Effect of Calcination Temperatures on Surface Properties of Spinel ZnAl₂O₄ Prepared via the Polymeric Citrate Complex Method—Catalytic Performance in Glycerolysis of Urea

Abstract

In this study, we investigated urea glycerolysis over ZnAl₂O₄ catalysts that were prepared by using a citrate complex method and the influence of calcination temperatures on the surface properties of the prepared catalysts by varying the calcination temperature from 550 °C to 850 °C. As the reciprocal substitution between Al³⁺ and Zn²⁺ cations led to the formation of a disordered bulk ZnAl₂O₄ phase, different calcination temperatures strongly influenced the surface properties of the ZnAl₂O₄ catalysts, including oxygen vacancy. The increase in the calcination temperature from 550 °C to 650 °C decreased the inversion parameter of the ZnAl₂O₄ structure (from 0.365 to 0.222 for AlO₄ and 0.409 to 0.358 for ZnO₆). The disordered ZnAl₂O₄ structure led to a decrease in the surface acidity. The ZnAl₂O₄-550 catalyst had a large specific surface area, along with highly disordered surface sites, which increased surface acidity, resulting in a stronger interaction of the Zn NCO complex on its surface and an improvement in catalytic performance. Fourier transform infrared and thermogravimetric analysis results of the spent catalysts demonstrated the formation of a greater amount of a solid Zn NCO complex over ZnAl₂O₄-550 than ZnAl₂O₄-650. Consequently, the ZnAl₂O₄-550 catalyst outperformed the ZnAl₂O₄-650 catalyst in terms of glycerol conversion (72%), glycerol carbonate yield (33%), and byproduct formation.

1. Introduction

In recent years, climate change has become increasingly severe, which has led to the promotion of the development of the biodiesel industry [1,2]. Large amounts of crude glycerol as a byproduct generated from biodiesel production have an economic impact as a cheap and commercially viable renewable feedstock [3]. Among various glycerol reaction pathways, glycerolysis into glycerol carbonate (GC) is an interesting research direction. GC is an important glycerol derivative that is considered an interesting value-added chemical due to its excellent properties such as being a non-hazardous low-vapor pressure liquid, as well as its biodegradability, its low toxicity, its non-flammability, its high boiling point, and its low volatility [4,5,6]. Because of its outstanding properties, GC has a wide range of applications, including solvents, electrolytes, surfactants, wetting agents, and as a chemical intermediate in polymer synthesis [5,7].

Traditionally, the chemical synthesis of GC is by glycerol reacting with phosgene; however, this reaction is very dangerous and not environmentally friendly [8]. Alternatively, carbonate sources with milder reaction conditions such as dialkyl carbonate [9,10,11], alkylene carbonate [12], and CO₂ [13,14] can be used. With special advantages such as being cheap, easy to obtain, and recyclable, urea is used, along with glycerol, to efficiently produce GC [15,16,17,18]. In our previous studies, catalytic performance was affected by the existence of Zn-containing intermediates, namely, zinc diamine diisocyanate (Zn(NH₃)₂(NCO)₂, abbreviated as a Zn NCO complex). The formation of the solid Zn NCO complex on the ZnAl₂O₄ phase offered many benefits for increasing the GC yield: acting as active sites for the heterogeneous GC pathway and interacting with an insoluble ZnAl₂O₄ phase [17].

The ZnAl₂O₄ spinel structure is represented with the typical formula (Zn²⁺)[Al³⁺₂]O₄, where Zn²⁺ cations occupy one-eighth of the tetrahedral sites, and Al³⁺ cations occupy half of the octahedral sites. However, a normal ZnAl₂O₄ spinel structure can be partially disordered, replacing some Al³⁺ cations with Zn²⁺ cations in the octahedral sites and some Zn²⁺ cations with Al³⁺ cations in the tetrahedral sites [16]. It has been reported that the disordered structure of the ZnAl₂O₄ spinel structure affected the formation of the Zn NCO complex. Despite many studies on the various types of reactions or methods for preparing the ZnAl₂O₄ spinel structure, only a few studies have been conducted to investigate the influence of calcination temperatures on the structural properties of ZnAl₂O₄ catalysts. The formation of the disordered ZnAl₂O₄ structure is highly dependent on the calcination temperature.

Hence, in this work, ZnAl₂O₄ spinel catalysts with a disordered structure (denoted as ZnAl₂O₄-X, where X represents calcination temperature in °C) were prepared by using a polymeric citrate complex method and calcined at different temperatures ranging from 550 °C to 850 °C to investigate the influence of calcination temperatures on the ZnAl₂O₄ spinel lattice structure, the acidity of the catalysts, and the formation of a solid Zn NCO complex on the catalysts. We also used the prepared catalysts in the glycerolysis of urea under vacuum (3 kPa) at 140 °C for 5 h. Characterizations, including Fourier transform infrared spectroscopy (FT-IR) and X-ray photoelectron spectroscopy (XPS), were employed to observe the disordered ZnAl₂O₄ spinel structure. Acidic sites were measured via the temperature-programmed desorption of NH₃ (NH₃-TPD), and the formation of a solid Zn NCO complex was demonstrated by FT-IR and thermogravimetric analysis (TGA).

2. Materials and Methods

2.1. Catalyst Preparation

Citrate complex ZnAl₂O₄ catalysts were prepared by using a modified citrate complex technique [19]. An aqueous solution of Zn(NO₃)₂.6H₂O (Sigma-Aldrich Korea, Gyounggi, Republic of Korea) and Al(NO₃)₃·9H₂O (Sigma-Aldrich Korea) with a molar ratio of Zn and Al in the precursor of 1:2 was prepared at room temperature, and the citric acid powder (citric acid to metal ratio of 2:1) (Sigma-Aldrich Korea) was added to the solution, followed by stirring with a magnetic bar. Stirring was continued at 70 °C to evaporate water until a yellow viscous gel solution was formed. The gel was further dried in an oven at 140 °C and spontaneously solidified by the emission of NO_x gases. Finally, all catalysts were calcined under a set temperature (550 °C–850 °C) in a furnace for 4 h. The catalysts were denoted as ZnAl₂O₄-X, where X indicates the calcination temperature.

2.2. Reaction Test

The reaction was performed in a round-bottom, three-neck 100 mL flask, in which one neck was connected to a vacuum line through a water condenser. Glycerol (Sigma-Aldrich Korea) (0.2 mol) was added to the flask at 80 °C under stirring by a magnetic bar to reduce the high viscosity of the glycerol. The reactor was connected to a vacuum pump through a HNO₃ (Sigma-Aldrich Korea) solution trap (to remove NH₃) and a cold trap (to remove order volatiles). After 10 min, 0.2 mol of urea (Sigma-Aldrich Korea) was poured into the flask to mix with glycerol in the solution. When all the urea was dissolved completely in glycerol and the solution was transparent, a certain amount of catalyst (5 wt% compared with the initial glycerol amount) was added to the flask. The reaction was performed under vacuum pressure (3 kPa) at 140 °C with constant stirring.

After the reaction tests, ethanol (Sigma-Aldrich Korea) was poured into the final products, and the liquid products were separated from the spent catalyst via filtration. The liquid product was quantitatively analyzed via gas chromatography by using a gas chromatography machine (Acme 6100 GC, YL Instrument Co., Ltd., Dongangu, Anyang, Republic of Korea) (a schematic diagram of the catalyst activity test device is shown in Figure S1 in the Supplementary material) with a flame ionization detector and a capillary column DB-Wax (30 m × 0.25 mm × 0.25 µm). The molar amount of each component was calculated by using the internal standard method, with tetraethylene glycol (Sigma-Aldrich Korea) as the internal standard chemical. Glycerol conversion, GC selectivity, GC yield, and byproduct selectivity were calculated by using the equations below. The amount of each chemical is on the mole unit.

$$\begin{gathered} \mathrm{Glycerol} \mathrm{conversion} (\%)=\frac{\mathrm{Initial} \mathrm{amount} \mathrm{of} \mathrm{glycerol} - \mathrm{Residual} \mathrm{amount} \mathrm{of} \mathrm{glycerol}}{\mathrm{Initial} \mathrm{amount} \mathrm{of} \mathrm{glycerol}}\times 100 \\ \mathrm{GC} \mathrm{yield} (\%)=\frac{\mathrm{Amount}\mathrm{of}\mathrm{GC}}{\mathrm{Initial} \mathrm{amount} \mathrm{of} \mathrm{glycerol}}\times 100 \\ \mathrm{GC} \mathrm{selectivity} (\%)=\frac{\mathrm{G}\mathrm{C} \mathrm{yield} (\%) }{\mathrm{Glycerol} \mathrm{conversion} (\%)}\times 100 \\ \begin{array}{c}\mathrm{Byproduct} \mathrm{selectivity} (\%)\\ =\frac{\mathrm{Amount} \mathrm{of} \mathrm{byproduct}}{\mathrm{Initial} \mathrm{amount} \mathrm{of} \mathrm{glycerol} - \mathrm{Residual} \mathrm{amount} \mathrm{of} \mathrm{glycerol} }\times 100\end{array} \end{gathered}$$

FT-IR spectra of the liquid products were obtained by using a Thermo Scientific™ Nicolet™ iS™5 FT-IR spectrometer (Thermo Fisher Scientific, Waltham, MA, USA) by dropping a liquid sample between KBr (Sigma-Aldrich Korea) plates.

2.3. Catalyst Characterization

The Chemical Composition Zn/Al Molar Ratio was measured by using an Agilent Technologies 5110 ICP-OES (Agilent, Santa Clara, CA, USA) instrument. The surface characterizations were measured via N₂ adsorption isotherm analysis on a Micromeritics ASAP 2020 (USA) apparatus. The surface area was calculated by using the Brunauer–Emmett–Teller (BET) method. X-ray diffraction (XRD) patterns for fresh catalysts were obtained by using a Rigaku RAD-3C diffractometer (Rigaku Corp., Tokyo, Japan) with Cu Ka radiation (λ = 1.5418 Å) at a scattering angle (2θ) scan rate of 2°/min, operating at 35 kV and 20 mA. The fresh and spent catalysts were analyzed by using a Thermo Scientific Nicolet iS5 FT-IR spectrometer (Thermo Fisher Scientific). XPS data were surveyed by using a Thermo Scientific K-Alpha XPS spectrometer (Thermo Fisher Scientific). The numbers of acidic and basic sites were measured on the basis of TPD-NH₃/CO₂ on a MicrotracBEL BELCAT-M instrument (MicrotracBEL Corp., Osaka, Japan). An amount of 100–200 mg of fresh catalysts was placed into the quartz sample tube of the instrument. Initially, the sample was pretreated under helium flow (100 mL/min) at 600 °C for 1 h and then cooled down to 50 °C. After that, a flow of NH₃ or CO₂ (50 mL/min) was injected for chemisorption. Finally, the temperature was increased to 600 °C with a ramping rate of 1.5 °C/min; the desorbed species were removed via helium flow (30 mL/min) and analyzed via thermal conductivity detection. The TGA of the spent catalysts was performed by using a TGA Q50 apparatus (TA Instruments, New Castle, DE, USA).

3. Results and Discussion

3.1. Characterizations of Fresh Catalysts

The textural properties of the fresh catalysts, which were calculated from N₂ adsorption–desorption measurements, are summarized in Table 1, and the N₂ adsorption–desorption isotherms and pore size distributions are shown in Figure S2. As indicated in Figure S2a, all ZnAl₂O₄ catalysts exhibited type IV isotherm and type H3 hysteresis loops (based on the IUPAC classification [20]), which is typical for mesoporous materials, indicating slit-shaped pores. As the calcination temperature increased, the hysteresis loop became narrower, with a lower BET surface area and a lower average pore volume in the order of ZnAl₂O₄-850 < ZnAl₂O₄-750 < ZnAl₂O₄-650 < ZnAl₂O₄-550. In contrast, the average pore diameter exhibited an opposite trend. Increasing the calcination temperature led to the collapse of small pores to generate bigger ones. The relationships between textural properties and calcination temperatures are depicted in Figure 1.

Table 1.

Textural properties, ZnAl₂O₄ crystallite sizes, and the degree of orderliness of the ZnAl₂O₄ spinel of fresh catalysts.

Catalyst	Textural Properties			Chemical Composition	Order Parameter c	ZnAl2O4 Crystallite Size (nm) d
	SBET (m2/g) a	Vpore (cm3/g) a	Dpore (nm) a	Zn/Al Molar Ratio b
ZnAl2O4-550	37.65	0.21	14.09	2.07	0.4784	20.29
ZnAl2O4-650	25.29	0.14	15.31	2.10	0.4922	25.50
ZnAl2O4-750	19.17	0.11	16.87	2.11	0.5074	29.21
ZnAl2O4-850	9.45	0.10	27.16	2.06	0.5153	34.06

^a Measured via N₂ isothermal adsorption–desorption at 77 K; surface area was calculated with BET method; pore volume and pore size were obtained with the BJH method. ^b Determined from ICP-OES. ^c Calculated via the fraction intensity of ZnAl₂O₄ (331) and [ZnAl₂O₄ (331) + ZnAl₂O₄ (400)] from XRD patterns. ^d Calculated via Debye–Scherrer equation applied to XRD ZnAl₂O₄ (311) peak.

Figure 1.

Relationship between textural properties and calcination temperatures.

Figure 2 depicts the XRD patterns of the fresh ZnAl₂O₄-X catalysts. Characteristic peaks appearing at 2θ = 31.2°, 36.8°, 44.8°, 49.1°, 55.7°, 59.3°, 65.2°, 74.1°, and 77.3° can be attributed to the crystalline phase of ZnAl₂O₄. All XRD patterns show ZnAl₂O₄ peaks, which indicates the formation of ZnAl₂O₄ crystallite after high-temperature calcination by using the polymeric citrate complex method. The final decomposition temperature of the precursor in the derivative TGA (DTGA) result is below 550° (Figure S3), confirming that the ZnAl₂O₄ crystallite structure is completely formed at a calcination temperature above 550 °C. With increasing calcination temperature, the characteristic diffraction peaks become sharper, narrower, and more intensive, indicating an increase in the crystallinity and particle size of ZnAl₂O₄.

Figure 2.

XRD patterns of fresh catalysts: (a) ZnAl₂O₄-550, (b) ZnAl₂O₄-650, (c) ZnAl₂O₄-750, and (d) ZnAl₂O₄-850.

In addition, the average crystallite size (D) of single-phase spinel samples can be estimated by using the Debye–Scherrer equation [21,22]:

$$\mathrm{D}=\frac{0.9\mathsf{\lambda }}{\mathsf{\beta }\mathrm{Cos}\mathsf{\theta }},$$

where D represents the average crystallite size, λ denotes the wavelength of the X-ray source (Cu Kα, 1.54 Å), β denotes the integral breadth of the (311) diffraction peak, and θ denotes the Bragg’s diffraction angle. The calculated results are presented in Table 1. In normally ordered spinel structures, the intensity of all odd reflections (e.g., 331) shows higher intensity than those of all even reflections (e.g., 400). The XRD peak intensities of (400) and (331) for various calcination temperatures are shown in Figure S4. It was observed that the peak intensities of (400) and (331) changed significantly with increasing calcination temperature. The degree of the order parameter in spinel samples can be calculated by using the following equation [23]:

$$\Delta \mathrm{I}=\frac{{\mathrm{I}}_{\mathrm{o}}}{{\mathrm{I}}_{\mathrm{o}}+{\mathrm{I}}_{\mathrm{e}}},$$

where I_o and I_e denote intensities corresponding to odd and even reflections, respectively. The disordered spinel phase can be formed because some fraction of Al³⁺ cations occupies the tetrahedral site, and Zn²⁺ cations occupy the octahedral site [24,25]. The degree of the order parameter in the spinel samples is shown in Table 1. The order parameters of ZnAl₂O₄-550 and ZnAl₂O₄-650 are 0.4784 and 0.4922, respectively. These values prove that the ZnAl₂O₄ spinel structure obtained at the calcination temperatures of 550 °C and 650 °C is a disordered structure, where Al³⁺ exists in tetrahedral positions and Zn²⁺ in octahedral positions. When the calcination temperature is raised above 750 °C, the degree of the order parameters is above 0.5, indicating that the ZnAl₂O₄ spinel is a normally ordered structure.

The FT-IR spectra of the fresh catalysts are shown in Figure 3A. The two vibration bands at 661 and 554 cm⁻¹ can be attributed to the stretching mode of the octahedrally coordinated Al-O (AlO₆), with another assignment of the peak at 495 cm⁻¹ to the bending mode of AlO₆. A broad shoulder band from 704 to 900 cm⁻¹ can be assigned to tetrahedrally coordinated Al-O (AlO₄) [16,26,27,28]. The enlarged spectra (from 600 to 1000 cm⁻¹) focusing on the vibration bands of AlO₄ and AlO₆ are depicted in Figure 3B. The relative intensities of AlO₄ to AlO₆ decrease as the calcination temperature increases. The AlO₄ vibration bands for the ZnAl₂O₄-750 and ZnAl₂O₄-850 catalysts are weak. In contrast, the shoulder of the AlO₄ vibration for ZnAl₂O₄-550 and ZnAl₂O₄-650 is strong and observable. The presence of Al³⁺ in the tetrahedral position reflects a partial inversion of the ZnAl₂O₄ spinel structure. The FT-IR results are consistent with the XRD results and the inversion parameter obtained from the XPS measurement, which is shown later.

Figure 3.

(A)FT-IR spectra of fresh catalysts: (a) ZnAl₂O₄-550, (b) ZnAl₂O₄-650, (c) ZnAl₂O₄-750, and (d) ZnAl₂O₄-850. (B) Enlarged FT-IR spectra.

The deconvoluted XPS results of the fresh ZnAl₂O₄-550 and ZnAl₂O₄-650 catalysts are depicted in Figure 4. The Al2p XPS spectrum is illustrated in Figure 4A. The peak at a binding energy of 71.0 eV can be assigned to Al³⁺ occupying the tetrahedral sites (AlO₄) [29], and a peak of approximately 74.2 eV can be assigned to Al³⁺ occupying the octahedral sites (AlO₆) [29,30,31]. In Zn2p_3/2 spectra, as shown in Figure 4B, the larger peak at a lower binding energy (approximately 1022–1022.2 eV) is attributed to Zn²⁺ occupying the tetrahedra sites (ZnO₄), and a smaller peak at a higher binding energy (approximately 1024.5 eV) is attributed to Zn²⁺ occupying the octahedra sites (ZnO₆) [29,30,32]. The normally ordered ZnAl₂O₄ spinel structure contains Al³⁺ cations at the octahedral sites and Zn²⁺ cations at the tetrahedral sites. When some Al³⁺ cations substitute Zn²⁺ cations in the tetrahedral positions and some Zn²⁺ cations substitute Al³⁺ cations in the octahedral positions, this results in the formation of a partially inversed spinel structure. The inversion parameter of a spinel structure can be defined by surface XPS measurements [30]. Table 2 shows the inversion parameter of the fresh ZnAl₂O₄-550 and ZnAl₂O₄-650 catalysts by calculating the ratio of AlO₄/(AlO₄ + AlO₆) and ZnO₆/(ZnO₄ + ZnO₆), and the inversion parameter values follow the order of ZnAl₂O₄-550 > ZnAl₂O₄-650. These results help us demonstrate that the degree of disorder of the spinel ZnAl₂O₄ structure decreases with increasing calcination temperature. The deconvoluted patterns of O1s can be fitted to three peaks. The peak O_a at the highest binding energy (approximately 543.5–534.8 eV) can be assigned to the oxygen weakly bonded with the surface of the catalysts (such as adsorbed H₂O and O₂ from the atmosphere) [16,33,34]. The peak at approximately 533.4–533.6 eV (O_b) can be assigned to the oxygen-deficient regions or oxygen vacancy (O_v) [33,34]. Finally, the peak at the lowest binding energy (approximately 531.4–531.7 eV) can be assigned to the lattice O in the ZnAl₂O₄ phase [16]. In a partially inversed spinel structure or a disordered structure, the substitution of a Zn²⁺ cation with an Al³⁺ cation to generate a Zn²⁺ octahedral site and Al³⁺ tetrahedral site is the main reason for the oxygen vacancy formation on the catalyst surface. An oxygen vacancy was formed to balance the positive charge of the cation caused by the substitution of Zn²⁺ for Al³⁺ at an octahedral site. The intensities of oxygen vacancy follow the order of ZnAl₂O₄-550 > ZnAl₂O₄-650, which are listed in Table 2.

Figure 4.

Deconvoluted XPS results of fresh catalysts: (A) Al2p XPS results, (B) Zn2p_3/2 XPS results, and (C) O1s XPS results.

Table 2.

The change in the number of acidic sites and oxygen vacancy is based on the ratio of AlO₄/(AlO₄ + AlO₆) and ZnO₆/(ZnO₄ + ZnO₆).

Catalyst	Intensity Ratio of AlO4/(AlO4 + AlO6) a	Intensity Ratio of ZnO6/(ZnO4 + ZnO6) b	Acidic Sites (mmol/g)	The Fraction of Ob (%) c
ZnAl2O4-550	0.365	0.409	0.45	15.7
ZnAl2O4-650	0.222	0.358	0.32	13.6
ZnAl2O4-750	-	-	0.29	-
ZnAl2O4-850	-	-	0.1	-

^a Intensity ratios of the AlO₄ (tetrahedral site) and AlO₆ (octahedral site) were obtained from the deconvolution of XPS results by using the Origin software (OriginPro 2021 9.8.0.200). ^b Intensity ratios of the ZnO₆ (octahedral site) and ZnO₄ (tetrahedral site) were obtained from the deconvolution of XPS results by using the Origin software. ^c The fraction of O_b was calculated via the percentage of intensity of O_b in total intensity of deconvolution O1s XPS results.

The NH₃-TPD measurements of the fresh catalysts are shown in Figure S5, and the acidic sites are summarized in Table 2. The number of acidic sites decreases in the order of ZnAl₂O₄-550 > ZnAl₂O₄-650 > ZnAl₂O₄-750 > ZnAl₂O₄-850. The relationship between the calcination temperature and the total number of acidic sites is shown in Figure 5. In this study, the presence of AlO₄ and ZnO₆ in the ZnAl₂O₄ spinel structure indicates the partially disordered structure of the catalyst. Furthermore, the partially disordered structure ZnAl₂O₄ spinel is capable of producing surface acidity on catalysts, which is related to the XPS intensity of O_a (in Figure 4) [16]. The high-intensity peak of O_a in the ZnAl₂O₄-550 catalyst and its dependency on the calcination temperature follow the same trend as the number of acidic sites.

Figure 5.

Relationship between total number of acidic sites and calcination temperatures.

The relationship between the number of acidic sites of the fresh catalysts and the inversion parameter (intensity ratio of AlO₄/(AlO₄ + AlO₆)), the XPS intensity of oxygen vacancies (O_v), and the inversion parameter (intensity ratio of ZnO₆/(ZnO₄ + ZnO₆)) are shown in Figure 6. In the normally ordered ZnAl₂O₄ spinel structure, Al³⁺ cations occupy the octahedral sites (AlO₆) with low acidity. Meanwhile, Al³⁺ cations occupying tetrahedral sites (AlO₄) in the partially inversed ZnAl₂O₄ spinel structure can result in higher surface acidity [16,27,35,36,37,38]. The higher Lewis acidity of AlO₄ than AlO₆ can be explained by the lower energy acceptor orbital of the AlO₄ sites. When AlO₄ contacts the surface, the removal of one of the four oxygen atoms results in the formation of three-coordinated Al with higher Lewis acidity strength [39]. Figure 6 depicts the positive relationship between the inversion parameter (AlO₄/(AlO₄ + AlO₆)) and total surface acidity. With an increase in the calcination temperature, the disordered ZnAl₂O₄ structure decreases, indicating that there are fewer sites in which Al³⁺ cations occupy the tetrahedral position (AlO₄), which decreases the total surface acidity. Another factor increasing surface acidity is the Lewis acidity of surface oxygen vacancies. Substitution of Zn²⁺ cations for Al³⁺ cations at the octahedral site and the formation of oxygen vacancies (O_v) to balance the positive charge of the cations can increase surface acidity. Mefford et al. [40] reported that the formation of surface hydroxyl groups was related to the Lewis acidity of surface oxygen vacancies, and evidence was provided from XPS O_a peaks of H₂O or O₂ adsorbed from the atmosphere. As the number of ZnO₆ sites decreases with increasing calcination temperature, fewer oxygen vacancies are formed, lowering the surface acidity of the ZnAl₂O₄-650 catalyst compared with the ZnAl₂O₄-550 catalyst. From the above findings, we conclude that at the calcination temperature of 550 °C, a partially disordered ZnAl₂O₄ spinel structure is formed, in which some Al³⁺ cations occupy tetrahedral positions, and Zn²⁺ occupy octahedral positions, resulting in the formation of oxygen vacancies, which increases the surface acidity of the catalyst. With increasing calcination temperature, the structure of ZnAl₂O₄ gradually returns to normal, and the number of AlO₄ and ZnO₆ sites and oxygen vacancies decreases, lowering the surface acidity of the catalyst.

Figure 6.

Relationship between the number of acidic sites vs. inversion parameter (AlO₄/(AlO₄ + AlO₆) and O1s XPS intensity (oxygen vacancy) vs. inversion parameter (ZnO₆/(ZnO₄ + ZnO₆).

3.2. Catalytic Activity

The reaction results (yield, conversion, and selectivity) calculated from gas chromatograms (a typical gas chromatogram is shown in Figure S6, see Supplementary Materials) for the glycerolysis of urea over the ZnAl₂O₄-550 and ZnAl₂O₄-650 catalysts are summarized in Table 3 and Figure 7A. During the reaction, besides the formation of the main product GC, there are also byproducts such as chemicals (2), (3), and (5). After the 5 h experiment, no other by-products are collected, indicating that the carbon balance between reagents and products is close to 100% (Table S1, see Supplementary Materials). The relationship between the number of acidic sites vs. GC yield and the number of basic sites (Figure S7, see Supplementary Materials) vs. byproduct (2) selectivity for each catalyst is depicted in Figure 7B. At a reaction time of 5 h, the ZnAl₂O₄-550 catalyst exhibited higher GC yield and lower (2) selectivity (GC yield = 46%, (2) selectivity = 34%) than the ZnAl₂O₄-650 catalyst with GC yield = 41% and (2) selectivity = 41%. Previous studies have reported the existence of the intermediate isocyanate (NCO) complex of Zn, which is an active site for the reaction [18,41]. Moreover, it has been proven that the Lewis acidic site acts as an active site for the ammonia (NH₃) group of Zn NCO complex adsorption [16]. The ZnAl₂O₄-550 catalyst has a higher GC yield and more surface acidic sites than the ZnAl₂O₄-650 catalyst.

Table 3.

Analysis of liquid products obtained from the glycerolysis of urea over various mixed oxide catalysts prepared at different calcined temperatures. (Reaction temperature = 140 °C, reaction time = 5 h, reaction pressure = 3 kPa, glycerol/urea ratio = 1:1). (2): 2,3-dihydroxypropyl carbamate, (4): 4 (hydroxymethyl)oxazolidin-2-one, and (5): (2-oxo-1,3-dioxolan-4-yl) methylcarbamate.

Catalyst	Glycerol Conversion (%)	GC Yield (%)	Selectivity (%)				Acidic Sites (mmol/g)	Basic Sites (mmol/g)
			GC	(2)	(4)	(5)
ZnAl2O4-550	72.2 ± 0.7	33 ± 0.3	45.8 ± 0.5	34.3 ± 0.1	12.4 ± 0.2	7.5 ± 0.1	0.45	0.23
ZnAl2O4-650	72.2 ± 0.8	30 ± 0.3	40.9 ± 0.7	40.9 ± 0.3	10.9 ± 0.2	7.3 ± 0.1	0.32	0.13

Figure 7.

Catalytic performance of ZnAl₂O₄-550 and ZnAl₂O₄-650 catalysts in the glycerolysis of urea (A), and the relationship between the number of acidic sites vs. GC yield and the number of basic sites vs. (2) selectivity (B).

The formation of the Zn NCO complex on the solid phase can be confirmed by the FT-IR and TGA of the spent catalysts. The FT-IR spectra of the spent catalysts are shown in Figure 8. Besides the bands for AlO₄ and AlO₆ being detected at 704 and 661 cm⁻¹, respectively, the vibration band of NCO in the Zn NCO complex on the solid phase is detected at 2350 cm⁻¹ [42]. There is no information about the NCO band in the fresh catalysts. However, after a 5 h reaction, this band can be detected by FT-IR. The intensity of the NCO peak for ZnAl₂O₄-550 is higher than that for ZnAl₂O₄-650 because of the higher number of acidic sites in this catalyst. The same trend is presented by TGA and DTGA profiles in Figure 9. The DTGA profiles of the spent catalysts (Figure 9B) show the decomposition of the Zn NCO complex at 273 °C for ZnAl₂O₄-550 and 254 °C for ZnAl₂O₄-650 [16]. The higher decomposition temperature for the ZnAl₂O₄-550 catalyst can be attributed to a stronger crystalline Zn NCO complex with higher thermal stability. The DTGA results are consistent with the FT-IR results. The higher intensity of the Zn NCO decomposition peak in ZnAl₂O₄-550 demonstrated that the ability to adsorb the ammonia group in the Zn NCO of AlO₄ (Lewis acidic site) is more significant in the ZnAl₂O₄-550 catalyst than in the ZnAl₂O₄-650 catalyst.

Figure 8.

FT-IR results of fresh and spent catalysts: (a) ZnAl₂O₄-550 fresh, (b) ZnAl₂O₄-550 spent 5 h, (c) ZnAl₂O₄-650 fresh, and (d) ZnAl₂O₄-650 spent 5 h.

Figure 9.

(A) TGA and (B) DTGA of spent catalysts after 5 h reaction.

Another notable finding is the relationship between surface basicity and byproduct (2) selectivity. The basic site values are shown in Table 3 and Figure 7B, showing that the relationship between the number of basic sites and (2) selectivity is inversely proportional. That is, if the number of basic sites decreases, more byproducts (2) are formed, and the GC yield reduces. Fernandes et al. [43] explained that surface basicity acted as an active site for the (2) byproduct decomposition process. This study shows the same trend: when using ZnAl₂O₄-550 with more basic sites for the reaction, (2) selectivity (34%) is lower, whereas it is higher with ZnAl₂O₄-650 (41%).

Catalytic reaction routes over a disordered ZnAl₂O₄ spinel structure are depicted in Scheme 1. As mentioned above, there are several disordered sites over a partially inverted ZnAl₂O₄ spinel: AlO₄ (symbolized as Al*), ZnO₆ (symbolized as Zn*) and oxygen vacancies (symbolized as O_v). ZnAl₂O₄ catalysis follows the heterogeneous reaction pathways. The catalysts form the Zn NCO complexes via the reaction with urea. The AlO₄ site works as a Lewis acidic site where the ammonia (NH₃) group of the Zn NCO complex can adsorb. The presence of the Zn NCO complex adsorbed on the solid phase is confirmed by the FT-IR and TGA measurements.

Scheme 1.

An illustration of the reaction pathways involved in the urea-glycerol carbonylation on the surface of the disordered ZnAl₂O₄ spinel structure.

In Table 4, the catalytic reaction results in this work are compared to those recently published in the literature. Because the ZnAl₂O₄ catalyst follows only the heterogeneous catalysis pathway, the glycerol conversion and GC yield values are relatively low. However, it can still be compared with some previous studies. Zhang et al. prepared the zinc glycerolate (ZMG) and ZnO from zinc glycerolate (ZnO from ZMG) via the calcination method. Glycerol conversion and GC yield of the ZMG are about 65 and 55%, respectively. The authors have demonstrated through reaction tests that catalysts with acidic and basic properties favor the synthesis of glycerol carbonate.

Table 4.

Summary of catalytic reaction performance of glycerol carbonylation with urea over several catalysts.

Catalyst	Mol Ratio Glycerol/Urea	Weight % Catalyst/Glycerol	Glycerol Conversion (%)	GC Yield (%)	Reference
ZnO pure	1:1.5	5	~60	~35	[44]
ZMG	1:1.5	5	~65	~55	[44]
Co3O4/ZnO	1:1.5	5	69	97	[45]
La2Cu0.5Fe0.5O4	1:1	5	49	82	[46]
La(OH)3	1:1.5	5	42.3	55.6	[47]
MoO3/SnO2	1:3	10	69	97	[48]
MgO	1:1.5	-	71	100	[43]
ZnO	1:1.5	5	67	49	[49]
[HOEMIm][PF6]	1:1.5	-	62	47	[50]
ZnAl2O4-550	1:1.5	5	72.2	33	This work
ZnAl2O4-650	1:1.5	5	72.2	30	This work

4. Conclusions

In this study, the disordered ZnAl₂O₄ catalysts were successfully prepared by using the citrate complex method at different temperatures ranging from 550 °C to 850 °C. The ZnAl₂O₄-550 catalyst exhibited a more disordered ZnAl₂O₄ structure than the other catalysts calcined at a higher calcination temperature, resulting in more disordered sites of AlO₄ and more oxygen vacancies in ZnAl₂O₄-550. The ZnAl₂O₄-650 catalyst having Al³⁺ cations at the octahedral sites (AlO₆) exhibited low acidity, whereas the AlO₄ sites in ZnAl₂O₄-550 resulted in high surface acidity. From the XPS intensities of O_a (H₂O or O₂ adsorbed) and O_b (oxygen vacancies), ZnAl₂O₄-550 contained more surface oxygen vacancies than ZnAl₂O₄-650, which contributed to the increased surface acidity of ZnAl₂O₄-550. The high surface acidity of ZnAl₂O₄-550 resulted in strong interactions with the Zn NCO complex on its surface, improving catalytic performance. The intermediate product of urea glycerolysis was 2,3-dihydroxypropyl carbamate (2), which had higher selectivity for ZnAl₂O₄-650 than ZnAl₂O₄-550, lowering the GC yield.

Supplementary Materials

The following supporting information can be downloaded at https://www.mdpi.com/article/10.3390/nano13131901/s1, Figure S1. A schematic diagram of the catalyst activity test device with the gas chromatographic analysis conditions; Figure S2: N₂ adsorption–desorption isotherms (A) and pore size distribution curves (B): (a) ZnAl₂O₄-550, (b) ZnAl₂O₄-650, (c) ZnAl₂O₄-750, and (d) ZnAl₂O₄-850; Figure S3: DTGA curves of ZnAl₂O₄ xerogel precursor prepared by using polymeric citrate complex method; Figure S4: Enlarged XRD patterns of fresh catalysts: (a) ZnAl₂O₄-550, (b) ZnAl₂O₄-650, (c) Znl₂O₄-750, and (d) ZnAl₂O₄-850; Figure S5: NH₃ profiles of fresh catalysts: (a) ZnAl₂O₄-550, (b) ZnAl₂O₄-650, (c) ZnAl₂O₄-750, and (d) ZnAl₂O₄-850; Figure S6: Typical gas chromatograms of ZnAl₂O₄-550 (A) and ZnAl₂O₄-650 (B). (2): 2,3-dihydroxypropyl carbamate, (4): 4-(hydroxymethyl) oxazolidine-2-one, and (5): (2-oxo-1,3-dioxolan-4-yl) methyl carbamate; Figure S7: CO₂ profiles of fresh catalysts: (a) ZnAl₂O₄-550, (b) ZnAl₂O₄; Table S1: The initial amount of each reactant; the amount of glycerol and products after reaction and carbon balance.

Author Contributions

H.N.-P.: methodology, validation, investigation; N.P.-N.: data curation, writing—original draft preparation; E.W.S.: conceptualization, writing—review and editing, supervision. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement

The data presented in this study are available on request from the corresponding author.

Conflicts of Interest

The authors declare no conflict of interest.

Funding Statement

This research was supported by the Core Research Institute Basic Science Research Program (2021R1A6A1A03038858) and “Regional Innovation Strategy program” (No. 2021RIS-003) through the National Research Foundation of Korea (NRF) funded by the Ministry of Education and the National Research Foundation of Korea (NRF) grant funded by the Korea government (MIST) (No. 2021R1A2B5B01001448).