Introduction to Gas Treating
When talking of gas treating, it is most often implied that natural gas is the focus. The natural gas industry is one of the World’s largest. However, there is also treatment of gas in the synthesis gas industry, and there are a number of processes used for this that are similar to those used for natural gas treatment.
Finally, there is a large industry devoted to separate air to make nitrogen, oxygen and argon, and to an extent, krypton and xenon. Such plants would be cryogenic distillation outfits for large capacities while adsorption and membranes are also in use for smaller units. There is also an emerging interest in CO2 removal from flue gases caused by the focus on CO2’s role in global warming.
The term gas treating is normally used to cover CO2 removal, H2S removal, water removal, hydrocarbon dew pointing and gas sweetening. Gas sweetening is a generic term for sulfur removal. Sometimes the term ‘gas conditioning’ is used instead of gas treating.
The key question is ‘why treat gas?’
This is a multi-faceted issue. The gas is produced from a well at a location where there is usually only a negligible market for it. Transport of the gas to the market is the first challenge. Traditionally this has been achieved by pipelines.
A lot could be said about pipelines, but here it will suffice to say that these are constructed in some steel material, and the properties of these materials are such that the presence of certain gases must be kept low to ensure the integrity of the pipeline. Hydrogen sulfide is a key component as it may cause stress corrosion cracking.
Pipeline specifications may vary, but its content is commonly kept below 3–5 ppm. In the US the number 0.25 grain per 100 SCF is often used, but there is no standard in this matter. It must be remembered that the flow velocity in gas pipelines will be too high to allow integrity for a protective sulfide film on
the inner pipe wall. For flow assurance reasons the dew point of the gas must be engineered before entering the pipeline. If the temperature is reduced, both water and hydrocarbons may condense. Water could form hydrate crystals with methane and these could block the flow of gas. Such hydrates are hard and time consuming to get rid of.
Clearly no pipeline operator would want this to happen. Liquid water could also cause corrosion when acidified by CO2 that is likely to be present in the gas. This is also undesirable. Finally hydrocarbon condensate could amass to quantities that would cause flow problems if left unchecked.
It must be remembered that pipelines follow landscapes where its elevation goes up and down repeatedly. CO2 is usually also kept below a certain limit, say 1–4%, depending on the local situation.
There is also another reason other than the pipeline considerations to treat gas. Downstream of the transport system, that could be complex, there is a multitude of customers that will use the gas. Their equipment will have been made with certain gas specifications in mind. Here, the gas heating value will an issue, theWobbe number is often specified and there will be limits on H2S and CO2. Corrosion issues apart, H2S would end up as SO2 in the flue gas and this would be an environmental problem.
There is limited attention paid to processing gases in a typical chemical engineering curriculum. This is, however, a huge field where many chemical engineers find employment. In general terms, there are four principally different main methods that may be used to separate gases. They include (in alphabetical order):
• Cryogenics: liquefaction and distillation
• Membrane permeation.
Ab- and adsorption are often mixed up in write-ups, probably because their spelling is so similar. Process-wise there is a huge difference though. Adsorption is essentially a surface phenomenon while absorption involves something being dissolved.
Cryogenics involves gases being cooled until they condense after which they may be separated by distillation. Some such processes could also be argued to lean towards absorption and/or desorption. A nitrogen wash unit, sometimes used for synthesis gas treatment, is an interesting case with respect to that kind of discussion.
Membrane technology used for gas separation is in general based on so-called dense membranes that separate gases based on different permeation rates. Small volume niche
products within inorganic membranes may be different, but a discussion of this is beyond
the present scope.
Absorption is a much used process for separating gases, removing undesired gas components or to prevent pollution from stacks. The process is, by its nature, run at supercritical temperature with respect to the main gas component(s). There is no boiling like that seen in distillation columns. The mass transfer process is generally rate controlled. All components
are in principle undergoing mass transfer between gas and liquid, but all does not need to be accounted for and/or may be neglected. Mass transfer rates and mass transfer coefficients may differ in different directions for different components.
If a lot of gas needs to be absorbed, large absorbent flows will be needed. This represents an operational cost that, in the end, may be a show stopper for using absorption.
It is a very interesting process, and is in many ways the main focus for this treatise. A separate sub-chapter is dedicated to a preliminary discussion of absorption into alkanolamines in view of these absorbents’ commercial importance.
Practical adsorption processes use a granulated material with affinity for the component, or components that are desired to be removed. This material is referred to as the adsorbent,
while the material adsorbed is referred to as the adsorbate. There are four categories of adsorbents commonly used:
- Molecular sieve zeolites
- Activated alumina
- Silica gel
- Activated carbon.
There is also a carbon molecular sieve that is used for making moderate quantities of nitrogen from air, but we shall leave that aside. Also liquids may be treated by adsorption. In gas treating with absorbents, there is usually an adsorption treatment of this absorbent.
Regeneration of adsorbent may be done by both pressure swing and thermal swing, or a combination. Pressure swing alone is a commercial process that is applied to air separation and at least to hydrogen recovery from streams of synthesis gas. In gas treating contaminant removal by a combination process involving both temperature and pressure variation is mainly used. It could be used to remove water from the gas, and it is used as pretreatment
upstream of liquefied natural gas (LNG) trains to ensure sufficiently low dew points.
Pressure swing implies that the pressure is changed, and temperature swing implies that the temperature is changed.
Adsorption processes are semi-continuous. By this we mean that they continuously treat the gas without a buffer volume, the discontinuity comes from the need to switch between two or more parallel units. The unit not adsorbing is being regenerated offline at a lower pressure and increased temperature with a heat carrying dilution gas flowing through.
This dilution gas may need to be a part of the product gas that most likely will need to be recirculated. In big units more than two parallel columns are often used, sometimes in intricate process stages to emulate some counter-current action while regenerating.
Isotherms for commercial adsorbents are hard to come by. There used to be a couple of companies that handed out leaflets with such content, but such information is certainly not offered on their web sites. In this context it has to be kept in mind that these products are forever being developed such that isotherms may be improved. It is, however, nice to be able to make the odd order of magnitude estimate. This is by no means the result of a thorough
review, but the isotherms used have been published by a number of people and such publications are summarized in Table 2.1 to provide a quick reference as a starting point.
When both CO2 and water are adsorbed, it should be clear from Table 2.1 that water is significantly more strongly adsorbed and will push CO2 away as they compete for adsorption sites. In pretreatment of air for air separation plants this means that there will be a CO2 front moving through the adsorption bed in the direction of flow with a water front pushing from behind. A practical aspect of this is that there is no real need to check for water breakthrough.
It is actually easier to handle a CO2 detector and a water break-through would, in any case, be worse for a cryogenic plant.
On the practical side molecular sieve zeolites could catch 10 g water per 100 g of zeolite in a practical cycle. (Suggested as a quick first approach by a sales engineer a long time ago.
The capacity for CO2 is less. This has implications for adsorption column design. When the gas quantity to be treated is large and the contaminant concentration is significant, the amount of adsorbent needed could be become very large.
Adsorption is mostly used for trace quantity removal. Specific applications will be discussed as they arise. They could include water removal from gas, and a big application is combined water and CO2 removal from air feed to ASUs (ASU=Air Separation Unit).
There is a lot of research going on in the hope of finding a solution that may be used for CO2 removal from flue gas. Recent adsorbents have been reviewed by Hedin and co-workers (2013). They point out that rapid cycling is necessary and believe that some form of structured adsorbent is necessary for success. Treatment of flue gas by vacuum-pressure swing adsorption (VPSA) has been studied (Xiao et al., 2008).
They tested three-bed designs using up to 12 steps in the adsorption-desorption cycle to improve CO2 recovery and purity. Recoveries reported were in the range 70–82%, and purities 82–96%. In the air separation industry VSA is used when oxygen purity does not need to be high, typically 90% although higher can be provided. Argon follows oxygen and is one reason why the purity is that low for reasons of economics.
1. Gas Treating – Absorption Theory and Practice – DAG A. EIMER
2. Fundamentals of Natural Gas, Arthur J. Kidnay & William R. Parrish