MSW Management | From LFG to BTUs

SUBSCRIBE

COMMENT
ON THIS
ARTICLE

CREATE A LINK
TO THIS ARTICLE
ON YOUR SITE


	Photo: @istockphoto.com/guyerwood

No one technology is capable of performing the required task; but by combining commercially proven technologies, high-Btu gas can be achieved with very little technical risk.

By Gary J. Nagl

In the past, landfill gas (LFG) recovery projects consisted of the installation of gas gathering systems, followed possibly by desulfurization and siloxane removal and then by combustion in gas engines or turbine/generator sets for the production of electrical power or in boilers for the production of steam. These types of projects were very attractive since very little processing was required, even though the gas had low- to medium-heating values (300–500 Btus per cubic foot). Unfortunately, the power companies would not pay a very high price for the power.

Times have changed. With today’s high energy prices, LFG recovery projects to produce high-Btu gas, which can be injected into natural gas pipelines, are becoming very attractive even without tax breaks; however, a good deal more gas processing is required to achieve the 900-plus-Btus-per-cubic-foot heating value.

The real technical risks associated with LFG recovery systems for the production of high-Btu gas are mostly related to the operational and maintenance aspects of the equipment. Compared to normal landfill operations, gas recovery systems are relatively complex; however, with the substantial revenues that can be obtained from the recovered energy, these projects can easily support a higher grade of operating and maintenance personnel, thus eliminating the risks associated with the technical and operational aspects of producing high-Btu gas.

Upgrading landfill gas to high-Btu gas requires the almost complete removal of all compounds other than methane. This requirement necessitates removal of the following components from a landfill gas stream: hydrogen sulfide, carbon dioxide, non-methane organic hydrocarbons (NMOH), chlorinated hydrocarbons, fluorinated hydrocarbon siloxanes, nitrogen, oxygen, and moisture.

Unfortunately, there is no one process that can remove all of these components in one step. Consequently, a series of processes must be combined to achieve the desired results, which is a gas consisting of the following compositions:

Hydrogen sulfide	< 4 ppm (v/v)
Carbon dioxide	< 3% (v/v)
Siloxanes	0%
Oxygen	< 1%
Moisture	< 5 poounds/MMCF
Heat content	900 + Btus/SCF

A discussion of the processes available to remove the various contaminants follows.


	Figure 1

Hydrogen Sulfide Removal
There are a variety of commercially proven processes available to remove hydrogen sulfide from gas streams. These processes can be divided into wet and dry processes. The dry processes are generally nonregenerable and are characterized by having low capital costs but high relative operating costs (dollars per pound of hydrogen sulfide removed). Wet processes are just the opposite in that they are usually regenerable but have high capital costs and low relative operating costs. The point at which the economics switch from the dry systems to the wet systems is usually around 300 pounds of hydrogen sulfide per day.

The oldest dry system still in use today is the iron sponge process, which consists of wood chips impregnated with hydrated ferric oxide. The hydrogen sulfide reacts with the ferric oxide to form iron sulfide. The spent media can be regenerated by passing air through the bed, which converts the iron sulfide back to ferric oxide and elemental sulfur.

This process has been successfully employed for many years on LFG; however, it has lost some appeal due to the environmental and safety problems associated with handling and disposing of the spent media.

Another dry system that has been successfully employed on LFG consists of iron oxide impregnated on a ceramic base. As the hydrogen sulfide passes over the media, it reacts with the iron oxide to form innocuous iron pyrite. The process is not regenerable, but it is attractive because the safety problems associated with iron sponge are eliminated. The two commercial processes employing this technology are Sulfur-Rite and SulfaTreat.

Activated carbon can also be employed for selectively removing hydrogen sulfide from LFG; however, the high cost of replacing and disposing of the carbon generally limits its use to polishing applications (very small amounts of hydrogen sulfide to be removed).

The only wet system in commercial use today for selectively removing hydrogen sulfide from LFG in one step is the Locat process. The process is based on liquid redox chemistry and employs an aqueous solution of chelated iron to directly convert the hydrogen sulfide to elemental sulfur. This technology has been in use at a landfill in Florida since 1995, and several other units are currently being installed.

All of the above processes are capable of removing more than 99.9% of the hydrogen sulfide and have been proved in LFG service for several years.

Hydrogen Sulfide and Carbon Dioxide
Several commercially proven, liquid absorption processes are available for the simultaneous removal of hydrogen sulfide and carbon dioxide. These technologies are further characterized by employing chemical or physical absorbents. Chemical absorbents selectively remove acidic compounds by reacting with them to form weakly bonded compounds, which disassociate upon the application of heat. Physical absorbents rely on the solubility of the acidic components in the absorbent, which are then released by merely reducing the pressure. When acidic gases constitute a major portion of the gas stream to be treated, such as LFG, physical absorbents become more attractive because of the lower energy requirement.

Chemical solvents, which are widely used in the natural gas–processing industry, consist of alkanolamines, such as mono-, di-, and methylethanolamine. A gas at elevated pressure is contacted with the amine in an absorber. The acidic components are absorbed and react with the amine. The amine solution is then depressured to release any absorbed hydrocarbons before entering the stripper column, in which the solution is heated to its boiling point. The acidic components disassociate and are released while the stripped solution is cooled and returned to the absorber column. Because of the large quantity of acidic components and the corresponding large heat requirements, chemical absorption systems are rarely employed in treating LFG.

The simplest physical absorption system consists of contacting the LFG with water at elevated pressures, which will preferentially absorb the acidic components. The effluent water from the absorber is then stripped with air to remove the absorbed components. This processing technique has been successfully employed in the treatment of LFG in Europe; however, its drawback is that a great deal of water must be circulated through the absorber resulting in high parasitic power consumptions.

The Selexol process is a proprietary technology that employs a physical solvent derived from a dimethyl ether of polyethylene glycol. The solvent has a strong affinity for carbon dioxide, hydrogen sulfide, water, and heavy hydrocarbons; however, it has a very low affinity for nitrogen and oxygen. If the gas to be treated contains oxygen, it is beneficial to remove the hydrogen sulfide prior to the Selexol unit, since oxygen will react with the hydrogen sulfide to form elemental sulfur, causing operational problems within the unit. One disadvantage of the Selexol process is the relatively high cost of the solvent. The Selexol process has been successfully employed in a number of LFG applications. The process will be further described in the example section of this report.

Acid gases can also be separated from LFG by utilizing membrane-separation techniques, which utilize the differential permeability of components through polymeric membranes to affect separation. For LFG applications, carbon dioxide and hydrogen sulfide have high permeabilities and will pass through the membrane (permeate), while methane has a very low permeability and will remain on the outside of the membrane. A single-stage membrane will produce a gas stream containing approximately 88% methane. To achieve a higher methane concentration, multiple stages at high pressure (approximately 575 psig) are required. All gaseous components have some permeability; consequently, the permeate stream will contain various amounts of all components contained in the landfill gas. Membranes have been successfully used to treat landfill gas since the early 1980s. To ensure long membrane life it is beneficial to process the LFG through activated carbon for removal of NMOCs prior to processing through the membranes.

Adsorption techniques (gas being adsorbed onto a solid) can also be employed for treating LFG. A molecular sieve consisting of a zeolite material is generally the adsorbent of choice for treating LFG. These materials are very porous and have very high surface areas, the majority of which are internal to the particles. Under pressure, the molecular sieves preferentially adsorb carbon dioxide, methane, and water, although all gaseous components will be adsorbed to some degree. When a bed of molecular sieves approaches saturation, it is taken offline and depressured, releasing the adsorbed material.

Because of this processing technique, these systems are called Pressure Swing Adsorbers, or PSAs. To achieve smooth operation, multiple vessels must be operated in parallel. Molecular sieves have been successfully employed for treating LFG for some time.

It is important to remember that all of these methods simply separate the acid gas components (carbon dioxide and hydrogen sulfide) from the LFG, producing an acid gas stream that will require further processing before being released to the atmosphere. At a minimum, incineration in a flare will be required and possibly hydrogen sulfide removal prior to incineration.

Heavy Hydrocarbon
A common method of removing heavy hydrocarbon, including halogenated organics, is with activated carbon. The heavy hydrocarbons are preferentially adsorbed onto the activated carbon, and when the carbon bed becomes saturated the carbon is replaced with fresh material. Generally, at least two carbon vessels are employed in a lead-lag arrangement to ensure that one vessel is always online. Activated carbon has been used to treat a variety of gas streams, including LFG, for quite
some time.


	Figure 2

Siloxane
Cosmetics, detergents, paper coatings, and textiles all contain silicon compounds, which degrade to volatile organo-siloxanes when deposited in a landfill.

Problems occur when these compounds are burned, since their product of combustion is silica, which will deposit on such cold surfaces in the combustion equipment as turbine blades and boiler tubes.

Many treatment schemes are available for removing siloxanes. The more common methods consist of passing the gas over a bed of activated carbon or activated graphite.

Both methods have been employed on LFG with good results.

Nitrogen and Oxygen
LFG will always contain some amount of air due to the infiltration of air into the landfill as LFG is withdrawn.

The amount of air infiltration depends on the position of the wells, with deeper wells having less air infiltration than shallow wells, and on the diligence of the operators balancing the gas-gathering system.

Unfortunately, nitrogen and oxygen are very difficult to remove from any gas. Oxygen is generally removed by passing the gas at elevated temperatures over a catalyst, which uses the oxygen to oxidize some methane to carbon dioxide and water. Membranes and PSAs can only remove approximately 45% of the oxygen.

Nitrogen is also very difficult to remove, mainly because its molecular size is very close to methane.

In the natural-gas industry, nitrogen is generally removed by cryogenic processes, which are very complicated and expensive and are not adaptable to LFG applications. Membranes and PSAs can only remove approximately 10% of the nitrogen. Consequently, the nitrogen content of landfill gas is generally controlled by adjusting the wells to minimize air intrusion into the landfill.

Recently the Engelhard Corp. developed a new molecular sieve called the Molecular Gate. This molecular sieve is unique in that the exact size of the pores can be controlled so that molecules with similar sizes in the 3.0–4.0 range (nitrogen/methane, argon/oxygen, and nitrogen/oxygen) can be separated. This technology is relatively new; however, it has been successfully employed on natural gas and coal-bed methane. Nitrogen and oxygen removals approaching 80% can be expected with this technology.

Physically, the Molecular Gate system is operated like a pressure-swing adsorber. The gas to be treated is passed over the Molecular Gate media until the media become saturated with the adsorbed components. The adsorber is then taken offline and depressured, releasing the adsorbed components.

As with a standard PSA, multiple vessels operating in parallel are required to achieve smooth operation.

Although the molecular gate technology has not been employed in processing LFG, the proposed processing scheme as described later includes a great deal of pretreatment of the gas prior to entering the Molecular Gate unit; consequently, the technical risk associated with a beta unit of this technology is greatly reduced. One common complaint of the molecular gate technology is that of cost.

PSA/Membrane/Molecular Gate System
Figure 1 illustrates a processing scheme employing a PSA, a membrane, and a Molecular Gate to achieve high-Btu gas from LFG. LFG is removed from the landfill via an LFG blower. A constant draft is maintained on the landfill well system by controlling the speed of the blower with a variable frequency drive. Other control methods can be utilized, such as maintaining a constant flow of LFG through the system. Since the LFG will contain varying amounts of oxygen, it is best to remove the hydrogen sulfide prior to further processing to prevent the formation of elemental sulfur in downstream processing equipment. For this application, a hydrogen-sulfide scavenger is being employed in a lead-lag arrangement, which in essence will remove all of the hydrogen sulfide. The gas is then compressed to approximately 215 psig, which is controlled via spillback to the suction of the compressor. The gas is then cooled to less than 125°F and then passed through coalescing filters to remove any entrained liquid droplets.

The gas then passes through a standard PSA, which removes a large portion (approximately 68%) of the carbon dioxide and all of the moisture. Methane loss is minimal (approximately 7.5%). The permeate stream from the PSA is the only outlet for the removed carbon dioxide. The permeate is directed to the flare, and the residue is passed through a lead-lag-activated carbon system, which removes nearly all of the siloxanes and heavy hydrocarbons. The effluent gas is then filtered to remove carbon dust before entering the membrane system. The membrane has a high selectivity for removing carbon dioxide; however, it also has relatively high methane loss (approximately 15%). To recover the lost methane, the permeate stream is recycled back to the suction of the compressor.

This initial processing scheme (PSA/activated carbon/membrane) is the Medal process as marketed by Air Liquide.

The methane-rich effluent from the membrane is still contaminated with nitrogen (approximately 16%), oxygen (approximately 1.7%), and carbon dioxide (approximately 3.8%), which lowers the heating value of the gas to approximately 785 Btus per cubic foot.

To increase the heating value to approximately 950 Btus per cubic foot, the gas is then passed through a Molecular Gate, which removes all of the remaining carbon dioxide and approximately 80% of the nitrogen and oxygen at the expense of a 10% loss of methane. To reduce this loss, 70% of the permeate stream is recycled back to the compressor suction.

The effluent tail gas from the Molecular Gate has a heating value of approximately 300 Btus per cubic foot. Consequently, this gas stream may be utilized in a gas engine or can be sent to flare. The final-product gas has a heating value of approximately 950 Btus per cubic foot, and the overall system has a methane recovery of approximately 88%.

Selexol Process
A schematic flow diagram of a Selexol system treating landfill gas is illustrated in Figure 2. As previously described, the Selexol solvent is a physical solvent, which means that components are absorbed by their relative solubility in the solvent. For LFG applications, all of the components contained in landfill gas have high solubilities in the solvent, with the exception of methane, nitrogen, and oxygen. Consequently, much of the processing to upgrade LFG to pipeline quality can be done in the Selexol process.

As in the previous example, the LFG is transferred from the landfill through a gas-gathering system via an LFG blower. The gas is then passed through a hydrogen sulfide scavenger system and then compressed to approximately 215 psig. The gas then passes counter-currently to the Selexol solvent in a Stage 1 Selexol absorber. The contaminant-laden solvent from this absorber is then directed into an air stripper for contaminant removal. The contaminant-laden air is sent to an incineration system prior to exhausting to the atmosphere. The gas then passes through a second-stage absorber before being directed to nitrogen and oxygen removal. The spent solvent is regenerated by three depressurization steps.

Summary
In conclusion, there are several technologies that have been demonstrated to be capable of upgrading LFG to pipeline-quality gas. The difficulties with upgrading landfill gas are not associated with the technical aspects of the project but rather with the operational and maintenance aspects of the systems. As illustrated in Figure 1 and Figure 2, the required processing schemes involve equipment and complexity that are common to the petroleum, gas processing, and chemical industries but are not common to a landfill operation. This is the main reason why some systems have failed in the past. However, with today’s energy prices, the recovery of LFG is no longer an environmental nicety that must be subsidized with tax credits. It presents a substantial amount of income and thus can support a higher grade of operating personnel. For example, recovering 5,000 scfm of landfill gas and upgrading it to pipeline-quality gas represents approximately $8,000,000 a year in revenue if it is sold at $8 per thousand Btus. Another situation, which may become a problem, is the operation of the LFG gathering system. All processing systems operate better if the composition and flow rate of the stream to be processed remains relatively constant. Variability in composition and flow rate will complicate the operation to some extent. There must be some coordination between the landfill operator and the operator of the gas recovery system.

Gary J. Nagl is vice president and general manager of Gas Technology Products.

MSW - January/February 2008