Effect of Calcium Doping Using Aqueous Phase Reforming of Glycerol over Sonochemically Synthesized Nickel-Based Supported ZrO₂ Catalyst

Abstract

The aqueous phase reforming (APR) of glycerol was studied using sonochemically synthesized 10%Ni-x%Ca/ZrO₂ catalysts (where x = 0, 0.5, 3, and 5) for the production of value-added liquid products. The APR reaction was performed in a batch reactor under the following conditions: 20 bar, 230 °C 450 rpm, and 1 h of reaction time. The synthesized catalysts were characterized using XRD, FESEM, BET, and H₂-TPR to observe the effect of Ca doping on the physio-chemical properties of the catalysts. The results revealed that, at higher Ca loading, the catalysts experienced serious particles’ agglomeration, which resulted in a larger particles’ size, smaller surface area, and smaller pore volume owing to uneven distribution of the particles. The characterization results of the catalysts confirmed that the Us catalysts have a slightly higher surface area, pore volume, and pore size, as well as highly reducible and fine crystalline structure, compared with WI catalysts. The catalytic performance of the catalysts shows that 1,3-propanediol (1,3-PDO) and 1,2-propanediol (1,2-PDO) were the two main liquid products produced from this reaction. The highest selectivity of 1,3-PDO (23.84%) was obtained over the 10%Ni/ZrO₂ catalyst, while the highest selectivity of 1,2-PDO (25.87%) was obtained over the 10%Ni-5%Ca/ZrO₂ catalyst.

1. Introduction

The utilization of fossil fuels as the main energy source has led to various environmental problems such as global warming, climate change, and air pollution. This might be caused by the release of greenhouse gases (carbon dioxide, methane, carbon monoxide, and others) into the environment during the burning of fossil fuels. Yumashev, et al. [1] agreed that human development such as urbanization growth, business in total energy consumption, and the level of socio-economic development has influenced fossil fuel consumption. Therefore, to tackle these problems, biodiesel was identified as an alternative new energy source to replace the current fossil fuels. Biodiesel is mainly produced from renewable resources such as vegetable oils, yellow grease, used cooking oils, or animal fats through the transesterification process [2]. During the last decade, the demand for biodiesel in the market has been continuously growing and it is expected that, by the year 2027, global biodiesel production will reach up to 4 billion liters/year [3]. Glycerol is produced as the major byproduct during biodiesel production, with approximately 10 kg of glycerol being produced for every 100 kg of biodiesel produced [4]. However, the abundant supply of this glycerol in the market has led to a sharp decrement in glycerol prices to $0.04 from $0.66/lb [5].

Thus, researchers are now working on the conversion of this glycerol into other valuable liquid products through various catalytic reactions such as fermentation [6], hydrogenation [7], and reforming [8,9]. In all such cases, heterogeneous catalysts have been developed for this application owing to several advantages such as recyclability, high thermal stability, and easy separation from the product [10]. In 2002, Dumesic and co-workers were the first groups to discover the production of hydrogen gas (H₂) and value-added chemicals via APR reaction utilizing oxygenated hydrocarbons. APR reaction operates at a relatively low temperature (210 to 250 °C) and pressure (1.5 to 5.0 MPa) compared with other reforming processes (steam reforming and autothermal reforming) [5]. Under these conditions, this APR reaction helps to minimize the undesirable decomposition reaction, which makes this process thermodynamically favorable. Besides that, this APR reaction can also be an alternative process to avoid the utilization of an external hydrogen source, as hydrogen is produced in situ during the reaction via WGS reaction [11,12].

Various metal-based catalysts such as Pt, Pd, and Cu have been studied for this APR reaction and, among them, the Pt-based catalyst is the most common metal catalyst used for this application thanks to its promising capabilities during the reaction [13]. Reynoso, et al. [14] proved that Pt/CoAl₂O₄ catalysts are highly active during the APR reaction, with glycerol conversion of 99%, and very stable over 100 h time on stream (TOS). Recently, researchers are moving to a cheaper metal-based catalyst, mainly Ni, for the APR reaction, in order to make this process more convenient and economically friendly for application in industry. Zhang and Xu [15] in their study proved that Ni-Al catalysts are highly selective toward the production of the desired product and are highly stable throughout the APR reaction. The addition of a promoter, mainly calcium metal, has recently captured interest, as it has been proven that this promoter enhances the performance of the catalysts during the activity [3]. The use of zirconia oxide (ZrO₂) support has been reported recently for APR reaction, and it has been reported that the Ni catalyst supported on ZrO₂ significantly improved the yield and selectivity of H₂ compared with the other metal oxide support [16].

The catalyst synthesis method plays an important role in improving the physicochemical properties and activity of the catalyst. Many studies in the literature reported on the synthesis of catalysts using a conventional method such as wet impregnation and co-impregnation [17,18]. Although this conventional method involved only a simple mixing and stirring process, this process may lead to a serious particle agglomeration owing to uneven particle distribution [19]. Therefore, to overcome this problem, catalysts synthesized by the sonochemical method have been introduced. This method involved the introduction of high intensity ultrasonic waves, which has been proven to improve the physicochemical properties of the catalysts [20,21]. Shahbudin, et al. [22] investigated the effect of introducing ultrasound irradiation on the physicochemical properties and catalytic activity of Mo-Ni/Al₂O₃ and Mo-Ni/H-Y zeolite for APR of glycerol and sorbitol for value-added chemicals’ production. They reported that ultrasound irradiation helped to improve the textural properties of the catalysts, including particle dispersion and surface area, as well as the catalytic performance of the catalysts during APR reaction.

Therefore, in the present study, the sonochemical method was selected as the method for catalysts’ synthesis in order to target high selectivity of value-added products as well as to reduce particles’ agglomeration, which can lead to catalysts’ deactivation. To the best of our knowledge, the APR of glycerol using sonochemically synthesized Ni-Ca-based catalysts supported on ZrO₂ has not been reported previously, which is the novelty of the present work. In the present study, the APR reaction was investigated using a series of sonochemically synthesized 10%Ni-x%Ca/ZrO₂ catalysts (where x = 0, 0.5, 3, and 5) with emphasis on the effect of Ca loading on the physicochemical properties of the catalysts and the distribution of the liquid products. The physicochemical properties and performance of these sonochemically synthesized catalysts were then compared to wet impregnation catalysts to observe the effect of ultrasound irradiation. The synthesized catalysts were characterized using XRD, FESEM coupled with EDX and dot mapping, BET, and H₂-TPR analysis. The performance of the catalysts during the reaction is investigated based on the selectivity of the liquid products and glycerol conversion.

2. Results and Discussion

2.1. Structural and Morphological Analysis

Figure 1 presents the XRD patterns of ZrO₂, 10%Ni/ZrO₂_WI, 10%Ni-0.5%Ca/ZrO₂_WI, 10%Ni-3%Ca/ZrO₂_WI, 10%Ni-5%Ca/ZrO₂_WI, 10%Ni/ZrO₂_Us, 10%Ni-0.5%Ca/ZrO₂_Us, 10%Ni-3%Ca/ZrO₂_Us, and 10%Ni-5%Ca/ZrO₂_Us catalysts. The detailed analysis of the XRD patterns shows that the diffraction peaks of ZrO₂ are observed at 2θ = 17.54°, 24.16°, 24.55°, 28.28°,31.57°, 34.49°, 35.99°, 38.65°, 41.46°, 44.92°, 45.60°, 49.34°, 50.64°, 51.40°, 54.33°, 55.96°, 57.40°, 58.48°, 59.86°, 60.29°, 61.43°, 62.22°, 63.08°, 64.34°, 65.44°, 68.98°, and 69.16° (JCPDS NO. No. 65-1022), which represents the monoclinic structure of ZrO₂ [23]. Nevertheless, the diffraction peaks of NiO are observed at 2θ = 37.49° and 43.1° (JCPDS No. 47-1049) [24,25]. Meanwhile, the CaO diffraction peaks are observed at 2θ = 37.4° and 44.4° (JCPDS NO. No. 82-1691), indicating the cubic structure of CaO [3]. However, the diffraction peaks of CaO are not clearly observed owing to the low content of Ca and overlapping of the CaO peaks with the NiO peaks for both US and WI methods, as agreed upon in the literature [3,26]. From the XRD patterns of all the catalysts, it is observed that the peak intensity of the metal oxides increased with the increase in metal loading. The increase in peaks’ intensity indicates the increase in the catalysts’ crystallinity [27].

For bimetallic Ni- and Ca-supported catalysts, clear and sharp peaks of NiO and CaO are observed for both sonochemical and wet impregnation catalysts. This proved the binary system of NiO-CaO presence in a separate metal oxide form rather than in a mixed oxide form [3]. Comparing the XRD spectra of Us and WI catalysts, it is observed that the sonochemically synthesized catalysts possess a lower peak intensity than wet impregnation catalysts. The low peak intensity indicates that the metal oxides are homogeneously dispersed on the support surface, and thus form a smaller crystallite size and prevent particles’ agglomeration [28]. Ahmadi, Haghighi and Ajamein [20] also reported that catalysts synthesized by the ultrasound-assisted method have slightly lower peak intensity, which indicates that the catalyst is less crystalline owing to the homogenous nucleation of the particles (Figure 1).

The specific surface area (SSA), pore volume, and pore size of WI and Us catalysts are reported in

Table 1. Specific surface area, pore volume, and pore size of the synthesized catalysts.

Catalyst	Specific Surface Area (m2/g)	Pore Volume (cm3/g)	Pore Size (nm)
ZrO2	24.00	0.42	5.08
10%Ni/ZrO2_Us	25.50	0.48	6.85
10%Ni/ZrO2_WI	15.50	0.39	4.58
10%Ni-0.5%Ca/ZrO2_Us	24.30	0.34	5.77
10%Ni-0.5%Ca/ZrO2_WI	14.40	0.31	3.64
10%Ni-3%Ca/ZrO2_Us	23.90	0.26	4.04
10%Ni-3%Ca/ZrO2_WI	12.50	0.24	2.21
10%Ni-5%Ca/ZrO2_Us	22.40	0.21	3.46
10%Ni-5%Ca/ZrO2_WI	11.70	0.20	1.01

. The SSA, pore volume, and pore size of ZrO₂ support were 24.00 m²/g, 0.42 cm³/g, and 5.08 nm, respectively. From the results, it is observed that the SSA of both WI catalysts Us catalysts decreased with an increase in metal loading. For Us catalysts, it is observed that the addition of Ni caused a slight increase in the specific surface area, pore volume, and pore size of the catalyst to 25.5 m²/g, 0.48 cm³/g, and 6.85 nm, respectively. Further, it increased in the amount of Ca doping from 0.5 to 5 wt.%, and decreased the SSA, pore volume, and pore size of the catalysts from 24.3 to 22.4 m²/g, 0.34 to 0.21 cm³/g, and 5.77 to 3.46 nm, respectively. However, for WI catalysts, the addition of Ni on the ZrO₂ support decreased the SSA, pore volume, and pore size to 15.5 m²/g, −0.392 cm³/g, and 4.58 nm, respectively. Further, it increased in the amount of Ca doping from 0 to 5 wt.%, and decreased the SSA, pore volume, and pore size from 14.4 to 11.7 m²/g, 0.305 to 0.204 cm³/g, and 3.64 to 1.01 nm, respectively.

The decrease in surface area with an increase in metal loading might be caused by the particle agglomeration, which resulted in a larger particle size. Meanwhile, the decrease in pore volume and pore size with an increase in metal loading might be caused by the blockage of the pores of the support by the metal/s [29]. These findings agree with those of Numpilai, et al. [30], who also reported a similar trend of decrease in specific surface area, pore volume, and pore size with an increase in metal loading due to particles agglomeration. Comparatively, it was observed that Us catalysts have slightly larger specific surface area compared with WI catalysts. This might be caused by the prominent effect of ultrasound irradiation, which caused a more homogeneous particle distribution, leading to smaller particle size. Ahmadi, Haghighi and Ajamein [20] also reported that high ultrasonic power enhances the nucleation rate, resulting in highly dispersed particles and a greater surface area of catalysts. Furthermore, it is also observed that Us catalysts have a slightly larger pore size and pore volume compared with WI catalysts, a phenomena that leads to a more homogeneous particle distribution [31].

However, the lower specific surface area, pore volume, and pore size of WI methods show that the WI catalysts experienced particles’ agglomeration. This result is supported with XRD and FESEM analysis, as discussed in the previous section (Table 1). The FESEM images, EDX spectra, and dot mapping analysis were performed to observe the morphology of the catalysts synthesized by Us and WI methods, as shown in Figure 2 and Figure 3, respectively. From the FESEM images, it is observed that the ZrO₂ support exists in the nano regime with spherical-shaped particles, and in some parts, the particles are highly agglomerated, which is evident by the EDX spectra and dot-mapping images. The addition of Ni on the supports slightly reduced the particle size for both Us and WI catalysts. Moreover, it is also observed that the addition of Ca as a promoter on the Ni-based catalysts has a significant impact on the particle size of the catalysts. It is observed that the increase in Ca loading resulted in larger particle size for both WI and US catalysts. This might be caused by the particles’ agglomeration at higher metal composition owing to uneven particles’ distribution, which is evident by the EDX spectra and dot-mapping images of the catalysts [32,33].

Comparatively, FESEM images of all the catalysts clearly show that sonochemically synthesized catalysts have slightly smaller particles size with homogenous particle distribution, as confirmed by the EDX spectra and dot mapping images. Meanwhile, FESEM images of wet impregnation catalysts show that the catalysts have slightly bigger particle size owing to particles’ agglomeration, which is evident by the EDX spectra and dot mapping images. Particle agglomeration is the major cause of catalysts’ deactivation, which can reduce the activity of the catalysts. Similarly, the decreased in particle agglomeration of catalysts synthesized by the Us method is mainly caused by the introduction of the high-intensity sound wave, which enhanced the nucleation rate of the catalysts and led to homogeneous particles’ distribution, as reported in the literature [20] (Figure 2 and Figure 3). This finding aligns with the XRD results, which confirmed that the Us catalysts have a smaller crystallite size with good metallic crystallites’ dispersion over all the supports compared with WI catalysts (Figure 1).

The reducibility of the catalysts was determined using H₂-TPR analysis, as shown in Figure 4. From the H₂-TPR spectra, it was observed that all the catalysts possess one broad peak in the range of 300–550 °C, which represents the reduction of NiO with weak interaction with the support [32]. The addition of Ca as a promoter significantly affects the catalyst reducibility, with shifting of the peaks to a higher temperature. For Us catalysts, it is observed that an increase in the amount of Ca doping from 0 to 5 wt.% increases the reduction temperature of the catalysts from 394 to 421 °C. Meanwhile, a similar trend is also observed in WI catalysts. The increase in the amount of Ca doping from 0 to 5 wt.% caused a slight increase in the reduction temperature from 399 to 471 °C. The increase in the reduction temperature at higher metal composition might be caused by the coverage of the Ni surface by Ca species, hindering the reduction of the catalysts [34]. Comparatively, Us catalysts have a lower reduction temperatures than the WI catalysts. This shows that the ultrasound irradiation helps inhomogeneous particle distribution, which improved the reducibility of the catalysts.

2.2. Performance of Catalysts for Glycerol APR

Table 2. Catalytic performance of various catalysts for glycerol APR.

Catalyst	Cglycerol (%)	Selectivity (%)								C Balance (%)
		1,3-PDO	1,2-PDO	EG	Propanol	Acetone	Ethanol	Methanol	Others Unidentified Compounds
10%Ni/ZrO2_Us	50.8 ± 0.2	23.8 ± 2.3	17.2 ± 1.3	10.5 ± 2.1	5.41 ± 2.2	7.32 ± 1.0	5.21 ± 1.3	6.43 ± 0.9	18.1 ± 0.8	94 ± 1.3
10%Ni/ZrO2_WI	49.5 ± 1.3	12.3 ± 1.3	10.2 ± 2.2	6.78 ± 1.2	3.54 ± 3.4	5.67 ± 0.7	4.93 ± 1.8	5.78 ± 1.2	38.8 ± 1.2	88 ± 4.0
10%Ni-0.5%Ca/ZrO2_Us	49.5 ± 0.9	19.8 ± 1.0	21.9 ± 1.0	12.2 ± 0.8	9.81 ± 1.2	8.82 ± 0.7	7.45 ± 2.3	8.67 ± 2.5	3.35 ± 3.5	92 ± 3.3
10%Ni-0.5%Ca/ZrO2_WI	48.2 ± 3.4	10.9 ± 2.5	12.3 ± 1.1	8.76 ± 1.7	5.21 ± 1.0	7.89 ± 3.0	6.98 ± 2.4	7.84 ± 3.2	30.1 ± 5.4	90 ± 1.2
10%Ni-3%Ca/ZrO2_Us	49.3 ± 1.3	15.8 ± 4.7	24.8 ± 3.4	14.2 ± 3.2	10.1 ± 1.2	9.43 ± 1.7	9.72 ± 3.4	9.67 ± 1.4	1.28 ± 2.9	95 ± 1.0
10%Ni-3%Ca/ZrO2_WI	42.7 ± 4.5	9.98 ± 3.4	14.2 ± 2.9	9.23 ± 2.4	7.89 ± 3.4	9.21 ± 2.3	8.98 ± 1.2	8.43 ± 3.2	16.1 ± 3.4	84 ± 1.2
10%Ni-5%Ca/ZrO2_Us	48.2 ± 0.6	14.2 ± 2.3	24.9 ± 2.8	14. 7 ± 1.7	12.3 ± 2.3	10.0 ± 3.2	9.91 ± 0.8	9.76 ± 4.1	1.23 ± 3.7	97 ± 3.4
10%Ni-5%Ca/ZrO2_WI	40.2 ± 3.4	7.89 ± 3.2	15.7 ± 4.0	11.4 ± 1.5	10.5 ± 1.3	9.98 ± 4.2	9.34 ± 0.7	8.98 ± 3.2	11.2 ± 1.0	85 ± 1.2

Reaction conditions: T = 230 °C, p = 20 bar, 10 vol.% glycerol in water, 450 rpm, 2 g catalyst, and 1 h reaction time. C_glycerol = glycerol conversion; 1,3-PDO = 1,3-propanediol; 1,2-PDO = 1,2-propanediol; EG = ethylene glycol.

shows the selectivity of liquid products produced via APR of glycerol, glycerol conversion, and carbon balance. The dataset present here is a replicate of three data points with a percentage error maintained below than 5%. The liquid product produced was identified as 1,3-propanediol (1,3-PDO), 1,2-propanediol (1,2-PDO, propanol, acetone, and others (unidentified peaks). Remón, et al. [35] summarized the three possible pathways during the APR of glycerol with the presence of suitable catalysts, as shown in Figure 5. In this research, 1.3-PDO and/or 1,2-PDO were identified as one of the main products produced via this APR of glycerol. However, the production of these two main products is a great challenge, as suitable catalysts are necessary to ensure that the reaction proceeds via the desired pathway. The production of 1,3-PDO via this APR reaction proceeds via Pathway 3, which involves dehydration of glycerol to 3-hydroxypropanal (3-HPA) and then hydrogenolysis of 3-HPA to 1,3-PDO in the presence of a Bronsted acid catalyst. 1,3-PDO can then be converted to propanol, which was also identified as one of the products produced via this reaction. Meanwhile, the production of 1,2-PDO (Pathway 2) proceeds via dehydration of glycerol to hydroxy acetone and then hydrogenolysis of hydroxy acetone to 1,2-PDO.

1,2-PDO can then be further converted into acetone or propanol, which were also identified among the products produced via this reaction, whereas ethylene glycol (EG) was produced only as a minor product via dehydrogenation of glycerol to 2,3-dihydroxypropanal and decarbonylation of 2,3-dihydroxypropanal to EG. From the results shown in Table 2, it is observed that 1,3-PDO and 1,2-PDO are the major liquid products produced during APR of glycerol over Ni-Ca/ZrO₂ catalysts synthesized by the Us and WI methods. The 10%Ni/ZrO₂ catalyst synthesized by both the Us and WI methods shows relatively high selectivity towards the production of 1,3-PDO. This might be caused by the acidic properties of ZrO₂ support that drove the reaction towards Pathway 3 (Figure 5) [29].

A significant effect on the selectivity of 1,3-PDO and 1,2-PDO is observed when Ca metals are doped on the Ni-based catalysts. For both Us and WI catalysts, it is observed that the increase in Ca doping from 0.5 to 5 wt.% slightly increases the 1,2-PDO selectivity from 21.9% to 25.9% and 12.3% to 15.7%, respectively. Meanwhile, the 1,3-PDO selectivity shows a decrement from 19.8% to 14.2% and 10.9% to 7.89% with an increase in Ca loading from 0.5 to 5 wt.% for Us and WI catalysts, respectively. Therefore, this shows that the addition of Ca on the Ni-based catalysts has relatively high selectivity towards the production of 1,2-PDO (Pathway 2). This might be caused by the basic properties of Ca metals, which reduced the acidity of the catalysts, and thus led the reaction via Pathway 2 [36]. Meanwhile, other liquid products such as EG, propanol, acetol, ethanol, and methanol were only produced as minor products in this APR of glycerol, with selectivity in the range of 12.31% to 2.21% (Table 2).

In the comparison of the performance of Us and WI catalysts, it is observed that the Us catalysts give a relatively high selectivity of the products produced via APR reaction compared with WI catalysts. Therefore, it can be concluded that the Us catalyst is highly active during the reaction compared with WI catalysts. This might be caused by the effects of ultrasound irradiation, which promotes homogeneous particles’ distribution, resulting in a larger surface area and more active sites for the reaction [25]. For the glycerol conversion, it is observed that all the catalysts have approximately 48 to 51% glycerol conversion, which shows that all the catalysts are slightly active during the reaction. This result is in line with the reported data for this APR of glycerol [28]. Reaction parameters such as temperature, pressure, glycerol concentration, and reaction times play a very important role to ensure high conversion of glycerol to liquid or gas products [35]. Besides that, it is also observed that the glycerol conversion decreased with an increase in metal loading for both the Us and WI methods (Table 2). This might be caused by the decrease in the catalyst activity due to particles’ agglomeration, which resulted in a smaller surface area and active sites for the reaction to occur [37].

3. Materials and Methods

3.1. Materials

Nickel nitrate hexahydrate (Ni(NO₃)₂·6H₂O (98.5%), calcium chloride dihydrate (CaCl₂·2H₂O) (99.5%), zirconium dioxide (ZrO₂) powder (Code: 230693, particle size of 5 µm, 99% trace metal basis), and glycerol (99%) were all purchased from Merck Company, Bandar Sunway, Malaysia, and were of analytical grade. All the chemicals used for analysis purposes such as 1,3-propanediol (99.99%), 1,2-propanediol (99.99%), ethylene glycol (99.98%), propanol (99.99%), and acetone (99.99%) were also purchased from Merck Company, Bandar Sunway, Malaysia. Purified hydrogen (99.99%) and nitrogen (99.99%) gases were supplied by Linde Malaysia (Bayan Lepas, PNG, Malaysia).

3.2. Synthesis of Catalyst

The 10%Ni-x%Ca/ZrO₂ catalysts (where x = 0, 0.5, 3, and 5) were synthesized via sonochemical (Us) and wet impregnation methods [25]. The 10 wt.% of the NiO is prepared by dissolving approximately 4.955 g of Ni(NO₃)₂·6H₂O salt in deionized water. Meanwhile, 0.5, 3, and 5 wt.% CaO are prepared by dissolving 0.183, 1.101, and 1.843 g of CaCl₂·2H₂O salts in deionized water, respectively. The amount of ZrO₂ support used to synthesize 10%Ni/ZrO₂, 10%Ni-0.5%Ca/ZrO₂, 10%Ni-3%Ca/ZrO₂, and 10%Ni-5%Ca/ZrO₂ catalyst is 9, 8.95, 8.7, and 8.5 g, respectively. Then, the aqueous solutions of nitrate and calcium salts were poured into a 500 mL beaker (diameter, D = 8 cm, height = 12 cm) containing ZrO₂ support. The resultant solution was then topped up with deionized water until it reached the maximum volume of 250 mL. For catalysts synthesized by the WI method, the resultant solution was stirred at room temperature for 4 h and 350 rpm using a hot plate. Meanwhile, for catalysts synthesized by the Us method, the resultant solution was then sonicated using an ultrasound sonicator (Model: Q700 Sonica; Fisher Scientific, Shah Alam, Malaysia) at 90 W for 45 min with 30 s pulse ON and 5 s OFF. The samples were then dried using a vacuum oven at 110 °C for 24 h and calcined using a furnace (model: Thermo Scientific FD1535m), with a dimension of 18 × 11 × 16.5 at 500 °C for 4 h in a static air environment.

3.3. Characterization of Catalyst

X-ray diffraction (XRD) analysis was performed using X’ Pert 3 Powder & Empyrean, PANalytical diffractometer (Mavern Panalytical, Petaling Jaya, Malaysia) equipped with a Cu Kα radiation source (λ = 1.5406Ӑ) operating at 45 kV and 40 mA [28]. The diffraction data were collected in the 2θ range of 20°–70° in continuous mode. Brunauer–Emmett–Teller (BET) surface area of the catalysts was determined from nitrogen adsorption–desorption isotherm performed at −196 °C using Micromeritics ASAP 2020 (Micromeritics, Petaling Jaya, Malaysia). The catalyst samples were first degassed at 200 °C for 4 h in a vacuum before the analysis [16]. Field emission scanning electron microscope (FESEM) was performed using VPFESEM, Zeiss Supra 55VP (Zeiss, Petaling Jaya, Malaysia), operating at 5 kV. Before the analysis, the catalysts were coated with a thin film of gold on their surface. The topographical and elemental information of the catalysts was analyzed using energy-dispersive X-ray (EDX) spectroscopy (Shah Alam, Malaysia) [28]. Hydrogen temperature programmed reduction (H₂-TPR) was performed using TPDRO, Model:1100, Thermo Scientific (Thermo Scientific, Shah Alam, Malaysia), equipped with thermal conductivity detector (TCD) to measure the H₂ consumption during the reduction. The catalysts were first degassed in a flow of nitrogen (N₂) gas at 300 °C for 1.5 h to remove the moisture and cooled to room temperature. Then, the catalysts were heated from room temperature until 950 °C at the rate of 10 °C/min in the presence of 5% H₂ balanced with N₂ with a holding time of 10 min [28].

3.4. Catalyst Performance

Aqueous phase reforming of glycerol was carried out in a PREMEX Autoclave High-Pressure Batch reactor, equipped with a 1000 mL reactor vessel (D = 15 cm, height = 20 cm), magnetic stirrer (type = streamline design, height = 15.5 cm, revolution rate = 450 rpm), pressure gauge, valve, and others (Premex Solution, Kuala Lumpur, KUL, Malaysia). Firstly, 2 g of the powder catalyst was loaded into the reactor and the catalyst was reduced under purified hydrogen at 4 bar, and its reduction temperature was determined from H₂-TPR analysis for 1 h. Then, 10 vol% glycerol solution was prepared by mixing 25 mL of pure glycerol with 225 mL distilled water. After the reactor had cooled down to 100 °C, 250 mL of 10 vol% glycerol solution was poured into the reactor. The reaction was conducted at 230 °C, 20 bar of purified N₂, and 450 rpm for 1 h. The presence of in situ hydrogen was confirmed by GC-TCD over the catalyst to proceed with further investigation into APR. The liquid product was analyzed using high-performance liquid chromatography (HPLC) equipped with a refractive index (RI) detector (Agilent Technologies, Bayan Lepas, PNG, Malaysia). The analysis was performed using Eclipse XDB C18 column (5 μm, 46 × 150 mm) with a total runtime of 35 min and sulfuric acid with a concentration of 0.005 M as mobile phase injected at a flow rate of 0.6 mL/min into the system. The catalysts’ performance was evaluated based on the selectivity of the liquid products, glycerol conversion, and carbon balance, as shown in Equations (1) to (3), respectively [22].

$$\mathrm{Selectivity} \mathrm{of} \mathrm{product} (\%)=\frac{\mathrm{Mole} \mathrm{of} \mathrm{product} \left(i\right)}{\mathrm{Total} \mathrm{mole} \mathrm{of} \mathrm{glycerol} \mathrm{consumed}}\times 100$$

(1)

$$\mathrm{Conversion} \mathrm{of} \mathrm{glycerol} (\%)=\frac{\mathrm{Mole} \mathrm{of} \mathrm{glycerol} \mathrm{consumed}}{\mathrm{Mole} \mathrm{of} \mathrm{glycerol} \mathrm{in} }\times 100$$

(2)

$$\mathrm{C} \mathrm{balance}=100-\left[{\left(\mathrm{mole} \mathrm{of} \mathrm{feed}\times \mathrm{n}\right)}_{\mathrm{in}}- \Sigma \frac{{\left({\mathrm{Mole} \mathrm{C}}_{\mathrm{i}}\times \mathrm{n}\right)}_{\mathrm{out}}}{{(\mathrm{Mole} \mathrm{of} \mathrm{feed} \times \mathrm{n})}_{\mathrm{in}}}\times 100\right]$$

(3)

4. Conclusions

The sonochemically synthesized Ni/ZrO₂ catalysts doped with different Ca loading have become a new development in the APR of glycerol for the production of value-added chemicals. The present study showed that the doping of Ca on the Ni/ZrO₂ catalysts significantly affects the physicochemical properties of the catalysts and the distribution of the liquid products. It is observed that the increase in Ca composition increases the agglomeration of the particles, which resulted in larger particles sizes and a smaller surface area. Besides that, the reducibility of the catalysts also decreased with an increase in metal loading owing to the strong interaction between metals and support, which required more H₂ consumption to reduce the metal oxide. For the effect of ultrasound irradiation on the physicochemical properties of the catalysts, it is observed that the particles of Us catalysts are homogeneously distributed, resulting in larger SSA, pore volume, and pore sizes compared with WI catalysts. Meanwhile, for the distribution of the liquid products, 1,3-PDO and 1,2-PRO were the two main liquid products produced via this APR of glycerol. The highest 1,3-PDO selectivity of 23.84% is obtained over the 10%Ni/ZrO₂ catalyst, while the highest selectivity of 1,2-PDO of 25.87% is obtained over the 10%Ni-5%Ca/ZrO₂ catalyst. The other products such as propanol, acetone, methanol, ethylene glycol, ethanol, and others (unidentified compounds) are only found as minor products during this APR of glycerol, with selectivity below 15%. All the catalysts were also proven to have a good catalytic performance during the APR reaction, with glycerol conversion above 40%.