Activated Carbon Adsorption Fundamentals

1. Adsorption Principles

Adsorption is a surface phenomenon in which molecules from a gas or liquid phase accumulate on the surface of a solid adsorbent. In midstream gas processing, activated carbon is used extensively to remove trace contaminants including hydrogen sulfide (H₂S), mercury (Hg), BTEX (benzene, toluene, ethylbenzene, xylenes), and volatile organic compounds (VOCs) from natural gas and process streams.

Physical vs Chemical Adsorption

Physical adsorption (physisorption) is driven by weak van der Waals forces between the adsorbate molecule and the carbon surface. It is reversible, non-specific, and favored by lower temperatures and higher pressures. Heat of adsorption is typically 2–10 kcal/mol. Physisorption allows the carbon to be thermally regenerated and reused.

Chemical adsorption (chemisorption) involves the formation of chemical bonds between the adsorbate and functional groups on the carbon surface or an impregnant. It is highly selective, often irreversible, and provides much higher capacity for specific contaminants. Impregnated carbons (e.g., KOH for H₂S, sulfur for mercury) operate primarily through chemisorption.

Activated Carbon Structure

Activated carbon derives its exceptional adsorptive capacity from an extremely high internal surface area, typically 800–1,200 m²/g. This surface area is created during activation, which develops an extensive network of micropores (diameter <2 nm), mesopores (2–50 nm), and macropores (>50 nm). The micropores account for most of the adsorptive surface area and are responsible for capturing small molecules, while the mesopores and macropores serve as transport channels that allow gas molecules to reach the micropore structure.

The raw material and activation process determine the pore size distribution and surface chemistry of the final product:

• Coal-based carbon: Produced from bituminous or anthracite coal; offers a broad pore size distribution with a good balance of micro- and mesopores; most commonly used in gas processing
• Coconut shell carbon: Predominantly microporous; very high hardness and abrasion resistance; excellent for trace contaminant removal where small molecules dominate
• Wood-based carbon: Predominantly mesoporous and macroporous; lower density; better suited for liquid-phase applications or large-molecule adsorption

Adsorption Isotherms

The equilibrium relationship between the amount of contaminant adsorbed and its concentration in the gas phase at constant temperature is described by adsorption isotherms. Two models are widely used in engineering design:

Langmuir isotherm: Assumes monolayer adsorption on a surface with a finite number of identical sites. Suitable when the carbon approaches saturation at higher concentrations.

Freundlich isotherm: An empirical model that assumes heterogeneous surface energies. More commonly used for gas-phase activated carbon applications because it better represents the multilayer adsorption behavior observed in practice.

Where q is the mass adsorbed per unit mass of carbon, C is the gas-phase concentration, and K_F and n are empirical constants determined from laboratory testing.

Key Parameters Affecting Adsorption

• Temperature: Lower temperatures increase adsorption capacity. Physical adsorption capacity decreases approximately 0.3% per °F above 80°F.
• Pressure: Higher operating pressures increase the partial pressure of contaminants, improving capacity. Gas-phase carbon beds in pipeline service (500–1,000 psig) have significantly higher capacity than atmospheric beds.
• Molecular weight: Heavier molecules with higher boiling points are adsorbed preferentially. BTEX compounds (MW 78–106) are much more readily adsorbed than methane (MW 16).
• Humidity: Water vapor competes for adsorption sites. Keeping the gas above its dew point prevents pore blockage by condensed water.

Carbon Types and Properties

2. Mass Transfer Zone and Breakthrough

Understanding the mass transfer zone (MTZ) is essential for designing an activated carbon bed that achieves the desired contaminant removal while maximizing carbon utilization. The MTZ concept governs bed sizing, predicts service life, and determines when carbon change-out is required.

The Mass Transfer Zone

When contaminated gas enters a fresh carbon bed, adsorption occurs in a narrow band called the mass transfer zone. Upstream of the MTZ, the carbon is fully saturated (equilibrium loading). Within the MTZ, the gas-phase concentration transitions from the inlet value down to essentially zero. Downstream of the MTZ, the carbon is still fresh and unused.

As gas continues to flow, the MTZ moves progressively through the bed from inlet to outlet. The rate of movement depends on the inlet concentration, gas velocity, and the equilibrium capacity of the carbon for that contaminant.

Breakthrough Curve

The breakthrough curve is the S-shaped plot of outlet concentration versus time (or total gas volume processed). It reflects the passage of the MTZ through the bed and is characterized by several key points:

• Stoichiometric time: The theoretical time to saturate the entire bed if the MTZ had zero length (a perfectly sharp front)
• Breakthrough point: The time at which the outlet concentration first reaches the target limit, typically defined as 1–5% of the inlet concentration. This is the practical end of the service cycle.
• Exhaustion point: The time at which the outlet concentration equals the inlet concentration, indicating the entire bed is fully saturated

Bed Utilization

Bed utilization is the fraction of the total carbon capacity that is actually used before breakthrough occurs. Because the MTZ occupies a finite length of the bed, the carbon within the MTZ is only partially saturated (approximately 50% on average) at breakthrough. The bed utilization is therefore:

Where L_MTZ is the length of the mass transfer zone and L_bed is the total bed length. For a bed that is 3× the MTZ length, utilization is approximately 83%. For a bed that is only 1.5× the MTZ length, utilization drops to 67%, meaning one-third of the carbon capacity is wasted at each change-out.

Factors Affecting MTZ Length

• Particle size: Smaller particles reduce the MTZ length by providing more external surface area for mass transfer, but increase pressure drop
• Gas velocity: Higher velocities increase the MTZ length because molecules have less contact time with the carbon
• Inlet concentration: Higher concentrations can increase MTZ length due to the steeper concentration gradient
• Adsorption kinetics: Contaminants with slower adsorption rates (e.g., chemisorption on impregnated carbon) produce longer MTZ lengths
• Temperature: Higher temperatures reduce adsorption rates and increase MTZ length

The fundamental design rule is that the total bed length must be at least 2–3 times the MTZ length to achieve acceptable bed utilization (greater than 67–83%). Beds shorter than 2× MTZ result in poor utilization and frequent change-outs.

Typical MTZ Lengths by Contaminant

3. Bed Sizing and Design Velocity

Proper bed sizing balances contaminant removal performance against capital cost and pressure drop. The key design parameters are superficial velocity, contact time, bed geometry, and carbon weight. All of these are interrelated, and the design process typically begins with the contaminant loading and required service life, then works backward to determine the vessel dimensions.

Superficial Velocity

The superficial velocity (also called face velocity) is the volumetric gas flow rate divided by the cross-sectional area of the bed. It determines the residence time in the bed and directly affects both the MTZ length and pressure drop.

For gas-phase applications, the recommended superficial velocity range is 15–30 ft/min. Lower velocities (10–15 ft/min) are preferred for mercury removal, where the slower chemisorption kinetics require longer contact time. Velocities above 30 ft/min can cause excessive pressure drop, carbon attrition, and poor bed utilization due to elongated MTZ.

Contact Time

The empty bed contact time (EBCT) is the ratio of bed volume to volumetric flow rate. It represents the theoretical residence time of the gas in the bed if the bed were empty (no carbon present). The actual gas residence time is shorter because the carbon particles occupy approximately 60–65% of the bed volume.

Minimum recommended EBCT values are 2–5 seconds for most gas-phase applications. Mercury removal typically requires 4–8 seconds due to the slower chemisorption kinetics on sulfur-impregnated carbon.

Bed Geometry

• Minimum bed height: 4–6 ft is required for adequate MTZ development. Beds shorter than 4 ft often exhibit premature breakthrough because the MTZ cannot fully form within the available bed depth.
• L/D ratio: A length-to-diameter ratio of 2:1 to 4:1 is typical for vertical vessels. Ratios below 2:1 increase the risk of channeling and poor gas distribution. Ratios above 4:1 can result in excessive pressure drop.
• Gas distribution: An inlet distributor plate and outlet support grid are essential for uniform flow through the bed. Poor distribution leads to localized breakthrough and wasted carbon capacity.

Carbon Weight Calculation

The required carbon weight is determined by the mass balance between contaminant loading and the carbon’s adsorptive capacity:

Where the capacity is the equilibrium or working capacity of the selected carbon for the target contaminant at operating conditions. Working capacity is always less than equilibrium capacity because of the unused carbon in the MTZ at breakthrough.

Pressure Drop

Pressure drop through a packed carbon bed is calculated using the Ergun equation, which accounts for both viscous and inertial flow resistance. For typical 4×10 mesh activated carbon at gas processing velocities, the pressure drop ranges from 0.5–2.0 psi per foot of bed depth. Factors that increase pressure drop include smaller particle size, higher gas velocity, higher gas density (higher operating pressure), and carbon fines accumulation.

Design Guidelines Summary

4. Carbon Types and Impregnation

The choice of activated carbon type is determined by the target contaminant, required capacity, regeneration requirements, and disposal considerations. Carbon manufacturers offer a range of virgin and impregnated products, each optimized for specific service conditions.

Virgin Activated Carbon

Virgin (non-impregnated) activated carbon removes contaminants through physical adsorption only. It is the preferred choice for BTEX and VOC removal from natural gas streams because these organic compounds are readily physisorbed in the carbon micropores. Virgin carbon can be thermally regenerated by passing hot inert gas or steam through the bed to desorb the captured contaminants, restoring much of the original capacity. Typical regeneration temperatures are 400–600°F.

Caustic-Impregnated Carbon (KOH/NaOH)

Caustic-impregnated carbon is the workhorse product for H₂S removal in midstream gas processing. The carbon is treated with potassium hydroxide (KOH) or sodium hydroxide (NaOH), which reacts chemically with H₂S to form potassium sulfide or sodium sulfide. This chemisorption mechanism provides much higher H₂S capacity than virgin carbon—typically 15–25 wt% versus 1–3 wt% for virgin carbon.

Because the reaction is irreversible, caustic-impregnated carbon cannot be regenerated and must be disposed of after exhaustion. The spent carbon is typically classified as non-hazardous waste and can be landfilled, though testing for TCLP metals may be required depending on the gas composition.

Sulfur-Impregnated Carbon

Sulfur-impregnated carbon is specifically designed for elemental mercury removal. The impregnated sulfur reacts with mercury vapor to form mercuric sulfide (HgS), which is chemically stable and insoluble. This irreversible chemisorption provides very high mercury removal efficiency (>99%) down to outlet concentrations below 0.01 µg/Nm³.

Sulfur-impregnated carbon is non-regenerable. Spent carbon containing mercury is classified as hazardous waste under RCRA and requires specialized disposal at permitted facilities, which significantly increases operating costs.

Catalytic Carbon

Catalytic carbon is manufactured with enhanced surface chemistry (typically through high-temperature steam activation) that promotes oxidation reactions on the carbon surface. In the presence of oxygen and moisture, catalytic carbon converts H₂S to elemental sulfur, which deposits in the carbon pores. This mechanism provides higher H₂S capacity than virgin carbon while allowing partial regeneration—the deposited sulfur can be washed from the pores with water or a mild solvent, partially restoring capacity.

Catalytic carbon is most effective when the gas contains a small amount of oxygen (0.5–2 vol%) and moisture. It is commonly used in wastewater treatment and low-pressure gas applications but is less common in high-pressure natural gas service.

Selection Criteria

The selection of the appropriate carbon type depends on several interrelated factors:

• Contaminant type and concentration: Determines whether physisorption or chemisorption is needed
• Required outlet specification: Tighter specifications may require impregnated carbon or multiple beds in series
• Regeneration requirements: If regeneration is economically or operationally necessary, only virgin or catalytic carbon is suitable
• Disposal considerations: Mercury-laden carbon requires hazardous waste disposal; caustic-impregnated carbon is typically non-hazardous
• Cost: Impregnated carbons cost 2–5× more per pound than virgin carbon, but their higher capacity can offset the price premium

Carbon Selection Guide

5. Operating Considerations

Successful operation of an activated carbon system requires attention to bed configuration, moisture management, temperature control, and timely change-out procedures. The operating strategy directly impacts both contaminant removal performance and long-term operating costs.

Lead-Lag Configuration

The lead-lag (or series) configuration uses two carbon beds arranged in series. The upstream (lead) bed performs the bulk of the adsorption, while the downstream (lag) bed acts as a polishing guard to catch any contaminant that breaks through the lead bed. This arrangement offers several advantages over a single-bed system:

• The lead bed can be operated to full exhaustion (100% utilization) because the lag bed provides backup protection
• When the lead bed is exhausted, it is taken offline for carbon change-out. The lag bed becomes the new lead bed, and the refreshed vessel becomes the new lag bed.
• Continuous operation is maintained without any period of unprotected service
• Overall carbon utilization is maximized, reducing operating costs by 15–30% compared to a single-bed system

Single-Bed Configuration

A single carbon vessel is simpler and less expensive but requires the system to be bypassed during carbon change-out. This configuration is acceptable only for non-critical service where brief periods without treatment are tolerable (e.g., BTEX removal for emission control where regulatory exceedances during change-out can be managed). For H₂S or mercury removal in pipeline-quality gas service, lead-lag configuration is strongly preferred.

Moisture Management

Excess moisture is the most common cause of poor carbon bed performance. Water vapor condenses in the carbon pores, blocking adsorption sites and dramatically reducing capacity for the target contaminant. Liquid water can also cause channeling and uneven flow distribution through the bed.

• Maintain the inlet gas temperature at least 10–15°F above its water dew point
• Install a knockout drum or coalescing filter upstream of the carbon bed to remove entrained liquids
• Monitor the differential pressure across the bed—a sudden increase can indicate liquid accumulation
• In humid environments, consider a gas heater upstream to prevent condensation during temperature swings

Temperature Effects

Adsorption capacity decreases as temperature increases because the equilibrium favors desorption at higher temperatures. For physical adsorption, the capacity reduction is approximately 0.3% per °F above 80°F. The optimal operating range for most gas-phase carbon applications is 60–100°F. At temperatures above 120°F, capacity losses become significant and should be factored into bed sizing calculations.

High temperatures can also pose safety risks when treating H₂S-laden gas. The exothermic reaction between H₂S and caustic impregnant can generate localized hot spots, particularly at high H₂S concentrations (>500 ppmv). Temperature monitoring thermocouples should be installed at multiple points in the bed to detect hot spots early.

Spent Carbon Disposal

The disposal classification of spent activated carbon depends on the contaminants it has adsorbed:

• Mercury-laden carbon: Classified as hazardous waste under RCRA (D009). Must be retorted or stabilized at a permitted facility. Disposal costs are typically $0.50–$2.00 per pound.
• H₂S-laden (caustic-impregnated) carbon: Generally non-hazardous. Can be landfilled at municipal solid waste facilities in most jurisdictions. Disposal costs are typically $0.05–$0.15 per pound.
• BTEX/VOC-laden virgin carbon: May be classified as hazardous if benzene exceeds TCLP limits (0.5 mg/L). Thermal regeneration is preferred over disposal when volumes justify the equipment cost.

All spent carbon should be tested using the Toxicity Characteristic Leaching Procedure (TCLP) before disposal to confirm the appropriate waste classification.

Bed Change-Out Procedure

1. Isolate: Close inlet and outlet valves; switch to lag bed (lead-lag systems) or bypass (single-bed systems)
2. Depressure: Slowly vent the vessel to atmospheric pressure through appropriate relief systems
3. Purge: Purge with nitrogen or inert gas to remove residual hydrocarbons. Confirm LEL <10% before entry.
4. Remove: Vacuum out the spent carbon through the vessel manway. Use proper PPE for potential H₂S or mercury exposure.
5. Inspect: Inspect the vessel internals, support grid, distributor plate, and thermocouple wells
6. Repack: Fill with fresh carbon through the top manway. Minimize drop height to reduce carbon breakage and fines generation.
7. Pressure test: Close the vessel and pressure test to confirm leak-free integrity
8. Purge and commission: Purge with process gas at low rate, then bring online gradually

Cost Optimization

The total cost of ownership for an activated carbon system involves a trade-off between capital cost (vessel size, piping, instrumentation) and operating cost (carbon replacement, disposal, labor). Larger beds require more capital but extend the change-out interval, reducing annual operating costs. The economic optimum depends on:

• Carbon cost per pound (varies by type and quantity)
• Disposal cost per pound (especially significant for mercury service)
• Labor and downtime cost per change-out event
• Target change-out interval (typically 6–24 months for gas processing)

Operating Best Practices

References

1. GPSA, Chapter 21 — Hydrocarbon Treating
2. Calgon Carbon Corporation — Activated Carbon Adsorption Technical Bulletins
3. ASTM D3860 — Standard Practice for Determination of Adsorptive Capacity of Activated Carbon by Aqueous Phase Isotherm Technique
4. EPA 40 CFR Part 261 — Identification and Listing of Hazardous Waste (TCLP)