Biogas Dry Reforming for Hydrogen through Membrane Reactor Utilizing Negative Pressure

Abstract

Biogas, consisting of CH₄ and CO₂, is a promising energy source and can be converted into H₂ by a dry reforming reaction. In this study, a membrane reactor is adopted to promote the performance of biogas dry reforming. The aim of this study is to investigate the effect of pressure of sweep gas on a biogas dry reforming to get H₂. The effect of molar ratio of supplied CH₄:CO₂ and reaction temperature is also investigated. It is observed that the impact of p_sweep on concentrations of CH₄ and CO₂ is small irrespective of reaction temperature. The concentrations of H₂ and CO increase with an increase in reaction temperature _t. The concentration of H₂, at the outlet of the reaction chamber, reduces with a decrease in p_sweep. It is due to an increase in H₂ extraction from the reaction chamber to the sweep chamber. The highest concentration of H₂ is obtained in the case of the molar ratio of CH₄:CO₂ = 1:1. The concentration of CO is the highest in the case of the molar ratio of CH₄:CO₂ = 1.5:1. The highest sweep effect is obtained at reaction temperature of 500 °C and p_sweep of 0.045 MPa.

1. Introduction

Global warming problem is a serious issue in the world. Each country has set the goal to reduce CO₂ emission, by 2030 or 2050. For instance, the Japan had declared to reduce the amount of CO₂ emission up to zero practically by 2050. Though there are many ways to reduce the amount of CO₂ emission, such as a digested sewage sludge for CO₂ removal from biogas [1] and a production of high-quality hydrocarbon fuel from palm waste [2], a renewable H₂ is a promising candidate. Renewable H₂, which is called a green H₂, can assist to solve the global warming and construct an energy supply chain to meet the increasing energy demand. There are some innovative approaches for green H₂ production. H₂ production by the water splitting over Fe₃O₄ pellet at low temperature of 250 °C, 290 °C and 310 °C [3] and photothermal-photocatalytic materials under light illumination without additional energy [4] were reported. In addition, the anion exchange membrane (AEM) electrolysis for large-scale H₂ production was investigated to clarify the resistances involved in AEM electrolysis [5].

This study focuses on H₂ produced from a biogas dry reforming. Biogas is a gaseous fuel consisting of CH₄ (55–75 vol%) and CO₂ (25–45 vol%) [6], mainly produced from fermentation by the action of anaerobic microorganisms on raw materials such as, garbage, livestock excretion, and sewage sludge. Additionally, the conversion of biogas to H₂ can be said as a carbon neutral since the CO₂, which is a by-product in the biogas production process, can be absorbed by plants. It is known from the International Energy Agency (IEA) [7] that the biogas has been produced 59.3 billion m³ globally with an equivalent energy of 1.36 EJ in 2018, which is approximately five times as large as that in 2000. Therefore, a biogas is thought to be a promising energy source.

It is generally known that a biogas is used as a fuel for gas engine or micro gas turbine [8]. Biogas contains CO₂ of approximately 40 vol% which reduces the heating value compared to a natural gas, resulting in the efficiency of power generation decreasing. Considering it, this study focuses on a biogas dry reforming to produce H₂, which can be used as a fuel in the fuel cell system, such as solid oxide fuel cell (SOFC). Biogas dry reforming has been studied previously by several researchers [9,10,11,12,13,14,15,16,17,18,19,20,21,22,23,24,25,26,27]. The catalyst development, which is one of important issues, is investigated to improve the performance of biogas dry reforming. Several catalysts such as Co-Ni/Al₂O₃, Ce-Co-Ni/Al₂O₃ [9], Pt-Ni/YSZ [10], Ni/Ce/MgAl₂O₄ [11], Pt-Ni/Al₂O₃ [12], La-La₂O₃/-Ni [13], Pd-Ni-Mg/Ceria-Zirconia [14], Ni/γ-Al₂O₃ [15], Ni/SBA-15 [16], Ni/ZnO-Al₂O₃ [17], Ni/CeO₂-SiO₂ [18], Ni/ZrO₂ [19], Ni/CaZrNiOx [20], Ni/CaFe₂O₄ [21], Ni/Ce/TiO₂-Al₂O₃ [22], Ni nanoparticle [23], Ni/ZnO-CeO₂ [24], Rh/Ni [25], Ni-based bimodal porous catalyst [26] and Cu/Ni bimetallic catalyst [27] have been investigated, resulting that Ni-based catalyst is the most common catalyst for the biogas dry reforming process. Therefore, Ni catalyst is adopted in this study. In addition, a membrane reactor is also adopted to improve the efficiency of biogas dry reforming in this study. It is well-known that a membrane reactor is used to improve the CH₄ steam reforming by separating H₂ from the reaction space [28,29,30] as soon as it is produced. The synergetic effect of the membrane reactor on performing the reaction and the separation in the same unit should be analyzed. It requires advanced levels of automation and control to re-design process [31]. Pd based membrane is used to separate H₂ due to its high efficiency [31,32,33]. According to the literature survey, Pd membrane has been used to improve the performance of CH₄ dry reforming [34,35,36,37,38,39,40,41,42,43,44]. The coupled effect of membrane and catalyst on CH₄ conversion has been discussed earlier [31]. Alloy membrane such as Pd/Ag [34,35,36], Pd/Au [37,38] and Pd/Cu [39,40] have been used generally due to the stability at high temperature [32]. Hollow fiber membrane reactor can perform CH₄ conversion, which has higher by 72 % output than traditional fixed bed reactor [41]. The Pd/Au alloy membrane has been used in the two-zone fluidized bed reactor [33,42] and showed CH₄ conversion and H₂ selectivity to be higher compared to the conventional fluidized bed reactor. Reaction combination consisting of dry reforming and steam reforming has been carried out in a Pd/Ag membrane reactor, resulting that CH₄ conversion above 90% has achieved at 650 °C [36]. The effect of flow rate of sweep gas on CH₄ conversion has been investigated by two membrane reactors, equipped with the expensive dense H₂ selective membrane and the inexpensive porous Vycor glass membrane [43]. There has been a significant loss of the reactant with an increase in sweep gas flow rate, which in turn can decrease the CH₄ conversion. The other study [35], which has investigated the effect of flow rate of sweep gas and reported that CH₄ conversion and H₂ recovery have increased with an increase in the flow rate of sweep gas. In addition, it is also reported that the increase in reaction pressure has showed the negative effect on CH₄ conversion at the reaction temperature of 550 °C [35]. However, the other study [44] has reported a contrary conclusion that in CH₄ conversion, H₂ recovery and H₂/CO ratio have increased with an increase in reaction pressure at the reaction temperature of 800 °C. It [44] has also reported that H₂/CO ratio has decreased with the increase in CO₂/CH₄ ratio. Since the H₂ flux through membrane has increased with transmembrane pressure difference [34], it is expected that the performance of biogas dry reforming has been improved due to the shift of the reaction toward further conversion [31]. Though the impact of negative pressure of sweep gas on H₂ diffusion flux has been reported [45], and at this moment, there is no study investigating the effect of negative pressure of sweep gas on the performance of biogas dry reforming.

Therefore, the aim of this study is to understand the effect of pressure of sweep gas (p_sweep) on the performance of the biogas dry reforming process. In this study, the pressure difference was provided by vacuum pomp as the negative pressure. The effect of changing reaction temperature_, which means the initial temperature for dry reforming in this study, from 400 °C to 600 °C, and the molar ratio of CH₄:CO₂ by 1.5:1, 1:1 and 1:1.5 is also investigated. The molar ratio of CH₄:CO₂ = 1.5:1 simulates a biogas. Since a pure Pd membrane has relatively high solubility for carbon, it causes membrane degradation leading to the loss of permeability [46]. Therefore, the Pd/Cu alloy membrane is used in this study. The reaction scheme of CH₄ dry reforming is as follows:

CH₄ + CO₂ ↔ 2CO + 2H₂

(1)

The other reaction schemes which can be thought to be occurred in this study are as follows [32]:

CO₂ + H₂ ↔ CO + H₂O

(2)

CH₄ + H₂O ↔ CO + 3H₂

(3)

CO₂ + 4H₂ ↔ CH₄ + 2H₂O

(4)

where Equation (2) is the reverse water gas shift reaction (RWGS), Equation (3) is steam reforming of CH₄ and Equation (4) is the methanation reaction.

2. Experimentation

2.1. Experimental Set-Up

Figure 1 illustrates the schematic drawing of experimental set-up used in this study. The experimental apparatus is composed of a gas cylinder, mass flow controllers (S48–32; produced by HORIBA METRON INC., Shanghai, China), pressure sensors (KM31; produced by NAGANO KEIKI, Tokyo, Japan), valves, a vacuum pump, a Pirani gauge (SW-1; produced by ULVAC, Saito-City, Miyazaki, Japan) for measuring negative pressure, a reactor consisting of reaction chamber and sweep chamber, and gas sampling taps. The reactor is in the furnace. The temperature in the furnace is controlled by far-infrared heaters (MCHNNS1; produced by MISUMI, Tokyo, Japan). CH₄ gas whose purity is over 99.4 vol% and CO₂ gas whose purity is over 99.9 vol% are controlled by the mass flow controllers and mixed before the reactors. The pressure of the mixed gas is measured by pressure sensors. Ar gas whose purity is over 99.99 vol% is controlled by the mass flow controller, and the pressure of Ar gas is measured by the pressure sensor, which is supplied as a sweep gas. The exhausted gas at the outlet of the reactor is suctioned by a gas syringe via the gas sampling tap. The concentration of sampled gas is measured by FID gas chromatography and a methanizer whose minimum resolution of both FID gas chromatograph and methanizer is 1 ppmV. The gas pressure at the outlet of the reactor is measured by the pressure sensor. The gas concentration and pressure are measured at the outlet of the reaction chamber and sweep chamber, respectively. A valve is installed next to the pressure gauge at the outlet of the reaction chamber to investigate the effect of p_sweep on the performance of CH₄ dry reforming. When the valve is closed, the gas in the reaction chamber flows to the sweep chamber preferentially.

Figure 2 shows the detail and photo of the reactor. The reactor is composed of a reaction chamber, a sweep chamber, and anH₂ separation membrane. The reaction chamber and the sweep chamber are made of stainless steel, which has a size of 40 mm × 100 mm × 40 mm. The reaction space has a volume of 16 × 10⁻⁵ m³. Porous pure Ni catalyst is placed in the reaction chamber. The average pore diameter of the catalyst is 1.9 mm. The weight of the catalyst is 63.1 g. The Pd/Cu alloy membrane (Cu of 40 wt%; produced by Tanaka Kikinzoku Kogyo, Tokyo, Japan) installed between reaction and separation chambers can provide H₂ separation. The thickness of the Pd/Cu alloy membrane is 60 μm. The temperatures at the inlet, middle, and outlet of the reaction and sweep chambers are measured using K-type thermocouples. The measured temperature and pressure are collected by the data logger (GL240; produced by Graphtec Corporation, Kanagawa, Japan).

Table 1. Experimental parameters.

Reaction temperature (°C)	400, 500, 600
Pressure of supply gas (MPa)	0.10
Pressure of sweep gas, psweep (MPa)	0.10, 0.09, 0.045
Temperature of supply gas (°C)	25
Molar ratio of supplied CH4:CO2 (Flow rate of CH4 and CO2 (NL/min))	1.5:1, 1:1, 1:1.5 (1.088:0.725, 0.725:0.725, 0.725:1.088)
Feed ratio of sweep gas	1.0

lists the experimental parameters in this study. The molar ratio of the supplied CH₄:CO₂ is changed at 1.5:1, 1:1 or 1:1.5, where CH₄:CO₂ = 1.5:1 simulates a biogas. The feed ratio of sweep gas, defined as the flow rate of sweep gas divided by the flow rate of supply gas composed of CH₄ and CO₂ is set at 1.0 since the best performance of CH₄ dry reforming was obtained at the feed ratio of sweep gas at 1 according to the authors’ previous study [47]. The pressure in the sweep chamber is changed by 0.10 MPa, 0.09 MPa and 0.045 MPa by means of vacuum pump and Pirani gauge. The effect of molar ratio on the performance of CH₄ dry reforming has been investigated changing the reaction temperature of 400 °C, 500 °C and 600 °C. The gas concentrations in reaction and sweep chambers have been evaluated by FID gas chromatograph (produced by GL Science, Nishi Shinjuku, Japan) and methanizer (produced by GL Science). This study shows the average data of five trials for each experimental condition in the following figures. The distribution of each gas concentration has been below 10 %. H₂ selectivity and CO selectivity have also been evaluated.

2.2. Performance Evaluation of Proposed Reactor

The performance of the proposed reactor is evaluated by gas concentration at the outlet of reaction and sweep chambers, H₂ selectivity and CO selectivity. H₂ selectivity (S_H2) and CO selectivity (S_co) are defined as following:

S_H2 = C_{H2, out}/(C_{H2, out} + C_{CO, out}) × 100

(5)

S_CO = C_{CO, out}/(C_{H2, out} + C_{CO, out}) × 100

(6)

where S_H2 is H₂ selectivity (%), C_{H2, out} is the concentration of H₂ at the outlet of reaction and sweep chambers (ppmV), C_{CO, out} is the concentration of CO at the outlet of reaction chamber (ppmV), and S_CO is the CO selectivity (%).

In addition, this study also evaluates the performance of the proposed reactor by CH₄ conversion (X_CH4), CO₂ conversion (X_CO2) and H₂ yield (Y_H2). This study defines X_CH4 and X_CO2 following Equation (1). X_CH4, X_CO2 and Y_H2 are defined as follows:

X_CH4 = (2C_H2,,out)/(C_{CH4, in}) × 100

(7)

X_CO2 = (2C_{CO, out})/(C_{CO2, in}) × 100

(8)

Y_H2 = (1/2C_{H2, out})/(C_{CH4, in}) × 100

(9)

3. Results and Discussion

3.1. Effect of Pressure of Sweep Gas on the Performance of Dry Reforming under Different Reaction Temperatures

Figure 3, Figure 4 and Figure 5 compare relationship between the concentrations of CH₄, CO₂, H₂ and CO at the outlet of the reaction chamber changing reaction temperature, respectively. Reaction temperature is varied to 400 °C, 500 °C, or 600 °C. In these figures, the molar ratio of the supplied CH₄:CO₂ is 1.5:1. In addition, the error bars are also shown in these figures.

It is seen from Figure 3 that the impact of p_sweep on concentrations of CH₄ and CO₂ is a little, while the concentration of CH₄ at the reaction temperature of 600 °C decreases with a decrease in p_sweep. It is believed that H₂ separation is promoted with a decrease in p_sweep, resulting that CH₄ is consumed more. However, it is found from Figure 3 that the change in concentration of CO₂ with increase in reaction temperature is a small. Since the kinetic rates expressed by Arrehenius type of Equations (1), (2) and (4), which consume CO₂ in the reaction, are much smaller than that of Equation (3) [11], it is believed that the impact of the reaction temperature on the concentration of CO₂ is a small. Though this study focuses on CH₄ dry reforming, H₂O, which is the by-product in the reaction as shown in Equation (3), might be produced, resulting that the consumption of CH₄ is larger than that of CO₂ at the reaction temperature of 600 °C due to the higher reaction rate of Equation (3) [11].

According to Figure 4 and Figure 5, it is seen that the concentrations of H₂ and CO increase with incases in reaction temperature. According to the theoretical kinetic studies [12,48], the reaction rates of H₂ and CO increase with an increase in reaction temperature, resulting that the concentrations of H₂ and CO increase. This is the same tendency observed in the experimental study using Pd/Ag/Cu membrane changing the reaction temperature from 350 °C to 700 °C [32]. Since reaction temperature is too low, it is believed that H₂ is not produced at the reaction temperature of 400 °C. In addition, it is observed from Figure 4 and Figure 5 that the concentrations of H₂ and CO under negative pressure conditions are lower than those under the atmosphere pressure condition, i.e., p_sweep of 0.10 MPa. Since H₂ flux thorough Pd/Cu membrane is improved by increase in the pressure difference between reaction chamber and sweep chamber [45], it is thought that H₂ is moved from the reaction chamber to the sweep chamber. Therefore, the concentration of H₂ under negative pressure conditions is smaller than that under the atmosphere pressure condition. As to CO, it is believed that the following reaction with H₂O, which is by-product in the reaction shown in Equation (3), might have been occurred [49]:

CO + H₂O ↔ CO₂ + H₂

(10)

In addition, the following reactions on carbon deposition are thought to be occurred [32,50,51]:

2CO → CO₂ + C

(11)

CO + H₂ → H₂O + C

(12)

In this study, it is assumed that the concentration of CO decreases according to the combination of these reactions.

Table 2. Relationship between H₂, CO selectivity and p_sweep with change in reaction temperature.

psweep (MPa)	H2 Selectivity (%)			CO Selectivity (%)
	400 °C	500 °C			500 °C	600 °C
0.045	0	0.5	1.1	100	99.5	98.9
0.090	0	0.4	1.2	100	99.6	98.8
0.101	0	0.7	1.6	100	99.3	98.4

lists the relationship between H₂, CO selectivity and p_sweep, respectively. Reaction temperature is varied to 400 °C, 500 °C, or 600 °C.

According to Table 2, it is revealed that CO is produced mainly irrespective of p_sweep changing reaction temperature. Since reaction temperature is relatively low even the reaction temperature of 600 °C and H₂ selectivity increases with an increase in reaction temperature, especially over 600 °C [32], high CO selectivity is noticed in this study.

Table 3. Relationship between CH₄ conversion and p_sweep with change in reaction temperature.

psweep (MPa)	400 °C (%)	500 °C (%)	600 °C (%)
0.045	0.002	0.129	0.303
0.090	0.002	0.112	0.304
0.101	0.003	0.193	0.429

,

Table 4. Relationship between CO₂ conversion and p_sweep with change in reaction temperature.

psweep (MPa)	400 °C (%)	500 °C (%)	600 °C (%)
0.045	16.5	35.2	39.4
0.090	29.3	40.1	38.9
0.101	37.0	39.4	39.2

and

Table 5. Relationship between H₂ yield and p_sweep with change in reaction temperature.

psweep (MPa)	400 °C (%)	500 °C (%)	600 °C (%)
0.045	0.0006	0.0322	0.0758
0.090	0.0005	0.0281	0.0761
0.101	0.0007	0.0483	0.1072

list the relationship between CH₄ conversion, CO₂ conversion, H₂ yield and p_sweep, respectively. The reaction temperature is varied to 400 °C, 500 °C, or 600 °C.

According to Table 3 and Table 5, it is seen that CH₄ conversion and H₂ yield are small. It is thought due to low concentration of H₂ as shown in Figure 4. It is also seen from Table 3 and Table 4 that CO₂ conversion is larger compared to CH₄ conversion irrespective of p_sweep and reaction temperature, resulting from that the reaction shown by Equations (2) and (4) might be progressed well.

3.2. Effect of Pressure of Sweep Gas on the Performance of Dry Reforming under Different Molar Ratios

Figure 6, Figure 7 and Figure 8 compare the impact of concentrations of CH₄, CO₂, H₂ and CO at the outlet of the reaction chamber with changing the molar ratio of CH₄:CO₂. The molar ratio of CH₄:CO₂ is varied to 1.5:1, 1:1, 1:1.5. In these figures, the reaction temperature is 500 °C. In addition, the error bars are also shown in these figures.

According to Figure 6, it is found that the concentrations of CH₄ and CO₂ are changed with the molar ratio of CH₄:CO₂. In addition, it is seen from Figure 6 that the impact of p_sweep on the concentrations of CH₄ and CO is relatively small.

It is seen from Figure 7 that the concentration of H_2, in the case of the molar ratio of CH₄:CO₂ = 1:1 is the highest among different molar ratio conditions. Since the molar ratio of CH₄:CO₂ = 1:1 is the theoretical stoichiometric ratio according to Equation (1), it is believed that the reaction is progressed. Compared to the concentration of H₂ in the case of the molar ratio of CH₄:CO₂ = 1.5:1 with that in the case of the molar ratio of CH₄:CO₂ = 1:1.5, the former is higher than the latter. When the molar ratio of CH₄ is larger, it is believed that Equation (3) is progressed well after H₂O production by the reactions shown by Equations (2) and (4). In addition, it is observed from Figure 7 that the concentrations of H₂ under negative pressure conditions are lower than those under the atmosphere pressure condition, except for the case of the molar ratio of CH₄:CO₂ = 1:1. Since H₂ flux thorough Pd/Cu membrane is improved by increase in the pressure difference between reaction chamber and sweep chamber [45], it is thought that H₂ is moved from the reaction chamber to the sweep chamber. Therefore, the concentration of H₂ under negative pressure conditions is smaller than that under the atmosphere pressure condition. The amount of produced H₂ might be large in the case of the molar ratio of CH₄:CO₂ = 1:1, and the thickness of Pd/Cu membrane might not be sufficient, resulting that H₂ flux is not sufficient to extract H₂ by Pd/Cu membrane even p_sweep at 0.045 MPa.

It is seen from Figure 8 that the concentration of CO in the case of the molar ratio of CH₄:CO₂ = 1.5:1 is the highest among different molar ratio conditions. At the molar ratio of CH₄:CO₂ = 1.5:1, it is believed that Equation (3) is progressed well after H₂O production by the reactions shown by Equations (2) and (4). As a result, the concentration of CO increases in the case of the molar ratio of CH₄:CO₂ = 1.5:1. Additionally, it is observed from Figure 8 that the concentration of CO in the case of the molar ratio of CH₄:CO₂ = 1:1.5 is low at p_sweep of 0.10 MPa. Since the performance of dry reforming of CH₄ is worse, and it can not obtain the side-effect of CO production by H₂ separation due to no pressure difference between reaction chamber and sweep chamber, it is believed that the concentration of CO is low under this condition.

Table 6. Relationship between H₂, CO selectivity and p_sweep changing molar ratios of CH₄:CO₂.

psweep (MPa)	H2 Selectivity (%)			CO Selectivity (%)
	CH4:CO2 = 1.5:1	CH4:CO2 = 1:1	CH4:CO2 = 1:1.5	CH4:CO2 = 1.5:1	CH4:CO2 = 1:1	CH4:CO2 = 1:1.5
0.045	0.5	1.0	0.3	99.5	99.0	99.7
0.090	0.4	0.9	0.5	99.6	99.1	99.5
0.101	0.7	0.9	1.4	99.3	99.1	98.6

lists the relationship between H₂, CO selectivity and p_sweep. The molar ratio of CH₄:CO₂ is varied to 1.5:1, 1:1, 1:1.5. In this table, the reaction temperature is 500 °C.

According to Table 6, it is revealed that CO is produced mainly irrespective of p_sweep changing reaction temperature. Since reaction temperature is relatively lower even the reaction temperature of 600 °C and H₂ selectivity increases with an increase in reaction temperature, especially over 600 °C [32], high CO selectivity is obtained in this study.

Table 7. Relationship between CH₄ conversion, CO₂ conversion and p_sweep changing molar ratios of CH₄:CO_2.

Psweep (MPa)	CH4 Conversion (%)			CO2 Conversion (%)
	CH4:CO2 = 1.5:1	CH4:CO2 = 1:1	CH4:CO2 = 1:1.5	CH4:CO2 = 1.5:1	CH4:CO2 = 1:1	CH4:CO2 = 1:1.5
0.045	0.129	0.231	0.047	35.2	33.5	22.2
0.090	0.112	0.162	0.072	40.1	27.8	20.6
0.101	0.193	0.178	0.105	39.4	29.0	9.85

and

Table 8. Relationship between H₂ yield and p_sweep changing molar ratios of CH₄:CO₂.

Psweep (MPa)	H2 Yield (%)
	CH4:CO2 = 1.5:1	CH4:CO2 = 1:1	CH4:CO2 = 1:1.5
0.045	0.0322	0.0578	0.0117
0.090	0.0281	0.0405	0.0180
0.101	0.0483	0.0444	0.0262

list the relationship between CH₄, CO₂ conversion, H₂ yield and p_sweep changing moral ratio of CH₄:CO₂.

According to Table 7 and Table 8, it is seen that the CH₄ conversion and H₂ yield is the highest in the case of CH₄:CO₂ = 1:1, respectively. It is because the concentration of H₂ is the highest in the case of CH₄:CO₂ = 1:1, as shown in Figure 7. In addition, it is found from Table 7 that CO₂ conversion is higher than CH₄ conversion irrespective of molar ratio of CH₄:CO₂. Since the reactions shown by Equations (2) and (4) might be progressed well, this tendency is obtained.

3.3. Effect of Pressure of Sweep Gas on H₂ Flux from Reaction Chamber to Sweep Chamber

Figure 9 compares the impact of concentrations of H₂ at the outlet of the sweep chamber with changing reaction temperature. The reaction temperature is varied to 400 °C, 500 °C, or 600 °C. The error bars are also shown in this figure. It is seen from Figure 9 that the concentration of H₂ increases with a decrease in p_sweep irrespective of reaction temperature as expected. This is due to an increase in H₂ flux depending on the pressure difference between the reaction chamber and sweep chamber. In addition, the concentration of H₂ is the highest at the reaction temperature of 500 °C and p_sweep of 0.045 MPa. The reason is discussed as follows:

In this study, the pressures of reaction chamber and sweep chamber have been measured by pressure sensors. The permeation flux through Pd/Cu membrane is calculated using the measured pressures of reaction chamber and sweep chamber as follows [52]:

$$J=\frac{Pe\left(\sqrt{p_{react}}-\sqrt{p_{sweep}}\right)}{\delta }$$

(13)

where J is the permeation flux (mol/(m²·s)), Pe is the permeability and it is an inherent property of membrane (mol/(m·s·Pa^0.5)), p_react is the pressure of reaction chamber (MPa), p_sweep is the pressure of sweep chamber (MPa) and δ is the thickness of membrane (m). Pe of Pd/Cu membrane, used in this study, is listed in

Table 9. Pe of Pd/Cu membrane used in this study [53].

Temperature (°C)	Pe (mol/(m·s·Pa0.5))
400	1.0 × 10−8
500	5.0 × 10−9
600	1.0 × 10−9

.

Figure 10 shows the impact of J and p_sweep with changing reaction temperature and molar ratio of CH₄:CO₂.

It is observed from Figure 10 that J increases with decrease in p_sweep irrespective of reaction temperature and molar ratio of CH₄:CO₂. It is due to an increase in the pressure difference between the reaction chamber and the sweep chamber. Additionally, it is found that J decreases with an increase in reaction temperature irrespective of p_sweep and molar ratio of CH₄:CO₂. According to Table 9, Pe of Pd/Cu membrane decreases with an increase in temperature, resulting that J decreases with an increase in reaction temperature.

In this study, J is the highest at the reaction temperature of 400 °C, while the amount of produced H₂ is the smallest at the reaction temperature of 400 °C due to a low kinetic production rate depending on temperature [54]. On the other hand, the amount of produced H₂ is the highest at the reaction temperature of 600 °C, while J is the lowest at the reaction temperature of 600 °C. Considering these tendencies, it can be thought that the reaction temperature of 500 °C is the optimum temperature, as shown in Figure 9.

In this study, the performance of CH₄ dry reforming has not been high. According to the previous review paper on CH₄ dry reforming [55], CH₄ and CO₂ conversion were over 80%, though the reaction temperature was over 700 °C using several alloy catalysts, such as Ni/MgO, Ni/Ce_0.8Zr_0.2O₂, Ni/Co and Ni/CeO₂. The previous study also reported that not only CH₄ conversion, but also H₂ yield increased with an increase in the reaction temperature from 550 °C to 750 °C [56]. In addition, it was also reported that H₂ and CO yield increased with an increase in the reaction temperature from 450 °C to 700 °C, especially over 600 °C [32]. H₂ and CO yield attained at 30% and 45%, respectively at 700 °C [32]. Though the previous studies exhibited the better performance compared to this study, the reaction temperature is higher than this study. Reaction temperature is thought to be one essential parameter to decide the performance of CH₄ dry reforming, resulting that this study will investigate the higher reaction temperature condition in the near future. According to the numerical kinematic study on CH₄ steam reforming [57], which is one of the conventional approaches for H₂ production, the CH₄ conversion was approximately 65% at the reaction temperature of 650 °C and the pressure of 3.5 MPa with Ni/Al₂O₃ catalyst in a fixed bed reactor. In addition, the other experimental and numerical previous study on CH₄ steam reforming over NiO/Al₂O₃ catalyst [58] reported that the rate of CO formation increased with an increase in reaction temperature and almost all the CH₄ was converted to CO and H₂ at the reaction temperature of 800 °C. At 800 °C, CH₄ conversion, H₂ production rate and CO selectivity were 92.28%, 3.34% and 99%, respectively. It can be also known from these references on the conventional approach for H₂ production that reaction temperature is an important parameter to improve the performance of H₂ production.

It is necessary to improve the performance of CH₄ dry reforming from the view point of reaction kinetics and H₂ separation. Though only pure Ni catalyst has been investigated in this study, Ni-based alloy catalysts, such as Ni/Ce [6], Ni/La [6] and Ni/ZnO-Al₂O₃ [17] exhibited the better performance, and they are promising candidate catalysts for further studies. In addition, the characterization of H₂ separation membrane, such as composition, thickness, and metal type, is important to improve the efficiency of H₂ separation. Since this study has investigated one type of H₂ separation membrane only, i.e., Pd/Cu alloy membrane (Cu of 40 wt%) whose thickness is 60 μm, the different composition and thinner membrane will also be investigated in the near future.

In addition, the control of pressure difference between reaction chamber and sweep chamber is important to improve the performance of H₂ separation membrane. The numerical analysis on biogas dry reforming membrane reactor reported that the system efficiency, which was defined by the ratio of H₂ energy output to biogas and auxiliaries energy inputs, increased with an increase in the pressure difference between reaction chamber and sweep chamber, and H₂ production efficiency was attained around 69 % at the pressure difference of 20 bar (=0.2 MPa) [59]. Since the pressure difference provided by this study is up to 0.055 MPa, it might be desirable to investigate the higher pressure difference in order to improve the proposed membrane reactor.

Finally, this study suggests optimizing the catalyst, H₂ separation membrane and operation condition in order to obtain a progress compared to other similar studies.

4. Conclusions

This study has studied the effect of p_sweep on the performance of biogas dry reforming. This study has investigated the effect changing the reaction temperature from 400 °C to 600 °C, and the molar ratio of CH₄:CO₂ by 1.5:1, 1:1 and 1:1.5, where the molar ratio of CH₄:CO₂ = 1.5:1 simulates a biogas. As a result, the following conclusions are drawn from the obtained results:

(i)

Under the condition of the molar ratio of CH₄:CO₂ = 1.5:1 changing the reaction temperature by 400 °C, 500 °C, 600 °C, the impact of p_sweep on concentrations of CH₄ and CO at the outlet of reaction chamber is negligible. It is concluded that the concentrations of H₂ and CO at the outlet of reaction chamber increase with incase in the reaction temperature. It is revealed that the concentration of H₂ at the outlet of reaction chamber, under negative pressure conditions, is smaller than that under the atmosphere pressure condition. High CO selectivity is obtained, while H₂ selectivity is low.

(ii)

Under the condition of the reaction temperature of 500 °C changing molar ratio of CH₄:CO₂, the highest concentration of H₂ at the outlet of reaction chamber is obtained at the molar ratio of CH₄:CO₂ = 1:1. In addition, the concentrations of H₂ at the outlet of reaction chamber under negative pressure conditions are lower than those under the atmosphere pressure condition, except for the case of the molar ratio of CH₄:CO₂ = 1:1. The concentration of CO at the outlet of reaction chamber is the highest in the case of the molar ratio of CH₄:CO₂ = 1.5:1. High CO selectivity is obtained, while H₂ selectivity is low.

(iii)

CO₂ conversion is larger compared to CH₄ conversion irrespective of p_sweep, reaction temperature and molar ratio of CH₄:CO₂. CH₄ conversion and H₂ yield is the highest in the case of CH₄:CO₂ = 1:1, respectively.

(iv)

The concentration of H₂ at the outlet of sweep chamber is the highest at the reaction temperature of 500 °C and p_sweep of 0.045 MPa. It is due to the balance between the permeation flux and reaction kinetics.

According to the investigation by this study, it is necessary to improve the performance of CH₄ dry reforming from the view point of reaction kinetics and H₂ separation. Ni-based alloy catalyst is a promising catalyst in order to improve the performance of CH₄ dry reforming. In addition, it is important to clarify the impact of the characterization of H₂ separation membrane, such as composition, thickness, and metal type, on the efficiency of H₂ separation. Furthermore, it is also important to control the pressure difference between reaction chamber and sweep chamber in order to improve the performance of the H₂ separation membrane.