A study of the precipitation of cerium oxide synthesized from rare earth sources used as the catalyst for biodiesel production

: This work aimed to study the preparation of cerium oxide ( CeO 2 ) used as the catalyst for biodiesel production from palm oil. The precipitation method was used in the catalyst synthesis. The e ﬀ ects of oxalic con centrations and stirring rates in the precipitation process were investigated. Oxalic acid was added into cerium ( Ce ) in ethylenediaminetetraacetic acid solution to form Ce oxalate before the Ce oxalate was calcined to obtain CeO 2 . The results showed that oxalic concentrations and stirring rates slightly a ﬀ ect the morphology of CeO 2 . However, these parameters considerably a ﬀ ect the amount of basic sites of CeO 2 . The basicity of CeO 2 plays the main role in catalyzing the transesteri ﬁ cation reaction for bio diesel production. When CeO 2 was used as the catalyst in biodiesel production from palm oil under operating condi tions using a 5% catalyst, methanol - to - oil molar ratio of 30:1, reaction temperature of 150°C, 13.8 bars, and 3 - h reac tion time, CeO 2 obtained from 3% oxalic concentration and 400 rpm stirring rates in the precipitation process provided the highest %FAME in the range of 93.9 – 94.2% since it had higher basicity. In addition, the decrease in surface area of CeO 2 aftertheusewaslessseverethanthatofbasicitydueto catalyst deactivation.


Introduction
At the present, there is a potential for the problem of fossil energy shortages to occur in the future due to development in several sectors such as industries, transportation, and logistics. Furthermore, the use of fossil energy such as benzene and diesel can generate environmental problems, such as air pollution, global warming, and PM2.5 dust (particulate matter with the size of 2.5 µm or smaller) [1,2]. Therefore, it is necessary to find other energy sources which are renewable, sustainable, and environmental-friendly for environmental and energy security. There are several types of alternative energy, such as wind energy, solar energy, water energy, hydrogen energy, and biodiesel, which can be used instead of fossil energy. Biodiesel is one such alternative energy that has shown its suitability for use in agricultural countries such as Thailand. Biodiesel can be produced from various oil plants. For example, rice bran oil was proposed as a promising renewable source for biodiesel production as well as those from palm, jatropha, coconut, and soybean plants, which are abundant in nature [3][4][5][6][7][8]. However, some properties of biodiesel such as density and viscosity, which affect the fuel supply system and spray characteristics, need to be improved in order to be efficiently used in diesel engines [9]. The combustion of biodiesel does not emit sulfur dioxide, and biodiesel is also labeled as "carbon-neutral." The net amount of CO 2 emission from the combustion of biodiesel does not increase in the atmosphere since the oil plants consume CO 2 for photosynthesis and are used as raw materials for biodiesel production. Thus, biodiesel is green, eco-friendly, and sustainable energy that can be practically used.
Biodiesel can be synthesized using several methods. Biodiesel production on the industrial scale mostly uses NaOH or KOH as the catalyst. However, the use of NaOH and KOH, which are homogeneous catalysts, generates large amounts of wastewater, and the separation of catalysts for reuse was not easily carried out [10]. Therefore, heterogeneous catalysts are expected to solve these drawbacks from the use of homogeneous catalysts. Several heterogeneous catalysts such as MgO, CaO, and SrO were studied in biodiesel production [11][12][13]. However, the leaching problem was found when these catalysts mentioned earlier were used [14]. The reuse of these catalysts cannot be carried out to reduce the production cost. It might affect the purification process of biodiesel from catalyst leaching. Thus, it is necessary to study other catalysts which reduced the leaching problem. Therefore, cerium oxide (CeO 2 ) has gradually gained attention for use as the catalyst in transesterification reactions due to its good thermal stability [15]. This might reduce the leaching problem, and the catalyst can be reused several times.
CeO 2 was one type of heterogeneous catalyst which has been studied. CeO 2 has wide applications in various areas such as microelectronics, optoelectronics, fuel cell technologies, gas sensors, oxygen storage, ceramics, and biomedical applications [16,17]. As CeO 2 in nanoparticle form provides a good thermal property, large oxygen storage, and flexible capacity in valence transformation, it can be used as an additive that is added to biodiesel. The CeO 2 -containing biodiesel helps diesel engines to reduce toxic emissions such as soot, smoke opacity, NO x , CO, and HC [18]. In addition, CeO 2 can be used as the catalyst in several reactions such as photocatalytic reactions, oxidation reactions, and transesterification reactions [19][20][21]. For transesterification reaction in biodiesel production, CeO 2 as a heterogeneous catalyst has been more studied since CeO 2 showed strong durability, resulting in several reuses. For transesterification reaction in biodiesel production, CeO 2 has been studied for use as the catalyst. CeO 2 contained basic sites which are active sites to catalyze the transesterification reaction [10]. However, most works have studied mixed oxide between other elements and cerium (Ce) [22][23][24][25]. In this work, CeO 2 in the form of oxide was used to catalyze the transesterification reaction.
Ce is an element in the lanthanide group. Ce compounds can be found in natural minerals such as alanite, bastanite, monazite, cerite, and samarskite. The main source of Ce is bastanite and monazite [26]. CeO 2 synthesis can be carried out using several methods such as precipitation, hydrothermal, solvothermal, sonochemical, spray pyrolysis, microemulsification, sol-gel, plantmediated, and fungus-mediated methods [27][28][29]. In this work, the precipitation method was used to synthesize CeO 2 since this method is simple, cheap, and easy to scale up [30]. There are several Ce sources used as the precursor in the synthesis such as Ce nitrate hexahydrate, Ce carbonate, Ce hydroxide, and Ce chloride [29][30][31][32]. Here, the Ce source was obtained from the Thailand Institute of Nuclear Technology. The Ce source was in the form of Ce in ethylenediaminetetraacetic acid solution (Ce in ethylenediaminetetraacetic acid [EDTA] solution), which was obtained and then purified using the ion exchange method [33]. The Ce source in this work was originally from natural minerals such as monazite ore. After decomposition and purification of monazite ore, a large amount of Ce in EDTA solution was obtained. Therefore, this Ce source can be suitably used to prepare adequate amounts of CeO 2 catalyst for largescale production of biodiesel. In addition, the precipitation method was used for CeO 2 preparation, which is simple, cheap, and easy to scale up. Therefore, the catalyst cost could be economical when CeO 2 is used as the catalyst for commercial biodiesel production.
Oxalic acid was used as the precipitant. Oxalic acid is one of the chemicals which is commonly used to separate rare earth elements from solutions since it has an affinity with rare earth elements [34]. After the addition of oxalic acid into Ce in EDTA solution, Ce oxalate was formed and then calcined to obtain CeO 2 . In this work, the effect of oxalic concentrations and stirring rates in the precipitation which affected the properties of CeO 2 used as the catalyst in transesterification reaction for biodiesel production were studied.

Materials
Ce in ethylenediaminetetraacetic acid solution (Ce in EDTA solution) was obtained from the Thailand Institute of Nuclear Technology. Oxalic acid (C 2 H 2 O 4 ·2H 2 O, 99.5%, AR/ACS) was from Loba Chemie PVT. Ltd., and methanol (CH 3 OH, 99.5%, AR) was from QREC company. Palm oil was purchased from a commercial company (Morakot brand, Thailand).

Catalyst preparation and its characterization
Ce in the form of dissolved Ce in EDTA (called, Ce in EDTA solution) was obtained from the purification process of mixed rare earth ores using ion exchange resin [33]. The concentration of Ce in EDTA solution was measured using ICP-OES. The main rare earth elements in the solution were praseodymium (Pr) and Ce. The concentrations of Pr and Ce in the solution were 36.49 and 524.6 mg L −1 , and the pH of Ce in the EDTA solution was 8. For the precipitation process, the Ce in EDTA solution was filtered using a filter paper (No. 1, Whatman). Then, filtered Ce in EDTA of 1,400 mL was added to a 2,000 mL beaker. Next, the 3% w/v oxalic solution was dropped into the beaker at the feed rate of 20 mL·min −1 under a stirring rate of 400 rpm, using a magnetic bar. The addition of oxalic solution was stopped when the final pH of the solution was 4. After that, the precipitates were obtained and the mixture was left at room temperature (28°C) for 16 h (aging time). The precipitates were then filtered using filter paper (No. 1, Whatman) with the help of a vacuum pump. The precipitates were dried in an oven at 110°C for 12 h. CeO 2 was obtained after the precipitates were calcined at 900°C for 3 h with a heat rate of 5°C·min −1 . The oxalic concentrations (0.5% and 3% w/v) and the stirring rates (200, 400, 600, and 800 rpm) for the precipitation process were studied. The characterization of CeO 2 was carried out, using Brunauer-Emmett-Teller (BET), temperature-programmed desorption of CO 2 (CO 2 -TPD), X-ray diffraction (XRD), and scanning electron microscopy (SEM) measurements. The surface area of the CeO 2 catalyst was determined using N 2 -adsorption in BET measurement (BELSORP-mini II, MicrotracBEL). The CO 2 -TPD (Autochem 2910, Micromeritics) was used to determine the basicity of CeO 2 . For the SEM measurement (Nira 3, TESCAN), CeO 2 was covered with gold film before it was placed in the chamber of the instrument using an acceleration voltage at 15 kV. The crystal structure of CeO 2 was measured using an XRD instrument (D8 Advance, Bruker), equipped with Cu Kα radiation at 40 kV and 40 mA. The XRD pattern was recorded using 2-theta in the range of 10-90°.

Biodiesel production
The palm oil of 113 mL and 5 g CeO 2 were added in a Parr stirred batch reactor (Model 4568) with the stirring rate of 600 rpm under an N 2 atmosphere at 13.8 bars. The mixture was heated to 150°C. Then, the methanol of 142.6 mL was fed into the reactor. The reaction was carried out for 3 h. After that, the mixture was separated to obtain biodiesel, using a centrifuge at 3,000 rpm for 30 min.
Biodiesel was baked at 110°C for 12 h to remove the residue of methanol. The %FAME was measured using GC-FID (Clarus 600, PerkinElmer), following the EN14103-2011 standard method. The catalyst was washed two times using methanol and was then measured using BET, CO 2 -TPD, XRD, and SEM techniques.

Results and discussion
Oxalic acid is one of the chemicals which is commonly used as the precipitant to separate rare earth elements from the solution. In general, oxalic acid has an affinity with rare earth elements. The precipitation of dissolved rare earth elements using oxalic acid can be performed for all pH ranges [34]. In this work, Ce in EDTA solution was added to an oxalic solution. The white precipitates (Ce oxalate) were formed immediately after the oxalic solution was added. The precipitates were filtered and then calcined. After the calcination, Ce oxalate was transformed into CeO 2 , which was pale yellow. For the precipitation process, the addition of oxalic solution was stopped when the final pH of the solution was 4. At pH = 4 of the solution, CeO 2 provided the larger basic sites than those of pH = 3 and 2, respectively (see Supplementary material). The basic sites of CeO 2 play the main role in catalyzing the transesterification reaction for biodiesel production [35][36][37]. As a result, CeO 2 obtained from pH = 4 offered more %FAME than those of pH = 3 and 2. Therefore, the final pH = 4 of the solution in the precipitation process was suitable and chosen for the precipitation in the next experiment.
Although the use of the final pH = 4 in the precipitation showed a higher %FAME than those of lower pH, a smaller amount of CeO 2 was obtained. In addition, a larger amount of Ce in EDTA solution was consumed during the precipitation using the final pH = 4, in order to obtain the same amount of CeO 2 from those of lower pH. It was expected that the use of higher pH than pH = 4 consumed a greater amount of Ce in EDTA solution and obtained less CeO 2 . Therefore, it might not be suitable to use higher pH than pH = 4 in the precipitation for use on a commercial scale. Therefore, the final pH = 4 of the solution in the precipitation was chosen to be used in the next experiment.

The effect of oxalic concentration
In this section, the concentrations of oxalic concentration used were 0.5% and 3% w/v. The XRD patterns of CeO 2 synthesized using 0.5% and 3% oxalic concentrations are shown in Figure 1. The peaks were observed at 28.5°, 33.0°, 47.5°, 56.5°, 59.1°, 69.5°, 76.9°, 79.2°, and 88.7°, assigned to crystal planes of (111), (200), (220), (311), (222), (400), (331), (420), and (422), respectively [38]. Although the use of 0.5% and 3% oxalic provided different Ce to oxalic ratios, XRD patterns of CeO 2 from both samples were similar. The precipitates of Ce oxalate were calcined at a high temperature of 900°C [39]. All amounts of Ce oxalate were completely transformed into CeO 2 . Therefore, the different XRD patterns from both samples were not clearly observed. From Figure 2a and b, SEM micrographs of CeO 2 synthesized using 0.5% and 3% oxalic concentration are illustrated. The CeO 2 particles were spherical and were attached via a rod. As listed in Table 1, surface area and pore size of CeO 2 synthesized using 0.5% and 3% oxalic concentrations were not much different. It was reported that the solubility of Ce oxalate decreased in the presence of higher oxalic concentrations [40]. Therefore, the use of 3% oxalic concentration decreased the solubility of Ce oxalate more than that of 0.5% oxalic concentration in a shorter time, affecting the precipitate formation. It could be said that the use of 3% oxalic concentration lowers the ratio of Ce to oxalic acid more than that of 0.5% oxalic concentration. Therefore, the use of 3% oxalic concentration offered low supersaturation, and there were few nuclei in the system. The remaining ions can improve the growth of nuclei. However, the use of 0.5% oxalic concentration might lead to the agglomeration of Ceoxalate due to the higher ratio of Ce to oxalic acid. As a result, CeO 2 synthesized using 0.5% oxalic concentration might have a little larger size and less surface area than that of 3% oxalic concentration. The concentration of oxalic acid had little effect on the morphology of CeO 2 [41].   ) was considerably more than that of CeO 2 synthesized using 0.5% oxalic concentration (517.99 µmol·CO 2 ·g cat −1 ), as shown in Table 1. The peaks of desorption of CO 2 between 600°C and 800°C as shown in Figure 3 indicate that basic sites of CeO 2 synthesized using 0.5% and 3% oxalic concentration provided strong basic sites [42]. As the final pH = 4 of the solution was controlled, the use of 3% oxalic concentration showed less ratio of Ce to oxalic acid than that of 0.5% oxalic concentration in a shorter time. This might affect the structure of Ce oxalate particles during the formation of precipitates, resulting in different amounts of basic sites of CeO 2 as the final product [43,44]. The greater amount of basic sites of CeO 2 from the use of 3% oxalic concentration in precipitation provided 94.2% FAME, which was more than that of 82.7% FAME obtained from 0.5% oxalic concentration, as shown in Table 1. Therefore, 3% oxalic concentration is the suitable concentration and so was chosen for the next experiment due to the higher basicity of CeO 2 . The basicity of the catalyst plays the main role in the catalysis of transesterification reaction. The high activity of catalysts in transesterification reaction was attributed to high numbers of basic sites and the basic strength [35][36][37]. Therefore, the high %FAME can be obtained using high basicity catalysts in the transesterification reaction.
The 94.2% FAME obtained from CeO 2 using 3% oxalic concentration in precipitation was slightly lower than the standard value (96.5%) [45]. In comparison with other works, mixed oxide CaO-CeO 2 used as the catalyst provided %FAME in the range of 95-98% under mild operating conditions [46][47][48]. One work by Soodjit et al. [21] used CeO 2 obtained from the precipitation method and using the same Ce source of this work. Their results showed that 88.92% FAME was obtained under more crucial operating conditions, compared to this work. However, the precipitation conditions were not similar to this work. This showed that the precipitation conditions should be controlled to improve the catalytic activity of CeO 2 for transesterification reaction. Therefore, the other parameters in the precipitation conditions should be further investigated in order to add more potential to CeO 2 for use as catalysts in large-scale biodiesel production.

The effect of stirring rates in precipitation
In this section, the effect of stirring rates was studied using 200, 400, 600, and 800 rpm in the precipitation with a 3% oxalic concentration. The XRD patterns of CeO 2 from different stirring rates are shown in Figure 4 and are assigned to the crystal planes of CeO 2 , as mentioned above [38]. The morphology of CeO 2 using different stirring rates looked similar for all samples, as shown in Figure 5. Table 2 shows the surface areas and pore sizes of CeO 2 obtained using different stirring rates. The surface area slightly increased at higher stirring rates. Stirring improves mass transfer and prevents local supersaturation [39]. In addition, the use of high stirring  , and the surface area of CeO 2 increased to 5.54 m 2 ·g cat −1 at 600 rpm. At higher stirring rates, the shear force could break up the particles, leading to smaller particles and higher surface area [39,49]. However, the surface area of CeO 2 decreased to 5.35 m 2 ·g cat −1 at 800 rpm, possibly because high stirring rates facilitate particle collision to form particle aggregate [50]. The sizes of CeO 2 particles from all stirring rates were not much different from those observed in Figure 5. However, CeO 2 particles at 800 rpm stirring rate agglomerated more tightly than those at lower stirring rates. This resulted in less surface area at the higher stirring rate of 800 rpm. The pore sizes of CeO 2 decreased as stirring rates increased. As shown in Table 2, the basicity of CeO 2 was the highest at the stirring rate of 400 rpm.     The CO 2 desorption of CeO 2 at 400 rpm is illustrated in Figure 6, which shows the desorption peaks of strong basicity between 600°C and 800°C. The stirring rates might affect the morphology of CeO 2 which was mentioned above, leading to the change in the basicity of CeO 2 . The high basicity of the catalyst provided a high activity to catalyze transesterification reaction. Therefore, CeO 2 at 200, 400, and 600 rpm showed %FAME, which was similar as they provided high basicity. The highest 93.9% FAME was obtained from CeO 2 at 400 rpm. As CeO 2 at 800 rpm had low basicity, the lowest %FAME was offered at 83.4%, as given in Table 2.

The change in properties of CeO 2 before and after reaction
In this section, CeO 2 , after use as the catalyst for transesterification reaction, was measured and compared to CeO 2 before the reaction. The XRD patterns and morphology of CeO 2 using 3% oxalic concentration and 400 rpm stirring rate are presented in Figures 7 and 8. It was found that the XRD peaks and the morphology of CeO 2 before and after the reaction looked similar. The difference was not clearly observed from XRD and SEM results. The XRD patterns and the morphology of CeO 2 from other samples are presented in Supplementary information. However, the surface area and basicity of CeO 2 decreased after the reaction, as shown in Table 3. The surface areas of most samples were reduced moderately. In addition, most samples considerately lost basicity after the reaction. These might be the result of the leaching of the catalyst and the effect of surface poisoning such as the adsorption of compounds on the active sites of the catalysts [12,48].

Conclusions
In conclusion, oxalic concentrations and stirring rates used in the precipitation slightly affect the morphology but considerately influence the basicity of CeO 2 . As the basicity of catalysts is the main factor in catalyzing transesterification reaction, the greatest amount of basic sites from CeO 2 obtained using 3% oxalic concentration and 400 rpm stirring rates provided the highest %FAME in the range of 93.9-94.2%. After the reaction, the change in surface area and basic sites of CeO 2 were observed. It was found that the decrease in surface area was less severe than that of the basicity of CeO 2 . Therefore, the precipitation conditions in CeO 2 preparation are one way to improve CeO 2 properties to be suitable for use as the catalyst in biodiesel production. Other parameters in the precipitation such as aging times, precipitation temperatures, and calcination temperatures, including the reuse of CeO 2 and reaction kinetics, should be further studied, in order to obtain biodiesel qualities within the standard values and be able to enlarge the biodiesel production for commercial use in the future.
Acknowledgment: SP would like to thank Silpakorn University Research, Innovation and Creative Fund for financial support and Thailand Institute of Nuclear Technology for supporting Ce in EDTA solution as the Ce source. Colin Liddle was also thanked for the English correction.
Funding information: Silpakorn University Research, Innovation and Creative Fund.