Abstract
The Ni-ZSM-5 catalyst, under four different factors of Swida wilsoniana pyrolysis products of catalytic hydrogenation, GC-MS, FI-IR, and elemental analyzers, was used to identify the elements, carbon chain distribution, and composition of the products. The effects of reaction temperature, reaction pressure, and catalyst hydrogenation level on the conversion rate were investigated. The reaction pressure and the amount of catalyst are the main factors that affect the conversion of the pyrolysis products into biofuels. Using 1.05 wt.% Ni-ZSM-5 catalyst, the highest conversion rate was 98.10% at 173°C and 2.00 MPa. The results show that Swida wilsoniana-decomposed products can be converted to high-quality biofuels by catalytic hydrogenation of Ni-ZSM-5 and can be used as an alternative energy source.
1. Introduction
The development of contemporary human civilization and economy is based on fossil energy. However, the excessive development and use of human resources and the depletion of fossil energy reserves cause a series of problems such as the greenhouse effect, atmospheric pollution, and water pollution [1–3]. In this context, energy security is currently a hot spot of global concern [4]. Biomass energy has the advantages of reproducibility, recyclability, and no harm to the environment. Biomass energy is known as green energy, and it became one of the important contents of the energy strategies of countries in the world today [5, 6]. Researchers have paid close attention to the production of clean biohydrocarbon fuels from vegetable oils recently. Because of the versatility of biohydrocarbon fuels, it can be used as a substitute for fuel or as an additive for fossil fuel, and it also provides solutions for the future energy and environmental challenges [7].Vegetable oil can be used to prepare biohydrocarbon fuel through transesterification, esterification, pyrolysis, gasification, and catalytic cracking reactions [8]. The current process for preparing biofuels from vegetable oils is transesterification to produce biodiesel (fatty acid methyl ester), but it has been found that the biodiesel has the disadvantages of low calorific value, high kinematic viscosity, and low temperature stability [9, 10].
Corn oil and soybean oil are used to produce biodiesel, but they occupy the resources of edible oil [11, 12]. Therefore, the development of new resources to produce biodiesel has attracted the attention of many researchers, such as Mangiti oil, baobab oil, date palm oil, Jatropha seed oil, Manchurian apricot oil, and Siberian apricot Oil [13–20]. In a study, Xanthium strumarium plant stems have been pyrolyzed with catalysts (ulexite, colemanite, and borax) and without catalysts at temperatures of 350–550°C. The catalyst and temperature were found to be effective on the conversion. The highest liquid product yield is obtained with colemanite catalyst at 550°C as 27.97%. It has been identified that the liquid product comprised aliphatic, aromatic, and heterocyclic compounds [21]. Another study found that the liquid and solid products obtained by pyrolysis of Lactuca scariola under different catalysts and temperatures are products with higher energy value and can also be used as alternative energy sources [22]. As a result of the study, black cumin seed cake (BCSC) was converted into liquid and solid products with the pyrolysis method. The highest conversion rate was obtained in the presence of Al2O3 (78.91%) and SnCl4·5H2O (76.06%) catalysts at 500°C [23]. In these vegetable oils, Swida wilsoniana oil is a very useful oil which has high oil content [24, 25]. There are 77.68% the fatty acid oleic acid and linoleic acid in Swida wilsonian oil [26], so it can be used as a raw material for the production of biodiesel.
Swida wilsoniana oil is mainly prepared by thermal cracking, catalytic cracking, and transesterification [27, 28]. Swida wilsoniana oil from direct thermal cracking and catalytic cracking has low temperature fluidity and low calorific value, which limits its wide application [29–32]. At present, it has received widespread attention that the first-generation biodiesel converts to hydrocarbon-rich liquid fuel by catalytic hydrogenation. Ouyang et al. [33] used fatty acid methyl ester hydride-oxygenation to prepare second-generation biodiesel with a conversion rate of 99.52 %. Zuo Hualiang et al. [34] found that Ni-supported catalysts have higher hydride-oxygenation performance, while maintaining higher alkane selectivity. Dhanalaxmi et al. have successfully designed a porous organic polymer-encapsulated (PPTPA-1) nanohybrid magnetically retrievable Pd-Fe3O4 catalyst in a one-step solvothermal route with a Pd: Fe (1 : 9) ratio which exhibited an outstanding higher activity over their monometallic counterparts for selective LA hydrogenation to GVL, a key platform molecule in many biorefinery schemes [35]. An experimental data suggested that a novel porous organic network material (TpPON) can be synthesized via a one-step acid-catalyzed condensation reaction between 1, 3, 5-triformylphloroglucinol and triphenylamine. Ru NPs have been successfully deposited at the surfaces of TpPON to obtain the Ru nanoparticles-decorated Ru@TpPON, which showed excellent catalytic activity in the hydrodeoxygenation of various oils to furnish long-chain alkanes (biodiesel) [36]. Wang et al. [37] used soybean oil as the raw material supported Ni and Mo metals on the ZSM-5 carrier to prepare liquid fuel. Wang Fei et al. [38] found that when a molecular sieve is used as a carrier in the hydrogenation of oils and fats, a higher yield of isomerized hydrocarbon products could be obtained. The key to get a higher yield is to choose the right oil hydrogenation catalyst [39]. Ni as an inexpensive transition metal has received widespread attention due to the high hydrogenation activity [40, 41].
Therefore, the purpose of this research is to study the interaction between independent variables (catalyst addition level, reaction temperature, and reaction pressure) and dependent variables by the response surface methodology and optimize the preparation conditions. On the other hand, we found that biohydrocarbon fuels produced by hydrogenation will be closer to the existing fossil fuels.
2. Materials, Instruments, and Methods
2.1. Materials and Instruments
Swida wilsoniana pyrolysis products were prepared from the laboratory of the Hunan Academy of Forestry Science. The ZSM-5 zeolite molecular sieve (Si : Al = 23 : 1) was from Alfa Essa Chemical Co., Ltd., China. Hydrogen (purity: 99.99%) was purchased from Huazhong Special Gas Co., Ltd., Hunan. Ni (NO3)2·6H2O and other chemicals were of analytical grade and purchased from Sinopharm Chemical Reagent Co., Ltd., China.
A controller (Parr 4848, Parr Instrument Co., Ltd.); high-pressure reactor (Parr 4568, Parr Instrument Co., Ltd.); Fourier infrared spectrometer (IS5, Thermo Fisher, USA); electronic analytical balance (AUY-220, Shimadzu Co., Ltd.); single four-stage bar GC-MS (Scion-SQ, Bruker Co., Ltd.); and element analyzer (Vario EL-III, Elementar Co., Germany) were used.
2.2. Preparation of the Ni-ZSM-5 Catalyst
The Ni-ZSM-5 catalyst loaded with nickel metal was prepared by the equal volume impregnation method: First, 15 g ZSM-5 was weighed into a quartz crucible and activated in a muffle furnace at 400°C for a certain period of time. Then, the nickel nitrate hexahydrate solution was mixed and stirred at a constant temperature overnight. After being left to stand for 10.0 h, it was dehydrated and dried. It continued to be roasted at a certain temperature to a set time and, finally, waited for cooling, milling, sieving, and reduction.
2.3. Method for Preparing Swida wilsoniana Pyrolysis Products
Swida wilsoniana oil was used as the raw material for the catalytic pyrolysis reaction. The specific operation steps are as follows: 250 g of Swida wilsoniana oil and 1.0 wt.% La2O3 are introduced into a straight three-port glass reactor, mixed evenly, and sealed with a stopper and heated to 500°C, and the collected liquid fuel is condensed by using the condenser after 80 minutes, which is the product of the pyrolysis of Swida wilsoniana.
2.4. Hydrogenation Effect of Pyrolysis Products
The hydrogenation effect of the pyrolysis product of the Swida Wilsoniana was used to evaluate the conversion rate of unsaturated components by the change of the iodine value before and after the reaction. The iodine value was determined according to the Chinese National Standard GB5532-2008.where is the iodine value of the product after the reaction, g/100 g; is the initial iodine value of the Swida wilsoniana pyrolysis products, g/100 g.
2.5. Method for Catalytic Hydrogenation of Pyrolysis Products
2.5.1. Single-Factor Design for Conversion Rates
The factors influencing the hydrogenation of pyrolysis products are the addition of a catalyst, reaction temperature, reaction pressure, and reaction time. First, we add 100 g of the pyrolysis product and the Ni-ZSM-5 catalyst (0.4 wt.%, 0.6 wt.%, 0.8 wt.%, 1.0 wt.%, 1.2 wt.%) in a high-pressure reactor. Second, it is connected to nitrogen gas for 8 to 10 min. Third, the vent valve is closed, and the parameters of the high-pressure reactor are adjusted from the operating end: reaction pressure (1.0 MPa, 1.5 MPa, 2.0 MPa, 2.5 MPa, and 3.0 MPa), reaction temperature (110°C, 140°C, 170°C, 200°C, and 230°C), after the reactor reaches the corresponding conditions, and start timing (60 min, 90 min, 120 min, 150 min, and 180 min) of the catalytic hydrogenation reaction to the end of the reaction. Finally, the condensed gas collected by using the collector is the biofuel; Figure 1 represents the catalytic hydrogenation process flow.
2.5.2. Response Surface Methodology (RSM) Experimental Design
Software Design-Expert (Trial Version 10.0.1.0, Stat-Ease Inc., Minneapolis, USA) was employed for experimental design. Based on the single-factor test, the catalytic addition level, reaction temperature, and reaction pressure were selected as independent variables, and the conversion rate of unsaturated components was the dependent variable. The independent variables and their levels are presented in Table 1. Similarly, the results of the whole design comprising 17 experimental points performed in randomized order are presented in Table 2.
2.6. Product Characterization Methods
2.6.1. Elemental Analysis
A constant element analyzer was used to detect the content of carbon, nitrogen, and hydrogen in pyrolysis products and biohydrocarbon fuels, and, then, the oxygen content was obtained by subtraction.
2.6.2. Infrared Spectroscopy
The detector is a midinfrared DTGS detector with a wavenumber scanning range of 400–4000 cm−1 and a resolution of 4.0 cm−1.
2.6.3. GC-MS Analysis
The components of hydrocarbon biofuels analysis were performed by GC-MS with a Scion-SQ single quadrupole mass spectrometer and electron bombardment (EI) ion source; the electron energy is 70 eV, the quadrupole temperature is 150°C, the ion source temperature is 230°C, and the mass scanning range is 33 ∼ 350 u.
2.6.4. Performance Analysis
The product was analyzed according to the Chinese National Standard GB19147-2016 (vehicle diesel).
3. Results and Discussion
3.1. Catalytic Characterization Analysis
3.1.1. N2 Adsorption-Desorption Isotherms Analysis
Figure 2 shows the N2 adsorption-desorption isotherms of the ZSM-5 and Ni-ZSM-5. The pore structure parameters of ZSM-5 and Ni-ZSM-5 are shown in Table 3.
(a)
(b)
It can be seen from Figure 2(b) that the isotherm of ZSM-5 is of type IV, which changes strongly at a relative pressure of 0.4 to 0.8 and has an obvious H4 hysteresis loop. The appearance of the hysteresis ring indicates that ZSM-5 is a mesoporous molecular sieve. It can be seen from Figure 2(a) that the isotherm of Ni-ZSM-5 is also of type IV, indicating that the catalyst maintains the structural characteristics of the carrier ZSM-5 after supporting the nickel metal active material and has a stable crystal frame support.
It can be seen from Table 3 that ZSM-5 has a high specific surface area (268.825 m2/g) and an average pore diameter of 2.276 nm. However, compared with the ZSM-5 carrier, the Ni-ZSM-5 specific surface area and pore volume decrease, but the change is not obvious. This is because of the accumulation effect on the ZSM-5 carrier after loading nickel. The reason why the average pore diameter becomes larger is that the nickel metal enters the carrier ZSM-5’s skeleton to replace Si or Al, and it is verified that the nickel metal is supported on the carrier ZSM-5.
3.1.2. XRD Analysis
XRD characterization of ZSM-5 and Ni-ZSM-5 is shown in Figure 3. As can be seen from Figure 3, the characteristic peak of the Ni-ZSM-5 catalyst (a) appeared at 2= 8°, 9°, 23°, 24°, and 45°, but the strength changed, indicating that the basic skeleton structure of the Ni-ZSM-5 catalyst was not damaged after loading the nickel active substance. The XRD pattern of the Ni-ZSM-5 catalyst (a) showed an obvious characteristic peak of Ni at 2 = 44.5°, 51.8°, and 76.4°, indicating that the active component Ni in the Ni-ZSM-5 catalyst had small particle size and high dispersion state, and the catalyst Ni-ZSM-5 had high catalytic activity.
3.1.3. SEM Analysis
The results of SEM characterization of ZSM-5 and Ni-ZSM-5 (21wt. %Ni), respectively, are shown in Figure 4. According to A1 and A2 in Figure 4, it can be seen that the ZSM-5 carrier is a flake smooth crystal and the overall appearance is irregular and round, which is the characteristic morphology of a typical ZSM-5 molecular sieve. Compared with ZSM-5, the Ni-ZSM-5 catalyst supported with nickel metal active substances of B1 and B2 attached spherical particles with a diameter of less than 100 nm on the grain surface, which confirmed that Ni-ZSM-5 presented nickel agglomeration, multiple active points, and good loose dispersion.
3.1.4. FT-IR Analysis
FT-IR characterization of ZSM-5 and Ni-ZSM-5 catalysts was carried out, respectively, and the results are shown in Figure 5. ZSM-5 in 1085.85 cm−1 infrared transmission peaks can be ascribed to the skeleton structure of SiO4 or AlO4 tetrahedron internal antisymmetric vibration, 790.79 cm−1 transmission peak belongs to skeleton SiO4 or internal symmetry vibration AlO4 tetrahedron, 1219.15 cm −1 transmission peak corresponds to the SiO4 or stretching vibration of AlO4 tetrahedron, and 452.92 cm−1 transmission peak corresponds to the skeleton SiO4 or bending vibration of AlO4 tetrahedron. By comparing a and b spectra, it can be seen that similar infrared characteristic peaks appear at the same position on Ni-ZSM-5, and the intensity of the peaks changes, indicating that Ni-ZSM-5 maintains the original skeleton structure and Ni successfully enters into the molecular sieve skeleton.
3.1.5. NH3-TPD Analysis
The results of NH3-TPD characterization of ZSM-5 and Ni-ZSM-5, respectively, are shown in Figure 6.
As shown in Figure 6, ZSM-5 has two wide peaks at 150 ∼ 550°C. Among them, the low-temperature peak (180°C) is generated during the desorption of NH3 due to weak adsorption. The high-temperature peak (480°C) is generated by NH3 desorption adsorbed on the strong acid position of the molecular sieve. The area of the NH3 desorption peak can be used to measure the amount of acid in the catalyst. The Ni in Ni-ZSM-5 treated with nickel-loaded active material interacts with the acid site on ZSM-5 to form strongly acidic [Ni (OH)]+ groups and form stronger acid sites, thus increasing the area of the high-temperature desorption peak. The acid content of ZSM-5 and Ni-ZSM-5 was quantitatively analyzed, and the results are shown in Table 4. It can be seen from the table that the acid content (4.8370 mmol·g−1) of the catalyst loaded with nickel active substances was significantly greater than that of ZSM-5 (3.4085 mmol·g−1).
3.2. Single-Factor Experimental Analysis
3.2.1. Effect of the Ni-ZSM-5 Addition Level on the Conversion Rate of Unsaturated Components
The effect of the Ni-ZSM-5 addition level on the conversion rate of unsaturated components is shown in Figure 7. The Ni-ZSM-5 addition level refers to the mass fraction relative to the pyrolysis product. The conversion rate was determined by applying the Ni-ZSM-5 addition level ranging from 0.4 wt.% to 1.2 wt.% with the other reaction conditions as follows: the reaction temperature was 200°C, the time was 180 min, and the pressure was 2.0 MPa. When the catalyst addition level was 1.0 wt.%, the conversion rate reached the peak, and then, when the addition level of Ni-ZSM-5 was increased, there was no obvious effect on the conversion rate. Because the addition level of the catalyst used in the early stage was small, so was the number of active sites, and the catalytic reaction cannot be fully completed within the investigation period. The conversion rate will increase with the addition level of the catalyst. Considering the cost and conversion effect, the optimal condition in the present experiment should be ranged from 0.8 wt. % to 1.2 wt.%.
3.2.2. Effect of Reaction Temperature on the Conversion Rate of Unsaturated Components
Figure 8 shows that the reaction temperature is an important factor affecting the hydrogenation reaction. With the increase of temperature (120 ∼ 170°C), the hydrogenation conversion rate of the pyrolysis products of Swida wilsoniana oil significantly increased, and their value reached the maximum when the reaction temperature was 170°C. This is because catalytic hydrogenation is an exothermic reaction, and the preheating is to provide the initial starting energy of the reaction for the hydrogenation catalyst. Once the reaction starts, other means need to be used to release part of the thermal energy.
3.2.3. Effect of Reaction Pressure on the Conversion Rate of Unsaturated Components
By fixing the Ni-ZSM-5 addition level of 1.0 wt.%, reaction temperature of 170°C, and reaction time of 180 min, respectively, the effects of reaction pressure ranging from 1.0 to 3.0 MPa were studied. Figure 9 shows that as the reaction pressure increased, the conversion rate of unsaturated components increased significantly, and after 2.0 MPa, it showed a small increase. The reason was that the higher the pressure, the greater the solubility of the pyrolysis products of Swida wilsoniana. The contact area of hydrogen with the catalyst became larger, which in turn, accelerated the reaction rate and promoted the chemical equilibrium to proceed in the positive direction. However, too high pressure could also increase the equipment cost, so the reaction pressure was chosen to be 2.0 MPa.
3.2.4. Effect of Reaction Time on the Conversion Rate of Unsaturated Components
Under the conditions of the Ni-ZSM-5 addition level of 1.0 wt.%, reaction temperature of 170°C, and reaction pressure of 2.0 MPa, the influence of the reaction time in the range of 60 min to 180 min was studied. Figure 10 shows that as the reaction time increased, the conversion of unsaturated components in the pyrolysis products increased significantly. However, when the reaction time was 150 min, the reaction reached the equilibrium point, and then, when the reaction time was increased, the conversion rate was basically unchanged. Considering and simplifying the follow-up experiment comprehensively, the reaction time was chosen to be 150 min.
3.3. Statistical Analysis and Model Fitting
As shown in Table 5, through the analysis of the experimental results by Design-Expert software, the quadratic regression equation between the Ni-ZSM-5 addition level (A/%), reaction temperature (B/°C), and reaction pressure (C/MPa) factors and the conversion rate of unsaturated components in the process of the hydrogenation reaction of Swida wilsoniana pyrolysis products could be expressed by the following second-order polynomial equation:
The quality of the fit of the polynomial model equation was assessed by the coefficient of determination (R) and ANOVA. The significance of the regression coefficient was evaluated by checking the F value and value.
The absolute value of the corresponding coefficient of the factor in the model equation was the degree of influence of the factor on the reaction conversion rate, and the positive and negative coefficients reflect the direction of influence. According to the size of the coefficient, it can be seen that the Ni-ZSM-5 addition level (A)> reaction pressure (C)> reaction temperature (B), the Ni-ZSM-5 addition level had the most significant effect on the conversion rate, and there was an interaction between the three factors.
3.4. Analysis of the Response Surface
According to the abovementioned regression equation, through software analysis, the surface curve and contour map of the effect of multiple conversions of the interaction surface curve factors (Ni-ZSM-5 addition level (A)> reaction pressure (C)> reaction temperature (B)) were obtained.
As shown in Figures 11–13, the three interactions were all convex spherical surfaces, indicating that there was the highest point of the response value (conversion rate) within the range of conditions under investigation. The Ni-ZSN-5 addition level had the most significant effect on the conversion rate, which was characterized by a steep surface, followed by reaction pressure and temperature.
(a)
(b)
(a)
(b)
(a)
(b)
3.5. Optimization of Modification Conditions and Model Validation
The optimal conditions obtained from the response surface are the Ni-ZSM-5 addition level 1.04 wt.%, reaction temperature 173.30°C, and reaction pressure 2.04 MPa, and the optimal simulated value of the unsaturated component conversion rate under this condition is 98.12%. In order to facilitate the actual operation, the best process is modified to 1.05 wt.% Ni-ZSM-5 addition level, 173°C reaction temperature, and 2.00 MPa reaction pressure.
Repeated experiments verified the reliability of the results, and the conversion rate of unsaturated components was 98.10%, which was close to the simulated value. This indicated that it is reliable to use the response surface method to optimize the process conditions of the hydrogenation reaction of Swida wilsoniana pyrolysis products.
3.6. Reaction Product Analysis Results
3.6.1. Elemental Analysis
Figure 14 shows that, after the catalytic hydrogenation of the light-skinned pyrene-based pyrolysis products, the contents of C and O elements have changed significantly, and the C element content increased from 76.53% to 84.83%, but the O element content decreased from 11.50% to 2.39%, The C/H ratio is 6.639.
It can be seen from this that this study effectively promoted the hydrogenation and deoxygenation pathways of the pyrolysis product of the light-bark-based wood.
3.6.2. FI-IR Analysis
As shown in Figure 15, the infrared spectra difference of the pyrolysis products and biohydrocarbon fuel is mainly reflected in the 915 ∼ 955 cm−1, 1700 ∼ 1750 cm−1, and 2600 ∼ 3800 cm−1wave bands.
The-OH contraction vibration peak of the biohydrocarbon fuel disappeared at 915 ∼ 955 cm−1, and the C = O contraction vibration peak disappeared at 1700 ∼ 1750 cm−1, but the intensity of the saturated C-H stretching vibration peak increased at 2600–2800 cm−1, indicating that, after hydrogenation, the alkane content increased and some oxygen was removed.
3.6.3. GC-MS Analysis
GC-MS characterization was performed on Swida wilsoniana pyrolysis products and biohydrocarbon fuels. It can be seen from Figure 16 and Table 6 that the total amount of biohydrocarbon compounds increased significantly, up to 94.36%. The content of unsaturated components was reduced. At the same time, the content of carbon chain C3–C7 in the biohydrocarbon fuel increased by 3.95%, C8–C19 increased from 44.16% to 62.75%, and the carbon chain content after C19 decreased to 9.37%.
3.6.4. Performance Analysis
A series of fuel performance analysis were carried out on the products of Swida wilsoniana pyrolysis products after Ni- ZSM-5 catalytic hydrogenation. The results are shown in Table 7. The performance indicators of petrochemical diesel include the cetane number, density, oxidation stability, kinematic viscosity, acid value, condensation point, and calorific value. Through comparison, it was found that the biohydrocarbon fuel prepared by the hydrogenation reaction of the Swida wilsoniana pyrolysis products had a higher calorific value.
4. Conclusions
In this study, on the basis of the single-factor experiment, the response surface method was used to further optimize the hydrogenation process and the influence of the addition of catalyst, reaction temperature, reaction time, and reaction pressure on the conversion of Swida wilsoniana hydrocarbons. Different analysis methods (GC-MS, elemental, FT-IR, performance) were used to analyze the obtained liquid. According to the results, the highest conversion rate was 98.10% when the catalyst dosage was 1.05 wt.%, the reaction temperature was 173°C, and the reaction pressure was 2.00 MPa. The results of GC-MS, elements, and FT-IR used to characterize the products support each other. The results of the research show that Swida wilsoniana-decomposed products can be transformed into high-quality biofuels by the catalytic hydrogenation of Ni-ZSM-5 and can be used as an alternative energy source.
Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
This work was supported by the grant from National Key R&D Program of China (2019YFB1504001) and Forestry Science and Technology Demonstration Project Funds of Central Finance (2019[XT]004).