic methods for converting glycerol into value-added compounds [1]. So far, various approaches to valorize glycerol, such as acetylation [7], aromatization [8], dehydration [9], esterification [10], hydrogenolysis [11], etherification [12], and acetalization [13,14], have been developed via the heterogeneous catalysis route.
Acetylation of glycerol has recently piqued researchers' interest due to the industrial significance of the end products. Mono-, di-, and triacetins (MATs, DATs, and TATs) are the three main products of the process. These products show broad and promising applications in many industries. Specifically, MAT is used in manufacturing explosives (dynamite), tanning animal skin for leather, and as a solvent for dyes [15], and DAT is used as a plasticizer, a solvent for dyes, and a softening agent [16,17]. TAT is the most difficult The typical synthesis process of the xNiO/CeO 2 catalysts is schematically illustrated in Scheme 1. For the preparation of the 45NiO/CeO 2 catalyst, a mixture of 50.00 g of distilled water, 1.50 g of cerium (IV) oxide (Sigma Aldrich, 99.9%, St. Louis, MO, USA), and 20.00 g of NaOH (Sigma Aldrich, 98%, St. Louis, MO, USA) was first mixed (400 rpm) for 20 min. This solution was then added to another clear solution containing 2.85 g of hexadecyltrimethylammonium bromide (Sigma Aldrich, 99%, St. Louis, MO, USA) and 25.00 g of distilled water. The final solution was further stirred (400 rpm) for another 17 h at 80 • C. Next, the pH of the mixture was adjusted to 10.0 using an HNO 3 solution (39.00 g, 3 M) that contained 1.49 g of NiCl 2 (Merck, 98%, Rahway, NJ, USA). At this pH, hexadecyltrimethylammonium bromide molecules tend to self-assemble into micelles, creating an ideal environment for the subsequent integration of metal precursors for catalyst production with mesoporosity [34]. The mixture was kept agitated for 2 h prior to aging in an oven at 80 • C for 48 h. The resulting precipitate was separated via centrifugation (7000 rpm, 5 min), washed until pH 7, and dried (90 • C, 10 h) before being calcined (580 • C, 5 h) to eliminate the organic porogen template. A similar protocol was used to prepare other xNiO/CeO 2 catalysts via varying the NiO amount where x = 9 (0.30 g NiCl 2 ) and x = 27 (0.90 g NiCl 2 ).
The synthesis of xNiO/CeO2 catalysts (x = 9, 27, and 45 wt.% of NiO) was carried using the sol-gel technique according to the modified method of Andas et al. [33]. typical synthesis process of the xNiO/CeO2 catalysts is schematically illustrated in Sch 1. For the preparation of the 45NiO/CeO2 catalyst, a mixture of 50.00 g of distilled w 1.50 g of cerium (IV) oxide (Sigma Aldrich, 99.9%, St. Louis, MO, USA), and 20.00 NaOH (Sigma Aldrich, 98%, St. Louis, MO, USA) was first mixed (400 rpm) for 20 This solution was then added to another clear solution containing 2.85 g of hexadecy methylammonium bromide (Sigma Aldrich, 99%, St. Louis, MO, USA) and 25.00 g of tilled water. The final solution was further stirred (400 rpm) for another 17 h at 80 Next, the pH of the mixture was adjusted to 10.0 using an HNO3 solution (39.00 g, 3 that contained 1.49 g of NiCl2 (Merck, 98%, Rahway, NJ, USA). At this pH, hexadecy methylammonium bromide molecules tend to self-assemble into micelles, creating ideal environment for the subsequent integration of metal precursors for catalyst pro tion with mesoporosity [34]. The mixture was kept agitated for 2 h prior to aging i oven at 80 °C for 48 h. The resulting precipitate was separated via centrifugation ( rpm, 5 min), washed until pH 7, and dried (90 °C, 10 h) before being calcined (580 °C, to eliminate the organic porogen template. A similar protocol was used to prepare o xNiO/CeO2 catalysts via varying the NiO amount where x = 9 (0.30 g NiCl2) and x (0.90 g NiCl2).

Catalyst Characterization
The X-ray diffractograms (XRDs) were obtained using a PANalytical X'Pert Pro fractometer, which employed CuKα radiation with a wavelength of 1.5418 Å, a voltag 40 kV, and a current of 10 mA. The scanning range was from 2θ = 20° to 65°, with a s ning rate of 0.2° per min and a step size of 0.02°. The surface area and pore size distribu of catalysts were characterized using the N2 adsorption-desorption isothermal ana obtained at −196 °C through a Micromeritics ASAP 2010 analyzer. The catalyst pow Scheme 1. Synthesis route of xNiO/CeO 2 catalysts.

Catalyst Characterization
The X-ray diffractograms (XRDs) were obtained using a PANalytical X'Pert Pro diffractometer, which employed CuKα radiation with a wavelength of 1.5418 Å, a voltage of 40 kV, and a current of 10 mA. The scanning range was from 2θ = 20 • to 65 • , with a scanning rate of 0.2 • per min and a step size of 0.02 • . The surface area and pore size distribution of catalysts were characterized using the N 2 adsorption-desorption isothermal analysis obtained at −196 • C through a Micromeritics ASAP 2010 analyzer. The catalyst powder (0.0900 mg) was first degassed for 12 h at 300 • C and 10 −4 Pa in the degassing port of the analyzer. The surface area was computed using the Brunauer-Emmett-Teller (BET) method, while the pore size distribution was calculated using the Barrett, Joyner, and Halenda (BJH) method. The diffuse reflectance (DRS) UV-Vis spectra of catalysts were obtained on a Perkin Elmer Lamda 35 spectrometer. The field emission scanning electron microscopy (FESEM) and energy dispersive X-ray (EDS) analyses for morphological and elemental mapping investigations were performed on FEI's Quanta FEG 650 microscope equipped with an Oxford XMax 50 Silicon Drift EDS detector. The transmission electron microscopy (TEM) study was performed on a Philips CM-12 TEM microscope. The samples in methanol solvent were first sonicated for 10 min before being deposited on copper grids for imaging analysis. The acidity of catalysts was characterized by the ammonia temperature programmed desorption (TPD-NH 3 ) technique using a Thermo Electron TPDRO 1100. Prior to analysis, the sample (0.060 g) was degassed at 500 • C overnight, followed by NH 3 adsorption for 30 min. The ammonia probe molecule was then desorbed from the sample surface from 50 to 1000 • C at a heating rate of 10 • C min −1 . The thermogravimetric analysis (TGA) was performed on Mettler Toledo 851e equipment. The temperature was measured from 50 to 900 • C at a heating rate of 20 • C min −1 and air flow rate of 40 mL min −1 . The amounts of cerium and nickel in the samples were measured using a Perkin Elmer Optima 4300DV ICP-OES spectrometer. Before measurement, the sample powder was first dissolved in an aqua regia (HNO 3 :HCl = 1:3) and subjected to microwave digestion at 200 • C for 20 min (800 W) using a CEM MARS 6 microwave digester.

Catalytic Reaction Study and Products Analysis
The process of acetylating glycerol was conducted within a temperature range of 130-170 • C, utilizing a non-microwave instant heating reactor (Anton Paar's Monowave 50). Initially, 0.040 g of xNiO/CeO 2 catalyst was first activated at 250 • C for 1 h before it was added with 0.139 g of glycerol (QReC, 99.5%) and 0.900 g of ethanoic acid (Sigma-Aldrich, St. Louis, MO, USA) into a glass tube (10 cm 3 ). The mixture was sealed with a silicone cap and rapidly heated to a desired temperature (150-170 • C) and kept at a plateau for 5-60 min. After the reaction, the tube together with the mixture were subjected to highspeed centrifugation (10,000 rpm, 10 min) to separate the solid catalyst. The remaining liquid product was then analyzed using an Agilent's HP 6890 GC instrument with an FID detector. The glycerol conversion and products' selectivity were determined using the following equations (toluene as internal standard): Selectivity(%) = mol MAT or mol DAT or mol TAT mol Glycerol converted × 100 (2) where mol glycerol,0 and mol glycerol,t are the amount of glycerol in mol before and after acetylation reaction at t min. Meanwhile, mol MAT , mol DAT , mol TAT , and mol Glycerol are the amounts of MAT, DAT, TAT, and glycerol in mol, respectively. All the quantitative analyses were performed using the absolute calibration method where the known amounts of reactant (glycerol) and products (MAT, DAT, and TAT) are used for plotting of calibration curves. The recovered catalyst was soaked and washed three times with diethyl ether (30 mL), and air dried before being activated once more (250 • C, 1 h) for the next cycles of reaction.

Characterizations of xNiO/CeO 2 Catalysts
NiO/CeO 2 catalysts were first studied with the XRD technique, and the wide-angle XRD pattern is shown in Figure 1. As seen, all prepared solids show five major peaks at 2θ angles of 28. These peaks become more intense with increasing the NiO content. The presence of these peaks can be explained by the presence of NiO particles deposited on the sample surfaces compared to that of pristine CeO 2 . However, the XRD peak intensity of NiO is weak despite the high Ni content in the catalyst. This phenomenon could be explained by the very small crystallite size of NiO, which is distributed evenly on the CeO 2 support. As a result, it causes the peaks to spread and lose intensity, lowering the diffraction intensity. Moreover, a few impurities peaks due to NiCl 2 and Ni(OH) 2 are also found in the 45NiO/CeO 2 catalyst (at 2θ = 29.5 • , 31.9 • ), where the results are well corroborated with the JCPDF No. 002-0765 [36] and JCPDS No. 14-0117, respectively [37]. However, these peaks are not observed in 9NiO/CeO 2 and 27NiO/CeO 2 catalysts with less NiO content, suggesting only pure NiO and CeO 2 crystal phases are obtained in both solids.
NiO/CeO2 catalysts were first studied with the XRD technique, and the wide-angle XRD pattern is shown in Figure 1. As seen, all prepared solids show five major peaks at 2θ angles of 28.5°, 33.0°, 47.5°, 56.3°, and 59.2° attributed to the (111), (200), (220), (311), and (222) planes of CeO2, respectively [35]. The crystallography data also correspond to the cubic fluorite structure (Fm3m) of CeO2 (JCPDF No. . Additionally, several weak diffraction peaks due to (101), (111), (200), and (110) crystal planes of NiO are also detected at 2θ = 37.2°, 38.9°, 43.3°, and 62.9° by 9NiO/CeO2, 27NiO/CeO2, and 45NiO/CeO2. These peaks become more intense with increasing the NiO content. The presence of these peaks can be explained by the presence of NiO particles deposited on the sample surfaces compared to that of pristine CeO2. However, the XRD peak intensity of NiO is weak despite the high Ni content in the catalyst. This phenomenon could be explained by the very small crystallite size of NiO, which is distributed evenly on the CeO2 support. As a result, it causes the peaks to spread and lose intensity, lowering the diffraction intensity. Moreover, a few impurities peaks due to NiCl2 and Ni(OH)2 are also found in the 45NiO/CeO2 catalyst (at 2θ = 29.5°, 31.9°), where the results are well corroborated with the JCPDF No. 002-0765 [36] and JCPDS No. 14-0117, respectively [37]. However, these peaks are not observed in 9NiO/CeO2 and 27NiO/CeO2 catalysts with less NiO content, suggesting only pure NiO and CeO2 crystal phases are obtained in both solids.  Figure 2 shows the nitrogen adsorption-desorption isotherms for the xNiO/CeO2 solids. The CeO2 initially has a specific surface area (SBET) and a total pore volume (VTotal) of 69 m 2 /g and 0.15 cm 3 /g, respectively ( Table 1). The solid material demonstrates a type IV adsorption isotherm and H1 hysteresis loop, as per the IUPAC classification, indicating the presence of mesoporous structure having a small pore mouth and wide body (inkbottle pores) with an average pore size of 9.86 nm (Figure 2a). The closure point of the hysteresis loop at P/Po = 0.4 shows that adding CTABr to the synthesis of xNiO/CeO2  Figure 2 shows the nitrogen adsorption-desorption isotherms for the xNiO/CeO 2 solids. The CeO 2 initially has a specific surface area (S BET ) and a total pore volume (V Total ) of 69 m 2 /g and 0.15 cm 3 /g, respectively ( Table 1). The solid material demonstrates a type IV adsorption isotherm and H1 hysteresis loop, as per the IUPAC classification, indicating the presence of mesoporous structure having a small pore mouth and wide body (ink-bottle pores) with an average pore size of 9.86 nm (Figure 2a). The closure point of the hysteresis loop at P/P o = 0.4 shows that adding CTABr to the synthesis of xNiO/CeO 2 results in the generation of an extra new mesopore system ( Figure 2b). Yet, when the nickel level increases, the fundamental mesostructure identity of the solid gradually diminishes. For instance, the 9NiO/CeO 2 retains its type IV isotherm and H3 hysteresis loop features but has lower nitrogen uptake. As a result, the average pore size and total pore volume are slightly decreased. On the other hand, the specific surface area remains almost intact due to the generation of secondary mesopores by the CTABr mesoporogen. Upon further increasing the nickel content, the porosity gradually decreases (low surface area, micropore surface area, and total pore volume), leading to narrower average pore sizes. In addition, the initial ink-bottle pore shape also changes to a slit shape (for 27NiO/CeO 2 ) and finally the inner pores of the solid are fully covered (for 45NiO/CeO 2 ). Such a phenomenon is associated with partial or complete pore blockage by the NiO nanoparticles [38]. level increases, the fundamental mesostructure identity of the solid gradually diminishes. For instance, the 9NiO/CeO2 retains its type IV isotherm and H3 hysteresis loop features but has lower nitrogen uptake. As a result, the average pore size and total pore volume are slightly decreased. On the other hand, the specific surface area remains almost intact due to the generation of secondary mesopores by the CTABr mesoporogen. Upon further increasing the nickel content, the porosity gradually decreases (low surface area, micropore surface area, and total pore volume), leading to narrower average pore sizes. In addition, the initial ink-bottle pore shape also changes to a slit shape (for 27NiO/CeO2) and finally the inner pores of the solid are fully covered (for 45NiO/CeO2). Such a phenomenon is associated with partial or complete pore blockage by the NiO nanoparticles [38].  SEM analysis is performed to investigate the morphological features of samples after sol-gel treatment. As seen, the CeO2 consists of agglomerated irregular-shaped particles with a size of 125 ± 23 nm ( Figure 3a). Upon impregnating with various amounts of NiO, the density of bright nanoparticles also increases and they are uniformly distributed all over the surface of the samples (Figure 3b-d). The identity of bright spots was further investigated using EDX elemental mapping analysis where Ce, Ni, and O elements were mapped onto the SEM images ( Figure 4). For the pristine CeO2 sample, only Ce and O elements are detected due to the CeO2 support itself. For xNiO/CeO2 solids, the microanalysis reveals that the Ni element is enriched on the solid surface relative to the amount  SEM analysis is performed to investigate the morphological features of samples after sol-gel treatment. As seen, the CeO 2 consists of agglomerated irregular-shaped particles with a size of 125 ± 23 nm ( Figure 3a). Upon impregnating with various amounts of NiO, the density of bright nanoparticles also increases and they are uniformly distributed all over the surface of the samples (Figure 3b-d). The identity of bright spots was further investigated using EDX elemental mapping analysis where Ce, Ni, and O elements were mapped onto the SEM images ( Figure 4). For the pristine CeO 2 sample, only Ce and O elements are detected due to the CeO 2 support itself. For xNiO/CeO 2 solids, the microanalysis reveals that the Ni element is enriched on the solid surface relative to the amount of NiO introduced. In addition, the Ni element is also well-scattered on the solid surface without segregation, further proving the successful functionalization process. of NiO introduced. In addition, the Ni element is also well-scattered on the solid surface without segregation, further proving the successful functionalization process.  The microscopic visualization of NiO nanoparticles was further carried out using the TEM microscopy technique. Similar to the SEM/EDX study, uniform dispersion of NiO nanoparticles (approx. 2-3 nm) on the surface of CeO 2 solid is seen in the xNiO/CeO 2 samples (as dark spots) ( Figure 5). On the other hand, the morphology and size of CeO 2 crystals (ca. 165 nm in grain shape) is unaffected upon sol-gel and calcination treatments. As a result, the microscopy results are in accordance with the XRD findings, viz. no substantial NiO particles are dispersed on the CeO 2 surface.  The microscopic visualization of NiO nanoparticles was further carried out using the TEM microscopy technique. Similar to the SEM/EDX study, uniform dispersion of NiO nanoparticles (approx. 2-3 nm) on the surface of CeO2 solid is seen in the xNiO/CeO2 samples (as dark spots) ( Figure 5). On the other hand, the morphology and size of CeO2 crystals (ca. 165 nm in grain shape) is unaffected upon sol-gel and calcination treatments. As a result, the microscopy results are in accordance with the XRD findings, viz. no substantial NiO particles are dispersed on the CeO2 surface. The electronic states of NiO and CeO 2 in the xNiO/CeO 2 samples were studied using DRS UV-Vis spectroscopy. A peak at 257 nm is observed in all Ni 2+ -containing samples, indicating the occurrence of charge transfer (CT) transitions involving Ni 2+ ions (Figure 6b-d). These Ni 2+ ions from NiO exist in octahedral form in the crystal lattice and are surrounded by six oxygen ions [39]. On the other hand, a peak at the nearly same position is also observed for the pristine CeO 2 , which corresponds to the CT transition from Ce 4+ (4f ) to O 2− (2p) (Figure 6a) [40]. CeO 2 has a complicated electronic structure because it contains both Ce 3+ and Ce 4+ ions, both of which can undergo CT transitions with the oxygen ions in the crystal lattice [41,42]. Thus, this result suggests that there is a chemical interaction occurring between the NiO and CeO 2 support. A peak of maximum absorption resulting from the charge transfer of O 2− → Ce 4+ is identified at 290 nm, and a wide absorption edge of CeO 2 is found at 329 nm [43,44]. Moreover, the peak intensity increases with the quantity of NiO impregnated, and this signal can be associated with the Ni-O charge transfer transition [45]. In addition, the 9NiO/CeO 2 , 27NiO/CeO 2 , and 45NiO/CeO 2 exhibit a new and broad signal centered at 725 nm, which is attributed to the electronic transition of octahedrally coordinated Ni 2+ ions from the 3 A 2g (F) → the 3 T 1g (F) state [46,47]. This band is stronger for 45NiO/CeO 2 , followed by 27NiO/CeO 2 and 9NiO/CeO 2 . Thus, this band indicates a substantial covalent character in Ni-O support metal-ligand bonds after calcination [46]. The electronic states of NiO and CeO2 in the xNiO/CeO2 samples were studied using DRS UV-Vis spectroscopy. A peak at 257 nm is observed in all Ni 2+ -containing samples, indicating the occurrence of charge transfer (CT) transitions involving Ni 2+ ions ( Figure  6b-d). These Ni 2+ ions from NiO exist in octahedral form in the crystal lattice and are surrounded by six oxygen ions [39]. On the other hand, a peak at the nearly same position is also observed for the pristine CeO2, which corresponds to the CT transition from Ce 4+ (4f) to O 2− (2p) (Figure 6a) [40]. CeO2 has a complicated electronic structure because it contains both Ce 3+ and Ce 4+ ions, both of which can undergo CT transitions with the oxygen ions in the crystal lattice [41,42]. Thus, this result suggests that there is a chemical interaction occurring between the NiO and CeO2 support. A peak of maximum absorption resulting from the charge transfer of O 2− → Ce 4+ is identified at 290 nm, and a wide absorption edge of CeO2 is found at 329 nm [43,44]. Moreover, the peak intensity increases with the quantity of NiO impregnated, and this signal can be associated with the Ni-O charge transfer transition [45]. In addition, the 9NiO/CeO2, 27NiO/CeO2, and 45NiO/CeO2 exhibit a new and broad signal centered at 725 nm, which is attributed to the electronic transition of octahedrally coordinated Ni 2+ ions from the 3 A2g(F) → the 3 T1g(F) state [46,47]. This band is stronger for 45NiO/CeO2, followed by 27NiO/CeO2 and 9NiO/CeO2. Thus, this band The acidity of xNiO/CeO 2 solids was studied with the NH 3 -TPD, and the results are presented in Figure 7. It is shown that all the samples exhibit similar desorption tendency except for the bare CeO 2 , where the results can be divided into three desorption regions corresponding to the specific basic sites, namely mild acid sites (<300 • C), mild to strong acid sites (300-500 • C), and strong acid sites (>500 • C). The data reveal that 27NiO/CeO 2 displays a larger desorption peak area and stronger peak intensity than pure CeO 2 , 9NiO/CeO 2 , and 45NiO/CeO 2 catalysts, indicating the presence of larger quantity of acidic sites on the surface. The acidity data are also tabulated as in Table 1. From the data, it can be observed that the total acidity of the samples increases with increasing the NiO content. For pristine CeO 2 , it has acid sites with mild strength (129 µmol/g), which increases to 271 µmol/g and 320 µmol/g after the NiO content is increased to 9% and 27%, respectively, before slightly dropping to 235 µmol/g for 45NiO/CeO 2 . Similarly, the acidities with mild-to-strong and strong strengths also increase with increasing the NiO content. It should be noted that the 45NiO/CeO 2 has a trace amount of non-reacted NiCl 2 , as indicated by TG/DTG and XRD (Figures 1 and 8). indicates a substantial covalent character in Ni-Osupport metal-ligand bonds after calcination [46]. The acidity of xNiO/CeO2 solids was studied with the NH3-TPD, and the results are presented in Figure 7. It is shown that all the samples exhibit similar desorption tendency except for the bare CeO2, where the results can be divided into three desorption regions corresponding to the specific basic sites, namely mild acid sites (<300 °C), mild to strong acid sites (300-500 °C), and strong acid sites (>500 °C). The data reveal that 27NiO/CeO2 displays a larger desorption peak area and stronger peak intensity than pure CeO2, 9NiO/CeO2, and 45NiO/CeO2 catalysts, indicating the presence of larger quantity of acidic sites on the surface. The acidity data are also tabulated as in Table 1. From the data, it can be observed that the total acidity of the samples increases with increasing the NiO content. For pristine CeO2, it has acid sites with mild strength (129 µmol/g), which increases to 271 µmol/g and 320 µmol/g after the NiO content is increased to 9% and 27%, respectively, before slightly dropping to 235 µmol/g for 45NiO/CeO2. Similarly, the acidities with mildto-strong and strong strengths also increase with increasing the NiO content. It should be noted that the 45NiO/CeO2 has a trace amount of non-reacted NiCl2, as indicated by TG/DTG and XRD (Figures 1 and 8). Thermogravimetric analysis/differential thermal analysis (TGA/DTG) was performed under air atmosphere to evaluate the thermal stability and heat resistance of the xNiO/CeO2 solids (Figure 8). The pristine CeO2 demonstrates two distinct stages of weight loss. The initial stage (<200 °C) is attributed to the removal of physiosorbed water (10.4%) and the second step of weight loss from 200 until 500 °C (1.7%) can be attributed to chemisorption of water and condensation of surface Ce-OH groups into a Ce-O-Ce group [48].  Hence, the desorption of NH3 at around 670 °C by 45NiO/CeO2 is also accompani by the decomposition of non-reacted NiCl2 and Ni(OH)2, leading to the detection of t quantity of mild-to-strong and strong acid sites larger than the actual values. Among t three xNiO/CeO2 catalysts, 27NiO/CeO2 has the highest number of acidity (1564 µmol with the highest acid strength (373 µmol/g). Thus, the NH3-TPD data clearly show th the incorporation of NiO on CeO2 (particularly 27NiO/CeO2) enhances the number of ac sites, which would promote enhanced catalytic activity for acetylation of glycerol reactio

Catalytic Comparative Study of xNiO/CeO2 Catalysts
Acetylation of glycerol was catalyzed by CeO2, 9NiO/CeO2, 27NiO/CeO2, a 45NiO/CeO2 under instant conductive heating (150 °C and 170 °C for 15 min) using ethanoic acid:glycerol molar ratio of 10:1 and a catalyst loading of 0.04 g ( Table 2). T reaction products obtained are monoacetin (MAT), diacetin (DAT), and triacetin (TA Pure CeO2 support is inactive in this reaction, which only converts 20.0% of glycerol in 19.3% of MAT and 72.7% of DAT at 150 °C (Entry 1). Upon elevating the temperature 170 °C, the glycerol conversion exhibited an increase to 31.1%. The product selectivit were observed to be 17.2%, 72.6%, and 0% for MAT, DAT, and TAT, respectively. Howev the reaction conversion increases upon impregnating NiO nanoparticles on the CeO2, w the catalytic activity enhanced In the following sequence: CeO2 < 9NiO/CeO2 45NiO/CeO2 < 27NiO/CeO2. Among these solids, the 27NiO/CeO2 shows the best resu with the highest glycerol conversion of 84.3% and 95.3% at 150 °C and 170 °C, respective and selectivity to DAT (66.0% and 60.4%) (Entry 3). The high activity of 27NiO/CeO2 c be explained by its large amount and good dispersal of active NiO nanoparticles (w Thermogravimetric analysis/differential thermal analysis (TGA/DTG) was performed under air atmosphere to evaluate the thermal stability and heat resistance of the xNiO/CeO 2 solids ( Figure 8). The pristine CeO 2 demonstrates two distinct stages of weight loss. The initial stage (<200 • C) is attributed to the removal of physiosorbed water (10.4%) and the second step of weight loss from 200 until 500 • C (1.7%) can be attributed to chemisorption of water and condensation of surface Ce-OH groups into a Ce-O-Ce group [48]. Upon functionalizing with NiO, an additional step of weight loss at 245 • C is observed by 9NiO/CeO 2 , 27NiO/CeO 2 , and 45NiO/CeO 2 , and the degree of weight loss increases with the amount of NiO incorporated. Meanwhile, 45NiO/CeO 2 shows an additional weight loss centered at 670 • C (1.4%) due to the decomposition of non-reacted NiCl 2 and Ni(OH) 2 . The data are in line with the XRD data indicating the presence of little NiCl 2 and Ni(OH) 2 phases in the 45NiO/CeO 2 solid.
Hence, the desorption of NH 3 at around 670 • C by 45NiO/CeO 2 is also accompanied by the decomposition of non-reacted NiCl 2 and Ni(OH) 2 , leading to the detection of the quantity of mild-to-strong and strong acid sites larger than the actual values. Among the three xNiO/CeO 2 catalysts, 27NiO/CeO 2 has the highest number of acidity (1564 µmol/g) with the highest acid strength (373 µmol/g). Thus, the NH 3 -TPD data clearly show that the incorporation of NiO on CeO 2 (particularly 27NiO/CeO 2 ) enhances the number of acid sites, which would promote enhanced catalytic activity for acetylation of glycerol reaction.

Catalytic Comparative Study of xNiO/CeO 2 Catalysts
Acetylation of glycerol was catalyzed by CeO 2 , 9NiO/CeO 2 , 27NiO/CeO 2 , and 45NiO/CeO 2 under instant conductive heating (150 • C and 170 • C for 15 min) using an ethanoic acid:glycerol molar ratio of 10:1 and a catalyst loading of 0.04 g ( Table 2). The reaction products obtained are monoacetin (MAT), diacetin (DAT), and triacetin (TAT). Pure CeO 2 support is inactive in this reaction, which only converts 20.0% of glycerol into 19.3% of MAT and 72.7% of DAT at 150 • C (Entry 1). Upon elevating the temperature to 170 • C, the glycerol conversion exhibited an increase to 31.1%. The product selectivities were observed to be 17.2%, 72.6%, and 0% for MAT, DAT, and TAT, respectively. However, the reaction conversion increases upon impregnating NiO nanoparticles on the CeO 2 , with the catalytic activity enhanced In the following sequence: CeO 2 < 9NiO/CeO 2 < 45NiO/CeO 2 < 27NiO/CeO 2 . Among these solids, the 27NiO/CeO 2 shows the best results, with the highest glycerol conversion of 84.3% and 95.3% at 150 • C and 170 • C, respectively, and selectivity to DAT (66.0% and 60.4%) (Entry 3). The high activity of 27NiO/CeO 2 can be explained by its large amount and good dispersal of active NiO nanoparticles (with strong acidity) on the support that are highly accessible for the adsorption and activation of reactant molecules. In contrast, the 45NiO/CeO 2 catalyst exhibits poorer catalytic activity than the 27NiO/CeO 2 solid due to its lower surface area with fewer octahedral NiO species incorporated, as confirmed by the nitrogen adsorption isotherm and DRS UV-Vis spectroscopy results. Additionally, pore blockage could also happen due to the excessive amount of Ni species present on the CeO 2 surface, which inhibits molecular diffusion to the active sites, thereby causing poor catalytic performance [49]. Thus, the 27NiO/CeO 2 catalyst was chosen to further study the acetylation of glycerol involving reaction temperature, heating time, catalyst loading, glycerol to ethanoic acid ratio, and catalyst reusability.

Effect of Reaction Temperature and Duration
The acetylation reaction over the 27NiO/CeO 2 catalyst was investigated from 0 to 60 min at 150 to 170 • C. As seen in Figure 9A, the reaction temperature is the dominant factor affecting the glycerol conversion. For instance, at 150 • C, the conversion is 63.8% within 5 min and the conversion increases to 95.5% after 60 min. However, when the temperature increases from 150 • C to 170 • C, a substantial increase in the conversion from 63.8% to 87.0% is observed after 5 min. The reaction time extends to 60 min; the conversion (95.5%) at 150 • C increases to 97.6% and 98.8% for 160 • C and 170 • C, respectively. The observed phenomenon may be attributed to the increased frequency of effective collisions between glycerol and ethanoic acid at elevated temperatures, resulting in a high reaction rate [50]. At 170 • C, the conversion increases to 97.6% and becomes nearly stable after 30 min. The activation energies (E a ) of the reaction catalyzed with 27NiO/CeO 2 and without a catalyst are calculated using the Arrhenius equation to further understand the catalytic effect on the reaction kinetics where second-order rate constants are applied ( Figure 9B). The E a of the non-catalyzed reaction is 149.3 kJ mol −1 and the value decreases tremendously to 106.6 kJ mol −1 after catalyzing with 27NiO/CeO 2 . Thus, this indicates that the acetylation reaction has been activated, where 27NiO/CeO 2 provides another reaction route of lower activation energy. Based on the study, the optimum reaction temperature and time are 170 • C and 30 min, respectively.

Effect of the Amount of Catalyst
A catalyst is needed in a transformation reaction to ensure cost effectiveness of a pr cess [51]. Hence, the 27NiO/CeO2 catalyst loading ranging from 0 to 0.05 g was studi with other variables remaining unchanged ( Figure 10). Only 17.8% of glycerol is convert into 27.4% of MAT and 63.2% of DAT after 30 min at 170 °C. Upon introducing 0.01 g catalyst, the conversion reaches 45.2%. The conversion is again enhanced to 64.8% an 87.4% when the amount of catalyst is increased to 0.02 g and 0.03 g, respectively. T optimum catalyst loading is 0.04 g, with 97.6% conversion is achieved at the same reactio time. However, upon increasing the catalyst loading to 0.05 g, the glycerol conversio slightly decreases (95.1%) due to inefficient stirring and mixing resulting from excessi catalyst loading [52].

Effect of the Amount of Catalyst
A catalyst is needed in a transformation reaction to ensure cost effectiveness of a process [51]. Hence, the 27NiO/CeO 2 catalyst loading ranging from 0 to 0.05 g was studied with other variables remaining unchanged ( Figure 10). Only 17.8% of glycerol is converted into 27.4% of MAT and 63.2% of DAT after 30 min at 170 • C. Upon introducing 0.01 g of catalyst, the conversion reaches 45.2%. The conversion is again enhanced to 64.8% and 87.4% when the amount of catalyst is increased to 0.02 g and 0.03 g, respectively. The optimum catalyst loading is 0.04 g, with 97.6% conversion is achieved at the same reaction time. However, upon increasing the catalyst loading to 0.05 g, the glycerol conversion slightly decreases (95.1%) due to inefficient stirring and mixing resulting from excessive catalyst loading [52]. An increase in conversion is anticipated with catalyst loading as a result of the availability of more accessible active sites. However, the products' selectivity exhibits the opposite trend as the catalyst loading is altered. For instance, the selectivity towards DAT is predominant without a catalyst since the glycerol conversion is low, and the formation of TAT increases with the amount of catalyst. The 27NiO/CeO2 catalyst at a dosage of 0.04 g showed the best performance (97.6% conversion, 70.5% selective to TAT). To our knowledge, this result is the best catalytic performance thus far [22,23,53,54]. For the subsequent catalytic reaction experiments, 0.04 g of 27NiO/CeO2 was used.

Effect of Glycerol:Ethanoic Acid Molar Ratio
Acetylation reaction performance can be greatly affected by the quantity of reactants used since they govern the reaction chemical kinetics [55]. In order to obtain the highest conversion and TAT selectivity, the amount of glycerol and ethanoic acid (in molar ratio) was varied, with other reaction parameters remaining identical. The conversion is low (22.0%) and no TAT is obtained (MAT: 62.3%, DAT: 27.8%) when the molar ratio of glycerol to ethanoic acid is 10:1 (Figure 11). When the glycerol:ethanoic acid ratio is reduced to 5:1, a less viscous reaction mixture is formed. As seen, the conversion increases onefold (41.0%), and an increased selectivity of MAT and DAT of 51.0% and 37.4%, was recorded, respectively. This clearly indicates that a low glycerol concentration is beneficial for the reaction conversion as it eases reactant mixing and stirring, thereby improving product selectivity. An increase in conversion is anticipated with catalyst loading as a result of the availability of more accessible active sites. However, the products' selectivity exhibits the opposite trend as the catalyst loading is altered. For instance, the selectivity towards DAT is predominant without a catalyst since the glycerol conversion is low, and the formation of TAT increases with the amount of catalyst. The 27NiO/CeO 2 catalyst at a dosage of 0.04 g showed the best performance (97.6% conversion, 70.5% selective to TAT). To our knowledge, this result is the best catalytic performance thus far [22,23,53,54]. For the subsequent catalytic reaction experiments, 0.04 g of 27NiO/CeO 2 was used.

Effect of Glycerol:Ethanoic Acid Molar Ratio
Acetylation reaction performance can be greatly affected by the quantity of reactants used since they govern the reaction chemical kinetics [55]. In order to obtain the highest conversion and TAT selectivity, the amount of glycerol and ethanoic acid (in molar ratio) was varied, with other reaction parameters remaining identical. The conversion is low (22.0%) and no TAT is obtained (MAT: 62.3%, DAT: 27.8%) when the molar ratio of glycerol to ethanoic acid is 10:1 ( Figure 11). When the glycerol:ethanoic acid ratio is reduced to 5:1, a less viscous reaction mixture is formed. As seen, the conversion increases onefold (41.0%), and an increased selectivity of MAT and DAT of 51.0% and 37.4%, was recorded, respectively. This clearly indicates that a low glycerol concentration is beneficial for the reaction conversion as it eases reactant mixing and stirring, thereby improving product selectivity.
The acetylation is also studied by increasing the concentration of ethanoic acid. As equimolar reactants (5:5) were applied, the conversion was further enhanced to 52.0%, shifting the product distribution towards higher substituted product (15.7% MAT, 54.0% DAT, and 19.3% TAT). Thus, the results show that ethanoic acid activation is essential for enabling high catalytic activity [56]. In order to further prove this statement, the glycerol:ethanoic acid ratio was again increased to 1:10, where a significant reaction conversion enhancement was shown (up to 97.6%) and the TAT selectivity was doubled (70.5%) thanks to the high availability of activated ethanoic acid in the reaction medium. Beyond this ratio (1:15), the reaction performance was almost intact. Therefore, the optimum glycerol:ethanoic acid ratio is 1:10. The acetylation is also studied by increasing the concentration of ethanoic acid. As equimolar reactants (5:5) were applied, the conversion was further enhanced to 52.0%, shifting the product distribution towards higher substituted product (15.7% MAT, 54.0% DAT, and 19.3% TAT). Thus, the results show that ethanoic acid activation is essential for enabling high catalytic activity [56]. In order to further prove this statement, the glycerol:ethanoic acid ratio was again increased to 1:10, where a significant reaction conversion enhancement was shown (up to 97.6%) and the TAT selectivity was doubled (70.5%) thanks to the high availability of activated ethanoic acid in the reaction medium. Beyond this ratio (1:15), the reaction performance was almost intact. Therefore, the optimum glycerol:ethanoic acid ratio is 1:10.

Catalyst Recyclability Study
The deactivation of catalyst active sites is a crucial issue for a heterogeneous catalysis system [5]. Therefore, a reusability test was conducted on 27NiO/CeO2 to evaluate its stability and industrial feasibility. Figure 12 shows the glycerol conversion and products' selectivity for ten consecutive runs. The glycerol conversion decreases from 97.6% to 86.0% after ten cycles of reuse. Hence, the reduction in the conversion could have resulted from the deactivation of active sites and partial leaching of active metal species from the surface of the catalyst support [57]. In addition, the product distribution shifted towards lower substituted products after multiple reaction runs. For instance, the MAT and DAT selectivities increased from 3.6% to 7.3% and 14.7% to 14.9%, respectively, after the 1 0t h run. However, a slight reduction in the TAT selectivity was recorded from 70.5% to 65.2%, which indicates that the 27NiO/CeO2 catalyst is somewhat stable up to ten runs.

Catalyst Recyclability Study
The deactivation of catalyst active sites is a crucial issue for a heterogeneous catalysis system [5]. Therefore, a reusability test was conducted on 27NiO/CeO 2 to evaluate its stability and industrial feasibility. Figure 12 shows the glycerol conversion and products' selectivity for ten consecutive runs. The glycerol conversion decreases from 97.6% to 86.0% after ten cycles of reuse. Hence, the reduction in the conversion could have resulted from the deactivation of active sites and partial leaching of active metal species from the surface of the catalyst support [57]. In addition, the product distribution shifted towards lower substituted products after multiple reaction runs. For instance, the MAT and DAT selectivities increased from 3.6% to 7.3% and 14.7% to 14.9%, respectively, after the 10th run. However, a slight reduction in the TAT selectivity was recorded from 70.5% to 65.2%, which indicates that the 27NiO/CeO 2 catalyst is somewhat stable up to ten runs.

Thermodynamics and Kinetics Properties of Reaction
The production of ethyl levulinate in the presence of the 27NiO/CeO2 catalyst affects the kinetics and thermodynamic properties. Hence, in order to explain this phenomenon, the thermodynamic quantities (∆Hrxn, ∆Srxn and ∆Grxn) of acetylation of glycerol at 170 °C were calculated using the data provided in Refs. [58][59][60] and the following equations:

Thermodynamics and Kinetics Properties of Reaction
The production of ethyl levulinate in the presence of the 27NiO/CeO 2 catalyst affects the kinetics and thermodynamic properties. Hence, in order to explain this phenomenon, the thermodynamic quantities (∆H rxn , ∆S rxn and ∆G rxn ) of acetylation of glycerol at 170 • C were calculated using the data provided in Refs. [58][59][60] and the following equations: where enthalpy and entropy of vaporization (∆H vap ) values of ethanoic acid and water are also included. The energy profile of the reaction is summarized as in Figure 13. The calculation reveals that the acetylation of glycerol with ethanoic acid reaction is an exothermic reaction with ∆H rxn = −92.7 kJ/mol. Meanwhile, the ∆G rxn is calculated to be −137.4 kJ/mol, revealing that the acetylation reaction is very spontaneous at 170 • C. Although this reaction is thermodynamically favorable, it occurs at a very slow speed in the absence of a catalyst. This can be proven when mild conversion (50.2%) is observed at 170 • C after 60 min in the absence of a catalyst ( Figure 9A). Thus, the Gibbs free energy cannot address this particular concern (how fast does the reaction achieve high conversion). The chemical kinetics are hence the tools needed to address it. The reaction kinetics (or rate constants) are very reliable on the temperature. When high temperature is applied, the reaction takes place at a faster rate but the activation energy is still unchanged [59]. Hence, glycerol in mild conversion (~50%) is recorded at 170 °C after 60 min ( Figure 9A). The presence of the 27NiO/CeO2 catalyst is important in the acetylation of glycerol as it provides a different reaction route by allowing the diffusion, adsorption, and close proximity between reactant molecules on the active sites of the 27NiO/CeO2 catalyst before they are converted into TAT. As a result, this new reaction pathway lowers the activation energy from 149.3 kJ/mol to 106.6 kJ/mol and enhances the reaction kinetics.

Possible Mechanism of the Acetylation Reaction of Glycerol
The proposed mechanism is based on the current results and previous research [61]. The most plausible pathway for the acetylation of glycerol with ethanoic acid involves the activation of the carbonyl group of ethanoic acid by NiO in a 27NiO/CeO2 catalyst, which The reaction kinetics (or rate constants) are very reliable on the temperature. When high temperature is applied, the reaction takes place at a faster rate but the activation energy is still unchanged [59]. Hence, glycerol in mild conversion (~50%) is recorded at 170 • C after 60 min ( Figure 9A). The presence of the 27NiO/CeO 2 catalyst is important in the acetylation of glycerol as it provides a different reaction route by allowing the diffusion, adsorption, and close proximity between reactant molecules on the active sites of the 27NiO/CeO 2 catalyst before they are converted into TAT. As a result, this new reaction pathway lowers the activation energy from 149.3 kJ/mol to 106.6 kJ/mol and enhances the reaction kinetics.

Possible Mechanism of the Acetylation Reaction of Glycerol
The proposed mechanism is based on the current results and previous research [61]. The most plausible pathway for the acetylation of glycerol with ethanoic acid involves the activation of the carbonyl group of ethanoic acid by NiO in a 27NiO/CeO 2 catalyst, which increases the electrophilicity of the carbonyl carbon ( Figure 14). CeO 2 , on the other hand, is a redox catalyst and is capable of providing oxygen vacancies to aid the catalytic reaction process. The oxygen in glycerol attacks the carbonyl carbon, resulting in the formation of an intermediate species. The proton is transferred from the intermediate to the second hydroxyl group of glycerol, leading to the formation of an activated complex and a water molecule. Monoacetin (MAT) is formed subsequently. The aforementioned mechanism is augmented by the interaction of MAT with ethanoic acid, which forms diacetin (DAT) and triacetin (TAT). Besides increasing the glycerol conversion, the addition of both NiO and CeO 2 also accelerates and increases the selectivity of the glycerol acetylation reaction. The mechanism combines acid-base and redox chemistry, which is facilitated by the taskspecific properties of the incorporated NiO and CeO 2 support.

Conclusions
In summary, NiO incorporated onto CeO2-supported solids (xNiO/CeO2) was prepared, and these solids were used as highly selective catalysts for glycerol acetylation. The XRD, SEM, and TEM data confirm that the NiO nanoparticles are well dispersed on the CeO2 support, and the chemical interactions between the NiO nanoparticles and CeO2 support are proven by the DRS UV-Vis spectroscopy. Furthermore, enhanced conversion of glycerol to MAT, DAT, and TAT products with ethanoic acid was successfully per-

Conclusions
In summary, NiO incorporated onto CeO 2 -supported solids (xNiO/CeO 2 ) was prepared, and these solids were used as highly selective catalysts for glycerol acetylation. The XRD, SEM, and TEM data confirm that the NiO nanoparticles are well dispersed on the CeO 2 support, and the chemical interactions between the NiO nanoparticles and CeO 2 support are proven by the DRS UV-Vis spectroscopy. Furthermore, enhanced conversion of glycerol to MAT, DAT, and TAT products with ethanoic acid was successfully performed with all the prepared catalysts (CeO 2 , 9NiO/CeO 2 , 27NiO/CeO 2 , and 45NiO/CeO 2 ). 27NiO/CeO 2 appears to be the best catalyst in catalyzing the glycerol acetylation, with very high efficiency (97.6% glycerol conversion; 70.5% selectivity towards TAT) under optimum conditions, where the formation of by-products can be suppressed thanks to its strong acidity and high accessible surface area. Thus, this study demonstrates that 27NiO/CeO 2 is very promising for catalyzing glycerol within a short reaction time (<30 min), thanks to the instant heating mode of the Monowave 50 reactor. Furthermore, the catalyst is reusable (up to 10 times) with minimal effect on the reaction performance. We thus believe that 27NiO/CeO 2 will be a very promising heterogeneous catalyst and has high potential for synthesizing renewable biofuels in the petrochemical industry.