Introduction

Biodiesel is an environmentally friendly renewable energy that can replace fossil energy because of its non-toxic, sulfur-free, aromatics free, high cetane number and good combustion performance and so on [1,2,3,4,5,6]. At present, the esterification reaction of fatty acid with alcohol catalyzed by liquid acid catalyst is one of the most commonly used method to synthesize biodiesel in industry. Nevertheless, liquid acid catalyst often had the inevitable problems of recycling difficulty, equipment corrosion, serious equipment corrosion and environmental pollution [7,8,9,10,11]. So, the exploration of substitution environmentally friendly solid acids for liquid acid catalysts to achieve efficient synthesis of green biodiesel has received growing research attention in recent years [12, 13].

SO42−/MxOy solid acid has been recognized as the most development valuable green catalyst in catalytic esterification synthesis of biodiesel [14,15,16,17,18,19,20,21]. The traditional SO42−/MxOy solid acid is mainly composed of both active matrix (MxOy) and promoter (SO42−), which mainly focus on SO42−/ZrO2, SO42−/TiO2 and SO42−/SnO2 systems [22, 23]. Among them, SO42−/ZrO2 has attracted much attention owing to its characteristic advantages of unique strong acid properties, high catalytic activities and good selectivity for catalytic esterification synthesis of biodiesel. Regrettably, the traditional SO42−/ZrO2 still has some scientific problems of restricting its industrial application as follows: On the one hand, the traditional SO42−/ZrO2 solid acid has some shortcomings of fast inactivation, poor stability and short one-way life due to the loss of surface active sulfur and surface area carbon in the catalytic esterification reactions. On the other hand, ZrO2 crystalline form was easily interconverted between the active tetragonal phase and the inactive monochromic phase during the preparation of SO42−/ZrO2 solid acid [24, 25]. As a result, the traditional SO42−/ZrO2 solid acid inevitably faced with the unstable structure and the poor reusability. In order to resolve the above two scientific problems of the traditional SO42−/ZrO2 solid acid, many researchers have tried to explore the development and application of SO42−/ZrO2–MxOy composite solid acids by modified SO42−/ZrO2 with other metallic oxides of MxOy [26]. Numerous studies have demonstrated that the addition of other MxOy is beneficial to improve the catalytic performance of SO42−/ZrO2–MxOy composite solid acids by effective means of optimizing the acid properties or impeding the transition of ZrO2 from the active tetragonal phase to the inactive monoclinic phase. For example, Xu et al. [27] found that the incorporation of Fe2O3 into Pt/SO42−/ZrO2 enhanced Brönsted acidity and the catalytic activity for n-heptane hydroisomerization. Yu et al. [28] showed that the addition of Yb2O3 and Al2O3 in SO42−/ZrO2–Yb2O3–Al2O3 composite solid acids improved the specific surface area and the catalytic activity of the catalyst. Dussadee et al. [29] showed that the addition of 20 wt%La2O3 to SO42−/ZrO2 could efficiently promote both esterification and transesterification reactions of palm oil because of its dual strong basic and acid sites. Moreover, the 20 wt%La2O3–SO42−/ZrO2 catalyst maintained a relatively stable catalytic behavior. Li et al. [30] found that the appropriate addition of MoO3 in SO42−/ZrO2–MoO3 stabilized the metastable-state tetragonal ZrO2 crystalloid and inhibited its sintering during the calcination process, which was more efficient to increase the specific surface area and acid content of the catalyst. Fan et al.[31]. showed that the addition of CeO2 in S2O82−/ZrO2–CeO2 enhanced its thermal stability of tetragonal ZrO2 and correspondingly increased its acid strength. It is important to emphasize that the designed synthesis of SO42−/ZrO2–MxOy composite solid acids with the thermal stability and superior acid properties remains a challenging subject in recent years. So far, the efficient synthesis of green biofuels over SO42−/ZnAl2O4–ZrO2 composite solid acids have not yet been extensively reported in relevant literature. Compared with simple oxides of MxOy, compound oxide spinel of ZnAl2O4 has the unique advantage of single crystal shape and the more stable structure. Based on this selection of design ideas, the novel SO42−/ZnAl2O4–ZrO2 composite solid acids would be expected to obtain the higher structural stability and the superior acid properties.

In this paper, a novel series of SO42−/ZnAl2O4–ZrO2 composite solid acids were prepared by a simple sol–gel-impregnation method and applied to the typical esterification reaction of oleic acid with methanol for green biodiesel synthesis. The effectively design of SO42−/ZnAl2O4–ZrO2 composite solid acids were based on the single crystal advantage of ZnAl2O4 as well as the excellent acid forming capacity of ZrO2. The structure and acidic properties of SO42−/ZnAl2O4–ZrO2 composite solid acids were systematically investigated by the means of XRD, FE-SEM, TG, NH3-TPD, XPS, FT-IR, NH3 adsorption FT-IR spectra and acid–base titration. Additionally, we investigated the effect of both the calcination temperature and the different mass ratio of ZnAl2O4 to ZrO2 on the structures and the comprehensive acidic properties of SO42−/ZnAl2O4–ZrO2 composite solid acids in detail. In order to further evaluate the comprehensive catalytic performance of the optimal SO42−/ZnAl2O4–ZrO2 (8:2) composite solid acid for green biodiesel synthesis, we delved into its reusability, its kinetic study and its essential reasons for its high stability and its slight deactivation during the acid catalyzed esterification reaction. As a novel catalyst system for acid catalyzed reaction, SO42−/ZnAl2O4–ZrO2 composite solid acids have the prominent advantages of easy preparation, high structural stability, excellent acid catalytic activity and reuse stability, environmentally friendly and so on.

Experimental

Catalyst preparation

Preparation of ZnAl2O4

ZnAl2O4 spinel oxide was synthesized by sol–gel method. Al(NO3)3·9H2O and Zn(NO3)2·6H2O with the molar ration of 2:1 were dissolved in ethanol. Then, PEG-2000 (5 wt% of the total mass of nitrates) was added to the above solution and keep stirring for 4 h at room temperature. The obtained mixture solution was evaporated at 65 °C for 1 h to obtain the sol. The sol was dried at 120 °C to become a dried gel. The dried gel was grounded into a fine powder and was calcined at 600 °C for 5 h in air to obtain ZnAl2O4 spinel oxide.

Preparation of ZnAl2O4–ZrO2

According to the certain mass ratio of ZnAl2O4 to the ZrO2, the above obtained fine ZnAl2O4 powder was dispersed in ZrOCl2·8H2O aqueous solution with constant stirring. Then, aqueous ammonia was added to form the precipitate under pH of 8 ~ 9. The precipitate was aged for one day at room temperature. Afterwards, the precipitate was washed with distilled water to completely remove the chloride ions. At last, the precipitate was filtered and dried at 100 °C to obtain ZnAl2O4–ZrO2 composite active matrix. The composite active matrix with the mass ratio of ZnAl2O4 to ZrO2 of 1:0, 8:2, 6:4, 4:6 2:8 and 0:1, which were correspondingly designated as ZnAl2O4, ZnAl2O4–ZrO2 (8:2), ZnAl2O4–ZrO2 (6:4), ZnAl2O4–ZrO2 (4:6), ZnAl2O4–ZrO2 (2:8) and ZrO2.

Preparation of SO4 2−/ZnAl2O4–ZrO2 composite solid acids

A portion of ZnAl2O4–ZrO2 composite active matrix were impregnated with a certain volume (1 g/10 mL) of 1.5 M (NH4)2SO4 (ammonium sulfate) and stirred for 4 h. After filtration, the above sample was dried at 100 °C and calcined for 3 h at 600 °C to obtain SO42−/ZnAl2O4–ZrO2 composite solid acids, which were denoted as SO42−/ZnAl2O4, SO42−/ZnAl2O4–ZrO2 (8:2), SO42−/ZnAl2O4–ZrO2 (6:4), SO42−/ZnAl2O4–ZrO2 (4:6), SO42−/ZnAl2O4–ZrO2 (2:8) and SO42−/ZrO2.

Catalyst characterization

The structural characterization was completed by X-ray powder diffraction (XRD) performed on Bruker AXS D8-Focus. XRD were recorded in the range of 2θ = 10–70°. Field emission scanning electron microscopy (FE-SEM) measurements were performed on S4800. The dried samples were coated with gold. Thermogravimetric analysis (TG) were performed on a STA-409PC thermoanalyzer in the temperature range of 50~1000 °C with a heating rate of 10 °C min−1. NH3 Temperature Programmed Desorption (NH3-TPD) experiment was carried out using a TP-5076 TPD/TPR. X-ray photoelectron spectroscopy (XPS) was performed on a VG Multilab 2000. The Fourier transform Infrared Spectroscopy (FT-IR) spectra was recorded by a Nicolet 6700 IR spectrometer in the range of 400–4000 cm−1 and coupled with KBr pellet technique. KBr was pretreated by drying for 4 h at 100 °C before using and was ground to a powder under the infrared lamp. Then, the samples and KBr were mixed and pressed into pellet on a hydraulic tablet machine. The final FT-IR spectra of the sample was obtained by eliminating the background spectrum. The sample examined by FT-IR coupled with NH3 chemisorption was pre-treated in a He flow (20 mL min−1) at 300 °C for 1 h. Then, the dried sample as cooled to 60 °C and exposed to NH3 (20 mL min−1) at room temperature for 1 h, and followed by heated up to 120 °C and flowed by helium (20 mL min−1) for 1 h to remove gas-phase and physically adsorbed NH3. The obtained samples were examined by FT-IR by using KBr technique, which obtained the FT-IR of NH3 chemisorption. The shown spectra were obtained after normalized and subtracted by the FT-IR of samples without chemisorption of NH3. The acid site density of catalysts was determined by ion-exchange titration. 100 mg of the catalyst was added to 25 mL of 0.1 mol L−1 NaCl solution and stirred for 24 h at room temperature. The catalyst was separated by filtration. The filtrate was titrated by 0.10 mol L−1 NaOH solution using phenolphthalein as an indicator to obtain the corresponding acid values. The total acid densities of the catalysts were estimated by the acid values [32].

Catalytic activity test

The esterification reaction of oleic acid with methanol was tested in a three-necked round flask equipped with a magnetic stirrer, a thermometer and a refluxing condenser tube. The conditions of esterification reaction were as follows: the reaction temperature was 65 °C; the molar ratio of oleic acid to methanol was 1:25; the reaction time was 8 h, the amount of catalysts was 5 wt% (based on the mass of oleic acid). According to the references [33, 34] and the method of GB1668-81, the acid–base titration had been used to calculate the conversion of oleic acid on the basis of acid value. The detailed process was as follows: the 0.50 mL initial reaction mixture or the 0.50 mL final reaction mixture were diluted in 20.00 mL ethanol. Then, the diluted mixtures were titrated by 0.10 mol L−1 NaOH solution with an indicator of phenolphthalein. The conversion of oleic acid could be calculated from Eq. 1:

$$ {\text{Conversion~of~acetic~acid}}\left( {\text{\% }} \right) = \frac{{M_{0} - M_{1} }}{{M_{0} }} \times 100 $$
(1)

Here M0 represented the volume of NaOH consumed by the initial reaction mixture and M1 represented the volume of NaOH consumed by final reaction mixture.

In order to evaluate the reusability, the optimal SO42−/ZnAl2O4–ZrO2 (8:2) composite solid acid after finishing each catalytic evaluation was repeatedly used for the next new esterification reaction of oleic acid with methanol only through filtering and drying.

Results and discussion

Characterization

It is well known that the crystal structure of SO42−/MxOy solid acid is closely related to its acidity and its catalytic activity. Based on this consideration, we explored the influence of the calcination temperature on the crystal structure of SO42−/ZrO2 and SO42−/ZnAl2O4–ZrO2 (4:6). As shown in Fig. 1, only a weak peak of active tetragonal phase ZrO2 at 2θ = 30.4° was observed in SO42−/ZrO2 at the lower calcination temperature of 500 °C [35]. The characteristic diffraction peaks of ZrO2 active tetragonal phase became very obvious in SO42−/ZrO2 at the higher calcination temperature of 600 °C. Regrettably, the inactive monoclinic phase of ZrO2 at 2θ of 28.2° and 31.5° was also generated in SO42−/ZrO2 at the higher calcination temperature of 600 °C. This above result demonstrated that the higher calcination temperature resulted in the transformation of ZrO2 crystal form between the active tetragonal phase and the inactive monochromic phase, which would affect the activity of SO42−/ZrO2 solid acid [36]. It was worth emphasizing that the calcination temperatures had no evident influence on compound oxide spinel of ZnAl2O4 in SO42−/ZnAl2O4–ZrO2 (4:6). As shown in Fig. 1, the characteristic diffraction peaks of ZnAl2O4 spinel at 2θ of 31.5°, 37.1°, 45.1°, 49.3°, 55.9°, 59.6° and 65.5° (JCPDS No.05–0669) were all observed in SO42−/ZnAl2O4–ZrO2 (4:6) composite solid acid at the different calcination temperatures of 500 and 600 °C. Surprisingly, the diffraction peaks of ZrO2 monoclinic phase were almost undetected in SO42−/ZnAl2O4–ZrO2 (4:6). As expected, the transformation of ZrO2 from active tetrahedron to inactive monoclinic was successfully inhibited by the addition of ZnAl2O4 in SO42−/ZnAl2O4–ZrO2 (4:6) at the higher calcination temperatures of 600 °C. As a result, the better structural stability might be more beneficial to improve the reusability SO42−/ZnAl2O4–ZrO2 composite solid acid by compared with SO42−/ZrO2.

Fig. 1
figure 1

XRD patterns of SO42−/ZrO2 and SO42−/ZnAl2O4–ZrO2 (4:6) solid acids at the different calcination temperatures of 500 °C and 600 °C. (Operating conditions: Operating voltage: 40 kV, Scan range: 10°–70°, Scan speed: 10° min−1, Cu target)

Full size image

On the basis of experimental results in Fig. 2, we further investigated the effect of the mass ratios of ZnAl2O4 to ZrO2 on the crystal structure of SO42−/ZnAl2O4–ZrO2 composite solid acids at the calcination temperature of 600 °C. As shown in Fig. 2, the characteristic diffraction peaks of ZnAl2O4 spinel were still evident in both SO42−/ZnAl2O4 and SO42−/ZnAl2O4–ZrO2 composite solid acids owing to its single crystal shape and its stable structure. On the contrary, the mass ratios of ZnAl2O4 to ZrO2 had the obvious influence on the crystal structure of ZrO2. Both active tetragonal ZrO2 and inactive monoclinic ZrO2 coexisted in SO42−/ZrO2 and SO42−/ZnAl2O4–ZrO2 (2:8). However, it could be clearly observed that the diffraction peaks of inactive monoclinic ZrO2 were gradually reduced with increasing the mass ratio of ZnAl2O4 to ZrO2 in SO42−/ZnAl2O4–ZrO2 composite solid acids. Finally, the distinct inactive monoclinic phase almost disappeared and the active tetragonal ZrO2 still presented in SO42−/ZnAl2O4–ZrO2 (4:6) [35]. This above results further proved that the addition of ZnAl2O4 could effectively retard the crystal transformation of ZrO2 from the active tetragonal phase to the inactive monoclinic phase, which was in good agreement with the results of Fig. 1. As shown in Fig. 2, the characteristic diffraction peaks of ZnAl2O4 spinel became stronger and the characteristic diffraction peaks of ZrO2 tetragonal phase gradually became weaker with increasing the mass ratio of ZnAl2O4 to ZrO2. At last, the characteristic diffraction peaks of ZrO2 tetragonal phase almost disappeared in SO42−/ZnAl2O4–ZrO2 (6:4) and SO42−/ZnAl2O4–ZrO2 (8:2). The possible reasons for this result might be that both ZnAl2O4 and ZrO2 were well incorporated in the crystal lattice of SO42−/ZnAl2O4–ZrO2. On the other hand, ZrO2 might be covered with the surface of ZnAl2O4 in SO42−/ZnAl2O4–ZrO2 according to SEM results. Combining with the results of Figs. 1 and 2, we could draw an important conclusion that both the calcination temperature and the mass ratios of ZnAl2O4 to ZrO2 had a critical effect on the crystal structure of SO42−/ZnAl2O4–ZrO2 composite solid acids, which might be one of the essential reasons for their different acidic properties and their different catalytic activities. Therefore, the appropriate calcination temperature and the optimum mass ratio of ZnAl2O4 to ZrO2 were beneficial to successfully form the high performance solid acids. Besides, some weak peaks of Al2(SO4)3 and ZnSO4·H2O were observed in some samples because of the long-time impregnation and the interaction between excess SO42− and the metal ions, which might be ineffective for the catalytic activity [37].

Fig. 2
figure 2

XRD patterns of SO42−/ZnAl2O4–ZrO2 composite solid acids with the different mass ratios of ZnAl2O4 to ZrO2 at the calcination temperature of 600 °C. (Operating conditions: Operating voltage: 40 kV, Scan range: 10°–70°, Scan speed: 10° min−1, Cu target)

Full size image

SEM was performed to observe the surface morphology of SO42−/ZnAl2O4, SO42−/ZnAl2O4–ZrO2 (8:2) and SO42−/ZrO2. As shown in Figs. S1a and S1b, SO42−/ZnAl2O4 exhibited the typically sphere-like nanoparticles (~ 50 nm) and the nanoparticles assembled together. However, SO42−/ZrO2 showed the lager bulk morphology and its surface was very smooth in Figs. S1e and S1f. As shown in Figs. S1c and S1d, it was clearly observed that SO42−/ZnAl2O4–ZrO2 (8:2) composite solid acid performed the similar bulk morphologies with SO42−/ZrO2. In the meantime, the sphere-like ZnAl2O4 were highly dispersed on the surface of ZrO2 bulk. This result further demonstrated that both ZnAl2O4 and ZrO2 were successfully composited in SO42−/ZnAl2O4–ZrO2 (8:2), which was in good accordance with the result of XRD. Especially, SO42−/ZnAl2O4–ZrO2 (8:2) composite solid acid still kept the advantageous structures of ZrO2. Moreover, ZrO2 inhibited the aggregation of ZnAl2O4 particles and facilitated the dispersion of ZnAl2O4 on the surface of SO42−/ZnAl2O4–ZrO2 (8:2) composite solid acid. According to Figs. S1a, S1c and S1e, it could be found that the morphology of SO42−/ZnAl2O4–ZrO2 (8:2) relatively became the loosest among these catalysts, which was advantageous for heterogeneous reactions.

It is well known that the active acid center of SO42−/MxOy solid acid comes from the coordination adsorption of SO42− with the surface metal ions of MxOy. Correspondingly, the characteristic absorption peak associated with acid center structure will appear in FT-IR spectra. Based on this consideration, Fig. 3 shows the FT-IR spectra of SO42−/ZnAl2O4, SO42−/ZrO2 and SO42−/ZnAl2O4–ZrO2 composite solid acids with the different mass ratios ZnAl2O4 to ZrO2. The specific bands in the range of 900 and 1400 cm−1 for all the samples were formed by corresponding to their active acid structures on the surface of the samples, which was attribute to the strong interaction between the sulfuric groups and the metal ions [38]. The formation of the active acid center structure was the essential reason for their certain acid catalytic activities of the samples. Among them, the band at ~ 1226 cm−1 due to the stretching vibration of S=O was detected in all samples, indicating that a chelating bidentate structure was formed between the surface sulfate species and the metal ions [39, 40]. Such a bidentate structure is believed to be a driving force in the generation of many acidic sites on the surface of SO42−/MxOy solid acids, making the samples possess super acidity. Additionally, two bands at 1057 and 1152 cm−1 in SO42−/ZnAl2O4 were assigned to the symmetric and asymmetric stretching vibration of S–O, respectively. Nevertheless, these two bands were shifted to 1021 and 1107 cm−1 for SO42−/ZrO2, which might be owing to their different linked M–O species in SO42−/ZnAl2O4 and SO42−/ZrO2 [40, 41]. These above relevant bands attributed to the symmetric and asymmetric stretching vibration of S–O were all maintained in SO42−/ZnAl2O4–ZrO2 composite solid acids, demonstrating that both ZnAl2O4 and ZrO2 belonged to their active component and were involved in their formation of acid centers. It was important to note that the synergistic promotion between ZnAl2O4 and ZrO2 might benefit to adjust the comprehensive acidities of SO42−/ZnAl2O4–ZrO2 composite solid acids, which was also proved by the following experimental results of NH3-TPD and NH3 adsorption FT-IR spectra. In addition, the distinct bands at 505, 563 and 670 cm−1 ascribed to Zn–O–Al, Al-O and Zn–O vibrations for ZnAl2O4 spinel were observed in SO42−/ZnAl2O4 and SO42−/ZnAl2O4–ZrO2 composite solid acids, which was in good agreement with the XRD results [42, 43]. Two bands at 600 cm−1 and 725 cm−1 ascribed to Zr–O stretching vibrations were detected in SO42−/ZrO2 and partial SO42−/ZnAl2O4–ZrO2 [44, 45], which were well in agreement with the XRD analysis. Besides, the bands located at 1640 cm−1 and 3400 cm−1 resulted from the bending and stretching mode of the OH group of water molecules, respectively.

Fig. 3
figure 3

FT-IR spectra of SO4−/ ZnAl2O4–ZrO2 composite solid acids with the different mass ratios of ZnAl2O4 to ZrO2. (Experiment conditions: the calcination temperature of 600 °C; Operating conditions: Scan range: 400–4000 cm−1, KBr pellet)

Full size image

It was generally accepted that the surface acidic types of SO42−/MxOy solid acids were significant to their acid catalytic performances [46, 47]. NH3 is frequently used as the probe molecule in the FT-IR spectra to distinguish the Brönsted acid sites from the Lewis acid sites by the means of NH3 adsorption FT-IR spectra. NH3 interacts with the Brönsted acid sites to generate surface NH4+* and the asymmetric bending vibration of surface NH4+∗ will appear at ~ 1400 cm−1 in NH3 adsorption FT-IR spectra. The lone pair electrons on NH3 is coordinated with the Lewis acid sites to form NH3* and the symmetric bending vibrations of surface NH3∗ will be shown at ~ 1115 cm−1 in NH3 adsorption FT-IR spectra [48, 49]. As shown in the NH3 adsorption FT-IR spectra of Fig. 4, it was obviously observed that the strong bands at 1400 and 1115 cm−1 were all detected in SO42−/ZnAl2O4, SO42−/ZrO2, SO42−/ZnAl2O4–ZrO2 (2:8) and SO42−/ZnAl2O4–ZrO2 (8:2), indicating that both Brønsted acid sites and Lewis acid sites coexisted on the surfaces of the samples. In terms of SO42−/ZrO2, the relative strength of Brønsted acid sites was greater than that of Lewis acid sites. For SO42−/ZnAl2O4, the relative strength of Lewis acid sites was greater than that of Brønsted acid sites. It should be emphasized that the combination of ZnAl2O4 with ZrO2 was obviously good for regulating the ratio of Brønsted/Lewis acid sites in SO42−/ZnAl2O4–ZrO2. Among the samples, SO42−/ZnAl2O4–ZrO2 (8:2) showed the highest mass ratio Brønsted/Lewis acid sites and the most number of acid sites, which might be one of the essential reasons for its higher catalytic activity in acid catalyzed esterification of oleic acid with methanol. This result was manifested that the combination of ZnAl2O4 and ZrO2 might be more beneficial to adjust the acid type by compared with the traditional SO42−/ZrO2 solid acid.

Fig. 4
figure 4

The NH3 adsorption FT-IR spectra of SO42−/ZrO2, SO42−/ZnAl2O4 and SO42−/ZnAl2O4–ZrO2 solid acids. (Conditions: The sample was pre-treated in a He flow of 20 mL min−1 at 300 °C for 1 h. Then the dried sample as cooled to 60 °C and exposed to NH3 of 20 mL min−1 at room temperature for 1 h, and followed by heated up to 120 °C and flowed by helium of 20 mL·min−1 for 1 h to remove gas-phase and physically adsorbed NH3. Operating conditions: Scan range: 1000–1500 cm−1, KBr pellet)

Full size image

The acid strength distribution obtained from the NH3-TPD of SO42−/ZnAl2O4, SO42−/ZnAl2O4–ZrO2 (8:2) and SO42−/ZrO2 solid acids are shown in Fig. 5. Generally, the desorption temperature was closely related to the acid strength. The higher the desorption peak temperatures was, the stronger the acid strength was [50]. Meanwhile, the area of the desorption peak corresponded to the number of active acid centers. The bigger the area was, the more active acid centers were. According to the desorption temperature, the acid strength was divided into the weak (120 ~ 250 °C), the medium (250 ~ 400 °C) and the strong (> 400 °C) strength. As shown in Fig. 5, SO42−/ZnAl2O4–ZrO2 (8:2) showed the prominent broad desorption peaks in the range of 120 to 475 °C, which suggested the presence of the weak, the moderate and the strong acidic sites. According to Fig. 5, there were some weak acid sites and medium acid sites on the surface of SO42−/ZnAl2O4. However, there were only the strong acid sites on the surface of SO42−/ZrO2. In the meantime, the acid site density of these samples was further studied by the acid–base titration (shown in Table 1). Among them, SO42−/ZnAl2O4 showed the lowest acid site density. As shown in Table 1, it was worth mentioning that the mass ratio of ZnAl2O4 to ZrO2 had a significant effect on the acid site density. SO42−/ZnAl2O4–ZrO2 (8:2) had the highest density of acid sites, which might be related to the better dispersion of two components and the synergy between two components of ZnAl2O4 and ZrO2 [51]. These above results further demonstrated that both ZnAl2O4 and ZrO2 in SO42−/ZnAl2O4–ZrO2 were active components and participated in the formation of active acid centers. So, it might be an effective method to regulate the synthetic acid properties and the catalytic activities by additions of ZnAl2O4 to ZrO2 in SO42−/ZnAl2O4–ZrO2, which were also in good consistent with the results of NH3 adsorption FT-IR. However, the excess ZrO2 had a negative influence on the acid site density. The acid site density of SO42−/ZnAl2O4–ZrO2 decreased with increasing the mass ratio of ZrO2 to ZnAl2O4. SO42−/ZnAl2O4–ZrO2 (4:6) and SO42−/ZnAl2O4–ZrO2 (2:8) revealed the little lower acid site density by compared with SO42−/ZrO2. Combining the results of Figs. 3, 4, 5 and Table 1, we could draw an important conclusion that the suitable mass ratio of ZrO2 to ZnAl2O4 played a key role in the comprehensive acidic properties of SO42−/ZnAl2O4–ZrO2 composite solid acids, which was also reported in other SO42−/ZrO2–MxOy composite solid acids [52].

Fig. 5
figure 5

NH3-TPD profiles of SO42−/ZnAl2O4, SO42−/ZnAl2O4–ZrO2 (8:2) and SO42−/ZrO2 solid acids. (Conditions: The sample was pretreated in a He flow of 20 mL min−1 at 300 °C for 1 h. Then the dried sample as cooled to 60 °C and exposed to NH3 of 20 mL min−1 at room temperature for 1 h, and followed by heated up to 120 °C and flowed by helium of 20 mL min−1 for 1 h to remove gas-phase and physically adsorbed NH3. After removing the physically adsorbed NH3, measurements were started in helium flow of 20 mL·min−1 with a heating rate of 10 °C min−1)

Full size image
Table 1 The acid site density of SO42−/ZrO2, SO42−/ZnAl2O4 and SO42−/ZnAl2O4–ZrO2
Full size table

Figs. 6 and 7 give the TG analysis of SO42−/ZnAl2O4–ZrO2 (8:2) and SO42−/ZrO2, which was used to compare their thermostability. The first weight loss below 200 °C could be assigned to the removal of the physically adsorbed water. The second weight loss in the following temperature range of 200–500 °C was related to the removal of the structure water. The third weight loss at the higher temperature range between 500 and 1000 °C was due to the gradual decomposition of the sulfur species on the surface of the samples [53]. So, the third weight loss above 500 °C was used to estimate the sulfur content and the amounts of the acid sites on the surface of the samples. Accordingly, the more the weight lost, the more the sulfate groups existed. As shown in Fig. 6, SO42−/ZnAl2O4–ZrO2 (8:2) gave the mass weight loss of 32.8% above 500 °C. By contrast, SO42−/ZrO2 had the relatively little weight loss of 26.5% above 500 °C (as shown in Fig. 7). This above result revealed that the cooperation of ZnAl2O4 and ZrO2 was beneficial to improve the number of acid center and the surface resistance to sulfur loss. As a result, SO42−/ZnAl2O4–ZrO2 (8:2) had the higher catalytic activity.

Fig. 6
figure 6

TG curves of SO42−/ZnAl2O4–ZrO2 (8:2). (Operating conditions: Scan range: from room temperature to 1000 °C, Heating rate: 20 °C min−1)

Full size image
Fig. 7
figure 7

TG curves of SO42−/ZrO2. (Conditions: Scan range: from room temperature to 1000 °C, Heating rate: 20 °C min−1)

Full size image

The surface composition and the oxidation state of SO42−/ZnAl2O4–ZrO2 (8:2) composite solid acid were further investigated by the means of XPS. As shown in Fig. S2, Zn, Zr, Al, O, S and C elements were all detected in SO42−/ZnAl2O4–ZrO2 (8:2). The observed peak of C 1s might originate from the signal carbon in the instrument and was used for the calibration. The broad O1s peak was consisted of three distinct peaks at 530.1, 531.5 and 532.5 eV, which were ascribed to the lattice oxygen of the oxides contribution, the oxygen of OH species and the sulfate oxygen, [54].The Zn 2p3/2 and Zn 2p1/2 binding energies were located at 1021.5 eV and 1044.8 eV, which were close to the standard data for Zn2+. Fig. S2e shows the Al 2p peak of the catalyst at 75.3 eV, which was assigned to the Al3+ ion [55]. The peaks at 183.5 eV and 185.9 eV correspond to Zr 3d5/2 and Zr 3d3/2, respectively, indicating that Zr species existed as the formation of Zr(IV) [56]. With reference to the XRD result, the intensity of ZrO2 diffraction peaks became no apparent in SO42−/ZnAl2O4–ZrO2 (8:2). So, the detection of the Zr in XPS further evidenced that ZrO2 might be highly dispersed on the surface of the sample. Additionally, it was necessary to emphasize that the peak at 169.5 eV corresponding to S 2p binding energy was clearly observed in SO42−/ZnAl2O4–ZrO2 (8:2), which was attributable to the sulfur oxidation state of + 6 [57]. It is well known that S6+ contributes to the formation of the surface acid sites. The suction-induced complex S=O promotes the electron-accepting ability for the metal atoms, making the sample possess supper acid. Accordingly, IR spectra of SO42−/ZnAl2O4–ZrO2 (8:2) showed the special bands of the active acid structures in the range of 900–1400 cm−1. XPS analysis further confirmed that the surface active sulfur species, ZnAl2O4 and ZrO2 coexisted in SO42−/ZnAl2O4–ZrO2 (8:2) composite solid acid.

Catalytic activities

According to XRD results, the mass ratios of ZnAl2O4 to ZrO2 had a key influence on the crystal structure of SO42−/ZnAl2O4–ZrO2 composite solid acids, which is a very critical factor to affect their acid catalytic performance. Based on this consideration, the catalytic activities of SO42−/ZnAl2O4–ZrO2 composite solid acids with the different mass ratios of ZnAl2O4 to ZrO2 in the esterification reaction of oleic acid with methanol were shown in Fig. 8. It was clearly observed that all the samples displayed the certain catalytic activities in esterification reaction of oleic acid with methanol, which resulted from their formation of active acid center structure. Compare with SO42−/ZrO2 and SO42−/ZnAl2O4, the catalytic activities of SO42−/ZnAl2O4–ZrO2 composite solid acids were effectively modified by combination of ZnAl2O4 and ZrO2, which might be owing to their different acidic properties on the basis of the NH3 adsorption FT-IR spectra, TG, NH3-TPD and acid–base titration analysis. In view of its optimal synthetic acid properties among the samples, SO42−/ZnAl2O4–ZrO2 (8:2) exhibited the highest catalytic activities with more than 80% oleic acid conversion in the esterification reaction of oleic acid with methanol. On this basis, the kinetic profile of the esterification reaction over SO42−/ZnAl2O4–ZrO2 (8:2) composite solid acid at different reaction temperature of 60, 65 and 70 °C was obtained with the different reaction time. As shown in Fig. 9, it was obviously discovered that the conversion of oleic acid was increased with the increase of the reaction temperature, suggesting that the esterification reaction was be assigned to the kinetically controlled reaction. Take the boiling point of methanol into account, 65 °C was selected as the optimum temperature for the esterification reaction of oleic acid with methanol.

Fig. 8
figure 8

The catalytic activities of SO42−/ZnAl2O4–ZrO2 with the different mass ratios of ZnAl2O4 to ZrO2 at the calcination temperature of 600 °C. (Reaction conditions: the reaction temperature was 65 °C, the molar ratio of oleic acid to methanol was 1:25, the reaction time was 8 h, the amount of catalysts was 5 wt%)

Full size image
Fig. 9
figure 9

The catalytic activities of SO42−/ZnAl2O4–ZrO2 (8:2) at the different reaction temperature. (Reaction conditions: the reaction temperature range was 60–70 °C, the molar ratio of oleic acid to methanol was 1:25, the reaction time was 8 h, the amount of catalysts was 5 wt%)

Full size image

The esterification reaction of carboxylic acid [A] with alcohol [M] to form esters [E] and water [W] over catalysts is given as Eq. 2:

$$\text{A+M}\stackrel{{\text{K}}_{\text{s}}}{\iff }\text{E+W}$$
(2)

In order to simplify the kinetic model, the following assumptions were built [28]:

(1) The rate of reaction without catalyst was ignored; (2) The internal and external diffusion effects of matter were ignored; (3) The surface reaction was the rate control step; (4) Both the forward and backward reactions belonged to second-order reactions. The overall rate can be expressed as Eq. 3:

$$r={k}_{\mathrm{s}}\left[{\text{A}}\right]\left[\mathrm{M}\right]-{k}_{-\mathrm{s}}\left[\mathrm{E}\right]\left[\mathrm{W}\right]$$
(3)

The ks represented the rate constant for the forward reaction and the \({k}_{-\mathrm{s}}\) was the rate constant for the reverse reaction.

The following relationship between reactant and product concentrations was followed as Eq. 4:

$$\begin{array}{ccccc} &n = \frac{{{{\left[ {\rm{M}} \right]}_{\rm{0}}}}}{{{{\left[ {\rm{A}} \right]}_{\rm{0}}}}}\\ &\left[ {\rm{A}} \right] = {\left[ {\rm{A}} \right]_0}({\rm{1}} - {\rm{X}})\quad \left[ {\rm{M}} \right] = {\left[ {\rm{A}} \right]_0}({\rm{n}} - {\rm{X}})\\ &\left[ {\rm{E}} \right] = {\left[ {\rm{A}} \right]_0}X\quad \left[ {\rm{W}} \right] = {\left[ {\rm{A}} \right]_0}X \end{array}$$
(4)

Symbol n represents the molar ratios of alcohol to carboxylic acid. The X represented the conversion of the acid (< 1). The Eq. 3 can be expressed as in Eq. 5:

$$r={k}_{\mathrm{s}}{\left[{\text{A}}\right]}_{0}^{2}\left(1 -X\right)\left(n-X\right)-{k}_{-\mathrm{s}}{\left[{\text{A}}\right]}_{0}^{2}{X}^{2}$$
(5)

The forward reaction rate was much greater than the reverse reaction rate, so Eq. 5 can be simplified as Eq. 6:

$$r={{k}_{\mathrm{s}}\left[{\text{A}}\right]}_{0}^{2}(1 - X)\left(n-X\right)$$
(6)

Equation 6 can be integrated to obtain Eq. 7:

$$\mathrm{ln}\frac{n-X}{n\left(1-X\right)}={\left[{\text{A}}\right]}_{0}\left(n-1\right){k}_{\mathrm{s}}t$$
(7)

Equation (7) was written as Eq. (8) for nonlinear fitting:

$$\frac{n-X}{ n\left(1-X\right)}={e}^{{\left[{\text{A}}\right]}_{0}\left(n-1\right){k}_{\mathrm{s}}t}$$
(8)

As shown in Fig. S3, the good nonlinear fitting results indicated that the esterification reaction of oleic acid with methanol conformed to the second order kinetic model. Correspondingly, Table 2 gave the forward rate constants (ks) at different reaction temperature on the basis of these lines. The activation energy (Ea) could be calculated according to the Arrhenius equation of Eq. (9), which gave the dependence of the forward rate constant on the reaction temperature.

Table 2 Kinetic parameters of esterification reaction
Full size table
$$-{\text{ln}}{\text{k}}_{\text{s}}\text{=}\frac{{\text{E}}_{\text{a}}}{\text{RT}}-{\text{ln}}{\text{A}}$$
(9)

R was the gas constant (8.314 J mol−1 K−1), T was the thermodynamic temperature (K), Ea was the activation energy (kJ mol−1) and A was the pre-exponential factor (s−1).

According to the kinetic calculation, the activation energy was 37.5 kJ mol−1 for the esterification reaction of oleic acid with methanol, which was the relatively lower than H3PW12O40 and H2SO4 (5% and 10%w/w) with the activation energy over 50 kJ mol−1 [58]. Combining kinetic study and thermodynamic analysis, these above results strongly demonstrated that SO42−/ZnAl2O4–ZrO2 (8:2) composite solid acid could effectively catalyze the typical esterification reaction of oleic acid with methanol for the synthesis of green biodiesel.

It is well known that traditional SO42−/MxOy solid acid can perform the high initial activities. However, they always suffer from rapid deactivation and short lifetime owing to the loss of surface active sulfur, the deposition of surface carbon and the crystal transformation of active carrier. Based on these above considerations, SO42−/ZnAl2O4–ZrO2 (8:2) composite solid acid was recycled to study its reusability, which is presented in Fig. 10. Compared with SO42−/ZrO2, SO42−/ZnAl2O4–ZrO2 (8:2) composite solid acid obviously showed the better reusability. The conversion of oleic acid still remained above 75% after being used for four times, suggesting that the synergistic effect of both ZnAl2O4 and ZrO2 was beneficial to the improve the reusability of SO42−/ZnAl2O4–ZrO2 (8:2) composite solid acid.

Fig. 10
figure 10

Reusability of SO42−/ZnAl2O4–ZrO2 (8:2) and SO42−/ZrO2 for esterification reaction of oleic acid with methanol. (Reaction conditions: the reaction temperature was 65 °C, the molar ratio of oleic acid to methanol was 1:25, the reaction time was 8 h, the amount of catalysts was 5 wt%.The used catalysts were recovered by filtering and drying, after completing each reaction)

Full size image

In order to further explore the essential reasons for its better reusability and its slight deactivation, the fresh and used SO42−/ZnAl2O4–ZrO2 (8:2) composite solid acid were characterized by means of XRD, TG, direct and NH3 adsorbed FT-IR. As shown in Figs. S4, S5, S6 and S7, the used SO42−/ZnAl2O4–ZrO2 (8:2) composite solid acid had no evident changes in the intensity and characteristics peaks in XRD and FT-IR analysis by compared with the fresh catalyst. This above result indicated that used SO42−/ZnAl2O4–ZrO2 (8:2) still kept its major phase structure, its active acid structures and its acid type on the surface of the sample. As a result, SO42−/ZnAl2O4–ZrO2 (8:2) showed the better structural stability and the better stability of the surface active sites, which was the essential reason for its higher reusability after recovery process. Additionally, the peaks ascribed to the ZnSO4·H2O was obviously decreased in the used catalysts and the used catalyst still performed its higher activities, suggesting that ZnSO4·H2O was not the active component. Moreover, the slight deactivation reason for SO42−/ZnAl2O4–ZrO2 (8:2) during the acid catalyzed esterification was also deeply investigated in this paper. The used SO42−/ZnAl2O4–ZrO2 (8:2) had a slight decrease in the intensity of Brønsted and Lewis acid sites bands in NH3 adsorbed FT-IR spectra, indicating that the number of active acid centers decreased during the acid catalyzed esterification reaction. According to TG analysis, the surface sulfur weight percentages of the fresh and used SO42−/ZnAl2O4–ZrO2 (8:2) were 32.8% and 21.0%. The comprehensive analysis of TG and Table 1 results certified that the inevitable loss of surface active sulfur species occurred in SO42−/ZnAl2O4–ZrO2 (8:2), which might be the major reason for its slight deactivation in the process of the acid catalyzed esterification reaction.

Conclusion

A novel series of SO42−/ZnAl2O4–ZrO2 composite solid acid were successfully used to catalyze the typical esterification reaction of oleic acid with methanol for biodiesel synthesis. The addition of ZnAl2O4 successfully retarded the crystal transformation of ZrO2 from the active tetragonal phase to the inactive monoclinic phase. Moreover, both the calcination temperature and the mass ratios of ZnAl2O4 to ZrO2 had a critical effect on the crystal structure of SO42−/ZnAl2O4–ZrO2 composite solid acids, which resulted in their different acidic properties and their different catalytic activities. In the meantime, both ZnAl2O4 and ZrO2 acted as active components and participated in the formation of active acid center structure for SO42−/ZnAl2O4–ZrO2 composite solid acids. As a result, the suitable mass ratio of ZrO2 to ZnAl2O4 benefited to adjust the comprehensive acidic properties of SO42−/ZnAl2O4–ZrO2 composite solid acids, such as the acid type, the acid strength and the number of active acid centers. Among them, the optimal SO42−/ZnAl2O4–ZrO2 (8:2) composite solid acids exhibited the highest catalytic activities with more than 80% oleic acid conversion in esterification of oleic acid with methanol. Additionally, the kinetic study indicated that SO42−/ZnAl2O4–ZrO2 (8:2) composite solid acid showed the highly efficient for the esterification of oleic acid with methanol because of its lower activation energy value of 37.5 kJ mol−1. Particularly, the optimal SO42−/ZnAl2O4–ZrO2 (8:2) had the obviously better reusability with above 75% conversion of oleic acid after being used for four times in the esterification reaction of oleic acid with methanol owing to its excellent structural stability and its better stability of the surface active sites. It was worth mentioning that the used SO42−/ZnAl2O4–ZrO2 (8:2) had a slight decrease in the intensity of Brønsted and Lewis acid sites bands in NH3 adsorption FT-IR spectra, indicating that the number of active acid centers decreased during the acid catalyzed esterification reactions. The Further TG result showed that the inevitable loss of sulfate species on the surface of SO42−/ZnAl2O4–ZrO2 (8:2) resulted in its slight deactivation during the acid catalyzed esterification reaction. The obtained results would provide reference value for designing and synthesizing new composite solid acids with adjustable comprehensive acidities and excellent acid catalytic performance by combination of ZnAl2O4 with ZrO2, which might have a broad application prospect in the field of biodiesel synthesis by acid catalyzed esterification reactions.