Measuring two gases still means managing four hazards together
Biogas and landfill gas are not simply methane fuels. Raw gas from an anaerobic digester or landfill layer contains large proportions of CH₄ and CO₂, while components such as H₂S, water vapor, nitrogen, oxygen, siloxanes and volatile organic compounds vary with the facility and feedstock. The U.S. EPA’s basic information on landfill gasdescribes landfill gas as approximately 50% CH₄, 50% CO₂ and small amounts of non-methane organic compounds. The composition of farm digester gas may vary more widely depending on feedstock and operating conditions. These representative values must not be used as a guaranteed composition or alarm setpoint for a specific facility.
The key is not to treat H₂S and CO₂ merely as separate toxic substances. CH₄ indicates fire and explosion hazards and oxygen displacement; H₂S indicates acute toxicity and flammability; CO₂ indicates its own physiological effects and asphyxiation; and O₂ reveals air ingress and whether the atmosphere is breathable. Viewing all four values on the same timeline helps distinguish leakage, dilution and process changes. For example, if CH₄ and CO₂ fall together while O₂ rises, air ingress or sample dilution may be suspected. Conversely, if O₂ falls and CO₂ rises in an enclosed area, breathable air itself may be disappearing. These patterns are diagnostic hypotheses, however, not formulas that yield a definitive conclusion from a single measurement.
The hazards of CH₄, H₂S, CO₂ and O₂ overlap
CH₄ is a colorless, odorless, flammable gas, and NIOSH data give an approximate explosive range in air of 5–15%. In raw-gas piping, CH₄ may be above the upper limit, but it necessarily passes through the flammable range at the boundary where a leak mixes with air. Thus, saying that the gas inside the pipe is too rich to burn does not establish that the leaking cloud is safe. Ignition-source control, ventilation, suitability for hazardous locations and %LEL monitoring must be designed as one package.
H₂S is known for its rotten-egg smell, but odor cannot serve as an alarm. OSHA warns that the sense of smell may become dulled over time at low concentrations or rapidly at high concentrations. NIOSH H₂S IDLH datagive a recommended exposure limit of 10 ppm as a 10-minute ceiling and an IDLH value of 100 ppm. These are U.S. industrial hygiene criteria and do not replace statutory requirements or a site risk assessment in Korea. H₂S is somewhat heavier than air and may collect in low areas such as manholes, pump pits and condensate drains, but the buoyancy of warm gas and flows from blowers, doors and ducts alter its actual distribution.
CO₂ is also more than a simple oxygen-displacing gas. The NIOSH CO₂ Pocket Guidegives an 8-hour TWA of 5,000 ppm, a short-term value of 30,000 ppm and an IDLH value of 40,000 ppm. At high concentrations, difficulty breathing, dizziness, increased heart rate and loss of consciousness can occur even while some oxygen remains. An O₂ sensor therefore must not be used as a substitute for assessing CO₂ risk. CO₂ is heavier than air, but it is likewise affected by jets, temperature, ventilation and structures, so measurement at a single point on the floor is not sufficient.
O₂ is needed both for worker protection and process diagnosis. OSHA’s confined-space standard treats an atmosphere below 19.5% as oxygen-deficient and hazardous. Meanwhile, higher-than-expected O₂ in a raw-gas system may indicate air ingress through a cover, flange, condensate trap or pipe under negative pressure. This can affect not only formation of a flammable mixture but also biological desulfurization reactions, heat generation and gas quality. Facilities that intentionally introduce air, such as oxygen-dosed biological desulfurization processes, must follow their own operating limits and interlock logic.
Process analysis and worker protection differ in purpose from the outset
Process analyzers determine whether gas meets product specifications and equipment operating limits. Tracking CH₄, CO₂, O₂ and H₂S in the raw-gas header, H₂S before and after the desulfurizer, moisture or dew point before and after the dryer, and CH₄, CO₂ and O₂ after upgrading helps assess digestion conditions, air intrusion, adsorbent breakthrough, scrubbing performance and membrane-separation performance. Recording these values together with flow, pressure and temperature makes it easier to determine whether a concentration change reflects an actual material-balance change or a change in sample conditions. Siloxanes are important in managing contamination of engines, turbines and fuel cells, but an ordinary portable multi-gas detector does not measure them. Application-specific periodic analysis or a dedicated online analyzer is required.
Worker-protection detectors warn of immediate hazards in the air people occupy. Fixed H₂S, CO₂, %LEL and O₂ detectors should be combined with personal or portable multi-gas detectors, but an analysis value from process piping must not be treated as a workspace concentration. Conversely, a workspace reading of 0 ppm H₂S does not mean that the desulfurizer outlet meets specification. Range, response time, accuracy, cross-sensitivity, alarm delay and fault signals differ by purpose.
Before entry into a confined space, measurements must be taken from outside with a calibrated direct-reading instrument. The OSHA confined-space standardrequires testing in the order of O₂, flammable gases and vapors, and toxic air contaminants. This is because many catalytic-combustion flammable-gas sensors may not be accurate when oxygen is deficient, and an explosion hazard may be more immediately dangerous than toxicity. During entry, monitoring must continue at the work location. When a long probe is used, allow for the sample transit time calculated from the pump flow and hose volume, plus the sensor response time, and do not take a reading until the sample from each height has actually reached the sensor.
How condensate and siloxanes alter readings
When warm, saturated biogas passes through cold piping, water condenses. The EPA AgSTAR operator guiderecommends sloping piping and installing self-draining condensate traps. Accumulated condensate can obstruct the flow path and raise digester pressure; in a sampling hose, it can form a liquid plug that delays response or damages the analyzer. Components that interact with water, such as H₂S, may appear lower or more slowly than they actually are because of wet filters, condensate and long tubing surfaces. Values before and after drying therefore do not represent the same measurement point.
Siloxanes can enter from consumer products and become mixed into digester gas and landfill gas. U.S. DOE Better Buildings materialexplains that siloxanes can form silicate and silica deposits during combustion, causing problems in energy-use equipment such as engines. Moisture removal, sulfur removal and siloxane removal are distinct functions, so a single filter must not be assumed to treat every component. Because siloxanes, oil mist and sulfur compounds can affect analyzer and sensor filters, optical components and catalysts, maintenance intervals approved by the manufacturer and performance checks are necessary.
To obtain a representative sample, first define the sampling point and the pretreatment boundary. Measure before desulfurization and dehumidification to assess raw-gas quality; before and after each treatment stage to assess equipment-protection performance; and before compression or injection to assess final fuel quality. Keep sample lines as short as practical, confirm material compatibility and, when necessary, use temperature-controlled sample lines, particle filtration and condensate separation. If pretreatment is used, however, the result must record whether the basis is wet or dry, along with temperature, pressure, flow, filter type, removed condensate and analysis delay. Zero and span calibration and functional tests should verify leaks and delays across the entire actual sample path—including the probe, hose and filter—not merely the sensor body.
Place detectors according to leak scenarios, not a specific-gravity table
Initial candidate locations include the digester gas space, landfill-gas header, blower and compressor rooms, desulfurization, drying and upgrading skids, engine rooms, flare systems, and areas around condensate traps and drainage pits. Prioritize low pits and drainage points for H₂S and CO₂, and ceilings and upper stagnation zones for CH₄. Sensors are also needed at anticipated leak points on valves, flanges and seals, along supply and exhaust airflow paths, and around workers’ breathing zones and access and evacuation routes. Do not install them at only one height; use actual smoke tests, ventilation models and field measurements to identify blind spots at different elevations.
A fixed alarm should be an input that initiates action, not merely a device that produces sound and light. Distinguish warning and high alarms, sensor faults and ventilation failure, and link them—based on the site risk assessment—to gas isolation, ignition-source isolation or suspension of non-explosion-protected work, equipment shutdown, remote verification, evacuation and notification. For automatic ventilation, review the discharge location so gas is not moved toward ignition sources or occupied areas, and use equipment suitable for the hazardous location. A one-size-fits-all response of simply turning on a blower can be dangerous.
Minimum actions when an alarm sounds
Stop work and leave the alarm zone, taking account of the wind and designated evacuation route. Do not return merely because the odor has disappeared.
From a safe location, notify the control room and emergency-response organization and control access. Isolate the gas supply and ignition sources remotely or through the designed emergency-shutdown procedure only when it is possible to do so safely.
Trained personnel should check concentrations from outside using suitable equipment. Do not enter a confined space without a permit procedure, attendant, communications, rescue plan and continuous measurement.
Even if someone has collapsed, do not enter after them without appropriate protective equipment. OSHA requires rescuers in areas where H₂S may be present to have training and protection such as positive-pressure SCBA and a lifeline. An impulsive coworker rescue can lead to multiple fatalities.
Even after ventilation and isolation, recheck O₂, %LEL, H₂S and CO₂, and do not return to normal operation until the cause has been identified.
Validate alarm values through commissioning and management of change
Setting an alarm does not mean copying a single number from an online table. Worker-protection alarms should account for applicable regulations, toxicity criteria, evacuation time and instrument error, while process alarms should be based on the normal range, equipment-manufacturer limits, purification specifications and the outcome of linked actions. Even with the same H₂S sensor, a process alarm indicating desulfurizer breakthrough and a workplace alarm requiring immediate evacuation differ in purpose and range. Whether a high alarm should automatically start a ventilation fan, close a process valve or stop a blower must not be decided without a risk assessment and safety-integrity review.
During commissioning, establish baselines for conditions such as normal operation, startup, shutdown, feedstock changes, condensate draining and desulfurization-media replacement. The EPA’s farm anaerobic-digestion safety material recommends recording operating variables and air quality weekly during the initial 6–12 months to establish a baseline for normal conditions. This cannot be treated as the statutory interval for every facility, but it demonstrates why trend management matters more than one-off measurements at facilities with substantial seasonal and load variations.
Verification of the detection system should also be divided into layers. A functional test before each use determines whether the sensor responds to gas and the alarm actually sounds. Periodic calibration adjusts the indicated value using a standard gas of known concentration. A full-loop test confirms that the sensor signal reaches the control panel, warning lights, ventilation, isolation valves and remote notification. Testing detection performance by releasing test gas at an actual leak location requires a separate safety plan. Actual biogas must not be released arbitrarily for testing.
When equipment is changed, review the existing detector placement again. Changes to duct direction, walls, ceilings, blower capacity, process pressure, feedstock, desulfurization method or door-opening conditions also change which sensors the gas can reach. Alarm histories should record not only the peak value but also the sensor ID, time, process status, ventilation status, related CH₄, CO₂ and O₂ values, actions taken and the basis for return to service. Do not disable recurring low alarms as mere false alarms; distinguish among condensation, cross-sensitivity, air ingress, small leaks and sensor degradation and resolve the underlying cause.
Good simultaneous management does not end with adding more sensors. Process quality and human exposure must be measured separately, changes to wet raw gas along the sample path must be recorded, and each alarm must specify who stops what and where people evacuate. Only by interpreting the relationship among CH₄, H₂S, CO₂ and O₂ can a facility simultaneously manage the explosion, toxicity, asphyxiation and process abnormalities that a single number may conceal.