1

1-Membrane Separation In Natural Gas Processing

1.1-Introduction;
Membrane technology, as applied to gases, involves the separation of individual components on the basis of the difference in their rates of permeation through a thin membrane barrier. The rate of permeation for each component is determined by the characteristics of the component, the characteristics of the membrane, and the partial pressure differential of the gaseous component across the membrane. Since separation is based on a difference in the rates of permeation rather than on an absolute barrier to one component, the recovered component that flows through the membrane (the permeate) is never 100% pure. Also, since a finite partial pressure differential is required as the driving force, some portion of the permeating component remains in the residue gas, and 100% recovery is not possible. As these generalizations would suggest, the process is particularly suitable for bulk removal operations rather than for the removal of trace impurities from gas streams. . The residue gas product normally leaves the unit at a pressure close to that of the feed, while the permeate product, which must pass through the membrane, leaves at a much reduced pressure. The principal product may be either permeate (e.g., the production of hydrogen from dilute gas streams) or the residue gas (e.g., the purification of natural gas by the removal of excess carbon dioxide from high pressure feed), and the process may be considered either separation or purification.

1.2-Membrane Fundamentals:
1.2.1-Membranes:
Membranes are thin semipermeable barriers that selectively separate some compounds from others. This definition is necessarily broad because of the large variety of membrane materials separating an equally vast number of compounds in all phases.

The key requirements for a membrane to be used in an economical gas purification or separation process are: 1) High permeability for the component to be removed 2) High selectivity for the component to be removed in relation to other components

2 3) High membrane stability in the presence of all gas components which will come into con4) Uniformity--freedom from pinholes or other defect tact with the membrane 5) Low effective thickness of the active portion of the membrane to ensure a high permeation 6) Physical strength to withstand the required operating conditions
Applications include:

Ceramic membranes for gas purification in the semiconductor industry Palladium-based metallic membranes for hydrogen extraction Silicon rubber membranes for organic vapor recovery from air Polyvinyl alcohol-based membranes for ethanol dehydration

1.2.2-Principles Governing Membrane Technology;

The membranes used for natural gas removal do not operate as filters, where small molecules are separated from larger ones through a medium with pores. Instead, they operate on the principle of solution-diffusion through a nonporous membrane. Gas first dissolves into the membrane and then diffuses through it. Because the membrane does not have pores, it does not separate on the basis of molecular size. Rather, it separates based on how well different compounds dissolve into the membrane and then diffuse through it. Because carbon dioxide, hydrogen, helium, hydrogen sulfide, and water vapor, for example, permeate quickly, they are called fast gases. Carbon monoxide, nitrogen, methane, ethane and other hydrocarbons permeate less quickly and so are called slow gases. The membranes allow selective removal of fast gases from slow gases. For example, as CO2 is removed from a natural gas stream, water and H2S are removed at the same time; but methane, ethane, and higher hydrocarbons are removed at a much lower rate. Ficks law, shown below, is widely used to approximate the solution-diffusion process:

J is the membrane flux of Gas, that is, the molar flow of gas through the membrane per unit area of membrane.

3 k is the solubility of gas in the membrane. D is the diffusion coefficient of gas through the membrane. p is the partial pressure difference of gas between the feed (high pressure) and permeate (low pressure) side of the membrane. l is the membrane thickness. To simplify matters further, the solubility and diffusion coefficients are usually combined into a new variable called permeability (P). Ficks law can therefore be split into two portions: a membrane dependent portion (P/ ) and a process-dependent portion (p). To achieve a high flux, the correct membrane material and the correct processing conditions are needed. P/ is not a constant; it is sensitive to a variety of operating conditions such as temperature and pressure. The Ficks law equation can be equally written for methane or any other component in the stream. This set of equations leads to the definition of a second important variable called selectivity (). Selectivity is the ratio of the permeabilities of gas to other components in the stream and is a measure of how much better the membrane permeates gas compared to the compound in question. For example, most gas membranes provide a gas-to-methane selectivity anywhere between 5 and 30, meaning that gas permeates the membrane 5 to 30 times faster than methane. Both permeability and selectivity are important considerations when selecting a membrane. The higher the permeability, the less membrane area is required for a given separation and therefore the lower the system cost. The higher the selectivity, the lower the losses of hydrocarbons as gas is removed and therefore the higher the volume of salable product.

1.2.3-Membrane Module Configuration;

The most popular module configurations are the given as below Spiral-wound Configuration Hollow Fiber Configuration

i-Spiral-wound Configuration;
In the spiral-wound configuration an envelope is formed with two membrane sheets separated by a porous support material. Typically the module consists of several such envelopes. The material between the membranes (permeate channel spacer) supports them against the operating pressure and defines the permeate flow channel. The envelope is sealed on three sides. The fourth side is sealed to a perforated permeate collection tube, and the envelope is wrapped around the collection tube with a net-like spacer sheet that has two functions:

4 1) It keeps adjacent membranes apart to form a feed channel. 2) It promotes turbulence of the feed gas mixture as it passes through the module, thus reducing concentration polarization.

During operation, the feed gas mixture enters one face of the module, travels axially along the feed channel spacer and membrane surface, and exits the other face as a residue or retentate. The more permeable gases pass through the membranes and travel in a spiral path inward within the envelope through the permeate channel spacer until they reach the perforated collection tube and finally exit as permeate (Figure). The feed channel spacer is a key feature of the spiral-wound module and, as shown by Da Costa et al. (1991, 1994), its design significantly affects module performance. Typically the modules have about 1 ,OOO square feet of surface per cubic foot of volume and are 4-12 inches in diameter by 3642 inches long. Up to six modules may be housed in a single pressure vessel shell.

ii-Hollow Fiber Configuration;

The hollow fiber configuration consists of thousands of hollow fibers packaged in bundles mounted in a pressure vessel resembling a shell and tube heat exchanger. For high pressure applications the fiber diameter is usually on the order of 100 pm ID and 150-20 pm OD. The bundles are capped on one end and have an open tube sheet on the other end

The bundles typically have 3,000 square feet of membrane surface per cubic foot of module volume and the module dimensions range from 4-12 in. in diameter by 4-20 ft long. The feed gas is introduced on the shell side because hollow fibers are much stronger under compression than expansion. The faster permeating gases migrate into the fiber bore and exit via the open end of the bundle. For low pressure applications the fibers have a diameter greater than 400 pm and the feed gas enters the bore side while the permeate exits via the shell side. This configuration reduces pressure drop on the feed side. Not all membrane materials can be made into a thin selective layer on a porous substrate in a hollow fiber form.

Consequently, spiral-wound membranes, which can be made from a wider range of materials, usually have higher permeation rates. However, this is offset by the much

6 higher packing density of hollow fiber modules, resulting in similar overall productivity per unit module volume for the two configurations. This situation could change if developments in polymer science lead to more effective thin films in a hollow fiber form. Numerous mechanical designs of modules have been developed for both the spiralwound and hollow-fiber concepts. The various designs are aimed at optimizing such features as: membrane area per unit volume, gas flow distribution and pressure drop, seals and fastenings, and assembly technology. With either spiral-wound or hollow-fiber systems, large commercial installations normally require a large number of individual modules. This is evident in figure, which shows a UOP Advanced Membrane System for removing carbon dioxide from natural gas. A third module design-plate and frame-utilizes a stack of disk shaped membranes separated by sheets of porous filter paper and membrane support plates. The flow pattern is very much like that of a plate and frame filter. This design is not competitive for large commercial applications because of the relatively small membrane area per unit volume attainable. However, plate and frame designs are used for producing oxygen enriched air in small medical applications.

1.3-CO2 Removal from Natural Gas;

Many different types of membranes have been developed or are under development for industrial separations , but for CO2 removal, the industry standard is presently cellulose acetate. These membranes are of the solution-diffusion type, in which a thin layer (0.1 to 0.5 m) of cellulose acetate is on top of a thicker layer of a porous support material. Permeable compounds dissolve into the membrane, diffuse across it, and then travel through the inactive support material. The membranes are thin to maximize mass transfer and, thus, minimize surface area and cost, so the support layer is necessary to provide the needed mechanical strength. Commercial membrane configurations are either hollow fiber elements or flat sheets wrapped into spirally wound elements. Presently, about 80% of gas separation membranes are formed into hollow fiber modules (Baker, 2002), like those shown in Figure. The low-pressure, bore-feed configuration is a countercurrent flow configuration similar to a shell-tube heat exchanger with the gas entering on the tube side. It has the advantage of being more resistant to fouling because the inlet gas flows through the inside of the hollow fibers. However, the mechanical strength of the membrane limits the

7 pressure drop across the membrane. The configuration is only used in low-pressure applications, such as air separation and air dehydration (Baker, 2002). To handle high pressures, the permeate flows into the hollow fiber from the shell side. This feature makes the membrane much more susceptible to plugging, and gas pretreatment is usually required (Baker, 2002). The gas flow is cross current and provides good feed distribution in the module. This configuration is widely used to remove CO2 from natural gas. In the spiral wound element shown in Figure, two membrane sheets are separated by a permeate spacer and glued shut at three ends to form an envelope or leaf. Many of these leaves, separated by feed spacers, are wrapped around the permeate tube, with the open end of the leaves facing the tube. Feed gas travels along the feed spacers, the permeating species diffuse through the membranes and down the permeate spacers into the permeate tube, and the residue gas exits at the end. The gas flow is cross flow in this configuration.

1.3.1-Design Considerations;
Many process parameters can be adjusted to optimize performance depending on the customer and application needs. Optimization is most critical for larger systems where small improvements can bring large rewards. Some typical requirements are:
Low cost High reliability High on-stream time Easy operation High hydrocarbon recovery Low maintenance Low energy consumption Low weight and space requirement

Flow Scheme;
The simplest membrane processing scheme is a one-stage flow scheme (Figure 7). A feed gas is separated into a permeate stream rich in CO2 and a hydrocarbon-rich residual stream. In high CO2 removal applications, a significant amount of hydrocarbons permeate the membrane and are lost. Multistage systems attempt to recover a portion of these hydrocarbons. The twostep design shown in Figure 8 allows only a portion of the first-stage permeate to be lost. The rest is recycled to the feed of the first stage.

The portion of first-stage permeate that is lost is usually taken from the first membrane modules, where feed CO2, hence permeate CO2, is highest and hydrocarbons are lowest. The permeate that is recycled is at low pressure and must be repressurized before it can be combined with the feed gas.

Two-stage designs process the first-stage permeate in a second membrane stage, as shown in Figure 9. The permeate from the second stage, which has typically twice the CO2 content as the first-stage permeate, is vented. The residue is either recycled and combined with the feed gas. A compressor is required to repressurize the first-stage permeate before it is processed in the second stage. Two-stage designs provide higher hydrocarbon recoveries than two-step or one-stage designs but require more compressor power (because more gas must be compressed to be treated)

9 The percentage hydrocarbon recovery is plotted versus percentage CO2 removal for oneand two-stage systems at certain process conditions. The percentage hydrocarbon recovery is defined as the percentage of hydrocarbons recovered to the sales gas versus the hydrocarbons in the feed gas.

The hydrocarbon recovery of a two-stage system is significantly better than that for a single stage system. However, when deciding whether to use a single or multistage approach, the designer must also consider the impact of the recycle compressor. This impact includes the additional hydrocarbons used as fuel, which increases the overall hydrocarbon losses, as well as the significant capital cost of compressors and the difficulty of maintaining them in remote locations. For moderate CO2 removal applications, that is, below approximately 50%, single-stage membrane systems usually provide better economic returns than do multistage systems.

1.3.2-Operating Considerations: i-Flow Rate;

Because membrane systems are modular, an increase in flow rate causes a directly proportional increase in membrane area requirement, for a given separation. Hydrocarbon losses, that is, the flow rate of hydrocarbons lost to vent, also increases proportionally, but the percentage hydrocarbon losses (hydrocarbon losses feed hydrocarbons) stay constant. A maximum acceptable feed gas rate per unit area applies to the membrane, and required membrane area is directly proportional to the flow rate. Membrane units perform well at reduced feed rates, but their performance drops when design flow rates are exceeded.

ii-Operating Temperature;
An increase in feed temperature increases membrane permeability and

10 decreases selectivity. The membrane area requirement is therefore decreased, but hydrocarbon losses and the recycle compressor power for multistage systems are increased, as shown in Figure

Increased operating temperature increases permeability but decreases selectivity. Because membranes are organic polymers, they have a maximum operating temperature that depends upon the polymer used. Exceeding this temperature will degrade membrane material and shorten the useful life of the unit.

iii-Feed Pressure;
An increase in feed pressure decreases both membrane permeability and selectivity. However, the increased pressure creates a greater driving force across the membrane. A net increase in permeation through the membrane results and the membrane area requirement therefore drops. Compressor power increases slightly, and hydrocarbon losses decrease slightly

Because the membrane area requirement is so affected by pressure, while other variables are not, designers attempt to use the maximum operating pressure possible to achieve a cheaper and smaller system. A limiting factor is the maximum pressure limit for the membrane elements and the cost and weight of equipment at a higher-pressure

11 rating.

iv-Permeate Pressure;
The effect of permeate pressure is the opposite of the effect of feed pressure. The lower the permeate pressure, the higher the driving force and therefore the lower the membrane area requirement. Unlike feed pressure, however, permeate pressure has a strong effect on hydrocarbon losses

The pressure difference across the membrane is not the only consideration. Detailed analysis shows that an equally important factor in system design is the pressure ratio across the membrane. This ratio is strongly affected by the permeate pressure. For example, a feed pressure of 90 bar and a permeate pressure of 3 bar produce a pressure ratio of 30. Decreasing the permeate pressure to 1 bar increases the pressure ratio to 90 and has a dramatic effect on system performance. For this reason, membrane design engineers try to achieve the lowest-possible permeate pressure. This need is an important consideration in deciding how to further process the permeate stream. For example, if it must be flared, then flare design must be optimized for low pressure drop. If the permeate gas is to be compressed, for example, to feed it to a second membrane

v-CO2 Removal;
For a given sales-gas CO2 specification, an increase in feed CO2 increases membrane area requirement as well as hydrocarbon losses (more CO2 must permeate, and so more hydrocarbons permeate). This is shown in Figure The membrane area requirement is determined by the percentage of CO2 removal rather than the feed or sales-gas CO2 specifications themselves. For example, a system for reducing a feed CO2 content from 10 to 5% is similar in size to one reducing a feed from 50 to 30% or one reducing a feed from 1 to 0.5% all have a CO2 removal requirement of about 55%. This behavior is different from the way in which traditional

12 CO2 removal technologies operate. For these operations, a reduction in CO2 from 3 to 0.1% does not require a much larger system than that required for a reduction from 3 to 1%. For a membrane system, the large difference in CO2 removal (97 versus 70%) means that the system for 0.1% sales gas is about three times the size of the 1% system.

Traditional solvent- or adsorbent-based CO2 removal technologies have the opposite limitation, that is, their size is driven by the absolute amount of CO2 that must be removed. So a system for CO2 removal from 50 to 30% is substantially larger than one reducing CO2 from 1.0 to 0.5%. For this reason, using membranes for bulk CO2 removal and traditional technologies for meeting low CO2 specifications makes a lot of sense. Depending on the application, either one or both of the technologies could be used.

1.4-Membrane Processing in Pakistan;

The two largest CO2 removal membrane systems in the world are the SeparexTM units installed in Qadirpur and Kadanwari, in Pakistan. Both of these plants specified membranes as the CO2 removal technology to be used, because of their simplicity, ease of use, and high reliability, essential attributes for remotely located plants. These criteria have all been met, and significant lessons have been learned. For example, a major impetus in the development of UOPs advanced pretreatment system came from experiences with the Kadanwari unit.

1.4.1-Kadanwari SeparexTM Membrane System;

When this facility started up in 1995, it was the largest membrane-based natural gas processing plant in the world. It has been in operation for more than three years using UOPs Separex cellulose acetate membranes.
The Kadanwari system is a two-stage unit designed to treat 210 MM SCFD of feed gas at

13
90 bar. The CO2 content is reduced from 12 to less than 3%.

The pretreatment system for this plant was designed for a light gas with minimal C10+ content and a dew point about 50F, as provided in a customer-supplied gas analysis. After start-up, an analysis of the feed stream indicated a significant heavy hydrocarbon content, including C30s, with a dew point above 125F. As a result, UOP proposed enhanced pretreatment. Installation has been delayed because well production has been lower than expected. However, because production is expected to soon increase, the client has the enhanced pretreatment design under consideration.

1.4.2-Qadirpur SeparexTM Membrane System;

The Separex membrane system in Qadirpur, Pakistan, is the largest membranebased natural gas plant in the world. It is designed to process 265 MM SCFD of natural gas at 59 bar. The CO2 content is reduced from 6.5% to less than 2%. The unit was designed to also provide gas dehydration to pipeline specifications. The owner has plans to expand the system to approximately 400 MM SCFD of processing capacity within the next year. The Qadirpur membrane system was designed and constructed in two 50% membrane trains. Each membrane train consists of a conventional pretreatment section and a membrane section. The pretreatment section has filter coalescers, guard vessels, and particle filters. Membrane feed heaters are also included in this design to maintain stable membrane process conditions. Figure shows a view of the Qadirpur system. To give an idea of the size of the system, a person in the upper left corner of the picture is circled. This plant started up in 1995 and has been in operation for almost three years. UOP provided on site assistance from the loading of the membrane elements through start-up and remained until the customer was fully comfortable with the equipment. The plant continues to operate routinely, processing all gas available unless limited by pipeline demand. The Qadirpur system is proof of the ruggedness of UOPs Separex membrane systems and cellulose acetate membranes. The feed gas contains a significant heavy hydrocarbon content as well as polynuclear aromatics, which are known to damage other membranes. In spite of these contaminants, the unit is operating at design capacity.
JJJ

14

1.5-Advantages of Membrane System;

Membrane system has major advantages on the traditional CO2 methods. i-Low Capital Cost; Membrane systems are skid mounted, except for the larger pretreatment vessels, and so the scope, cost, and time taken for site preparation are minimal. Therefore, installation costs are significantly lower than alternative technologies, especially for remote areas. Furthermore, membrane units do not require the additional facilities, such as solvent storage and water treatment, needed by other processes.
ii-Lower Operating Costs:

The only major operating cost for single-stage membrane systems is membrane replacement. This cost is significantly lower than the solvent replacement and energy costs associated with traditional technologies. The improvements in membrane and pretreatment design allow a longer useful membrane life, which further reduces operating costs. The energy costs of multistage systems with large recycle compressors are usually comparable to those for traditional technologies.
iii-Operational Simplicity and High Reliability:

15 Because single-stage membrane systems have no moving parts, they have almost no unscheduled downtime and are extremely simple to operate. They can operate unattended for long periods, provided that external upsets, such as well shutdowns, do not occur. Items in the pretreatment system that could cause downtime, such as filter coalescers, are usually spared so that production can continue while the item is under maintenance.
iv-Good Weight and Space Efficiency: Skid construction can be optimized to the space available, and multiple elements can be inserted into tubes to increase packing density. This space efficiency is especially important for offshore environments, where deck area is at a premium, and is the reason why so many new offshore developments have chosen to use membranes for acid gas removal. v-Environmentally Friendly:

Membrane systems do not involve the periodic removal and handling of spent solvents or adsorbents. Permeate gases can be flared, used as fuel, or reinjected into the well. Items that do need disposal, such as spent membrane elements, can be incinerated.
vi-High Turndown: The modular nature of membrane systems means that low turndown ratios, to 10% of the design capacity or lower, can be achieved. Turnup and turndown increments can be set at whatever level is required during the design phase. vii-Adaptability:

Because membrane area is dictated by the percentage of CO2 removal rather than absolute CO2 removal, small variations in feed CO2 content hardly change the sales-gas CO2 specification. For example, a system designed for 10% down to 3% CO2 removal produces a 3.5% product from a 12% feed gas, and a 5% product from a 15% feed gas. By adjusting process parameters such as operating temperature, the designer can further reduce the sales-gas CO2 content.

16

viii-Design Efficiency:

The membrane and pretreatment systems integrate a number of operations, such as dehydration, CO2 and H2S removal, dew-point control, and mercury removal. Traditional CO2 removal technologies require all of these operations as separate processes and may also require additional dehydration because some technologies saturate the product stream with water.

1.6-Disadvantage of Membrane System;

Economy of scale: Because of their modular nature, they offer little economy of scale Clean feed: Pretreatment of the feed to the membrane to remove particulates and liquids is generally require Gas compression: Because pressure difference is the driving force for membrane separation, considerable recompression may be required for either or both the residue and permeate streams For natural gas systems, the following disadvantages also apply: Generally higher hydrocarbon losses than solvent systems H2S removal: H2S and CO2 permeation rates are roughly the same, so H2S specifications may be difficult to meet Bulk removal: Best for bulk removal of acid gases; membranes alone cannot be used to meet ppmv specifications

1.7-Conclusion;
Membrane systems are a solid and proven addition to the range of technologies for the removal of CO2. With correct pretreatment design, they are extremely reliable, efficient, and ideally suited to installation

17 in remote regions. Continuing enhancements in membranes, membrane systems, and membrane pretreatment makes membranes an even more natural choice in the future, especially for applications requiring higher levels of CO2 removal.

References;
Kohl, A.R., and Neilsen, R.B.: Gas Purification 5th ed. Kidnay, A.J., and Parrish, W.R.: Fundamental of Natural Gas Processing Dortnunt, D., and Dhoshi, K.: Recent Developments in CO2 Removal Membrane Technology Research Paper. Mokhatab, S., Speight, J.G.: Handbook of Natural Gas Transmission & Processing

2-Molecular Sieve In Natural Gas Processing

2.1-Molecular Sieve;
A molecular sieve is a material containing tiny pores of a precise and uniform size that is used as an adsorbent for gases and liquids. Often they consist of aluminosilicate minerals, clays, porous glasses, microporous charcoals, zeolites, active carbons, or synthetic compounds that have open structures through which small molecules, such as nitrogen and water can diffuse. Molecules small enough to pass through the pores are adsorbed while larger molecules are not. It is different from a common filter in that it operates on a molecular level. For instance, a water molecule may not be small enough to pass through while the smaller molecules in the gas pass through. Because of this, they often function as a desiccant. A molecular sieve can adsorb water up to 22% of its own weight

2.1.1-Working of Molecular Sieves;

18 Molecular sieves are activated by heat to drive off the water of crystallization. The crystals have a robust cubic structure, which does not collapse on heating, so that activation results in a geometric network of cavities connected by pores. The pores are of molecular dimensions and cause the sieving action of these materials. The small molecules diffuse through the porous structure of sieve while the larger molecules excluded.

2.1.2-Type of Molecular Sieve;

About 40 different zeolite structures have been discovered; however, only three synthetic types are widely used in commercial applications: 1. Zeolite A 2. Faujasite 3. Pentasil

1-Zeolite A;
The pores of Zeolite A are restricted by 8-membered oxygen rings. The free aperture for this structure is about 3.0 A" for the K+ form (3A), 3.8 A" for the Na+ form (4A), and 4.3 A" for the Ca++ form (5A).

2-Faujasite;
Faujasite is represented by two forms, X and Y, with pores restricted by 12membered oxygen rings. The pores of these materials are relatively large with a free aperture of about 8.1 A". The X and Y zeolites differ from each other only with regard to the SUA1 ratio which controls cation density, and therefore, affects adsorptive properties.

3-Pentasil;
Pentasil-type zeolite has pores restricted by a 10-membered oxygen ring, which gives a free aperture of about 6.0 A". This zeolite was first proposed by Mobil in the early 1970s and named ZSMJ. At about the same time, a pure silica analog with the same framework structure was prepared by Union Carbide and named Silicalite

2.1.3-Basic Type & Properties of Molecular Sieve;

19

2.2-Natural Gas Purification Using Molecular Sieves;

Molecular sieves play an important role in natural gas processing especially in dehydration and separation.

2.2.1-Dehydration of N.G;
Solid desiccants or absorbents are commonly used for dehydrating gases in cryogenic processes. The use of solid adsorbent has been extended to the dehydration of liquid. Solid adsorbents remove water from the hydrocarbon stream and release it to another stream at higher temperatures in a regeneration step. A flow diagram for adsorption is shown in Figure.

In a dry desiccant bed, the adsorbate components are adsorbed at different rates. A short while after the process has begun, series of adsorption zones appear. The distance between successive adsorption zone fronts is indicative of the length of the bed involved in the adsorption of a given component. Behind the zone, all of the entering component

20 has been removed from the gas; ahead of the zone, the concentration of that component is zero. Note the adsorption sequence: C1 and C2 are adsorbed almost instantaneously, followed by the heavier hydrocarbons, and finally by water that constitutes the last zone. Almost all the hydrocarbons are removed after 30-40 min and dehydration begins. Water displaces the hydrocarbons on the adsorbent surface if enough time is allowed. At the start of dehydration cycle, the bed is saturated with methane as the gas flows through the bed. Then ethane replaces methane, and propane is adsorbed next. Finally, water will replace all the hydrocarbons. For good dehydration, the bed should be switched to regeneration just before the water content of outlet gas reaches an unacceptable level. The regeneration of the bed consists of circulating hot dehydrated gas to strip the adsorbed water, then circulating cold gas to cool the bed down.

2.2.2-CO2 Removal from N.G by Molecular Sieves;

Molecular sieves are used in many installations for the simultaneous removal of carbon dioxide and water from gas streams which undergo low-temperature Liquefaction in a subsequent operation. Even relatively low concentrations of carbon dioxide and water can cause the formation of frost, or "rime," on the surfaces of cryogenic heat exchanger fins. One of the earliest COz removal applications of molecular sieves was for purifying the air feed to cryogenic air separation plants. In this service, continuous product gas purities of less than 1 ppm C02 and 1 ppm H20 are readily attained.

Front End Purification System (FEP);

The design of molecular sieve front end purification (FEP) systems for cryogenic air separation plants is described by Kerry (1991). In a typical design, the feed air is compressed to about 6-8 atm and cooled to about 8C (46F) before it enters the bed of molecular sieve. The adsorption step is stopped before breakthrough of C02 occurs. Since propane, ethylene, acetylene, and higher hydrocarbons as well as H20 are more strongly adsorbed than C02, only the inlet portion of the bed will be near saturation with regard to these impurities at the end of the adsorption step so they will be removed from the gas with very high efficiency.

21

Regeneration;
The regeneration of molecular sieve beds in FEP systems is accomplished by flowing heated nitrogen through the bed, in the reverse direction to adsorption, at approximately atmospheric pressure. Regeneration temperatures are modest, 90" to 100C (194" to 212"F), but an amount of nitrogen equal to about 20 to 30% of the inlet air is required. The bed is cooled with low temperature nitrogen before it is ready for the adsorption step.

2.2.3-Sulfur Compound Removal Using Molecular Sieves;

Molecular sieves are very effective for H2S removal and can produce gas containing extremely low levels of this impurity. Since H2S is adsorbed more strongly than COz, the molecular sieves also offer the capability for selectively removing H2S from gas streams that contain both impurities.

Process Description;
The gas feed to the dry bed and liquid absorption units is ratioed in such a manner that the final mixed product meets pipeline C02 specifications. The increased gas volume that results from leaving 3% C02 in the product gas provides a credit for gas sold on a

22 volume basis. Use of the dry-bed adsorption process on a portion of the gas has the further advantage of permitting the liquid absorber system to operate closer to maximum capacity, as a product gas which is slightly off specification with regard to H2S, and water can be tolerated for dilution with the very pure adsorber product.

2.4-Advantages of Molecular Sieve Processing;

Molecular Sieve processing is relatively expensive but they have several advantages. 1. They provide good capacity with gases of low relative humidity 2. They are applicable to gases at elevated temperatures 3. They can be used to adsorb water selectively 4. They can be used to remove other selected impurities together with water 5. . They can be used for adiabatic drying 6. . They provide extremely low dew points 7. They are not damaged by liquid water

2.5-Conclusion;
Molecular Sieves are very efficient adsorbent for removal of water from natural gas. In separation of impurities from natural gas molecular sieves are suitable for weakly adsorbed compounds such as CO2 and H2O but are not suitable for strongly absorbed compound as mercaptanes.

References;
Kohl, A.R., and Neilsen, R.B.: Gas Purification 5th ed.

23