Industry new
Selection and Loading of CO Purification Catalysts for Mine Refuge Chambers: Humidity-Prioritized Decision-Making and Engineering Design
I. Unique Operating Conditions for CO Purification in Refuge Chambers
1.1 Fundamental Differences from Conventional Industrial Scenarios
Catalysts for CO removal from industrial exhaust typically operate continuously under relatively stable conditions of temperature, humidity, and concentration, supported by scheduled maintenance and replacement plans. In contrast, the operating conditions within a refuge chamber exhibit four distinct characteristics:
Suddenness and Immediacy: CO leakage or smoke intrusion occurs within minutes; the purification unit must activate instantly, with no time available for preheating.
Enclosure: Air circulation within the chamber is poor, and CO does not disperse naturally; purification relies entirely on the active oxidation capability of the catalytic material.
Uncertain Operating Conditions: The underground environment is characterized by high humidity year-round, and the air may contain trace impurities such as sulfides and dust. Sulfides can accumulate on the metal active sites at low temperatures, directly inhibiting CO adsorption through a cumulative poisoning effect.
Long-term Standby: The equipment may remain unused for years; the catalytic material must retain its rated activity even after prolonged storage.
1.2 Key Design Parameters and Regulatory Requirements
According to the *Interim Provisions on the Construction and Management of Emergency Refuge Systems in Coal Mines*, the CO removal capacity of the hazardous gas treatment facility must ensure that the CO concentration drops from 0.04% (400 ppm) to 0.0024% (below 24 ppm) within 20 minutes. Refuge systems are typically designed to accommodate 6 to 100 people, with a rated protection duration of at least 72 hours (some designs require 96 hours). The life-support system comprises subsystems such as compressed-air self-rescue oxygen supply, CO₂ absorption units, CO catalytic purification equipment, and temperature/humidity control. Target conditions within the chamber are a temperature of ≤35°C and a relative humidity of ≤85% RH. Technical specifications for the CO catalytic purification equipment require an outlet CO concentration of ≤24 ppm, with levels not exceeding 25 ppm throughout the rated protection period.
A constraint requiring particular attention is the lower limit of O₂ concentration. The catalytic oxidation reaction of CO is represented by 2CO + O₂ → 2CO₂; treating 1 mole of CO consumes 0.5 moles of O₂. Based on a scenario with a rated capacity of 50 people and a refuge chamber volume of approximately 200 m³, if the initial CO concentration is 400 ppm and must be reduced to a safe level within 20 minutes, the total amount of CO to be treated is approximately 0.08 moles. The corresponding O₂ consumption is about 0.04 moles, which is negligible compared to the total O₂ inventory within the chamber. However, under extreme conditions—such as continuous massive influx of external CO combined with a nearly hermetically sealed chamber—the O₂ consumption associated with catalytic oxidation should still be factored into the oxygen balance calculations. It is recommended to install an O₂ concentration sensor upstream of the catalyst inlet; if the O₂ volume fraction in the chamber drops below 18%, priority should be given to oxygen supply rather than CO catalysis to prevent the catalytic process from further depleting O₂ levels. The compressed-air oxygen supply system must ensure that the per-capita airflow rate meets the requirements for rapid CO purification; relevant studies indicate that a per-capita compressed-air supply of 450 L/min is necessary to satisfy the rapid CO purification requirements stipulated in interim regulations. Additionally, the activated carbon layer within the CO catalytic purification unit may release trace amounts of heat while adsorbing acidic gases such as SO₂ and H₂S, and the adsorption process of desiccants (silica gel/molecular sieves) is also exothermic. Although the impact on the chamber's overall thermal balance is limited, this heat generation should be accounted for in the thermal control design for the enclosed space.
II. Selection and Performance Evaluation of CO Catalysts
2.1 Copper-Manganese-Based Catalysts
Copper-manganese-based catalysts are composed of active MnO₂ and CuO; they lower the activation energy of the CO oxidation reaction through electron transfer between Mn⁴⁺/Mn³⁺ and Cu²⁺, as well as the participation of lattice oxygen. The catalytic process proceeds in three steps: CO molecules adsorb onto active sites on the catalyst surface; adsorbed CO reacts with lattice oxygen to form CO₂, which then desorbs; and gas-phase O₂ fills the oxygen vacancies, regenerating catalyst activity.
When the Cu/Mn molar ratio is between 1:1 and 1:2, CO conversion exceeds 90% at room temperature, with an optimal gas hourly space velocity (GHSV) range of 10,000–50,000 h⁻¹. Crystal structure is also critical: samples with low crystallinity (below 15%)—obtained via calcination at 280–350°C—exhibit abundant oxygen vacancies and coordinatively unsaturated sites, resulting in a specific activity approximately 2.3 times higher than that of highly crystalline samples. A specific surface area of 120–220 m²/g is advantageous; samples with a mesopore fraction exceeding 60% demonstrate an apparent activity about 40% higher than those dominated by micropores.
A key drawback of Cu-Mn-based catalysts is their high sensitivity to water vapor. Water molecules compete with CO for active sites, preferentially occupying the active centers at low temperatures. For Hopcalite-type catalysts, activity can drop by more than 50% within a few hours when relative humidity exceeds 40%, and the effective service life falls below 500 hours when humidity exceeds 60%. Water molecules may also induce hydrolysis or hydration of the active components, altering the catalyst structure. The data above derive from controlled experimental conditions; in actual refuge chambers, factors such as temperature fluctuations, co-existing impurities, and uneven gas flow distribution may accelerate deactivation. Therefore, operational suitability data provided by the supplier or on-site validation results should be taken as the definitive reference.
Research based on multi-scale heat and mass transfer models indicates that, to meet the regulatory requirement of reducing CO levels from 400 ppm to below 25 ppm in refuge chambers, the most energy-efficient configuration utilizes 3 mm catalyst particles at a GHSV of 6,327 h⁻¹. In practical engineering applications, catalyst particle sizes typically range from 2 to 5 mm. 2.2 Precious Metal Catalysts
Precious metal catalysts supported on activated alumina or titanium dioxide exhibit superior low-temperature activity. Supported Pd catalysts are effective for low-temperature CO oxidation and have been utilized in CO catalytic systems for mining refuge chambers. Research indicates that the performance behavior of precious metal systems under varying conditions—such as high humidity, heat, and sulfur exposure—differs significantly from that of copper-manganese-based systems; while supported Pd catalysts show relatively slow performance degradation in humid, sulfur-containing environments, the type and loading of active promoters critically influence their moisture and sulfur resistance. Overall, precious metal catalysts offer better moisture resistance than copper-manganese-based systems but come at a significantly higher cost. Selection requires balancing factors such as CO concentration levels within the chamber, required protection duration, and equipment budget constraints.
2.3 Core Decision-Making Logic for Selection
The CO concentration range is the primary consideration. CO within the chamber originates from external intrusion and minor internal generation, typically ranging from tens to hundreds of ppm. Within this range, copper-manganese-based catalysts can achieve effective purification if properly designed; precious metal systems offer advantages primarily in operating conditions characterized by wider CO concentration fluctuations and stricter requirements for response speed.
Humidity is a decisive factor in catalyst selection. If the chamber's humidity cannot be maintained below 40%, copper-manganese-based catalysts require supplementary drying/pretreatment measures; ideally, the relative humidity of the air entering the catalytic layer should be kept below 10% to ensure a long service life. Precious metal catalysts are more adaptable to high-humidity conditions, though this entails higher costs.
Long-term storage stability assessment: Changes in catalyst performance following prolonged periods of equipment inactivity cannot be overlooked. When evaluating supplier products, the following data should be requested: accelerated aging test reports (CO conversion rate retention after aging for 30–90 days at temperatures of 40–60°C and relative humidity of 60%–90%) and post-storage activity test reports (activity data after 12, 24, and 36 months of storage). For copper-manganese-based catalysts, attention must also be paid to secondary particle size growth caused by nanoparticle agglomeration; for unloaded nanoparticles, the secondary particle size can increase to 300–500 nm after 30 days of storage, resulting in a 15%–20% decline in activity. It is recommended that suppliers provide activity data for a storage period of at least 24 months during the selection process and that functional spot checks be conducted every 12 months after the refuge chamber is put into operation.
III. Loading Structure and Process Design
3.1 Particle Size and Bed Design
Catalyst particle size directly affects the pressure drop of gas flowing through the bed and the effective contact area. Smaller particle sizes result in higher specific surface area utilization but increased pressure drop; larger particle sizes lead to more uniform airflow distribution but reduced effective contact area. To meet the regulatory requirement of reducing CO levels from 400 ppm to below 25 ppm in refuge chambers, research based on multi-scale heat and mass transfer models and multi-objective optimization algorithms indicates that the optimal energy-efficiency solution utilizes 3 mm catalyst particles. In practical engineering applications, catalyst particle sizes typically range from 2 to 5 mm.
Method for estimating catalyst loading volume: The loading volume is determined based on the definition of space velocity—Space Velocity (GHSV) = Airflow Rate (Q) / Catalyst Loading Volume (V). Taking a refuge chamber rated for 50 people as an example: if the design airflow rate is 10 m³/min (600 m³/h) and a space velocity of 10,000 h⁻¹ is selected (a reasonable value for copper-manganese-based catalysts under controlled humidity conditions), the required catalyst loading volume is approximately 0.06 m³. If the space velocity is increased to 20,000 h⁻¹, the packing volume can be reduced to 0.03 m³; however, the CO conversion rate may drop, necessitating experimental verification to ensure compliance with the requirement that outlet CO concentration remains ≤24 ppm. Experimental studies on CO purification in refuge chambers indicate that as airflow velocity and the quantity of purification agent increase, the CO purification rate initially rises and then falls, peaking at a space velocity of approximately 183,000 h⁻¹. These data derive from a scaled-down simulation model under specific experimental conditions; in actual engineering design, the optimal space velocity depends on parameters such as catalyst type, particle size, CO concentration, and humidity, and should be determined based on pilot-scale test data or supplier recommendations. The design thickness of the granular bed must be calculated based on the space velocity and CO concentration to ensure the gas residence time within the bed meets the requirements of the catalytic reaction.
3.2 Configuration of the Drying Pre-treatment Layer
Given the high sensitivity of copper-manganese-based catalysts to water vapor, the fundamental principle of the packing structure design is "drying followed by catalysis." Engineering practice typically employs silica gel as the primary (coarse) desiccant and 5A molecular sieve as the secondary (deep) desiccant; using them in series achieves superior dehydration capacity. A desiccant layer is installed at the catalyst inlet; moisture-laden air passes through the desiccants first, where water vapor is absorbed and removed, and the dried air subsequently flows through the CO catalyst layer for catalytic oxidation. A typical structural arrangement—ordered according to the direction of airflow—consists of a metal top mesh, an activated carbon layer, a silica gel layer, a molecular sieve layer, and finally, a Hopcalite catalyst layer.
Desiccant capacity estimation: Consider an operating scenario with a rated capacity of 50 people, a 96-hour protection duration, an inlet air relative humidity of 85%, and a temperature of 30°C. Based on an airflow requirement of 450 L/min per person, the total airflow is 22.5 m³/min. At 30°C, the moisture content of saturated air is approximately 30 g/m³; at 85% relative humidity (RH), this corresponds to approximately 25.5 g/m³. If the relative humidity of the outlet air is required to drop below 40% (with a moisture content of approximately 12 g/m³), about 13.5 g of water vapor must be removed per cubic meter of air. The total amount of water to be removed over 96 hours is approximately 22.5 × 60 × 96 × 13.5 / 1000 ≈ 1,750 kg. The static adsorption capacity of silica gel is approximately 20%–30% of its own weight, while that of 5A molecular sieve is approximately 15%–22%. If silica gel handles 60% of the moisture removal (approx. 1,050 kg), a mass of about 3,500–5,250 kg is required; if molecular sieve handles 40% (approx. 700 kg), a mass of about 3,200–4,700 kg is required. This estimate serves only as a reference for the order of magnitude; actual engineering designs must incorporate specific adsorption isotherms, fluctuations in inlet air humidity, airflow distribution efficiency, and adsorption breakthrough curves, while also providing a capacity margin of at least 20%.
3.3 Cartridge Structure and Modular Loading
Air purification units for refuge chambers typically employ a cartridge-based design. The cartridge features a hollow rectangular parallelepiped structure with metal mesh faces on the two opposing wider sides, and the interior is filled with catalyst granules. Multiple cartridges are arranged vertically within a support frame, allowing the airflow—after entering through the damper—to pass sequentially through the layers of cartridges. The advantages of this modular design include: the flexibility to adjust the number of cartridges based on the chamber's rated occupancy and required airflow rate; ease of operation, as only the specific module needs to be replaced when changing the catalyst; and independent sealing between modules, which facilitates transport and storage.
3.4 Airflow Distribution and Pressure Drop Control
The uniformity of airflow through the catalytic bed directly affects the effective utilization rate of the catalyst. During loading, care must be taken to avoid particle stratification or areas of excessive density or looseness, ensuring that the gas is evenly distributed across the cross-section of the bed. Pressure drop control strategies differ based on the type of catalyst bed: Honeycomb ceramic catalyst modules typically exhibit an initial airflow resistance of 200–500 Pa; the pressure drop arises primarily from channel friction, and the airflow distribution is relatively uniform, making them suitable for high-flow, low-resistance applications. For granular beds, the pressure drop can be estimated using the Ergun equation, which accounts for particle size, bed thickness, superficial gas velocity, and gas properties. Taking the example of 3 mm particles, a bed thickness of 200 mm, a superficial gas velocity of 0.5 m/s, a gas density of 1.1 kg/m³, and a viscosity of 1.9 × 10⁻⁵ Pa·s, the estimated pressure drop falls within the range of 800–1,500 Pa. Since the pressure drop across a granular bed is typically significantly higher than that of a honeycomb ceramic module, the actual pressure drop of the catalytic bed (rather than the reference value for honeycomb ceramics) must be factored into the total system resistance calculation to ensure sufficient pressure head margin when selecting the fan. It is recommended to provide a pressure head margin of at least 1.3 times the design pressure drop to accommodate increases caused by factors such as catalyst aging, dust accumulation, and the swelling of desiccants due to moisture absorption.
4. Operation, Maintenance, and Performance Assurance
4.1 Humidity Control and Desiccant Management
Desiccants have a finite adsorption capacity and lose their dehydration capability once saturated. A system for the scheduled replacement or regeneration of desiccants must be established during operation. In scenarios where chamber humidity cannot be maintained below 40%, an "intermittent use" strategy is recommended: once the CO concentration drops to a safe level, the CO catalyst bed should be removed from the airflow path to prevent catalyst poisoning caused by continuous exposure to high-humidity airflow; the catalyst bed should be returned to the purification unit only when CO treatment is required.
4.2 Catalyst Activity Monitoring and Replacement
The CO catalytic purification system should be interlocked with the CO sensors inside the chamber to enable automatic start-stop functionality—activating catalytic purification when the CO concentration exceeds the threshold and stopping it once the concentration falls to a safe range. A regular inspection protocol must be established to check the catalyst for signs of moisture, agglomeration, or pulverization, and to monitor for significant declines in CO conversion efficiency under consistent operating conditions. The criteria for catalyst replacement are as follows: the catalyst should be replaced if the CO concentration at the catalytic bed outlet consistently exceeds 24 ppm under normal operating conditions, or if the equipment pressure drop increases by more than 50% compared to its initial state.
Regarding the limits of regeneration applicability: Deactivation caused by competitive water vapor adsorption is partially reversible; drying the catalyst at 120–150°C for several hours can remove physically adsorbed water molecules and restore some activity. However, sulfates and sulfites formed through sulfide poisoning are difficult to decompose at low temperatures—sulfates only begin to desorb in the 350–400°C range—making thermal regeneration unfeasible under chamber conditions. Deactivation resulting from pore blockage by dust or crystal phase destruction caused by the hydrolysis of active components is irreversible. Therefore, in engineering practice, one should not assume that copper-manganese-based catalysts can be regenerated; low-temperature drying regeneration should only be considered if the cause of deactivation is confirmed to be solely water vapor adsorption. Catalysts that have suffered sulfur poisoning or physical blockage should be replaced directly.
4.3 Pre-start checks after prolonged standby
Before activating emergency equipment for the first time after a long period of standby, a functional check of the catalytic module must be performed: verify that the desiccant is effective, the catalyst granules show no signs of caking or pulverization, and the fan operates normally. Periodic short-term functional tests are recommended; running the catalytic system briefly verifies purification performance while preventing potential performance degradation caused by keeping the catalyst in a completely sealed state for extended periods.
4.4 Coordination with other chamber subsystems
CO catalytic purification is not an isolated process. Compressed air supplied by the oxygen supply system may contain moisture and requires drying before entering the catalytic module. The arrangement order of the CO₂ absorbent and the CO catalyst within the cartridge significantly affects the catalyst's exposure to humidity; specific analysis based on the absorbent type is required. The reaction of soda lime (primarily composed of Ca(OH)₂ and NaOH) with CO₂ is exothermic and produces water (Ca(OH)₂ + CO₂ → CaCO₃ + H₂O); if the CO catalyst is placed downstream of the soda lime, water vapor released during absorption enters the catalytic layer directly, thereby increasing the risk of moisture poisoning for the catalyst. While lithium-based absorbents (such as LiOH) also produce water during absorption, the quantity of water produced varies depending on the absorbent formulation and reaction pathway. Therefore, the CO catalyst should not be placed downstream of the CO₂ absorbent by default; instead, the arrangement should be determined based on the specific water-production characteristics of the absorbent. It is recommended to position the CO catalyst upstream of the CO₂ absorbent and downstream of the desiccant—following the airflow sequence: desiccant → CO catalyst → CO₂ absorbent—or to use separate airflow paths for CO catalysis and CO₂ absorption to minimize the impact of water vapor on the catalyst. Understanding the fundamental differences between refuge chambers and conventional industrial settings—while fully accounting for operating humidity during equipment selection, adhering to the "dry first, then catalyze" principle in the design of the filling medium, and establishing a system for regular inspection and replacement during operation—forms the foundation for ensuring the reliable performance of the CO purification system at critical moments.
