Carbon Capture and Storage
The release of carbon dioxide into the atmosphere is considered to be responsible for global warming, the heating of the global atmosphere through the greenhouse effect. This is propelling a worldwide movement to restrict emissions from coal fired power plants. Control can be achieved using carbon capture and storage technologies. There are three main technologies being explored, post combustion capture, pre-combustion capture – which is a form of coal gasification – and oxyfuel combustion where the fuel is burnt in pure oxygen instead of air. A fourth, chemical looping, is similar to oxyfuel combustion but carried out using solid state oxygen carriers. It is also possible to burn a proportion of biomass in a coal-fired power plant in order to reduce, but not eliminate net carbon dioxide emissions. Once captured the carbon dioxide must be compressed and then transported to a sequestration site where it can be stored in a way that prevents it ever reaching the atmosphere again. Underground storage offers the greatest potential for carbon sequestration.
Keywords
greenhouse effect; carbon capture; carbon dioxide emissions; post combustion capture; pre-combustion capture; oxyfuel combustion; chemical looping; sequestration; enhanced oil recovery; brine aquifer; biomass co-firing; sequestration site monitoring
There are currently no commercial carbon capture technologies that can be applied to large coal-fired power stations, although several configurations are being tested and most of the technology required is well established. However, legislation and incentives to limit carbon emissions are expanding with systems such as the carbon certificate policy and emissions trading system within the European Union where large emitters are required to have carbon certificates for every tonne of carbon they emit.
Faced with growing pressure to control their emissions, utilities and independent power generators are attempting to find strategies other than carbon capture that will reduce the amount of carbon dioxide they produce. Increasing plant efficiency is one strategy and there has been a slow global shift in the last decade or two towards supercritical and ultra-supercritical boilers in coal-fired plants. These produce less carbon dioxide per unit of electricity than a conventional sub-critical boiler.
A more common strategy, particularly in the developed world, has been to shift away from coal altogether and generate power from natural gas. The latter produces significantly less carbon dioxide for each unit of power than the best supercritical coal-fired plant. However, gas is both more expensive and exhibits greater price volatility than coal, making the economic risk associated with investing too heavily in gas-fired plants very high. That, coupled with the easy availability of coal, particularly in countries like China and India, means that coal plants will remain crucial to global power generation at least until the middle of this century, if not longer. Faced with this, the development of new strategies to reduce carbon emissions from coal plants is vital if atmospheric warming is to be limited. That means carbon capture and sequestration.
For carbon capture there are three main approaches available. The first, often called post-combustion capture, involves installing a plant similar to an FGD scrubber to the exhaust of the power plant. Reagents capable of capturing carbon dioxide when deployed from a spray tower are already available and this technology is likely to be one of the simplest and possibly the most economical to deploy. An alternative to this, called pre-combustion capture, involves pre-treating coal to remove the carbon before combustion. This is achieved using a modified version of coal gasification which leaves a fuel gas composed primarily of hydrogen. The gasification process also requires some form of carbon dioxide capture technology and this is likely to be via a scrubber system too, although in this case other solutions may be possible. The third way of tackling carbon dioxide emissions is to sidestep the difficulty of separating carbon dioxide from flue gases, which are usually a mixture of residual oxygen, nitrogen, carbon dioxide, and other trace gases. Instead, oxygen is separated from air first, and then coal is burnt in virtually pure oxygen. The scheme, called oxyfuel combustion, leads to an exhaust gas stream composed primarily of carbon dioxide, mixed with some water vapor and excess oxygen from which it is much easier to isolate the carbon dioxide than when the latter is mixed with nitrogen. In effect, the oxyfuel system replaces carbon dioxide separation with oxygen separation.
If carbon dioxide is captured from coal before or after power generation, the gas must then be stored in a way that prevents it ever returning to the atmosphere. This is called carbon sequestration. The most likely place for carbon sequestration is underground and some pilot scale projects have demonstrated that this is feasible. However, the process will have to be carried out at an extremely large scale so this technology needs to be perfected alongside carbon capture technologies if carbon capture and storage is to become feasible.
There is one last strategy that can be used to reduce the environmental impact of coal combustion and that is to replace some of the coal burnt in the coal plant with a biomass fuel such as wood chips. The biomass, provided it comes from a sustainable source, is essentially carbon neutral so while the combustion of the biomass still releases carbon dioxide, this should be recaptured from the atmosphere when further biomass is grown. The process, called biomass cofiring, cannot eliminate carbon emissions from coal plants and so does not offer a long-term solution – unless the coal plant converts to 100% biomass firing.
Biomass Cofiring
The cofiring of biomass in a coal-fired power plant involves replacing some of the coal with a biomass fuel. The fuel can be burned in any type of coal boiler including pulverized coal-fired power stations and fluidized plants. However, all plants will require some adaptation to make cofiring possible.
Cofiring will reduce the environmental impact of a coal plant by reducing the net amount of carbon dioxide added to the atmosphere. The biomass will generate carbon dioxide when burnt in the same way as coal but subsequent re-growth of biofuel will absorb carbon dioxide from the atmosphere again so that the net contribution is zero. This relies on supplying fuel from a sustainable and continuous biomass production source. Some fuel sources, particularly where the fuel is transported over long distances, provide a questionable environmental advantage. Nevertheless, if managed properly it does provide a potential environmental gain.
One of the main advantages of cofiring is that it allows biomass to be burnt in a power plant that operates at high efficiency. Most dedicated biomass combustion plants are small and have relatively low energy conversion efficiencies. Modern, large coal-fired plants on the other hand offer some of the best combustion efficiencies achievable. There is very little loss in efficiency when part of the coal is replaced with biomass, so burning biomass in this way allows significant gains compared to the use of more conventional biomass combustion plants.
Small amounts of biomass can be mixed with pulverized coal and introduced into the boiler furnace through the coal burners. This appears to be effective for up to around 10% cofiring, although it is normally only used for up to 5% cofiring. Up to this level, the biomass fuel can simply be mixed with the coal before it enters the power plant processing train allowing the two to be processed together in the coal mills and then fed to the plant burners.
When larger proportions of cofiring are planned a separate biomass preparation system is needed to pre-mill the biomass before it is mixed with the milled coal and injected into the boiler. With this type of system it appears possible to use up to 20% biomass cofiring without major modification to the plant burners and this is the scheme adopted in most coal-fired plants in Europe that have experimented with large-scale biomass cofiring. The alternative, when large-scale cofiring is being considered, is to fit separate burners to introduce the biomass fuel into the furnace.
For high cofiring ratios it may be possible to modify the biomass fuel by a process called torrefaction. This involved heating the fuel to around 200–300°C in a reducing atmosphere for around 1 h. The torrefaction alters the properties of the biomass, making it more like coal in terms of handling and preparation. The technique is still in the demonstration stage. Its aim is to allow biomass to be burnt in a coal-fired power plant with only minimal modification. This might allow 100% biomass firing without major plant modification.
There are alternative approaches to cofiring. One is to gasify the biomass fuel first and then burn the gaseous product in a boiler. However, this appears to be more expensive and may not be an economically viable approach, while adding significant complexity compared to the simple cofiring outlined above.
Fuels for cofiring include woods and grasses, both of which can be grown as dedicated biomass fuels. One potential problem with many biomass fuels is that they can cause fouling of boilers that have been designed to burn particular coals. Woods often have similar low-ash, low-alkali, and low-chlorine content as coal fuels and these should not present an increased problem. However, herbaceous fuels, such as grasses, may have higher levels of ash, alkali, and chlorine. These can lead to higher levels of corrosion within boilers.
One potential advantage of cofiring is that it can lead to lower nitrogen oxide production because the biomass fuel contains less fuel nitrogen than the coal it is replacing. While some gains have been seen from biomass cofiring, the relationship between the amount of nitrogen in the fuel and NOx production is not simple. In addition, high levels of alkali metals in biomass can lead to early poisoning of the catalyst in an SCR reactor, leading to the need for early replacement. Sulfur dioxide emissions may also be lowered as biomass fuels contain only low levels of sulfur. However, the effects on particulate emissions can be more complex.
There is one further interest in co-firing. When the technology is combined with carbon capture it can lead to a net reduction in the amount of carbon dioxide in the atmosphere. This is because carbon dioxide is being removed from the carbon neutral part of the combustion fuel, leading to a (long-term) net fall in the amount of carbon dioxide. If widely adopted in the future, this could offer a means of actually reversing the build up of carbon dioxide within the atmosphere.
Post-Combustion Capture of Carbon Dioxide
The post-combustion capture of carbon dioxide in a coal-fired power plant is conceptually the simplest scheme for controlling carbon emissions. The scheme, which involves adding a plant that will capture and remove carbon dioxide from the flue gases that emerge from the plant boiler, before they are released into the atmosphere, is also the easiest type of carbon capture plant to fit to existing coal-fired power plants. This is likely to be an important consideration if carbon capture from coal plants becomes mandatory.
The capture process is conceptually very similar to that used in FGD scrubbers to remove sulfur dioxide. A reagent capable of capturing carbon dioxide is sprayed from the top of a tall reactor vessel, the absorption tower, through which the flue gases pass from bottom to top. As the reagent cascades down the tower it mixes with the plant flue gases and carbon dioxide is removed by being bound chemically by the reagent. This type of process is capable of removing up to 90% of the carbon dioxide contained in the plant exhaust gases.
The spent reagent is collected at the bottom of the absorption tower while the cleaned flue gases are allowed to escape to the atmosphere. The reagent must now be treated to release the carbon dioxide, leaving the capture agent to be recycled through the system. This is an energy-intensive process because in this type of system the carbon dioxide molecules are relatively strongly bound to the reagent and it can lead to an overall fall in efficiency, using the materials available today, of close to 28%. This means that a supercritical coal-fired plant with an efficiency of 45% would see its overall efficiency drop to 33% or lower once carbon capture had been added. Once the carbon dioxide has been released, it must then be compressed ready for transportation, expected to be by pipeline, to a facility where it can be sequestered. A schematic of a power plant with post-combustion capture is shown in Figure 7.1.
The efficiency of carbon dioxide capture depends on the concentration of carbon dioxide within the flue gases of a power plant. For a typical coal-fired boiler the concentration is normally between 12% and 14% of the flue gas by volume. At this concentration the most effective way of capturing carbon dioxide is to use a reagent that bonds chemically with the carbon-containing gas. The chemical solvent will remove a large part of the carbon dioxide but, because it binds strongly with it, releasing the gas again is more difficult. In other situations where the concentration of carbon dioxide is much higher it is possible to employ a physical solvent into which the carbon dioxide simply dissolves. The bonds holding the gas in solution are much weaker than in the case of the chemical solvent and the energy required to release it again is much smaller.
The best solvents for capturing carbon dioxide chemically are amines, which bond with it strongly. Amines have been in use since the 1930s to produce food-grade carbon dioxide from gas streams with between 3% and 25% carbon dioxide. Of these, the most commonly used is monoethanolamine (MEA). The technology for MEA capture is well tested but it has not yet been tried at the level of a large coal-fired power station. An alternative to MEA is an ammonia-based solvent from which it is potentially easier to release carbon dioxide once it has been captured. However, capture efficiency will be the key to its success.
Other types of capture have been proposed for post-combustion capture. These include the use of ionic liquids which can, in principle, absorb carbon dioxide without binding to it very strongly. Cryogenic separation at a temperature of −120°C or lower may be capable of removing 90% of the carbon dioxide from the flue gas stream but cooling costs are the key consideration. Solid sorbents which adsorb carbon dioxide onto their surface are attractive if they can provide a high enough capture efficiency. Today they cannot. Membranes, similarly, would offer a simple solution if they could achieve the required efficiency.
Pre-Combustion Capture of Carbon Dioxide
The pre-combustion capture of carbon dioxide is essentially coal gasification as described earlier, with the gasification processes carried through so that the product gas is essentially carbon dioxide and hydrogen. To carry out gasification, coal is first heated in a special reactor in a mixture of either air or water and steam normally under pressure. Most will use oxygen so that the plant will need an air separation unit (ASU) to supply oxygen. The product of this reaction will be a mixture of compounds including hydrogen, carbon monoxide, carbon dioxide, and methane. The gas mixture, called synthesis gas, is then mixed with more steam and passed over a catalyst which promotes the reaction between carbon monoxide and water, the water shift reaction, producing carbon dioxide and hydrogen.
Various gasifiers have been developed for coal gasification including fixed bed gasifiers, fluidized bed gasifiers and entrained flow gasifiers. The last two are the most suitable for power plant use. Both are operated at around 25–30 bar. An entrained flow gasifier operates at a much higher temperature than a fluidized bed gasifier (it is more like a pulverized coal furnace) and requires more oxygen but also produces a higher-quality gas product. After gasification the product gases must be cleaned to remove dust, acid components such as hydrogen chloride and hydrogen sulfide,1 and any nitrogen oxide and carbon dioxide. This leaves a stream of hydrogen which can then be used in a power plant to generate electricity.
The most commonly proposed configuration for a coal gasification power plant is the IGCC power plant. In this type of plant the hydrogen generated from gasification is burnt in a gas turbine to generate electricity. Waste heat from the exhaust of the turbine is then used to generate steam in a heat recovery steam generator and this drives a steam turbine. The gasification and power generation elements of such a plant are tightly integrated so that waste heat from the gasification process is also utilized for steam generation. A schematic for a plant of this type is shown in Figure 7.2. Research suggests that it may be possible to achieve up to 55% energy conversion efficiency with an IGCC plant before carbon capture but the best efficiency demonstrated by IGCC plants in operation is closer to 40%. When carbon capture is added, the overall efficiency drops to 32%. It may be more efficient to utilize the hydrogen in a fuel cell instead.
The key to the pre-combustion plant, aside from the gasification process, is carbon capture. The product gas after the water shift reactor will often contain more than 50% carbon dioxide. This high concentration makes the gas easier to separate from the hydrogen. In addition the gas is already at high pressure, and there is little nitrogen present so the volumes of gas that have to be handled are smaller. This makes is possible to carry out separation using a physical absorption technique that is much less energy-intensive than the post-combustion amine capture technique.
Physical separation of carbon dioxide is common in both the natural gas and petrol refining industries and two commercial processes are in use. One is called the Rectisol process and uses refrigerated methanol to absorb carbon dioxide. The second, called Selexol, uses a glycol-based solvent manufactured by Dow Chemicals. Both processes can also remove hydrogen sulfide, which can be isolated separately after absorption, so that finally a stream of pure carbon dioxide is produced ready for compression and transportation. The overall energy loss for a Selexol-based system is around 20%.2 While these processes are well established, new solvents are being sought that might offer lower energy loss. Research is looking at ionic fluids and ammonia-based solvents as possible candidates.
Solvent capture requires the product gases be cooled before treatment. High-temperature absorption would avoid this energy loss and this is another avenue for development. Solid adsorbents offer one possible high-temperature solution as do membranes, but research into these is at an early stage.
Carbon Capture and Oxyfuel Combustion
Oxyfuel combustion, in which combustion air in a power plant is replaced with pure oxygen, is the third main approach to coal combustion and carbon capture being developed for coal plants. It offers a radically different approach to the problem. A conventional coal-fired power plant with post-combustion capture has to separate carbon dioxide from a flue gas mixture in which there is a large amount of nitrogen. The oxyfuel process, by removing the nitrogen at the start, so that coal burns in oxygen instead of air, sidesteps this problem because the flue gases from the combustion process are largely carbon dioxide. This makes is much easier to isolate.
This simplification is at the expense of an ASU, which the plant requires to provide oxygen. In addition, the exhaust gases from the plant are not pure. They will contain some residual nitrogen, unburnt oxygen, sulfur dioxide, nitrogen oxides, and particulate material. Much of this must still be removed to produce carbon dioxide that is pure enough for sequestration.
There is also a combustion problem. When coal burns in pure oxygen the flame temperature is much higher than when it burns in air. The peak temperature can reach 2500°C compared to 1700°C in a supercritical boiler. There are no boiler and burner construction materials today that can withstand such a high temperature. The solution to this problem is to take some of the carbon-dioxide-rich flue gas from the exit of the boiler and mix it with the oxygen fuel supplied to the burners. This dilutes the oxygen and reduces the flame temperature to a level similar to that found in a conventional air-blown plant. A schematic of an oxyfuel plant is shown in Figure 7.3.
One of the keys to the economics of an oxyfuel combustion plant is the ASU. Air separation is a well-developed technology and is normally based on cryogenic separation of the components of air but it is expensive. The overall efficiency of a supercritical boiler with oxyfuel combustion is expected to be around 26%. New air separation technologies might be able to reduce this cost but all are at an early stage of research.
Oxyfuel combustion could potentially be retrofitted to existing coal boilers. It is only likely to be economically effective on modern supercritical and ultra-supercritial plants and some adaptation is necessary to allow flue gases to be fed back to the oxygen feed system. However, it may prove an economical competitor to the retrofitting of post-combustion capture in the future.
Chemical Looping
Chemical looping is an advanced form of oxyfuel combustion that does away with the need for an expensive ASU. In its place it uses a solid-state oxygen carrier. The process requires two reactors. In the first the oxygen carrier is heated in the presence of air, when it preferentially captures oxygen, leaving a stream of nitrogen. The oxygen-rich solid carrier is then transferred to a second reactor where it is exposed to coal that has been gasified to produce syngas. The oxygen from the carrier will react with the combustible components of the syngas producing hot exhaust gases that contain a high concentration of carbon dioxide. This must then be purified before compression and transportation. A simple schematic of the chemical looping process is shown in Figure 7.4.
A number of materials have been identified that could potentially act as oxygen carriers in a chemical looping process. These include copper oxide, cobalt oxide, and di-manganese trioxide. Each of these oxides will capture oxygen at high temperature (800–1200°C) to produce a higher oxide of the metal CuO/Cu2O3, Mn2O3/Mn3O4, and CoO/Co3O4. The process is still in an early stage of development.
Carbon Dioxide Compression and Transportation
Once carbon dioxide has been captured using one of the cycles outlined above it has to be transported to a site where it can be stored securely. Transportation will probably be by pipeline, though in some cases it may involve bulk freight carriers by rail or sea. Whichever is used, the carbon dioxide must first be compressed. Compression is carried out in stages so that eventually the carbon dioxide reaches its critical point where its volume decreases significantly compared to the normal gaseous phase. Typical compression is to 10–15 MPa. Dehydration is normally carried out during compression because the presence of water can lead to pipeline corrosion as well as problems with solid hydrate formation.
Gas compression is a well-tested technology but it has an energy cost. Estimates suggest that the electrical energy required to compress the carbon dioxide from a large coal plant would use around 7–12% of the plant output. In order to reduce the cost new technologies such as refrigeration and pump compression, or shockwave compression are being explored.
Transportation of the compressed gas will most commonly be via pipeline. Ideally the distance the carbon dioxide has to be transported to a sequestration site is as short as possible but even so it is likely to be tens or hundreds of kilometers in many cases. Transportation by pipeline is carried out regularly in the oil industry, particularly in the USA where 40 m tons are pumped annually through 6000 km of pipeline.
Carbon Dioxide Sequestration
The capture and storage of carbon dioxide from power stations is going to involve massive quantities of the gas. An estimate based on one of the International Energy Agency future energy consumption scenarios puts the quantity that will need to be captured and stored from power plants by 2050 at around 80 Gtons.3 There are only two places that such large amounts could feasibly be stored, firstly within or below the world’s oceans, or secondly under the ground. Ocean storage is currently considered to carry too high a risk to be practical so storage will take place underground.
Three principal types of underground storage site have been identified. The easiest to access immediately are oil and gas wells. Carbon dioxide injection is already carried out by the hydrocarbon industry for enhanced oil recovery (EOR) from oil and gas fields. Evidence from such sites suggests that the gas remains underground once it has been pumped there. Continued EOR will consume a small amount of captured carbon dioxide. More importantly, spent oil and gas fields in many parts of the world will offer a simple solution for initial carbon dioxide sequestration projects by providing a storage capacity that will often have pipeline access. The capacity that these can accommodate is limited, however, and they cannot be found everywhere, so other sites will also be required when carbon capture and storage starts to be deployed widely.
Another option is to store carbon dioxide in coal-beds that are too deep to be mined. These deep coal-beds often contain methane. If carbon dioxide is pumped down into a coal-bed containing methane it will selectively displace the gas, which can be recovered for use as an energy source above ground. The recovery of methane will offset the cost of carbon dioxide sequestration but sites of this type are relatively rare so they cannot provide for all the carbon dioxide that needs to be sequestered either.
Over the longer term the main solution to the problem of carbon dioxide sequestration is likely to be the third, and potentially the most important type of underground site, called a brine aquifer. A brine aquifer is created when a cap of impermeable rock is formed deep underground. This cap prevents water or gas seeping upwards through it and so a hot, concentrated brine collects within the more porous rock beneath the cap. This is the brine aquifer. The brine can be displaced by carbon dioxide if the latter is pumped down into the aquifer. Experimental sites where carbon dioxide has been stored in brine aquifers indicate that the gas should remain securely stored away and it may eventually be trapped permanently by a reaction with the rock in which it is held to make carbonates.
It should be noted, when considering the large-scale capture and storage of carbon dioxide, that the gas can be highly toxic. The normal concentration in the atmosphere is 0.04%. If the concentration rises to 3%, inhalation begins to effect vision and hearing. At 10%, the gas acts as an asphyxiant and when the concentration rises to 20% inhalation leads to rapid death. It is vital, therefore, that large quantities of the gas, once stored, can never be released into the atmosphere.
For this reason, as well as to verify that storage is permanent and that carbon dioxide that has been stored is not returning to the atmosphere, storage sites will need to be monitored for tens, if not hundreds of years, to verify that the storage is secure. Regulations to ensure this happens are already being mapped out. Operational sites will most likely be inspected regularly but remote monitoring is likely to become normal once a site has been sealed.
Other types of sequestration are being explored too. One of the potentially most interesting is biological sequestration. Plants capture carbon dioxide from the atmosphere during photosynthesis and this process might be harnessed to absorb captured carbon dioxide too. Research has indicated that algae might be able to process carbon dioxide rapidly, but development is at an early stage. Another option is to try to convert carbon dioxide directly into a solid mineral. Magnesium hydroxide solution will absorb carbon dioxide, producing magnesium bicarbonate. However, like biological sequestration, this remains at a very early stage.
A number of projects around the world have demonstrated that underground storage of carbon dioxide is feasible. Several of these involve brine aquifers, while others are based in EOR projects where the fate of the underground carbon dioxide has been monitored. However, as will all current aspects of carbon capture and storage, this has to be demonstrated successfully at the scale of a large coal-fired power station before the technology can be considered mature.