Raw natural gas from the well consists of methane as well as many other smaller fractions of heavier hydrocarbons, and various other components. The gas has to be separated into marketable fractions and treated to trade specifications and to protect equipment from contaminants.
Many upstream facilities include the gathering system in the processing plant. However, for distributed gas production systems with many (often small) producers, there is little processing at each location and gas production from thousands of wells over an area instead feed into a distributed gathering system. This system in general is composed of:
- Flowlines: A line connecting the wellpad with a field gathering station (FGS), in general equipped with a fixed or mobile type pig launcher.
- FGS is a system allowing gathering of several flowlines and permits transmission of the combined stream to the central processing facility (CPF) and measures the oil/water/gas ratio. Each FGS is composed of:
- Pig receiver (fixed/mobile)
- Production header where all flowlines are connected
- Test header where a single flow line is routed for analysis purposes (GOR Gas to oil ratio, water cut)
- Test system (mainly test separator or multiphase flow meter)
- Pig trap launcher
- Trunk line – pipeline connecting the FGS with the CPF. Equipped with a pig receiver at the end.
When gas is exported, many gas trains include additional equipment for further gas processing to remove unwanted components such as hydrogen sulfide and carbon dioxide. These gases are called acids and sweetening/acid removal is the process of removing them.
Natural gas sweetening methods include absorption processes, cryogenic processes, adsorption processes (PSA, TSA and iron sponge) and membranes. Often hybrid combinations are used, such as cryogenic and membranes.
Gas treatment may also include calibration. If the delivery specification is for a specific calorific value (BTU per scf or MJ per scm), gas with higher values can be adjusted by adding an inert gas, such as nitrogen. This is often done at a common point such as a pipeline gathering system or a pipeline onshore terminal.
Raw natural gas from the well consists of methane as well, as many other smaller fractions of heavier hydrocarbons and various other components.
at 101 kPa
at 20 °C approx.
Tcri t−82.6 °C
@ 4,6 MPa
Higher order HC
Data source: Wikipedia, Air Liquide Gas Encyclopedia
Natural gas is characterized in several ways dependent on the content of these components:
- Wet gas is raw gas with a methane content of less than 85%.
- Dry gas is raw or treated natural gas that contains less than 15 liters of condensate per 1,000 SM3. (0.1 gallon per 1000 scf).
- Sour gas is raw gas with a content of more than 5.7 mg hydrogen sulfide (H2S) per scm (0.25 grains per 100 scf); this is about 4 ppm.
- Acid gas has a high content of acidic gases such as carbon dioxide (CO2) or H2St. Pipeline natural gas specification is typically less than 2% CO2. Acid gas fields with up to 90% CO2 exist, but the normal range for sour raw gas is 20-40%.
- Condensates are a mixture of hydrocarbons and other components in the above table. These are normally gaseous from the well but condense out as liquid during the production process (see previous chapter). This is a refinery and petrochemical feedstock.
Raw gas is processed into various products or fractions:
- Natural gas in its marketable form has been processed for a specific composition of hydrocarbons, sour and acid components, etc., and energy content. Content is typically 90% methane, with 10% other light alkenes.
- Natural gas liquids (NGL) is a processed purified product consisting of ethane, propane, butane or some higher alkenes separately, or in a blend. It is primarily a raw material for petrochemical industry and is often processed from the condensate.
- Liquefied petroleum gas (LPG) refers to propane or butane or a mixture of these that has been compressed to liquid at room temperature (200 to 900 kPa depending on composition). LPG is filled in bottles for consumer domestic use as fuel, and is also used as aerosol propellant (in spray cans) and refrigerant (e.g., in air conditioners). Energy to volume ratio is 74% of gasoline.
- Liquefied natural gas (LNG) is natural gas that is refrigerated and liquefied at below -162 °C, for storage and transport. It is stored at close to atmospheric pressure, typically less than 125 kPa. As a liquid, LNG takes up 1/600 of the volume of the gas at room temperature. Energy to volume ratio is 66% of gasoline. After transport and storage it is reheated/vaporized and compressed for pipeline transport.
- Compressed natural gas (CNG) is natural gas that is compressed at 2-2,2 MPa to less than 1% of volume at atmospheric pressure. Unlike higher alkenes, methane cannot be kept liquid by high pressure at normal ambient temperatures because of a low critical temperature. CNG is used as a less costly alternative to LNG for lower capacity and medium distance transport. Methane for vehicle fuel is also stored as CNG. Energy to volume ratio is typically 25% of gasoline.
Raw natural gas must be processed to meet the trading specifications of pipeline and gas distribution companies. As part of the purification other components such as NGL is produced, and pollutants extracted.
The diagram shows an overview of a typical gas plant. Marketable products are listed in blue and the production process is shown in grey as it is not considered part of the gas plant.
Acid gases such as carbon dioxide and hydrogen sulfide form acids when reacting with water, and must be removed to prevent corrosive damage to equipment and pipelines. Hydrogen sulfide is also toxic and total sulfur content is normally regulated.
The main removal process can be based on several principles:
Absorption allows acidic gases to be dissolved in a solvent, to be released by regeneration in a later stage. Amine absorption (as shown on the right) is the most common process for acid gas removal. Monoethanolamine (MEA) dominates for CO2 removal. Solutions with inorganic solvents based on ammonia are under development. Ill: Wikipedia
A typical amine gas treating process (as shown in the flow diagram) consists of an absorber unit, a regenerator unit and accessory equipment. In the absorber, a "lean" amine solution absorbs H2S and CO2 from the upflowing sour gas to produce a sweetened gas stream as a product. The "rich" amine solution contains the absorbed acid gases and is routed into the regenerator (a stripper with a reboiler). The stripped overhead gas from the regenerator is concentrated H2S and CO2.
Adsorption relies on the molecules to bind to the surface of certain solids. After a certain time the material must be regenerated to release the gas. Principles used include pressure swing adsorption (PSA), temperature swing adsorption (TSA) and electric swing adsorption (ESA).
Cryogenic removal uses a turbo expander: A gas turbine is driven by the expanding gas which then cools to below the dew point for the gas to be removed.
The inlet gas to the compressor is precooled by the acid gas removed. Cryogenic removal is most often used when the content of carbon dioxide is high, typically around 50%.
Membrane based removal is based on certain materials that allow the acid gases, but not the hydrocarbons, to diffuse through the membrane. This procedure can be performed alone or in combination with absorption liquid.
Sulfur Unit. The H2S-rich stripped gas stream is then fed to a Claus process – a multistage process with two main sections: A thermal section fires H2S with air or oxygen to produce SO2 and elemental sulfur, which is released when cooled. A catalytic section allows more H2S to react with SO2 with alumina or titanium dioxide (TiO2) to produce water and elemental sulfur (the Claus reaction: 2H2S + SO2 → 3S + 2H2O). The Claus process can recover 95-97% of the sulfur in the feed gases.
A tail gas treatment unit serves to reduce the sulfur content to below 250 ppm, corresponding to a total sulfur recovery of 99.9%. More complex solutions can reduce total sulfur down to 10 ppm. Some important processes include SCOT (Shell Claus offgas treatment) which removes SO2 by combustion with hydrogen over catalysts to produce H2S and water. H2S is recycled to the Claus unit. Other solutions are the Beavon sulfur removal process (BSR), based on amine solvent and catalysts.
Dehydration is either glycol-based scrubbers as described in chapter 4.3.2 or based on pressure swing adsorption (PSA). Newer processes also use membranes.
Mercury removal is generally based on molecular sieves. A molecular sieve is a substance containing a material with tiny pores to achieve a large surface area, such as activated carbon. The surface of the material allows certain molecules to bind by surface tension. The molecules can later be extracted and the sieve material regenerated by heating, pressure and/or purging with a carrier gas.
A molecular sieve is commonly cyclic with one active unit and one (or more) units in regeneration.
Excessive nitrogen is removed by cryogenic distillation and higher concentrations are removed by absorption with lean oil or another special solvent if a smaller fraction is detected. (See acid gas removal for both principles). Cryogenic removal also permits production of helium, if present, as a valuable byproduct.
Remaining NGLs are recovered from the gas stream in most modern plants by a cryogenic turbo expander-based process followed by a fractionating process. This process leads the cooled NGLs though distillation columns called de-ethanizer, de-propanizer and de-butanizer, to extract ethane, propane and butane respectively and leave a residual stream of pentane and higher hydrocarbons.
The final step is to remove mercaptans (smelly organic gases, e.g., CH3SH) if present, in a sweetening process based on molecular sieves adsorption or catalytic oxidization such as Merox mercaptan oxidization or Sulfrex, where the main difference is the type of catalyst.
The exact sales gas specification is specified by pipeline operators and distributors. Typical standard sales gas requirements use the following parameters:
Volume is measured in standard cubic meters (scm) defined as 1 m3 at 0 ºCand 101.325 kPa or standard cubic feet (scf) as 1 ft3 at 60 °F (16 °C) and 14.73 PSIA.
Calorific value specifies the total amount of energy per unit generated during combustion of the gas. The value is used to calculate the amount of energy delivered. Several values are listed:
- Gross calorific value or gross heat of combustion is the heat released when a specific quantity of fuel in mixture with air is ignited and the end products have returned to the initial temperature, normally 25 ºC. EU specifications are typically 38.8 MJ (10.8 kWh) ±5% per scm. In the US 1030 BTU ±5% per scf.
- Net calorific value or net heat of combustion is the net heat generated when the water vapor in the gas does not condense (water forms during combustion) and can be 10% lower.
Wobbe index measures the heating effect that a burner is exposed to during combustion. A higher value means a greater thermal load on the burner. Different gases with the same Wobbe index will impose the same load on the burner. An excessively high value is a safety hazard, as it can lead to burner overheating and to excess production of carbon monoxide during combustion.
Calorific value and Wobbe index can be adjusted by blending gas from different sources as well as by addition or removal of nitrogen (N2).
Methane number is a value similar to octane value for gasoline, and is important when the gas is used for internal combustion engines (as CNG).
Hydrogen sulfide and overall sulfur content: Both hydrogen sulfide (H2S) and total sulfur must be reduced. H2S is toxic as well as corrosive for the pipeline, as it forms sulfuric acid (H2SO4) and should be kept as low as possible. Typical maximum values are 5 mg per scm of H2S and total sulfur at 10 mg per scm.
Mercury should be kept below 0.001 ppb (parts-per-billion) which is its detectable limit. The goal is to limit emissions and to prevent damage to equipment and pipelines by mercury amalgamation, which makes aluminum and other metals brittle.
Dew point is a temperature below which some of the hydrocarbons in the gas can condense at pipeline pressure, forming liquid slugs that can damage the pipeline. The gas must also be clear of all water vapor to prevent the formation of methane hydrates within the gas processing plant or within the sales gas transmission pipeline.
Particles and other substances must be free of particulate solids and all liquids to prevent erosion, corrosion or other damage to the pipeline and satisfy limits on carbon dioxide, nitrogen, mercaptans, etc.
Additives: When the natural gas is intended for domestic use, tetrahydrothiophene (THT) is added so that the otherwise odorless natural gas can be detected in the event of a gas leak. The sulfurous-smelling substance added is equal to a sulfur content of 4-7 mg per scm.
Pipeline installations consist of driving compressors and pumps, valve stations, pig receive/launch facilities, where the pig is used for cleaning or inspecting the pipeline. A pipeline SCADA system and pipeline management system is required to control and operate the pipeline.
The pipeline terminal includes termination systems for the pipeline. A pig launcher and receiver is a minimum requirement, allowing insertion of a pipeline pigging device used to clean or inspect the pipeline on the inside. Essentially, it is a large chamber that can be pressurized and purged to insert and remove the pig or scraper without depressurizing the pipeline.
The pig is often driven by pipeline flow, either directly or through a pump or turbine arrangement that also drives wheels that roll against the walls and mechanisms to rotate brushes and other scraping mechanisms. Intelligent pigs also contain instrumentation for remote control and cameras, etc.
One or more compressor stations are needed to keep the required gas flow in the pipeline. Internal friction will cause a pressure drop along the pipeline that increases with flow. Thus, the starting pressure must be high enough to maintain design capacity flow up to the final terminal. If this is not practically possible, additional compressor stations are needed along the total length. Typical starting pressure is about 150-250 bar (15-25 MPa). The final pressure can be as low as 50 bar (5 MPa) at the pipeline terminal end.
The compressors are driven in the same way as explained in the compressor chapter under production (see Chapter 4.3.3).
As an example, about 150 MW in compressor power is required to transport 70 mill scm/day though the 1166 km Langeled pipeline with a starting pressure of 250 Ormen Lange Nyhamna. The initial section is 42 in, which increases to 44 in at Sleipner, a little more than halfway (627 km from Nyhamna, 540 km from Easington, UK), where the intermediate pressure is 155 bar maximum.
The 1200 km Northstream pipeline from Russia (Portovaya, Vyborg) to Germany (by Hannover), has two parallel pipes of almost the same dimensions, pressure and compression power each as Langeled.
Block valve stations (or vine valve stations) are required to allow closure of off flow at regular intervals to limit accidental release of gas in case of pipeline rupture. This will be determined by the maximum permissible released gas volume. A valve station is normally operated by remote telemetry and in addition to the closing valve, at least contains pressure, temperature and some level of flow measurement. These will feed back into the pipeline modeling system. Even if methane (CH4) is lighter than air and will rapidly rise, other gases such as propane (C3H8) are present in up to 8% by volume, and are heavier than air. They will sink, and rapidly create a large volume of combustible gas mixture.
In cases where multiple producers and pipelines feed into a larger pipeline grid, the pressure balancing to maintain required flow and observe contractual volume allocations back to individual production sites presents additional challenges for the pipeline balancing system.
Liquid crude or product pipelines are handled much the same way, with pump stations and block valve stations along the pipeline. However, when the pipeline goes up or down over hills and mountains, they behave differently. Liquid, due to the much higher specific gravity, will experience much higher pressure drops uphill, and increases downhill, than gas. This often requires additional pumping capacity uphill and corresponding pressure-reducing turbines (or braking stations) downhill. In case electrical power is used, the braking power from the turbine can be fed back into the grid.
Block valve stations (or line valve stations) are used with many of the same functions as gas pipelines. It is important to limit accidental spills in case of rupture of the pipeline, and placement will be determined both by maximum leak volume as well as city, river or valley crossings and wherever it is particularly important to prevent spills.
Pipeline management is used to maintain operation and integrity of the pipeline system. It is usually handled by a SCADA system, which interfaces with remote locations and collects data and controls valves and process setpoints. The pipeline management system may include functions such as:
Supervisory control to oversee the operation of the entire pipeline system.
Demand Forecasting to model demand for the coming days across the system based on parameters like contractual volumes, forecasts from consumers and producers and meteorology to determine necessary supply rates and corresponding pressures at various points in the network. Since the transport delay can be considerable during the off-peak season, delays can be compensated by pre-charging the line in anticipation of increases, or allowing pressure to be bled off if a reduction is expected.
Pipeline modeling models the entire pipeline system to account for pressure, temperature and flow at major checkpoints. Based on this model the management system can perform:
- Pressure balancing to make certain that pressure setpoints are correct to meet demand forecasts and avoid potential overload conditions.
- Production allocation, which ensures that producers are able to deliver their contractual volumes into the network.
- Leak detection, which compares actual measured data against dynamic data predicted by the model. A discrepancy indicates a leak (or a failing measurement). Simple liquid systems only calculate basic mass balance (in-out), while an advanced modeling system can give more precise data on size and position of the leak within a certain response time.
- Pig or scraper tracking is used to track the position of the pig within the pipeline, both from pig detection instruments and the pressure drop caused by the pig in the pipeline.
In case of liquid pipelines transporting batches of different products, a batch transfer system is needed. Based on information on when each product is injected into the pipeline, and gravity measurement at the receiving end, it is possible to sequentially transfer different products, such as gasoline and diesel in the same pipeline. Depending on product characteristics, there will be an interface section between the two products that widens as the product moves along the line. This “off spec” product must be discarded at the receiving end to avoid product degradation. It is often disposed by mixing with larger volumes of low grade fuel products. This system is often used with countrywide refined product distribution to terminals.
Safety systems are used, as for other process plants, to ensure that the systems shut down in case of malfunctions and out-of-bounds conditions. Of particular importance is the high integrity pressure protection system (HIPPS). This is a highly reliable system that is needed to maintain protection against overpressures, and manage shutdowns. The strategy used by normal safety system to isolate and depressure, or simply fail to safe condition, often cannot be applied to pipelines due to the large volumes of product in the pipelines. Simply opening or closing a valve would potentially cause overpressure conditions or overload safety devices such as flares HIPPS monitors and executes these responses in a specific sequence.
LNG is a gas transport product. The gas, which is primarily methane (CH4), is converted to liquid form for ease of storage or transport, as its volume is about 1/600th the volume of natural gas in the gaseous state. It is produced close to the production facilities in an LNG liquefaction plant, stored, transported in cryogenic tanks on an LNG carrier, and delivered to an LNG regasification terminal for storage and delivery to a pipeline system.
LNG carriers are used when the transport distance does not justify the cost of a pipeline. The main drawback is the cost of the liquefaction, calculated as how much of the total energy content of the gas is used for liquefaction. About 6% of energy content is used to produce LNG in a large modern plant, due to overall thermal efficiency. More than 10% could be consumed with smaller, gas turbine-driven trains. This compares to losses of about 0.9% per 1,000 km of transport distance for large pipeline systems.
The LNG feedstock comes from a gas plant as outlined above. It must satisfy sales gas specifications. Ethane, propane and butane all have freezing points of less than -180 °C and can be part of the LNG, but the concentration of methane is generally above 90%. Some NGLs are also needed as refrigerant for the cryogenic process.
LNG processes are generally patented by large engineering, oil and gas companies, but are generally based on a one- two- or three-stage cooling process with pure or mixed refrigerants. The three main process types of LNG process with some examples of process licensors are:
- Cascade cycle:
- Separate refrigerant cycles with propane, ethylene and methane (ConocoPhillips)
- Mixed refrigerant cycle:
- Single mixed refrigerant (SMR) (PRICO)
- Single mixed refrigerant (LIMUM®) (Linde)
- Propane pre-cooled mixed refrigerant: C3MR (sometimes referred to as APCI: Air Products & Chemicals, Inc.)
- Shell dual-mixed process (DMR) (Shell)
- Dual mixed refrigerant (Liquefin Axens)
- Mixed fluid cascade process (MFCP) (Statoil/Linde)
- Expander cycle
- Kryopak EXP® process
Each process has different characteristics in scalability, investment cost and energy efficiency. For smaller installations, e.g., to handle stranded gas or isolated small gas fields, a single cycle process is preferable due to its low CAPEX (and possibly lower weight for floating LNG), even if energy efficiency is significantly lower than the best cascade or DMR processes, which cost more but also allow the largest trains typically, 7.8 million tons per annum and lowest energy consumed per energy unit LNG produced. Most processes use a mixed refrigerant (MR) design. The reason is that the gas has a heat load to temperature (Q/T) curve that, if closely matched by the refrigerant, will improve stability, throughput and efficiency (see the figure below). The curve tends to show three distinct regions, matching the precooling, liquefaction and sub-coiling stages. The refrigerant gas composition will vary based on the individual design, as will the power requirement of each stage, and is often a patented, location-specific combination of one or two main components and several smaller, together with careful selection of the compressed pressure and expanded pressure of the refrigerant, to match the LNG gas stream.
Typical LNG train power use is about 28 MW per million tons of LNG per annum (mtpa), corresponding to typically 200 MW for a large trains of 7.2 mtpa, or 65 MW per stage for three cycles. In addition, other consumers in gas treatment and pre-compression add to total power consumption and bring it to some 35-40 MW per mtpa, and over 50 for small LNG facilities well under 1 mtpa capacity.
Some examples are given here. (Please note that these process flow diagrams are simplified to illustrate the principle and do not give a complete design.) All designs are shown with heat exchangers to the sea for comparison. This is generally needed for high capacity, but for smaller plants air fin heat exchangers are normally used.
A triple cycle mixed refrigerant cascade claims to have the highest energy efficiency. It is represented here by the Linde design, co-developed with Statoil.
The actual design varies considerably with the different processes. The most critical component is the heat exchanger, also called the cold box, which is designed for optimum cooling efficiency. Designs may use separate cold boxes, or two or three cycles may combine into one complex common heat exchanger. This particular deign uses the patented Linde coil wound heat exchanger, also called the “rocket design,” due to its exterior resemblance to a classic launch vehicle.
For each train, the cooling medium is first passed through its cooling compressor. Since pressure times volume over temperature (PV/T) remains constant, it results in a significant temperature rise which has to be dissipated, typically in a seawater heat exchanger as shown in the figure above (indicated by the blue wavy line). It then goes though one or more heat exchangers/cold boxes before it expands, either though a valve or a turbo-expander, causing the temperature to drop significantly. It is then returned to cool its cold box before going on to the compressor.
The pre-cooling stage cools the gas to a temperature of about -30 to -50 ºC in the precooling cold box. The cooling element is generally propane or a mixture of propane and ethane and small quantities of other gases. The precooling cold box also cools the cooling medium for the liquefaction and sub cooling stage.
The liquefaction process takes the gas down from -30 ºC to about -100-125 ºC, typically based on a mixture of methane and ethane and other gases. It cools the LNG stream as well as the refrigerant for the final stage.
Sub-cooling serves to bring the gas to final stable LNG state at around 162 ºC. The refrigerant is usually methane and/or nitrogen.
The ConocoPhillips optimized cascade process was developed around 1970. It has three cycles with a single refrigerant gas (propane, ethylene and methane) in each.
The dual cycle mixed refrigerant (DMR), developed by Shell and others, may look simpler but the overall design will be similar in complexity as multistage compressors are typically needed. It is shown on the left with the C3MR on the right for comparison.
For small and micro LNG, single cycle designs are often preferred. There are literally hundreds of patented solutions, but only a handful of mainline licensors, that have solved the challenge of achieving single cycle
refrigeration. However, this means multiple internal stages in the process flow and the heat exchanger itself. The PRICO SMR is shown on the left and the Linde LIMUM® (on the right).
5.5.2 Storage, transport and regasification
Storage at the terminals and on LNG carriers is done in cryogenic tanks at atmospheric pressure or slightly above, up to 125 kPa. The tanks are insulated, but will not keep LNG cold enough to avoid evaporation. Heat leakage will heat and boil off the LNG. Therefore LNG is stored as a boiling cryogen, which means that the liquid is stored at its boiling point for its torage pressure (atmospheric pressure), i.e., about -162 ºC. As the vapor boils off, heat of vaporization is absorbed from and cools the remaining liquid. The effect is called auto-refrigeration. With efficient insulation, only a relatively small amount of boil-off is necessary to maintain temperature. Boiloff gas from land-based LNG storage tanks is compressed and fed to natural gas pipeline networks. On LNG carriers, the boil-off gas can be used for fuel.
At the receiving terminal, LNG is stored in local cryogenic tanks. It is regasified to ambient temperature on demand, commonly in a sea water heat exchanger, and then injected into the gas pipeline system.
Cove point LNG terminal