Innovative Fixed-Bed Reactor Integrated with Heat Transfer System for Lean Methane Mixture Removal

Abstract

A new type of compact, portable fixed-bed reactor integrated with a heat transfer system was developed for the removal of volatile and flammable air pollutants such as lean methane and volatile organic compounds (VOCs). The reactor may operate in catalytic or thermal combustion conditions with the purpose of achieving autothermal processes with the possibility of energy recovery. An excess heat recovery point was designed behind the reactor bed at the place where the gas temperature is the highest to enable its usage. The mathematical model is presented together with a number of simulation calculations performed for the assessment of the developed reactor. The case study in this paper was for catalytic methane oxidation at a temperature of 400 °C, a methane concentration between 0.1% and 2% by weight, a gas flow rate of 1 m³/s STP, and a heat exchange surface for the assumed plate exchanger from 10 to 200 m². The calculations show that the thickness of the insulation is of little importance for the operation of the equipment, and a sufficient thickness was about 20–50 mm. The optimal area for the considered case is 80–100 m². It was found that recovery of thermal energy is possible only for higher methane concentrations, above 0.3% by weight. Using an appropriate surface for the exchanger, it is possible to recover even 50% of the combustion enthalpy at a methane concentration of 0.45% by weight. For an exchanger area below 50 m², the recoverable energy drops rapidly. It was found that the exchanger area is the most important equipment parameter under consideration.

1. Introduction

Growing greenhouse gas (GHG) emissions together with growing volatile organic compound (VOC) emissions have been recognized as one of the major hazards for both the environment and human health [1,2].

These emissions come from diverse sources. Methane global warming potential (GWP) shows that it absorbs 37 times more heat than carbon dioxide [3]. Important sources of world methane emissions on global scale estimated by IEA are agriculture (40%), energy (37%), waste (20%), and other (3%) [4]. A common feature of all these emissions is low methane concentrations in the air.

A large group of VOCs encompasses numerous hydrocarbons, including aromatic ones, as well as other organic compounds, like alcohols, aldehydes, and odors (e.g., mercaptans). Many of them are considered precursors of photochemical smog [5,6]. Some VOCs are carcinogenic and toxic compounds dangerous to human health [7,8]. The emission sources are as diverse as the VOCs themselves; these may be industrial plants, chemical or petrochemical plants, pharmaceutical plants, paint shops, the plastics industry, and many others.

In order to limit the harmful impact of these emissions, effective technologies and solutions are needed [9,10]. Various removal technologies have been developed, among which thermal and catalytic oxidation appear to be effective methods able to completely degrade VOCs or methane emissions. These methods are characterized by high economic feasibility and low pollutant generation [11,12,13].

Numerous research works related to catalytic and thermal combustion of atmospheric emissions of methane and VOCs have been presented in the literature [14,15,16,17,18]. They relate mainly to low concentrations of pollutants for which the heat of combustion is insufficient to maintain the process of autothermy; thus additional heating of the inlet gas is sometimes necessary.

A low concentration of gaseous pollutants is a common feature of the majority of methane and VOC emissions mentioned above. In the case of so-called lean streams, the thermal (or catalytic) combustion is not the main problem, but heat management. The—frequently large—streams of inert gases (usually air) have first to be heated to the appropriate temperature, at which the oxidation reaction starts (also referred to as the light-off temperature). The light-off temperature is distinctly lower for catalytic combustion compared with thermal combustion. Therefore, for the catalytic process, the heat losses are lower as well. Some VOCs are combustible species and heat is released during their oxidation equal to the reaction enthalpy; this heat is, however, usually low due to low concentrations. The heat of reaction is sometimes insufficient to cover thermal losses, i.e., to achieve autothermy in the process, despite the high efficiency of heat recovery from exhaust gases. In such a situation, thermal combustion is not sufficient, and the catalytic process may be necessary. However, the catalytic process also has its weaknesses; the catalyst is sensitive to so-called catalytic poisons that may be present in the gas. At slightly higher VOC concentrations, or with fluctuations in these concentrations, the enthalpy reaction may cause overheating of the catalyst and, consequently, its thermal deactivation; in such cases, the thermal process is preferable.

In spite of significant advances in catalyst development for methane or VOC oxidation, achievement of autothermal operation requires an appropriate design of the reactor. The reactor feed stream temperature is usually close to the ambient one, while the process requires an appropriate temperature, of at least the light-off temperature, for the reaction to occur. A good idea is to integrate the reactor with a heat exchanger into a single unit to recover heat from the exhaust gases, to heat the feed stream, and to reduce heat losses [18,19].

For the combustion of lean emissions, different reactors, operating in non-stationary conditions, are proposed in the literature, like the auto-cyclic packed bed reactor [17,18] or flow–reverse-flow reactor [20,21].

In this paper, a new type of compact, modular, portable reactor with an integrated heat transfer system was developed. The developed reactor may operate with either catalytic or thermal-type combustion. The main goal was to achieve an autothermal process [19] for the lowest possible concentrations of pollutants, with the option of additional heating of the gas feed (if necessary) or removing the excess heat from the exhaust stream (if possible). What is new is the location of the excess heat recovery point directly behind the reactor bed, at the point with the highest gas temperature. This solution slightly complicates the process control, but it provides heat with the highest temperature potential, which may—under certain conditions—enable its use, e.g., for the production of electricity.

Combustion of lean methane mixtures with air is the case study with total conversion assumed, neglecting the reaction nature (thermal or catalytic oxidation). The model derived for the thermal energy transfer within the integrated reactor enables us to assess the impact of heat exchanger size and insulation thickness on the apparatus’s functioning and optimization. The model derived may also be applied to the oxidation of VOCs, taking into account different concentrations, reaction enthalpy, and possibly lower final conversion.

2. Apparatus Design Conception

The goal is to design a highly efficient reactor to combust lean mixtures of VOCs or methane with air. The sketch of the apparatus is presented in Figure 1. The reactor can realize either catalytic or thermal combustion, and it is integrated with a complete heat exchange system that includes three heat exchangers. The whole device is thermally insulated to reduce heat losses.

The fixed-bed reactor 1 (Figure 1) is filled with, e.g., a dumped bed of particles or a ceramic monolith. The reactor filling may contain a reaction catalyst if a catalytic process is carried out, or it may be inert in the case of thermal afterburning. The choice of post-combustion method, catalytic or thermal, depends on the specific case. The catalytic process usually takes place at lower temperatures, resulting in lower heat losses and a lower initial reaction temperature. In this case, it is easier to achieve autothermy in the process, which is not without significance due to the high dilution of the oxidized substances. However, the catalyst can often undergo chemical deactivation in the presence of catalytic poisons; the price of the catalyst is also important. Most organic substances have a high enthalpy of combustion, which—by increasing the temperature of the reactants—may lead to thermal deactivation of the catalyst, especially in the case of significant concentration fluctuations. Thermal afterburning is free from such problems, but requires the use of much higher temperatures—hence greater energy losses and problems with achieving autothermy. There is often a need to use more expensive materials that are resistant to high temperatures. However, the details of the reactor functioning are not considered here. As the processes of combustion, both catalytic and thermal, are very fast, the assumed reaction conversion is 100%. Based on our earlier research, reactors a dozen centimeters long can achieve such a conversion [22].

One major problem to overcome in the case of lean mixtures is to achieve the autothermal process. For this purpose, a reactor and a system of three heat exchangers were integrated into one device. A compact design with good insulation helps to reduce heat loss. The main heat exchanger (2) (Figure 1) allows for energy recovery of hot waste gases to preheat the inlet gas. This exchanger was designed as a plate heat exchanger, with plates perpendicular to the apparatus axis. In this situation, the rectangular shape of the device was chosen as the most convenient. The preheated inlet gas flows through the central tube (5) to the inlet part of the reactor (7), where—if necessary—it can be additionally heated to the required temperature. The start-up exchanger (6) is used for this purpose; in Figure 1, it is presented as a burner (one possible implementation), and it could also be an electric heater or an exchanger powered by any hot medium, if available. This exchanger (6) is mainly used during reactor start-up.

The gases leaving the reactor have the highest temperature; it is often possible to use part of their thermal energy and transfer it outside the reactor and use it for other purposes, e.g., to produce electricity. The heat exchanger (8) was designed for this purpose. The energy discharged in this place has the highest temperature potential possible in the device under consideration. This exchanger (8) can be implemented as a tubular exchanger, with water evaporating in the pipes.

However, such a design of energy recovery complicates the control of the equipment. The key problem is to heat the inlet gas stream to the desired temperature T₁ (arbitrarily established here at 400 °C) in the reactor inlet part (7); the heat exchanger (2) is designed for this purpose. However, the inlet gas temperature should not be too high, in order to avoid catalyst overheating. The most convenient solution would be to remove excess heat in the inlet part of the reactor (7). However, from the energetic point of view, the best solution is to discharge the excess energy in the region of the highest temperature, i.e., directly behind the reactor (temperature T₂). To ensure the required temperature T₁, only enough energy should be received in the exchanger (8) so that the temperature T₃ is only slightly higher than the desired T₁ value (by the amount resulting from the heat losses and the efficiency of the exchanger (2)). Note the same gas amount flows in the exchanger (2) in both directions; thus the temperature increase (T₁ − T_in) is almost equal to the decrease (T₃ − T_out). Therefore, the key challenge of controlling the process is to keep the temperature T₃ at the desired level despite the fluctuations of gas flow and organic pollutants concentration. Modern advanced controllers can meet this challenge.

3. Integrated Reactor Model

The designations of the model parameters are shown in the reactor scheme in Figure 1. The apparatus considered here consists of a packed-bed reactor (catalytic or inert in the case of thermal afterburning) and three heat exchangers. The first exchanger is used to heat the inlet gas with the enthalpy of the outlet gases. The second one is used for possible additional heating of the gas before entering the reactor and is mainly used during equipment start-up; this exchanger is not included in the model presented here. The third exchanger is used to receive some thermal energy Q_en from the hot gases leaving the reactor. The value of Q_en is calculated in the model so that the gas flowing into the reactor has the required temperature T₁. This solution complicates the model and automation of the equipment, but allows for recovery of heat with the highest temperature potential, which is beneficial for obtaining mechanical or electrical energy in a possible heat engine

The reactor model assumptions are as follows:

• Steady state conditions
• Methane conversion of 100%
• Reactor inlet temperature constant and equal to T₁ = 400 °C

Heat flux losses Q_L due to heat conduction through the insulation are

$$Q_L=F_R\left(T_R-T_0\right){\displaystyle \frac{{\lambda }_i}{s_i}}$$

(1)

where

• T_R—average temperature of the apparatus (reactor) surface, $T_R={\displaystyle \frac{1}{F_R}}{\int }_0^{F_R}TdF$;
• T₀—ambient temperature;
• λ_i—insulation heat conduction;
• s_i—insulation thickness;
• F_R—whole apparatus external surface area.

As stated before, to maintain the reactor inlet temperature T₁ = 400 °C, some energy Q_en has to be received in the exchanger 8:

$$Q_{en}=G{c}_p\left(T_2-T_3\right)$$

(2)

Heat flux is supplied by the reaction of methane combustion in the reaction gas:

$$Q_R=G{x}_{{\mathrm{C}\mathrm{H}}_4}{\displaystyle \frac{{\mathsf{\Delta }H}_R}{M}}$$

(3)

The heat exchanger balance is calculated thus:

$$G{c}_p\left(T_3-T_{out}\right)=G{c}_p\left(T_{12}-T_{in}\right)+Q_L$$

(4)

where, for the steady state, T₁₂ = T₁. During equipment start-up, the heat flux supplied by the second exchanger amounts to

$$Q_{start}=G{c}_p\left(T_1-T_{12}\right)$$

(5)

and this is a temporary value, gradually decreasing until the assumed value T₁₂ = T₁ is reached.

The balance equation of the heat exchanger (4) results in

$$T_{12}=T_1=T_{in}-T_{out}+T_3-{\displaystyle \frac{Q_L}{G{c}_p}}$$

(6)

The heat flux of the methane combustion reaction Q_R increases the gas temperature from T₁ up to T₂:

$$Q_R=G{c}_p\left(T_2-T_1\right)$$

(7)

Combining equations (2) and (7) for T₁₂ = T₁ gives

$$T_3=T_1+{\displaystyle \frac{Q_R-Q_{en}}{G{c}_p}}$$

(8)

Equations (6) and (8) give

$$Q_{en}=Q_R-Q_L-G{c}_p\left(T_{out}-T_{in}\right)$$

(9)

Hence, the energy Q_en corresponds to the value of the methane combustion energy minus the heat loss Q_L and the enthalpy lost from the gas leaving the apparatus (last term of Equation (9). This enthalpy is always non-negative because the outlet gas temperature T_out is always higher than the inlet gas temperature T_in. Due to the possible variability of the methane concentration (and therefore the Q_R value) and the difficulty in determining the heat losses Q_L in the actual reactor, the amount of heat Q_en will be adjusted by the automatic control system so that the temperature of the gas flowing into the reactor has the assumed value.

The case study in this paper was catalytic methane oxidation. The reaction light-off temperature was 400 °C. The assumed methane concentration in the mixture with air was between 0.1% and 2% by weight, and the gas flow rate was 1 m³/s STP. Due to the low content of methane and its combustion products in the mixture, it was assumed that the properties of the mixture were the same as those of clean air.

It was assumed that the heat exchanger would be a plate exchanger with a single-pass connection layout, with dimensions of 1 m × 1 m and a length depending on the number of plates. The heat exchange surface considered was 10 to 200 m², and the heat transfer coefficient for the gas–gas flow was assumed to be 19 W/(m²K).

The calculations were carried out iteratively using MATLAB R2024a. Calculations were carried out until the relative error and absolute error were less than 10⁻⁶.

4. Results and Discussion

Based on the mathematical model presented in the previous chapter, a number of simulation calculations were performed for the most comprehensive assessment of the functioning of the developed reactor. Particular attention was paid to the issue of achieving autothermy of the process, including the thermal insulation of the apparatus (in order to reduce heat losses) and the basic heat exchanger (item 2 in Figure 1), enabling the recovery of thermal energy from the gases leaving the apparatus.

Figure 2 shows the dependence of the gas temperature at the device outlet (T_out, Figure 1) on the surface of the heat exchanger F_ex. Interestingly, while the T_out temperature is significantly influenced by the surface of the F_ex heat exchanger, the curves for all considered methane concentrations coincide. The influence of the thickness of the device’s thermal insulation is also not visible here.

During the operation, such an amount of heat Q_en is received in the exchanger at 8 that the temperature at the inlet to the reactor T₁ is at the set level, in this case 400 °C. As a result, the amount of heat exchanged in the exchanger at 2 is basically the same, regardless of the concentration of methane at the inlet to the apparatus. Of course, the exchanger at 8 must have adequate capacity (i.e., heat exchange surface) to remove the excess energy of combustion of the higher methane content.

Figure 3 shows the influence of the insulation thickness s and the heat transfer area F_ex on the thermal losses Q_L of the equipment for the assumed methane concentration of 0.5% by weight. The total surface area of the apparatus slightly depends on the heat transfer surface F_ex; hence a certain increase in heat losses for larger surfaces F_ex. On the other hand, a larger exchanger surface reduces energy losses by lowering the outlet gas temperature. The optimal range is F_ex = 50–100 m² and for an insulation thickness of not less than 50 mm.

Figure 4 shows the range of autothermal operation of the device for the assumed lack of heat recovery in exchanger 8 (Q_en = 0), depending on the methane concentration x_CH4, the surface of exchanger 2 F_ex, and the insulation thickness s. The graph clearly shows that the fundamental influence on achieving autothermal has a methane concentration x_CH4 and the heat transfer surface F_ex. The influence of insulation thickness is small; in fact, even 20 mm-thick insulation seems to be sufficient. However, a methane concentration of 0.25% by weight seems to be the lowest at which autothermal conditions can be maintained. This requires a heat exchange surface of F_ex = 80 m². For lower concentrations, a significant increase in the exchanger surface would be necessary, which would be economically unacceptable.

In Figure 5, the ratio of the recovered heat Q_en to the heat of reaction Q_R is presented; thus the Q_en/Q_R ratio vs. methane concentration x_CH4 and heat exchanger surface area is F_ex. The insulation thickness s = 20 mm is assumed according to the above paragraph. Formerly, a heat transfer surface area F_ex = 80 m² was recommended (see description of Figure 4). For such a surface area F_ex, a practical methane concentration higher than 0.3% by weight is the lowest limit to remove the excessive heat in the exchanger 8. For F_ex = 80 m² and methane content of 0.4% by weight, the ratio Q_en/Q_R = 0.45; thus a significant amount of energy can be utilized.

Figure 6 shows the dependence of the received energy Q_en on the exchanger surface F_ex and methane concentration. It is clearly visible that for an exchanger area below 50 m², the energy recovery Q_en decreases rapidly even for high methane concentrations. On the other hand, the increase in recoverable energy for areas exceeding 100 m² is moderate for all methane concentrations considered here. An exchanger area in the range of 80–100 m² seems optimal; for concentrations from 0.4% by weight, it is possible to recover significant amounts of energy. Please note that the line corresponding to Q_en = 0 is not at the very bottom of the graph; the chart also covers the range Q_en < 0.

Figure 7 shows the spatial dependence of the recovered energy Q_en on the surface of the exchanger F_ex and the methane concentration x_CH4. The graphs in Figure 5 and Figure 6 are projections of the surface in Figure 7 onto the appropriate planes of the spatial coordinate system. The conclusions from Figure 5, Figure 6 and Figure 7 are complementary.

The computational results presented here show that the surface of the heat exchanger F_ex has the greatest impact on the operation of the equipment in terms of energy, in particular on its autothermy and the possibility of energy recovery. The influence of the thermal resistance of insulation is small. Basically, the designer has no influence on the content of methane (or other substances) in the gas supplied to the reactor. The minimum concentration of methane necessary for ensuring the autothermal process was determined here at 0.25% by weight, and energy recovery is practically profitable from 0.4% by weight. The area of the heat exchanger for a given case should be 80–100 m².

5. Summary and Final Conclusions

The paper presents the concept of a reactor used for post-combustion (catalytic or thermal) of highly diluted mixtures of volatile organic compounds and methane with air, integrated with a system of three heat exchangers. The research focused on the energy problems of the reactor, with particular emphasis on achieving (maintaining) autothermal operating conditions and recovery of excess thermal energy at higher concentrations of post-burning substances. However, the reactor model and calculations regarding its operation were not presented. Experience from previous research and the literature shows that the post-combustion processes of organic substances are fast, and often only a few centimeters of reactor length are sufficient to achieve complete conversion. It is much more difficult to achieve the temperature required for the reaction, the so-called light-off temperature, for significantly dilute mixtures of organic compounds. For this purpose, a number of computational results are presented.

The calculations show that the thickness of the insulation used is of little importance for the operation of the equipment, including maintaining the autothermy of the process. It follows from Figure 3 and Figure 4 that the thickness of the insulation used is of little importance for the operation of the equipment, including maintaining the autothermy of the process. Figure 3 and Figure 4 show that typical thermal insulation (e.g., mineral wool) with a thickness of 20–50 mm is sufficient; for larger thicknesses, the results are almost identical.

What is essential for the functioning of the integrated reactor is the heat exchange surface F_ex of the exchanger at 2 (Figure 1). Figure 2 and Figure 4 show the important role of this element of the equipment. However, the analysis performed indicates that the optimal area for the considered case is 80–100 m². Larger surfaces are not economically justified, and it seems cheaper to reheat the gas, e.g., with a gas burner 6 (Figure 1), in the case of very low concentrations of post-burning pollutants.

The scope of autothermy of the process is shown in Figure 4. The key factor here is the concentration of methane (or other combustion substance) in the air. Of course, as the designer has no influence on this concentration, the only possibility is to increase the surface of the F_ex exchanger. Figure 4 shows that it is possible to obtain autothermal conditions with a methane concentration of 0.25 wt%. and area 80 m². Lower concentrations require additional heating with a burner (6) (Figure 1); increasing the heat exchange surface is economically irrational.

What is important is the possibility of recovering thermal energy Q_en in the exchanger at 8 (Figure 1). This is only possible for higher methane concentrations, above 0.3% by weight, with 0.4% being the profitability limit. Energy recovery possibilities are shown in Figure 5, Figure 6 and Figure 7. By using an appropriate surface of the exchanger F_ex, say 80 m² for the case study, it is possible, for example, to recover approximately 50% of the combustion enthalpy at a methane concentration of 0.45% by weight (Figure 5).

Figure 6 shows that for an exchanger area below 50 m², the value of recoverable energy Q_en drops rapidly, even below zero (which means the need for additional heating of the gas before the reactor). The area of heat exchanger 2 (Figure 1), F_ex, is the key and most important parameter that should be determined for the case under consideration. This is basically the only parameter of the equipment that can be determined by the designer.