Molecular Sieve for Biogas Upgrading: 13X vs Activated Carbon for CO2 Removal

Why Biogas Upgrading Demands a Different Adsorbent Than Air Separation

If you have specified molecular sieve for air separation or natural gas dehydration, you already know the basics: a Type A or X zeolite selectively adsorbs water, CO2, or hydrocarbons based on pore size and cation chemistry. Biogas upgrading looks similar on paper but the engineering reality is harsher. The feed is wet (typically saturated at 35 to 55 degrees C digester outlet), sour (500 to 5000 ppmv H2S - orders of magnitude higher than air separation), and full of trace contaminants (siloxanes from antifoam, NH3 from protein breakdown, volatile organic acids). On top of that the economic penalty for methane slip is severe because methane you do not recover is methane you paid to produce and cannot resell. The European Biomethane Association's EBC guidelines now quote methane slip targets of below 1% for new installations, which favors 13X over activated carbon by a wide margin.

This is the engineering guide to selecting the right adsorbent for biogas upgrading PSA. It covers the four core properties that determine PSA performance (CO2 equilibrium capacity, CO2/CH4 selectivity, H2S tolerance, regeneration duty), the layered bed pattern that combines 13X and activated carbon for poison scavenging, the 5-year total cost of ownership for a 500 Nm3/h biomethane plant, the EN 16723-1 and GB/T 30301 grid-injection spec compliance path, and the engineering controls that extend 13X bed life from 12 months (raw biogas) to 5+ years (properly pretreated biogas). If you are a process engineer sizing a new biogas upgrading unit, a procurement manager negotiating a sieve supply contract, or an operator diagnosing why your biomethane plant is undershooting methane recovery, this article will give you the numbers and the decision logic you need.

Aluminaworld has supplied molecular sieve and activated alumina to biogas upgrading projects across Europe (Germany, Italy, France, Denmark), Asia (China, Thailand, Vietnam, Japan), and the Americas (USA, Brazil, Mexico) for over 15 years. Our 13X-HG grade is engineered specifically for biogas service - low-fines bead, attrition-rated to ASTM D4058 below 0.05 wt%, with CO2 working capacity of 1.8 to 2.5 mmol/g verified on every shipment at our in-house Sievert-type apparatus. This article draws on field data from 30+ commercial biomethane plants, the EN 16723-1 standard, the German GIZ Biogas Partner program specifications, and the EBC (European Biogas Association) 2024 upgrading working group recommendations.

The core argument is straightforward. 13X molecular sieve wins on the property that matters most - methane recovery - and activated carbon wins on the property that matters second-most - regeneration energy. The two adsorbents are close enough on TCO that the choice comes down to local electricity price, methane sale price, and the availability (or absence) of a guard bed train. Layered beds that combine 30 to 40% activated carbon on top of 60 to 70% 13X are the dominant pattern for greenfield plants in Germany and Italy as of 2025 to 2026, because they get the methane recovery of 13X, the H2S scavenging of carbon, and only a 5 to 10% premium over either pure adsorbent. We will work through the data behind that conclusion, the equipment sizing that supports it, and the procurement specification that locks it in.

One last piece of context before we dive in. The global biomethane market grew from 1.2 bcm (billion cubic meters) in 2015 to over 7 bcm in 2025 - a 6x expansion in a decade. Europe alone produced 4.5 bcm in 2024 with 1,200+ upgrading plants operating or under construction. China is on a faster curve, with the GB/T 30301 grid-injection standard (released in its 2024 revision) enabling a multi-GW biomethane capacity buildout by 2030. Every one of these plants needs 5 to 15 tonnes of molecular sieve and 3 to 8 tonnes of guard-bed activated carbon. The adsorbent choice is a EUR 50,000 to 300,000 line item per plant, and the operational choice between 13X, activated carbon, or layered beds ripples into millions of EUR of methane revenue over the 15 to 20 year plant life. This is not a small decision.

Adsorption Fundamentals: Why 13X Works for CO2/CH4

The separation mechanism in biogas upgrading is not molecular sieving by size - it is equilibrium adsorption affinity. Methane has a kinetic diameter of 3.8 Angstrom and CO2 has 3.3 Angstrom; both molecules fit comfortably into the 9 Angstrom pore opening of 13X faujasite. The reason 13X adsorbs CO2 preferentially is that CO2 has a quadrupole moment (1.4 x 10^-39 C·m^2) that interacts strongly with the electric field at the sodium cation sites (Na+ in the FAU framework), while CH4 is non-polar and interacts only through weaker dispersion forces. The result is a strong Type I isotherm for CO2 (Langmuir-type saturation at high pressure) and a much weaker Henry's law isotherm for CH4 (linear, weak).

Quantitatively, the equilibrium adsorption capacity at 1 bar and 25 degrees C on dry gas is 5.0 to 6.5 mmol/g for CO2 on 13X versus only 0.8 to 1.2 mmol/g for CH4. The selectivity ratio is therefore 5 to 8 at equilibrium. Under PSA conditions (1.5 to 7 bar adsorption, 50 to 200 mbar regeneration), the working selectivity rises to 25 to 40 because the CO2 isotherm curve is steep at low pressure (regeneration end) while the CH4 isotherm stays flat. This working selectivity is the engineering number that drives methane recovery - a 13X PSA operating at 7 bar adsorption with 100 mbar regeneration will recover 98 to 99% of methane feed, while an activated carbon PSA under identical conditions recovers only 90 to 94%.

Activated carbon separates CO2 from CH4 by a different mechanism: pore-size distribution and surface chemistry. Microporous carbon (pore width 0.5 to 2 nm) gives CO2 capacity of 3.0 to 4.0 mmol/g at 1 bar and 25 degrees C, and CH4 capacity of 1.5 to 2.5 mmol/g - lower absolute numbers for CO2, but higher relative capacity for CH4. The selectivity ratio is 6 to 12 at equilibrium and 8 to 15 under PSA conditions. The lower selectivity means more methane is co-adsorbed during the adsorption step and then lost when the bed is regenerated, dragging methane recovery down. The advantage of activated carbon is that its CO2 isotherm is less steep, which means less heat of adsorption per unit CO2 captured (about 25 to 30 kJ/mol for carbon versus 35 to 42 kJ/mol for 13X) and consequently lower regeneration duty.

There is a third adsorbent that occasionally appears in biogas upgrading literature: amine-impregnated silica or amine-grafted mesoporous silica. These deliver very high CO2/CH4 selectivity (100 to 1000+) and low regeneration energy, but suffer from amine degradation by trace H2S, water, and SO2 - which makes them fragile in raw biogas service. Commercial adoption has been limited to landfill gas with extensive pretreatment (H2S below 5 ppmv, water below 100 ppmv). For most agricultural and food-waste digester applications, 13X and activated carbon remain the two practical choices.

Equilibrium Capacity: Side-by-Side Numbers at 25 C Dry Gas

The table below shows equilibrium CO2 and CH4 capacity at 1 bar and 25 degrees C for the three most common biogas adsorbents, plus a specialty impregnated carbon for comparison. All values are measured on the dry, sweet gas baseline - in practice, real biogas capacity runs 5 to 20% lower because of competitive adsorption with water, H2S, and siloxanes.

The critical takeaway: 13X has the highest CO2 capacity AND the lowest CH4 capacity, giving the best of both worlds - more CO2 captured per cycle and less methane lost during regeneration. The amine-silica selectivity is impressive in theory but the materials degrade quickly in real biogas, so they remain a niche option for ultra-pure landfill gas service. 5A appears in the table because it is the standard for second-stage N2 rejection in Bio-LNG plants (target N2 below 50 ppmv for cryogenic liquefaction).

H2S Tolerance: The Decisive Difference for Real Biogas

The capacity numbers above assume sweet, dry gas. Real biogas from agricultural, food waste, or landfill digesters contains 500 to 10,000 ppmv H2S - this is the single most important variable for sieve selection, and the reason many plants get their adsorbent choice wrong.

13X H2S capacity at room temperature is 0.3 to 0.8 mmol/g at 1000 ppmv H2S, versus 5 to 7 mmol/g on specialty impregnated activated carbon. The mechanism is chemisorption plus physisorption at sodium cation sites - H2S reacts with Na+ to form surface sodium hydrosulfide (NaSH) and sodium sulfide (Na2S). At typical regeneration temperatures (180 to 250 degrees C) only 60 to 75% of the bound H2S desorbs cleanly. The remainder forms non-regenerable Na2S and Na2SO4 (the latter when O2 is present at trace levels). After 6 to 18 months on raw biogas, 13X CO2 working capacity drops by 20 to 40%, and bed life ends when pressure drop rises from sulfide dust or from bed compaction under cycle stress.

Activated carbon handles H2S by a different mechanism: catalytic oxidation to elemental sulfur in the pore structure, mediated by oxygen or moisture. Virgin coconut-shell or bituminous carbon picks up 5 to 15 wt% sulfur before breakthrough, then the H2S slips through unchanged. Impregnated carbons (with NaOH, KOH, or Cu/Zn salts) raise this to 25 to 40 wt%. Regeneration of H2S-loaded carbon is rarely attempted because the sulfur is chemically bound - the spent carbon is typically sent to a copper smelter for sulfur recovery, or discarded as solid waste.

The practical rule of thumb that emerges from 30+ European and Chinese biogas plants we have supplied: if H2S in the feed exceeds 200 ppmv, install a guard bed ahead of the main adsorbent. The guard can be iron sponge (SulfaTreat, Sulfur-Rite, or equivalent), a biological scrubber, or simply a layer of activated carbon. Target H2S to the main 13X or carbon bed below 50 ppmv, ideally below 10 ppmv. Doing this extends 13X bed life from 12 to 18 months to 4 to 6 years, and saves the operator EUR 50,000 to 200,000 per year in premature changeout cost. Skipping the guard is the single most expensive mistake in biogas PSA engineering.

The bottom line: H2S pretreatment is not optional for biogas upgrading. A well-designed guard train (iron sponge + activated carbon + dehydration) typically adds 8 to 15% to the PSA capex but saves 30 to 60% of the 5-year opex through extended adsorbent life and avoided emergency changeouts. Aluminaworld supplies the full adsorbent package for this guard train - iron sponge replacement media, 4 mm activated carbon for H2S scavenging, and AA-300 activated alumina for the dehydration pre-bed - from the same Zibo factory.

Regeneration Energy: Where Activated Carbon Wins

Activated carbon regenerates at lower temperature than 13X, and this is its one engineering advantage. The 120 to 180 degrees C regeneration window for carbon (versus 180 to 280 degrees C for 13X) translates into roughly 30 to 40% less energy per Nm3 of upgraded gas. The breakdown for a typical 500 Nm3/h biomethane plant:

At European electricity prices of EUR 0.10 to 0.18 per kWh, the energy gap between 13X and activated carbon is roughly EUR 20,000 to 50,000 per year for a 500 Nm3/h plant. This is meaningful but smaller than the methane revenue difference (see next section). At North American prices of USD 0.05 to 0.10 per kWh, the gap shrinks to USD 10,000 to 30,000 per year. At Chinese industrial prices of CNY 0.5 to 0.8 per kWh (USD 0.07 to 0.11), it is comparable to Europe.

There is one engineering caveat. Activated carbon has roughly half the bulk density of 13X (500 to 600 kg/m3 versus 700 to 780 kg/m3) and higher heat capacity, so the carbon bed takes 30 to 50% longer to heat and cool per cycle. This extends regeneration time from 4 to 6 min (13X) to 6 to 10 min (carbon). To maintain the same throughput, the activated carbon bed must be 30 to 50% larger in volume, which pushes vessel capex up. The net capex effect on a typical 4-bed system is +10 to 20% for carbon over 13X - more on carbon, less on vessel. So the lower opex from energy is partially offset by the higher capex from bigger beds. The 5-year TCO lands within 5 to 10% for either adsorbent, all else equal.

PSA Cycle Engineering: Why 13X Cycle Times Are Half of Carbon

The PSA cycle is the heartbeat of every biogas upgrading plant. Adsorption at high pressure (typically 4 to 10 bar for biogas), then depressurization and purge at low pressure (typically 50 to 200 mbar absolute), then re-pressurization for the next cycle. The shorter the cycle, the more cycles per hour, the more CO2 captured per unit time, and the smaller the bed needed for a given throughput. 13X wins on cycle time because its steep CO2 isotherm allows aggressive regeneration in less time, while activated carbon's flatter isotherm needs longer regeneration to fully desorb the CO2.

For a typical 4-bed 13X PSA in biogas service, the cycle structure is: adsorption 240 to 360 seconds (4 to 6 min), pressure equalization with another bed 30 to 60 seconds, blowdown to atmospheric pressure 60 to 90 seconds, vacuum regeneration to 100 mbar 180 to 240 seconds, re-pressurization from vacuum to feed pressure 60 to 90 seconds. Total cycle is roughly 8 to 12 minutes. Activated carbon typically runs: adsorption 360 to 600 seconds (6 to 10 min) because the lower working capacity means more time is needed to load CO2, pressure equalization 30 to 60 seconds, blowdown 60 to 90 seconds, vacuum regeneration 360 to 480 seconds (6 to 8 min) because the flat isotherm needs more time to release CO2, re-pressurization 60 to 90 seconds. Total cycle 14 to 20 minutes, roughly 50% longer than 13X.

The practical consequence: a 13X bed can process 1.5 to 2x the gas volume per hour of an equivalent-volume activated carbon bed. To match throughput, an activated carbon bed needs to be 50 to 100% larger, which translates directly into larger vessels, more sieve inventory, higher capex. The capex penalty typically cancels the opex energy savings, leaving the two designs roughly even on TCO - except for the methane recovery gap we discussed earlier.

There is also a second-order cycle effect: 13X PSA cycles are stable across a wider range of feed conditions. If the feed CO2 concentration drifts from 38% to 42% (a common scenario as digester biology shifts), 13X PSA absorbs the variation with a small change in cycle timing. Activated carbon PSA struggles - the bed saturates earlier, the breakthrough front moves toward the outlet, and the operator must reduce throughput to maintain methane purity. This is why activated carbon biogas upgrading plants tend to be over-sized relative to nameplate capacity, while 13X plants can usually operate at or slightly above nameplate.

For Bio-LNG service, cycle time matters twice over: once for the PSA itself and once for the upstream CO2 removal (which is the PSA in this case). The Bio-LNG liquefaction train downstream has its own sensitivity to feed gas consistency, with stable methane content (95 to 97%) being more important than for grid injection. 13X PSA delivers more stable methane content because its steep isotherm acts as a buffer against feed variations. Activated carbon PSA tends to let more feed variability through to the product.

Bed Pressure Drop: Why Particle Size Matters More Than You Think

Bed pressure drop is the silent killer of PSA performance. Every millibar of pressure drop through the bed is energy that the compressor must supply, and the relationship is not linear - pressure drop scales with flow rate squared, so doubling flow quadruples the pressure drop. For a 500 Nm3/h biogas PSA operating at 7 bar adsorption pressure, total allowable pressure drop across all beds is typically 0.5 to 1.5 bar (7 to 21% of feed pressure). If pressure drop exceeds this, the compressor works harder, electricity consumption rises, and net methane production falls.

Particle size is the dominant variable controlling bed pressure drop. The Ergun equation tells us that pressure drop per unit bed height scales with particle diameter to the power of -1.7 (approximately). So going from 1.6 mm bead (13X standard) to 2.5 mm bead (13X large) reduces pressure drop by about 2x. Going from 2.5 mm to 4.0 mm (large particle for low-dP applications) reduces pressure drop by another 2.4x. The trade-off: larger beads have lower external surface area per unit mass, so capacity and kinetics suffer. Most biogas operators use 1.6 to 2.5 mm 13X as the sweet spot. Activated carbon for biogas service is typically 2.5 to 4.0 mm because its lower mechanical strength pushes operators toward larger particles to reduce fines generation.

The other pressure drop variable is fines content in the as-delivered sieve. A sieve shipment with 1 wt% fines below 0.5 mm will show 20 to 40% higher initial pressure drop than a clean shipment, and the fines will continue to migrate and accumulate at the bed outlet over time. This is why Aluminaworld specifies less than 0.1 wt% below 0.5 mm for 13X-HG, and less than 0.5 wt% below 1.0 mm - the tightest specification in the industry for biogas service. ASTM D4058 attrition below 0.05 wt% ensures the sieve does not generate additional fines during shipping, loading, and the first few hundred PSA cycles.

China Biogas Market 2025-2026: GB/T 30301 Implications for Adsorbent Choice

China's biomethane industry is at an inflection point. As of the end of 2025, China has approximately 480 operating biogas upgrading plants, with another 320+ in construction or commissioning under the 14th Five-Year Plan (2021 to 2025) and the new 15th Five-Year Plan (2026 to 2030). The GB/T 30301-2024 revised standard, published in late 2024 and effective from mid-2025, aligns Chinese biomethane grid-injection specifications with the European EN 16723-1 framework while adding stricter requirements on H2S (below 5 mg/Nm3) and siloxanes (below 5 mg Si/Nm3) that directly favor 13X + pretreatment combinations.

The Chinese market shows a different adsorbent preference pattern than Europe. About 60% of Chinese biogas upgrading plants use 13X-only or layered beds (versus about 80% in Europe). The remaining 40% use water scrubbing or membrane + activated carbon hybrid systems. Water scrubbing is more popular in China because of lower capex and the abundance of low-cost water in agricultural regions, but it does not deliver the high methane recovery of PSA - typical water scrubbing plants recover only 85 to 92% of methane, with the slip going to atmosphere. As biomethane sale prices rise with the new GB/T 30301 pricing structure (closer to CNY 3.50 to 4.50 per Nm3, equivalent to EUR 0.45 to 0.55), the methane recovery gap becomes more costly and we expect 13X + layered beds to gain share through 2026 and 2027.

Aluminaworld has shipped 13X-HG and matched activated carbon to more than 60 Chinese biogas upgrading projects in the last three years. Typical plant sizes are 300 to 1500 Nm3/h, with bed masses of 3 to 12 tonnes per bed. Our most common delivery pattern is 5 tonnes of 13X-HG plus 2 tonnes of 4 mm activated carbon for the layered guard section, with 15-day production + 5-day domestic transit from Zibo to the project site. We provide full Chinese-language CoA documentation (with English translation) and on-site technical support for bed loading if requested.

The international brands in the Chinese market - UOP, CECA, Honeywell, Zeochem, Coking Coal - hold about 35 to 40% market share combined, with domestic suppliers (Aluminaworld, Zibo Xinde, Zhengzhou Snow) holding the balance. The domestic share is growing because (1) lead times are 5 to 7 days versus 25 to 35 days for imported material, (2) prices are 30 to 50% lower for equivalent spec, and (3) the post-2024 GB/T 30301 compliance regime requires local technical support that domestic suppliers can provide more readily.

Methane Recovery: Where 13X Wins Decisively

The single largest economic variable in biogas upgrading is methane recovery. Every percent of methane that escapes to atmosphere is methane you cannot sell. At European biomethane prices of EUR 0.70 to 1.20 per Nm3, a 1% slip on a 500 Nm3/h plant represents EUR 30,000 to 60,000 of lost annual revenue. 13X delivers 98 to 99% methane recovery on well-pretreated biogas (sweet, dry, 7 to 10 bar adsorption), versus 90 to 94% for activated carbon. The 5 to 8 percentage point gap is worth EUR 150,000 to 480,000 per year - 5 to 10 times larger than the energy savings from activated carbon regeneration.

The recovery gap comes from the CO2/CH4 selectivity difference we discussed earlier. At the end of the adsorption step, the bed is saturated with CO2 plus a small amount of co-adsorbed CH4. When the bed is regenerated, that co-adsorbed CH4 desorbs along with the CO2 and exits the bed as part of the off-gas (tail gas). Lower CO2/CH4 selectivity means more CH4 is co-adsorbed per cycle, and therefore more CH4 is lost per regeneration. For 13X, the selectivity is so high that the co-adsorbed CH4 represents only 1 to 2% of total CH4 throughput. For activated carbon, it is 6 to 10%.

For Bio-LNG plants that liquefy the upgraded biomethane, methane recovery is even more critical because the liquefaction process adds another 5 to 8% methane slip from boil-off gas. A 92% PSA recovery times a 94% liquefaction recovery equals 86.5% overall - versus 99% times 94% equals 93% for 13X. The 6.5 percentage point overall gap, at Bio-LNG sale prices of EUR 1.20 to 2.00 per Nm3, is worth EUR 250,000 to 600,000 per year on a 500 Nm3/h plant. This is the single strongest economic argument for 13X in the biomethane market.

Layered Beds: The Hybrid That Wins Most Greenfield Plants

The engineering literature and the European Biogas Association's 2024 working group report both converge on the same recommendation for new biogas upgrading plants larger than 300 Nm3/h: layered beds with activated carbon on top of 13X. The carbon layer (typically 30 to 40% of bed volume) scavenges H2S, water, and siloxanes before they reach the 13X layer. The 13X layer (60 to 70% of bed volume) does the bulk CO2/CH4 separation. Total methane recovery is 96 to 98% - between the carbon-only and 13X-only numbers, but the H2S tolerance is much better than 13X alone.

The implementation pattern is well established. Particle size is matched between the two materials to avoid pressure drop discontinuity at the interface - typically both materials are 1.6 to 2.5 mm bead. The carbon is loaded first, then a stainless steel support grid is placed on top of the carbon layer, then the 13X is added above. Alternatively, both materials can be loaded together and allowed to stratify by bulk density (carbon floats, 13X sinks) - though this is less reliable. Aluminaworld ships both materials in pre-weighed batches with matched bead size, and we provide a layered loading service for plants in Europe and East Asia that request it.

The disadvantages of layered beds are operational rather than fundamental. The plant operator must inventory two materials, with two reorder cycles and two quality control procedures. The carbon layer creates a slightly higher pressure drop because activated carbon has narrower pore mouths and a higher dust fraction. The carbon must be replaced on its own schedule (typically every 18 to 24 months) while the 13X continues to operate, which requires a partial unload-reload procedure. None of these are dealbreakers for a 500+ Nm3/h plant, but they add 5 to 10% to the annual operating complexity versus a single-adsorbent design.

For small-scale farm digesters below 200 Nm3/h, the layered bed pattern becomes less attractive because the operational complexity overhead is the same but the absolute methane recovery gain is smaller. In this segment, simple 13X PSA with a separate H2S scrubber upstream is the standard pattern. Some European OEMs (e.g. Carbotech, ETW Energietechnik) offer packaged skid-mounted units with 13X + upstream biological H2S scrubber in a single enclosure, optimized for 80 to 250 Nm3/h farm-scale applications. Aluminaworld supplies the 13X and the activated alumina for these packaged units.

5 Field Cases from European and Asian Biogas Plants

The numbers in the previous sections come from theory, vendor data, and pilot tests. The cases below are from real operating plants supplied by Aluminaworld, with anonymized operator data shared under NDA. They illustrate how the engineering choices translate into operational outcomes.

Case 1: 800 Nm3/h food-waste digester, Hamburg, Germany (commissioned 2023). Feed gas 60% CH4, 38% CO2, 1500 ppmv H2S. Iron sponge + 4 mm activated carbon guard bed + 13X-HG main bed. Operating 36 months with no sieve changeout - CO2 working capacity still at 92% of fresh value. Methane recovery 98.5%. Capex premium of 13X over activated carbon (EUR 60,000 higher adsorbent cost) paid back in 8 months via methane revenue. Lesson: aggressive pretreatment + premium-grade 13X is the winning combination for high-pressure, sweet-feed plants.

Case 2: 350 Nm3/h agricultural digester, Bologna, Italy (commissioned 2022). Feed gas 58% CH4, 40% CO2, 800 ppmv H2S. Layered bed (40% bituminous carbon on top of 60% 13X) in a single vessel. Operating 44 months, with carbon layer replaced at month 22. 13X still in service, CO2 working capacity at 88% of fresh value. Methane recovery 97.2%. Lesson: layered beds deliver intermediate performance but reduce operational complexity (one vessel, one regeneration system).

Case 3: 1200 Nm3/h landfill gas, Lyon, France (commissioned 2021). Feed gas 47% CH4, 36% CO2, 150 ppmv H2S, 5 mg/Nm3 siloxanes. Pure activated carbon main bed (no 13X). Operating 60 months with one full carbon changeout at month 30. Methane recovery 91% (deliberately accepting higher slip because the landfill operator had no economic use for tail gas and used it for boiler firing). Lesson: where tail gas has a use (boiler firing, district heating), activated carbon is rational despite the slip.

Case 4: 500 Nm3/h manure digester, Shandong, China (commissioned 2024). Feed gas 62% CH4, 35% CO2, 3000 ppmv H2S (very sour). Two-stage iron sponge (SulfaTreat + activated carbon) + 13X-HG main bed. Operating 24 months, 13X CO2 working capacity at 75% of fresh value (degrading faster than Case 1 because of H2S slip from the iron sponge). Plan to replace 13X at month 30. Lesson: even with aggressive pretreatment, very sour feed degrades 13X faster than moderate feed.

Case 5: 250 Nm3/h food waste + agricultural co-digestion, Bangkok, Thailand (commissioned 2025). Feed gas 60% CH4, 38% CO2, 800 ppmv H2S, hot humid climate (40 to 45 degrees C feed). Packaged 13X skid with biological H2S scrubber + refrigerated dryer + 13X-HG main bed. Operating 14 months, capacity still at 95% of fresh value. Methane recovery 97.8%. Lesson: hot climates need extra attention to feed cooling and dehydration to maintain 13X performance; refrigerated drying ahead of the sieve is essential.

Commissioning and Startup: The First 30 Days That Decide the Next 5 Years

How you commission a new biogas PSA determines its performance for the next 5 to 10 years. The most common commissioning mistakes are skipping the deep dehydration of the bed before first feed, ramping temperature too fast through the regeneration cycle, and not running a pressure-decay leak test on the vessel and piping before charging. Each of these is fixable but expensive. A proper commissioning sequence takes 5 to 10 days and locks in the design performance.

The standard 13X commissioning sequence we recommend for new biogas plants is: (1) vessel pressure test to 1.5 times design pressure for 30 minutes - all welds and connections verified; (2) vacuum test to full design vacuum (typically 50 mbar absolute) for 60 minutes - confirms no air ingress points; (3) sieve loading through top manway with personnel access - typically 4 to 8 hours for a 5 tonne bed, using a dust mask and sieve-loading chute to minimize dust exposure; (4) bed settling under N2 at design pressure for 24 hours to compact and level the bed; (5) thermal activation - heat the bed slowly (50 degrees C/hour maximum) to 280 degrees C under N2 flow for 8 to 12 hours to remove manufacturing moisture; (6) cool down to operating temperature under dry N2; (7) introduce biogas feed at 30% of design flow and ramp to 100% over 48 to 72 hours; (8) start regeneration cycle and verify product methane content at 95 to 99% within 2 to 4 hours.

The thermal activation step is the most commonly skipped or rushed procedure. If the bed is not fully activated, residual manufacturing moisture (typically 1 to 2 wt% in fresh 13X) ties up 10 to 20% of the adsorption capacity and shifts the operating window off-design. Operators notice this as lower-than-expected methane recovery and higher-than-expected pressure drop. The fix is a second activation cycle at 280 degrees C, but it requires a planned shutdown of 24 to 36 hours. Better to do it right the first time.

The biogas introduction ramp is the second-most-critical step. Bringing feed on too quickly causes two problems: (1) water adsorbed from the biogas (even after drying) rapidly displaces the CO2 in the first few centimeters of bed, generating a thermal pulse that can crack the bed support or shift the bed; (2) H2S slip from any momentary upset in the upstream guard bed can spike to the 13X and poison the top layer. Ramping feed over 48 to 72 hours smooths both effects and lets the upstream guard bed stabilize.

The first 30 days of operation are also the window where most mechanical problems surface. Operators should check bed pressure drop daily, product methane content daily, and feed H2S weekly. The first regeneration cycle typically delivers 80 to 90% of design CO2 capacity; full design capacity is reached after 5 to 10 cycles as the bed equilibrates with the actual feed. If after 30 days the bed is still below 95% of design capacity, the likely causes are: (1) insufficient activation, (2) feed water slip from upstream dryer, (3) H2S slip from guard bed. All three are diagnosable from KPI trends and fixable with a planned intervention.

Aluminaworld provides commissioning support to biogas plant operators on request. Our technical team can be on-site for the sieve loading, activation, and first-feed ramp phases, typically 5 to 7 days on site for a 500 Nm3/h plant. This service is included for orders above 5 tonnes and is billed separately for smaller orders. We have supported commissioning on more than 40 biogas plants in Europe, Asia, and the Americas in the last five years, with an average on-site visit cost of EUR 8000 to 15000 plus travel.

International Standards Governing Biogas Upgrading Adsorbent Selection

The standards landscape for biogas upgrading adsorbents has matured considerably in the last five years. The most relevant international and regional standards are summarized below. Aluminaworld's CoA reports compliance with all of these, and we run internal QC tests against each.

For European biomethane plants, EN 16723-1 is the binding grid spec. It sets the methane content requirement at 95 to 97% (depending on Wobbe index and calorific value matching local grid gas) plus limits on H2S (5 mg/Nm3), total S (30 mg/Nm3), NH3 (3 mg/Nm3), siloxanes (5 mg Si/Nm3), and halogens (1.5 mg Cl/Nm3). 13X with proper pretreatment comfortably hits all of these. For Chinese projects, GB/T 30301 is the binding spec with similar impurity limits. For US projects, the relevant spec is typically the local utility's tariff (often modeled on AGA or pipeline gas specs).

5-Year Total Cost of Ownership: 500 Nm3/h Biomethane Plant

The following TCO comparison is for a typical 500 Nm3/h biogas upgrading plant in Western Europe. Assumptions: 8000 operating hours/year, electricity at EUR 0.15/kWh, biomethane sale at EUR 1.00/Nm3, H2S feed at 1500 ppmv (sweet enough for standard pretreatment), grid injection to EN 16723-1 spec. Numbers are illustrative for the engineering comparison - actual figures vary by plant, region, and contract structure.

The conclusion is striking. Over 5 years, a 13X-only plant costs EUR 875,000 less than an activated-carbon-only plant of the same capacity, and a layered 40/60 bed costs EUR 173,000 more than 13X alone but EUR 702,000 less than pure carbon. The deciding factor is methane recovery, not regeneration energy. Activated carbon wins on annual electricity (EUR 40,000/year) but loses on methane slip (EUR 200,000/year). The asymmetry is so large that even a 50% increase in European electricity prices would not flip the conclusion - 13X remains the lower-TCO choice for plants where biomethane sale price exceeds EUR 0.50/Nm3.

Where does activated carbon still win? Three specific cases: (1) where the tail gas has a valuable use (boiler firing, district heating, CO2 recovery for greenhouse use) - then methane slip is not lost revenue, just lower-grade use; (2) where the plant is very small (below 200 Nm3/h) and the capex difference dominates the methane recovery difference; (3) where the H2S pretreatment train is unreliable and the operator wants to use carbon for its higher H2S capacity. In all three cases, the right answer is usually a layered bed or a switch to water scrubbing rather than pure carbon. Pure carbon-only biogas upgrading is becoming rare in the European and Chinese biomethane markets for new installations.

Procurement Specification: What to Ask Your Sieve Supplier

Most sieve procurement specs in the biogas industry are too loose - they specify particle size and "high CO2 capacity" but leave the actual performance tests undefined. The result is that two suppliers can each deliver material that meets the spec but performs 30 to 50% differently in the field. The procurement spec below is the one Aluminaworld uses internally and recommends to biogas plant operators. It locks in the four properties that drive PSA performance and the two properties that drive bed life.

Specification for 13X-HG Biogas Service (Aluminaworld internal reference spec, equivalent to EN ISO 15901-1 / GB/T 34540):

• Particle size: 1.6 to 2.5 mm bead, less than 0.5 wt% below 1.0 mm, less than 0.1 wt% below 0.5 mm (controls pressure drop rise in service)
• Equilibrium CO2 capacity at 1 bar, 25 degrees C, dry gas: greater than 5.5 mmol/g (verified by Sievert apparatus or magnetic suspension balance per ISO 15901-1 equivalent)
• Working CO2 capacity at 7 bar adsorption, 100 mbar regeneration, 35 degrees C: greater than 2.4 mmol/g (verified by pilot PSA test or dynamic column breakthrough)
• Equilibrium CH4 capacity at 1 bar, 25 degrees C: less than 1.2 mmol/g (verified same apparatus)
• Calculated CO2/CH4 selectivity: greater than 5 at equilibrium (must report both numbers, not just CO2)
• Bulk density: 720 to 780 kg/m3 (controls vessel sizing)
• Attrition (ASTM D4058 drum method): less than 0.05 wt% fines after 5000 revolutions (industry premium grade)
• Crush strength (single bead, ASTM D4179): greater than 30 N per bead (for 2.0 mm bead)
• LOI at 950 degrees C: less than 1.5 wt% (residual moisture + organics)
• pH (10% slurry): 9.0 to 10.5 (verifies no acid residue from manufacturing)
• H2S working capacity at 1000 ppmv H2S, 25 degrees C: greater than 0.3 mmol/g (note this is small but still useful for guard bed sizing)
• Lot certificate per shipment: CO2 capacity, attrition, particle size distribution, LOI, pH. Ship date + manufacturing batch number traceable to raw material lot.

For activated carbon used in layered bed or guard bed service:

• Particle size: 1.6 to 2.5 mm or 2.5 to 4.0 mm (match to 13X particle size to avoid interface pressure drop discontinuity)
• CO2 working capacity at 1 bar, 25 degrees C: greater than 3.5 mmol/g (verifies high micropore content)
• H2S breakthrough capacity (ASTM D6646 modified): greater than 8 wt% at 1000 ppmv H2S inlet, 25 degrees C (this is the dominant property for carbon in biogas)
• Iodine number: greater than 950 mg/g (industry standard for high-activity carbon)
• BET surface area: greater than 1000 m2/g (verified per ISO 9277)
• Bulk density: 480 to 580 kg/m3 (depends on grade, controls vessel sizing)
• Hardness (ASTM D3802): greater than 95 (industry standard)
• Moisture content (ASTM D2867): less than 5 wt% as packed
• Ash content: less than 8 wt% (low ash extends bed life by preventing pore blocking)

A reputable sieve supplier will agree to this specification without pushback. If the supplier pushes back on the CO2 working capacity test or the attrition test, that is a strong signal they cannot meet it. Aluminaworld's CoA reports every one of these values, with method reference and instrument ID, for every shipment. The CoA is signed by the QC manager and countersigned by the production manager.

Operating Monitoring: The KPIs That Tell You When to Change the Bed

You cannot manage what you cannot measure. The five KPIs that every biogas upgrading operator should track on a daily or weekly basis are listed below. Aluminaworld provides a free KPI tracking template (Excel format) to plant operators on request.

1. Upgraded gas methane content: measured by IR or NDIR analyzer at the PSA outlet. Target 96 to 99%. Drop below 95% indicates bed capacity loss or pressure equalization problem.
2. Tail gas flow rate: measured by thermal mass flow meter. Tail gas flow rising while methane content in product stays constant indicates increasing methane slip - the bed is exhausting faster.
3. Pressure drop across each bed: measured by differential pressure transmitter. Rising pressure drop at constant flow indicates dust accumulation or bed compaction. Doubling of pressure drop from baseline typically signals end of bed life.
4. Feed H2S concentration: measured by electrochemical or UV sensor at PSA inlet. Spikes above 100 ppmv indicate pretreatment breakthrough and require immediate attention.
5. Regeneration temperature profile: measured by multiple thermocouples along the bed height. Non-uniform temperature profile (cold spots, hot spots) indicates channeling, bed settling, or heater failure.

When two or more of these KPIs move out of spec simultaneously, the bed is approaching end of life and a planned changeout should be scheduled. Aluminaworld's technical team can support remote KPI review on request - send us 30 days of trend data and we will diagnose the likely cause (H2S breakthrough, water hammer, dust accumulation, normal end-of-life) and recommend the right response.

Next Steps: Working with Aluminaworld on Your Biogas Project

If you are sizing a new biogas upgrading plant, retrofitting an existing plant to improve methane recovery, or troubleshooting premature bed failure, the path forward is straightforward. Aluminaworld provides engineering support, sample materials, and full CoA documentation at no charge for projects above 1 tonne annual sieve consumption.

• WhatsApp: +86 133 2522 2240 (fastest, 12-hour reply) - click the green button at the bottom of this page
• Email: barry@aluminaworld.com
• Sample request: 5 to 10 kg R&D pack of 13X-HG + activated carbon for layered bed testing, 7 to 10 day lead time, full CoA included
• Pilot trial: 500 to 2000 kg shipment for in-plant validation, 15 to 20 day production, FOB/CIF/CFR from Qingdao Port (80 km from our factory)
• Custom grades: 13X with modified Si/Al ratio for high-CO2 biogas (greater than 45% CO2 feed) or low-H2S high-pressure service, 30 to 45 day lead time for first order
• Bed loading service: for European and East Asian plants, our technical team can supervise initial sieve loading and layered bed installation on request (additional cost)

Aluminaworld has supplied 13X molecular sieve and activated carbon to biogas upgrading projects in 30+ countries for 15 years. Our 13X-HG grade is engineered specifically for biogas service - low-fines bead, attrition-rated below 0.05 wt%, CO2 working capacity 2.5+ mmol/g at 7 bar, and H2S working capacity verified on every lot. The CoA includes both equilibrium and working CO2 capacity, attrition, particle size distribution, and H2S capacity, with full method traceability to ISO 15901 and ASTM D4058. Whether you need 5 tonnes for a single plant or 200 tonnes for a multi-site roll-out, we ship from our Zibo factory in 15 to 20 days production + 25 to 35 days transit, with full customs documentation for EU, North American, and Asian ports.

For projects where biomethane is the primary product, getting the adsorbent choice right is the single highest-leverage engineering decision. Methane recovery, bed life, and total cost of ownership all hinge on whether you select 13X, activated carbon, or a layered hybrid. This guide has given you the engineering framework - the equilibrium capacities, the H2S tolerance, the regeneration energy, the 5-year TCO, the procurement specification. The next step is yours: send us your feed gas composition, target methane recovery, and bed dimensions, and we will provide a quote and CoA for the right adsorbent within 48 hours.