Acid Gas and Tar Removal from Syngas of Refuse Gasification by Catalytic Reforming

Abstract

The existence of acid gas and tar in syngas of municipal solid waste gasification limits its downstream utilization as a clean energy source. Here, we investigated the catalytic removal of HCl and tar. The key parameters affecting the catalytic reaction, including space velocity, temperature, the amounts of active metals in the catalyst and the carrier material, were studied, targeting optimized operating conditions for enhanced syngas purification. The morphology, mineral phases, surface area and pore size before and after the reaction were investigated to understand the mechanism to dominate the reaction. The results showed that the removal rate of CaO adsorbent and HCl reached 96% at 400 °C. When the space velocity ratio was 1.0 and the temperature was 400 °C, HCl removal (97%) by NaAlO₂ was even better. Nevertheless, clogging was observed for NaAlO₂ via the BET test after reaction to jeopardize its durability. A level of 25% Ni doping on Zr_1-x(Ce_x)O₂ support provides high stability for tar removal. This is because the Zr_1-x(Ce_x)O₂ carrier has higher carbon deposition resistivity than the Al₂O₃ carrier. The EDX results confirmed that a large amount of C (79.3%) was accumulated on the commercial catalyst surface supported by Al₂O₃ (25% Ni-based). As for the temperature, a temperature higher than 800 °C could not enhance the efficiency of tar removal, likely due to catalyst deactivation. Carbon deposition and agglomeration are the two main causes of catalyst deactivation. At 800 °C, 25% Ni-based synthetic catalyst can convert 48.5 ± 19.4% tar to low molecular weight organic compounds. By contrast, such a conversion rate under the same temperature only accounted for 5.0 ± 6.8% based on a commercial catalyst. These insights point to the important role of catalyst support materials.

1. Introduction

Incineration and pyrolytic gasification are the current mainstream methods of thermal treatment of municipal solid waste (MSW). Incineration refers to the vigorous oxidative combustion reaction of MSW mixed with excess air to generate water, carbon dioxide and ash. Although a volume reduction of MSW can be achieved, pollutants such as dioxins, heavy metals, nitrogen oxides, sulfur oxides and acid gases generated during the incineration process endanger the ecological environment and human health. Pyrolytic gasification refers to the thermal treatment technology of incomplete combustion of MSW under the condition of insufficient oxygen (excess air coefficient α < 1) to produce syngas, coke, and tar. Compared with traditional incineration technology, the advantages of pyrolytic gasification technology mainly include low pollutant emission levels, high power generation efficiency, and diversified products (combustible gas, coke, and other products) [1,2].

However, syngas also contains impurities such as tar, particulate matter, and acid gas [3]. In syngas, the acid gases contained are mainly HCl, which may cause corrosion of metal equipment and environmental problems, and can harm human health when exposed to acid gases. In the gas phase, acid gases are likely to form new pollutants with other substances, resulting in pollutant deposition and blockage [4]. Therefore, in the whole purification process, the removal of acid gases should be given priority to avoid the deactivation of catalysts caused by them in the subsequent high-temperature gasification process. In addition, alkali metal oxide catalysts can be effectively retained in the solid phase by means of physical adsorption and chemical adsorption in the gasification process, which is a relatively effective catalyst to eliminate acid gases [5,6]. Transition metal catalysts are also widely used to remove pollutants from syngas because of their stable performance and anti-carbon deposition ability [7].

Compared with particulate matter and acid gas, the concentration of tar in syngas is higher, usually around 1–100 g Nm⁻³ [8]. Tar is an inevitable by-product in the gasification process, and generally refers to all products with a molecular weight greater than benzene. The existence of organic pollutants and tar will seriously restrict the application of syngas, such as causing corrosion to gas turbines, internal combustion engines and downstream pipelines. Tar is carried by the syngas flow, and gradually condenses in the pipelines and related equipment in the process of transporting syngas, forming a viscous liquid substance. The production of tar reduces gasification efficiency. The energy of the tar product in gasification can account for 5–15% of the calorific value of the raw material [9], and this part of the energy is difficult to be utilized together with the combustible gas at low temperature. Some polycyclic aromatic hydrocarbon (PAH) substances with high content in tar components are toxic, which poses a threat to people who may come into contact with the gasification system.

According to the principle of pollutant removal, it can be generally divided into physical and chemical methods. The physical method uses physical means (such as water washing, filtration, cyclone separation, electrostatic decoking, etc.) to transfer pollutants from the gas phase to the condensed phase, so as to achieve the purpose of removing them [10]. Although the physical method is simple and less costly, it cannot fundamentally remove specific contaminant such as tar, which will also cause certain pollution to soil and water resources. Chemical removal is a method in which pollutants are converted into harmless matter by chemical reaction by changing the gasification temperature and adding a gasification agent or catalyst [11].

For tar removal, catalytic cracking is an ideal way. Compared with thermal cracking, it can achieve high-efficiency removal of tar at a lower temperature. Taking steam reforming as an example, the catalytic reforming mechanism of tar is as follows. First, the compound molecules in the tar are adsorbed onto the metal active sites of the catalyst surface which undergoes a dehydrogenation reaction to generate intermediate products and hydrogen radicals [12]. At the same time, the water vapor adsorbed on the catalyst is dissociated into hydroxyl radicals and hydrogen radicals, and the hydroxyl radicals will then hydroxylate the catalyst surface. At an appropriate reaction temperature, hydroxyl radicals gradually migrate to the center of the metal active site, oxidizing hydrocarbon intermediates and surface carbon to carbon monoxide and hydrogen [13].

Among the gas cleaning systems currently known for gas production, the hot gas cleaning system is considered superior to the condensation technology because it helps to solve the heat loss problem associated with condensation [14]. This approach promises higher cooling efficiency but also reduces harmful by-products and wastewater treatment problems [3]. As a result, hot gas cleaning systems have attracted the most attention in the last decade, especially for removing tar, particulate matter and chlorine, etc. However, the main limitation of the hot gas purification is the rapid inactivation of the catalysts during reaction [15].

The key to adopting the hot gas purification method for domestic waste pollutants is to select a catalyst suitable for the reaction and with high catalytic activity. At present, the catalysts commonly used in the reforming process mainly include the following three types: natural ore catalysts, alkali metal catalysts and transition metal catalysts [16].

Catalysts where the active component is Ni, Zn, Cu, Fe have gradually become a promising alternative for catalysts made of precious metals due to their wide sources, low cost, easy preparation and good catalytic activity. Common carriers of Ni-based catalysts include metal oxides, natural ores, molecular sieves, biochar etc. The carrier can effectively control the metal particle size and improve the uniformity of its dispersion, thereby enhancing the catalytic activity [17]. Metal oxides are the most used supports. The metal compound supports that can be used as transition metal catalysts usually include Al₂O₃, ZrO₂, MgO, and CaCO₃, and common additives include CeO₂ [18]. Among them, Al₂O₃ can provide more active sites for the catalyst due to its higher specific surface area. The addition of CeO₂ may improve the performance of catalysts through a redox mechanism. Low-valence Ce may adsorb water and dissociate it to generate oxygen radicals (·O) or hydroxyl (·OH), which react with carbon on the catalyst surface to generate CO, CO₂, and H₂. Zhang et al. [19] studied the effect of CeO₂-doped Ni/olivine catalysts on tar in the steam reforming process and found that CeO₂-doped CeO₂ had special effects in both catalytic activity and coking resistance.

The aim of this study is to investigate the performance of HCl and tar removal using hot gas cleaning process on the basis of different catalysts, explore the best process parameters, and confirm whether it can meet the standard of syngas use.

2. Results and Discussion

2.1. HCl

2.1.1. CaO Adsorbent

This experiment is mainly to study the effect of gasification temperature on the removal of hydrogen chloride in domestic waste (Figure 1). Under the condition that the airspeed ratio is determined, the optimum temperature of hydrogen chloride removal can be determined by changing the temperature of hydrogen chloride gasification. Mura and Lallai [20,21] studied the reaction between calcium oxide and hydrogen chloride and determined that the activation energy of the reaction was 45 kJ·mol⁻¹. Weinell and Jensen analyzed the reaction of HCl with lime and limestone at 60–1000 °C. They showed that lime and limestone are able to capture most of the HCl at 500–600 °C and then, beyond 500 °C, the chemical balance between gas and solid reduces the adsorption capacity. Therefore, we set the gasification temperature of 300 °C, 400 °C and 500 °C, to determine the best temperature for HCl purification of Shanghai municipal solid waste.

By analyzing the change in HCl concentration over time, we can see that the removal efficiency of HCl at different temperatures finally stabilized above 90%. This shows that the reaction between CaO and hydrogen chloride is particularly intense at medium and high temperatures, and the adsorption effect is excellent. When the gasification temperature is at 500 °C for the first 30 min, the removal efficiency obviously increases. Through the experiments of removing hydrogen chloride with CaO at different gasification temperatures, it can be found that the removal efficiency of HCl with CaO at 400 °C is the highest, which is up to 95.62%, conforming with the results of Weinell et al. [22]. At 300 °C, the efficiency of hydrogen chloride removal decreases slightly with the increase in temperature. This may be due to the increase in temperature, surface morphology and structure of CaO adsorbent changing before and after the reaction, resulting in reduced HCl removal efficiency.

2.1.2. NaAlO₂ Adsorbent

This experiment is mainly aimed at the optimal temperature for the removal of HCl by CaO, to study the influence of airspeed ratio on the removal of hydrogen chloride (Figure 2). The effect of space velocity ratios on catalytic efficiency is often discussed in gas–solid reactions. Reaction space velocity refers to the amount of gas treated by adsorbent/catalyst per unit volume of catalyst per unit time under specified conditions. In the calculation, it can be considered as the ratio of the carrier gas flow rate and the catalyst stack volume. Generally, the airspeed ratio should be less than 5 s⁻¹. When the carrier gas flow rate is unchanged, the more adsorbent/catalyst added, the greater the airspeed ratio, and the higher the pollutant removal rate.

To obtain the best space velocity ratio of adsorbent, the same catalytic temperature (400 °C) and the same catalyst (NaAlO₂) were set. The flow rate of N₂ was 148 mL/min, and the removal efficiency of HCl was studied by changing the amount of adsorbent (0.5 g, 0.9 g and 1.8 g).

Within 3 h of the experiment, the removal rate and HCl adsorption of 0.5 g adsorbent was relatively low. Both 0.9 g and 1.8 g NaAlO₂ could absorb more than 95% of HCl. In the whole reaction process, the cumulative adsorption rate of 1.8 g adsorbent for HCl was higher than that of 0.9 g. However, the adsorption efficiency of 1.8 g adsorbent decreased when the reaction was carried out for 60 min. The reason may be that HCl did not fully contact and react with the adsorbent below after fully reacting with NaAlO₂ on the surface. Therefore, there was a process of decline. As time went on, the reaction between HCl and NaAlO₂ became more sufficient, and the removal efficiency increased again. At 180 min, the conversion rate of 1.8 g was slightly lower than 0.9 g, which may have led to the change in chemical bonds in the adsorbent due to the long running time, thus affecting the adsorption efficiency. In general, in the two experiments with changed space velocity ratios, the removal rate of 1.8 g adsorbent was slightly better than that of 0.9 g adsorbent.

2.1.3. Characterization of NaAlO₂ before and after the Experiment

BET

The type Ⅲ isotherm indicates that the material interaction between adsorbent molecules is strong, and it is difficult to adsorb the adsorbent at the initial stage. With the progress of the adsorption process, the adsorption appears as a self-accelerating phenomenon (Figure 3). In this section, BET representation of NaAlO₂ before and after the reaction was made, as shown in

Table 1. BET characterization of NaAlO₂ before and after reaction.

NaAlO2	Before Reaction	After Reaction
BET surface area (m2/g)	0.8188	0.9314
Pore volume (cm3/g)	6.77 × 10−3	4.58 × 10−3
Average diameter (nm)	44.61	26.09

. Among them, the specific surface area and pore size of the fresh adsorbent are both low, and the specific surface area of the adsorbent after the reaction increases from 0.8188 m²/g to 0.9314 m²/g, which may be due to the collapse of its structure during the reaction, resulting in the appearance of new micropores on the surface of the sample, so it has a larger specific surface area than before the reaction. After the HCl removal experiment, the mesoporous volume of the adsorbent was reduced from 0.0067 cm³/g to 0.00458 cm³/g. This may be because during the experiment, the metal oxides inside the adsorbent void reacted with HCl gas, and the resulting products plugged the void of the adsorbent.

SEM

Figure 4 represents SEM photos before and after the adsorbent reaction. As can be seen from the figure, the surface of the sample presents a loose and porous morphology, and has a certain degree of agglomeration phenomenon, forming a large aggregate structure. Compared with the adsorbent before the reaction, the structure of NaAlO₂ after the reaction showed obvious shrinkage and collapse, the agglomeration phenomenon was more serious, and some voids were blocked.

XRD

In this section, the XRD analysis of the crystal phase of NaAlO₂ before and after the reaction was carried out. The results are shown in Figure 5. For fresh adsorbents, Na₂ZnSiO₄ can be clearly identified. By comparing the adsorbents after reaction, only Na₂CO₃ and NaClO₃ can be observed, indicating that most of the substances in sodium metalate reacted with HCl and were reduced. In general, the XRD characterization results of NaAlO₂ before and after the reaction very reliably show that NaAlO₂ can effectively absorb HCl.

2.2. Tar

2.2.1. Determination of the Flux of Naphthalene

Naphthalene was used as a model tar compound, so it was necessary to ensure a stable naphthalene vapor flow rate for each experiment. The vapor flow rate of naphthalene during the experiment was required to stabilize at 4.2 ± 0.4 mg/L, which is the typical concentration of tar in wet biomass and MSW syngas (based on the mean standard deviation of 4 measurements). The actual flow rate of naphthalene vapor was measured in blank operation state (no catalyst added), and the reducing environment was maintained during the reforming process (the N₂ flow rate was controlled at 200 mL/min). Test the vapor flow of naphthalene at different temperatures, and determine the evaporator temperature when the vapor flow of naphthalene is close to 4.2 mg/L. The results are shown in Figure 6.

Vapor flow of naphthalene was tested at four temperatures, 57 °C, 62 °C, 69 °C, and 74 °C, respectively. The corresponding vapor flow of naphthalene increased with the rise in temperature, indicating that high temperature promotes the melting of naphthalene and then into a gaseous state into the reactor. When the temperature is 62 °C, the vapor flow rate of naphthalene is stable at 4.3 mg/L (the average value of four parallel experiments), which is closest to the targeting value. Therefore, evaporating temperature was set to 62 °C.

2.2.2. The Effects of Space Velocity on Tar Catalytic Reforming

Reaction space velocity refers to the amount of gas processed per unit volume of catalyst per unit time under specified conditions. In the process of gas–solid catalytic reaction, the effect of space velocity ratio on catalytic efficiency is often discussed. When calculating, it can be considered as the ratio of the flow rate of the carrier gas to the volume of the catalyst stack. In a typical syngas scrubbing process, the space velocity ratio is often less than 5 s⁻¹ [11,23]. When the flow rate of the carrier gas remains unchanged, the more catalyst added, the larger the space velocity ratio, and the higher removal rate of naphthalene. In the study, to obtain the best space velocity ratio of the catalyst, the same catalyst temperature (750 °C) and the same catalyst (25% Ni-based synthetic catalyst) were compared. Since the carrier gas flow rate was controlled at 200 mL/min, by varying the amount of catalyst (0.5 g and 1.0 g), the effect of the space velocity ratio (3.78 and 1.89 s⁻¹, respectively) on the gas components (CO, CO₂ and CH₄) was obtained (Figure 7).

For the application of 0.5 g of catalyst (reaction time = 2 h), the removal rate of naphthalene highly fluctuated in the range of 80.1–97.5% (the lowest was 80.1% at 80 min, and the highest was 97.5% at 40 min), which would be unfavorable for sustainable naphthalene cracking. We postulate that as the operation time extended, carbon deposition more likely occurred under the low space velocity ratio resulting in deteriorating tar-cracking performance. For the application of 1.0 g of catalyst, the removal rate of naphthalene fell within the lower range of 77–87%. It yields the cumulative conversion of naphthalene corresponding to 0.5 g catalyst was higher than that of 1.0 g during the whole reaction process. However, it should be noted that the conversion rate demonstrated a rising trend with time when more catalyst was present in the reaction system, suggesting that there could be a longer adaptation period before the reaction reaches full-scale speed. In addition, studies [24] have shown that increasing the catalyst loading can provide a larger reaction area for the catalytic reforming of naphthalene. More nickel surface active sites can greatly reduce the activation energy required for naphthalene cleavage, thereby reducing the reaction time. At the same time, increasing the loading height can prolong the contact time between naphthalene and the catalyst, thereby improving the reforming efficiency of naphthalene. In this regard, 1.0 g catalyst was selected in the following experiments. It is also worth noting that excessive Ni leads to the sintering and agglomeration of NiO particles, which reduces the specific surface area of the catalyst and inhibits the catalytic activity.

2.2.3. The Effects of Temperature on Tar Catalytic Reforming

The effects of reaction temperature on the catalytic reforming of naphthalene were investigated, and 1.0 g of 25% synthetic catalyst was introduced, with reaction temperatures set at 750 °C, 800 °C, 850 °C and 900 °C, respectively. The results are shown in Figure 8.

Under 750 °C, the removal rate of naphthalene was stable in the range of 77.3–86.1% (the lowest was 77.3% in 20 min, and the highest was 86.1% in 120 min), which was similar to the rate under 900 °C (76–84%). The conversion of naphthalene was maintained at a low level. Under the working condition of 800 °C, the removal rate of naphthalene fluctuated greatly, reaching 110.8% in 20 min. This compared with removal rates of 77.3%, 107.8%, and 80.7% in 20 min at 750 °C, 850 °C, and 900 °C, amounting to an increase of 33.6%, 3.0%, and 30.2%, respectively. At all other reaction times, the catalytic reforming efficiency of naphthalene and the cumulative amount at 800 °C were the highest (except in 100 min at 850 °C). When the reaction temperature rose moderately (up to 800 °C), the active sites on the catalyst surface were activated, and the π electron cloud of the polyaromatic ring of the tar precursor was adsorbed on the active sites to be readily destroyed. C-C and C-H bonds are prone to be broken first, and the activation energy required for naphthalene reforming was reduced, thereby increasing the conversion rate of naphthalene. When temperature rose further up to 850–900 °C, however, catalytic efficiency started to decrease. The reason may be ascribed to carbon deposition, resulting in the reduction of reduced catalyst activity and even deactivation. This suggests that 800 °C is the optimal temperature for catalytic reforming of naphthalene.

2.2.4. The Effects of Different Catalysts and Associated Loading Rates on Tar Removal

The catalytic effect of 25%-Ni synthetic catalyst is better than that of 15%-Ni synthetic catalyst. It shows that higher nickel loading provides more nickel surface active sites for promoted catalytic reforming and simultaneously reduces the activation energy required for naphthalene cracking. Compared to commercial catalysts (25% Ni), the naphthalene conversion was comparable (87–110% and 89–102%, respectively). However, in terms of the total conversion of naphthalene, the synthetic one is slightly better. It can be seen from the SEM-EDX characterization (Figure 9) that the carbon content of the 25%-Ni commercial catalyst after the reaction increased by 59.2% compared with the fresh counterpart, and significant coking was observed in the SEM image, which indicates that the commercial catalyst is easier to deposit carbon, which leads to the reduction in its catalytic activity. In fact, the structure of alumina support in the commercial catalyst tends to collapse (to form agglomerates) under high temperature [25]. Therefore, 25% Ni-based synthetic catalyst is an ideal catalyst with higher durability and thermal resistance.

2.2.5. Gas Chromatography Analysis of Tar-Reforming End Products

For commercial catalysts (25% NiO), the content of C10 group and those greater than C10 (referring to higher C-containing compounds than naphthalene (C = 10) after catalytic reforming) in the liquid phase at 750 °C is the highest (Figure 10a). Since the reaction temperature is not optimized, as noted earlier, only a small amount of naphthalene (12.2%) is broken down into smaller molecular organics (designed as <C10 group). As the temperature increased from 800 °C to 900 °C, only 6.7–14.8% of naphthalene residue was collected (Figure 6). More naphthalene was either cracked (48.5–60.2%) or polymerized into high-molecular organic matter (25.7–44.7%). However, the phenomenon of polymerization was less pronounced with increased temperature. In another words, cracking activities became significantly facilitated with the incremental rise in temperature. Under 900 °C, the content of these three groups of compounds in the liquid phase leveled off, indicating the achievement of an equilibrium state in which raising the temperature posed limited effects on the promotion of decomposition or polymerization. Multiple factors are responsible for this, including the space velocity ratio, the available surface-active sites of the catalyst, and the coking effect. For self-made catalysts (25% NiO), temperature played a significant role in the formation of <C10 and C10 groups, while the influence towards >C10 groups was comparable among different temperatures. With the increase in temperature, a greater amount of tar tended to be cracked. The cracking effect likely peaks (at around 60% of <C10 groups) after 850 °C, corresponding to an optimized catalyst performance. Comparatively, self-made catalysts perform better than commercial ones in respects of tar-cracking ability, as shown in Figure 10c.

2.2.6. Characterization of Various Catalysts before and after Tar Reforming for Mechanism Investigation

XRD Characterization

For the freshly synthetic catalyst, the characteristic peaks of NiO, ZrO₂ and CeO₂ were clearly identified (Figure 11). Surprisingly, the crystal phase of Ni was not observed. All Ni appeared in its oxide format, which indicated that the catalyst was manufactured successfully. In contrast, for the catalyst after the reaction, only elemental nickel can be observed (no characteristic peaks of nickel oxide), indicating that most of the nickel oxides had been reduced to elemental nickel during the catalytic reforming process. Nevertheless, a small amount of nickel oxide could not be ruled out. The XRD characterization results of the catalysts before and after the reaction were very reliable to illustrate the fact that the nickel oxide was reduced in the catalytic reforming.

SEM Characterization

Figure 12 shows the SEM-EDX scanning results of the 25%-Ni commercial catalyst and 25%-Ni synthetic catalyst, before and after the tar reforming process. For the synthetic catalyst, the distribution between nickel and the carrier is loose and largely uniform. As the precipitated nickel particles have much smaller volume compared to the carrier particles, the SEM image shows a scattered distribution of Ni. For the commercial catalyst, due to its larger particle size, its special preparation process makes its appearance flatter, and the pores are not as obvious as those of the synthetic catalyst. For both commercial and synthetic catalysts, the semi-quantitative results of EDX elements after the reaction show fewer oxygen elements than before the reaction, which is attributed to the consumption of oxygen in the nickel oxide during the catalytic reforming reaction to form CO, CO₂, etc. It is further echoed by the EDX mapping results (according to the different density of oxygen elements presented before and after reaction), and a small amount of oxygen elements remaining after the reaction may originate from the carrier part.

For the EDX results, obvious peaks (Ni, Zr and Al) can be seen for both catalysts in the spectrum, and the Ce peak is not easily observed due to its small content. It is worth noting that a large amount of C (79.3%) was observed on the surface of the commercial catalyst after reaction, while no significant carbon content was noticed in the fresh one. With surface carbon precipitation, it blocks pores of catalyst and prevents naphthalene from diffusing into the interior part of the catalyst during the high-temperature reaction, thus resulting in incomplete tar removal. This phenomenon also suggests the weaker durability of the commercial catalysts.

BET Characterization

The type IV isotherm with an H1 hysteresis loop at higher relative pressure demonstrated that the synthetic catalyst is a mesoporous material (Figure 13a) [18]. Among them, the fresh one had a higher specific surface area (3.03 m²/g) and a lower pore size, which was due to the surface area provided by zirconia for the dispersion of the catalyst and the smaller pore size of the impregnated catalyst (

Table 2. BET characterization of synthetic catalysts before and after catalytic reforming process.

25% Ni-Based Synthetic Catalyst	Before Reaction	After Reaction
BET surface area (m2/g)	3.03	4.84
Pore volume (cm3/g)	2.74 × 10−2	4.82 × 10−2
Average diameter (nm)	36.90	39.80

). After catalytic reforming, the specific surface area, pore volume and average pore size of the catalyst became larger, which was attributed to the consumption of oxygen element. As a result, the catalysts became macro-porous solid materials and no longer had adsorption characteristics, which was supported by the III type isotherm in Figure 3b.

3. Materials and Methods

3.1. Catalyst Preparation

The commercial catalyst was a Ni-based spherical particle with Al₂O₃ as a carrier, purchased from a chemical engineering company in Shanghai, China. The Ni-based catalyst particle had a loading of 25% Ni by weight, with the particle diameter at 0.5–1 mm. Ni-based synthetic catalysts were synthesized using ceria-stabilized zirconia as a support. The carrier was purchased from a company in Beijing, China, and the addition of cerium oxide was 8% to enhance the stability of the carrier at high temperature [19]. Ni-based synthetic catalysts with Ni loadings of 15% and 25% were prepared. The carrier was added to a deionized aqueous solution containing Ni(NO₃)₂∙6H₂O and mixed. The solvent water was evaporated using a rotary evaporator at 82 °C. The material was then dried in an oven at 105 °C for 12 h, and finally placed in a muffle furnace for 2 h air roasting at 500 °C (with a heating rate of 2 °C/min). The calcined samples were ground and sieved, and the powder with a particle size of 0.053–0.1 mm was collected and stored in a sealed container before the experiments began.

A series of characterizations were performed on the prepared catalysts. An X-ray fluorescence spectrometer (XRF) was used to determine the type and content of oxides in the catalyst. An X-ray diffractometer (XRD) was used with Cu Kα radiation (40 kV, 35 mA), where the θ ranges from 10° to 80° with a step size of 0.02° to analyze the crystal structure of the catalyst. The morphology and surface elements were determined by environmental scanning electron microscopy (SEM). The specific surface area, total pore volume and pore size of the oxygen carrier were analyzed by nitrogen adsorption–desorption isotherm at −196 °C using an automatic specific surface area and pore size distribution analyzer (BET).

3.2. Experimental Setup

In this experiment, the catalysts and the effect of additives in the tube furnace on syngas absorption, which was pyrolytically gasified by domestic waste, were studied by using a fixed-bed pyrolysis reaction device in a vertical tube furnace. The experimental device, shown in Figure 14, was mainly composed of three parts: (1) Gas supply: it mainly includes the gas source of the high-pressure gas cylinder, the gas flow controller needed to maintain the reaction atmosphere in the gasification process, and the device needed to simulate pollutants; (2) Main element: including heating temperature control device, vertical tube furnace and quartz tube; (3) Sample collection: mainly air bag collection and condensation collection. The main part of the quartz tube diameter was 15 mm, the outer diameter was 20 mm, the length was 65 cm; 32 cm away from the lower end we placed a sand core plate aperture of 50–90 μm. Contaminants were exposed to the sample on the core board through a quartz tube and then removed.

In the HCl experiment, we first put the adsorbent in the quartz reactor, wrapped the heating belt on the pipes, and opened the standpipe furnace and heating belt to heat it to a preset temperature. When the temperature was stable, the hydrochloric acid solution was injected with a syringe pump with the injection rate at 0.042 mL/min to ensure that the hydrochloric acid solution could be vaporized at a stable rate. The whole experiment was carried out with nitrogen as the carrier gas at a flow rate of 148 mL/min. Hydrogen chloride was exposed to an adsorbent placed on a quartz tube sand core plate, and then the chlorine was adsorbed and removed. To prevent condensation of hydrogen chloride in the pipeline during the experiment, heating bands were wrapped on the tubes before and after the reactor, and the heating band temperature was controlled at 120–130 °C. After heating the heating belt to the evaporation temperature and stabilizing for 30 min, we started the timing. Hydrogen chloride gas was captured in 0.1 mol/L NaOH glass washing bottles for tail collection, and chloride ions collected in tail washing bottles were detected by ion chromatography (IC). Reaction samples were collected every 30 min and glass bottles were replaced every half an hour. All experiments were repeated twice, and the results were expressed as the average of the two experiments.

The experiment for catalytic tar reforming used naphthalene as a model tar compound [26]. Naphthalene vapor was produced by vaporizing naphthalene followed by N₂ purging. A certain temperature was set to ensure that the naphthalene was evaporated at a constant flux in the evaporator, and brought into the catalytic reforming device through N₂ as a carrier gas. Naphthalene contacted the catalyst pre-placed on the sand core plate through the quartz tube, and then catalytic reforming occurred. It generated small molecular carbon-based compounds and synthesis gas (H₂, CO and a small amount of CH₄ and CO₂). To avoid condensation of naphthalene vapor from the pipeline, the tetrafluoride pipelines at the front and rear ends of the reactor were constantly heated (at 85–95 °C), the naphthalene vapor was captured by an isopropanol trap in the end, and the isopropanol solution was analyzed by GC-MS, while the tailing gas was collected with a PTFE gas bag and passed through a GC-Thermal Conductivity Detector (TCD)/Flame Ionization Detector (FID) The carbon-containing gas content was measured to calculate the catalytic efficiency. In a typical operation, the catalyst was first loaded into the quartz reactor, and the entire experiment was fluidized with 200 mL/min of N₂ gas. After the experiment began, the reactor was heated to a preset temperature with a vertical tube furnace. We then turned on the heating belts at the front and rear ends of the naphthalene evaporator and the reactor, allowing the naphthalene evaporator to heat up to the designated temperature and stabilizing for 30 min. The gas was passed through two isopropanol solutions to capture the residual naphthalene vapor, and was then passed through anhydrous sodium sulfate to remove moisture. The gas samples were collected every 20 min, lasting 5 min at a time. Naphthalene collection bottles were replaced every half hour, and the exhausting gas was analyzed by GC-FID/TCD. All experiments were triplicated, and the average value was reported.

3.3. Calculations and Characterization

3.3.1. Data Analysis

HCl conversion rate (%) can be calculated as follows:

$$\mathrm{HCl} \mathrm{conversion} \mathrm{rate} (\%)= \frac{n_{HCl}^{\prime }-n_{HCl}}{n_{HCl}}\times 100\%$$

(1)

where $n_{HCl}^{\prime }$ (mg/L) refers to the concentration of HCl after reaction, and $n_{HCl}$ refers to the inlet concentration.

Naphthalene conversion rate was calculated as follows:

$$\mathrm{Naphthalene} \mathrm{conversion} \mathrm{rate} (\%)=\frac{n_{C{O}_2}+n_{CO }+n_{C{H}_4}}{n_{nap}\times 10}\times 100\%$$

(2)

where $n_{C{O}_2}$, $n_{CO}$ and $n_{C{H}_4}$ refer to the molar concentration of CO₂, CO and CH₄ generated during naphthalene reforming (mol/min), respectively. $n_{nap}$ (mol/min) refers to the concentration of naphthalene.

Reaction space velocity was calculated as follows:

$$GHSV=\frac{Q_{Nap}}{V}$$

(3)

where $GHSV$ (/h) stands for gas hourly space velocity, $Q_{Nap}$ refers to the volumetirc flow rate of naphthalene (L/h), and V refers to the volume of catalyst (L).

3.3.2. Characterization

Nickel in fresh catalysts is usually in an oxidized state. The contents of oxides in the catalysts were tested by XRF, and the characterization results of catalysts are shown in

Table 3. XRF characterization results of Ni-based catalysts.

Metal Oxides (%)	Commercial (25%)	Self-Made (15%)	Self-Made (25%)
Al2O3	71.896	0	0
ZrO2	0	71.05	63.21
CeO2	0	-	-
NiO	24.252	17.43	27.83
SO3	1.157	0.79	0.69
SiO2	2.219	0	0
P2O5	0	0	0
CaO	0.228	0	0
Ta2O5	0.089	0	0
Fe2O3	0.092	0	0
Cr2O3	0.021	0	0
K2O	0.027	0	0
HfO2	0	1.45	1.23
MoO3	0	0.62	0.62
Nb2O3	0	0.31	0.16
Na2O	0	0	0.13

.

The nickel content of the commercial catalysts was close to the corresponding theoretical values, whereas that of the self-made catalysts (17.43% and 27.83%) was slightly higher than the theoretical values (15% and 25%). It is noteworthy that cerium could not be detected in the XRF spectrum, likely due to its low sensitivity for detection. In addition to active component and support components, several types of trace metal oxides were also detected. Commercial catalysts tend to be mixed with more SO₃, CaO, Fe₂O₃ etc., which may be the admixture of impurities in the preparation process. Self-made catalysts, on the other hand, exhibit doping with trace amounts of noble metal oxides because of the impurities inherent from the support.

4. Conclusions

With the improvement in living standards, the disposal of municipal solid waste has become a serious challenge. Due to the advantages of incineration in the aspects of “Recycle, Reuse and Reduce”, the proportion of these three disposal methods is increasing. Therefore, in the context of sustainable energy, this research successfully developed a household-waste synthetic-gas purification system for toxic and harmful substances in modern urban household waste. The removal efficiency and conversion amount of hydrogen chloride and tar in syngas were measured. The results of this study are summarized as follows:

(1)

CaO adsorbent shows high efficiency of combining with HCl. At 400 °C, the removal rate of CaO adsorbent with HCl reaches 95.62%, and the adsorption capacity of the adsorbent may decrease when the temperature exceeds 500 °C. The reaction between NaAlO₂ adsorbent and HCl is more intense, but with the increase in the amount of adsorbent, HCl cannot easily enter the interior of the adsorbent and react with it. Therefore, for NaAlO₂, the removal efficiency of HCl is the highest when the reaction space velocity ratio is 1.0.

(2)

Experiments show that both the prepared nickel-based industrial catalysts for tar catalytic reforming and the synthesized catalysts can provide more than 80% tar removal. When the space velocity ratio is 1.89, a larger reaction area can be provided for naphthalene catalytic reforming. At 800 °C, 25% Ni-based synthetic catalyst can convert tar to low-molecular-weight organic compounds (compounds with a carbon number greater than 10). While commercial catalysts may have similar tar removal effects, the greater carbon precipitation potential hinders the sustainability of their long-term applications.

(3)

For tar catalytic reforming experiments, no matter whether for commercial or homemade catalysts, the oxygen of nickel oxide in the catalyst is consumed in the catalytic reforming process to form CO, CO₂, etc., which indicates that a small amount of oxygen left after the reaction may come from the carrier. Naphthalene has difficulty reaching the catalyst in the process of high temperature reaction, thus forming an incomplete reaction to yield fixed carbon, covering the reactor and the surface with catalyst particles, thus preventing further catalytic reaction. Carbon deposition and coking are the two main causes of catalyst deactivation.

(4)

The syngas purification system used in this study has strong potential application value in removing tar and acid gas produced by gasification. However, the larger scale of domestic waste purification technology needs to be discussed in future studies, and more attention should be paid to the possibility of converting tar into small molecular organic compounds for reuse and the treatment of deactivated catalysts.