Separation of carbon dioxide and other acid gases from flue gas using physical solvents and alkali

Abstract

Flue gas is compressed to high pressures, which raises the partial pressure of carbon dioxide thus facilitating its absorption into physical solvents (e.g., water, salt solutions, methanol). When added onto an existing CO2-producing process, the compression is performed using multiple stages with intercooling. After the high-pressure carbon dioxide dissolves into the physical solvent, it flows through a series of pressure letdowns using turbines that allow the dissolved carbon dioxide to evolve from the solvent. Through a series of multiple compression stages, the evolved gas is compressed for transport, utilization, or storage. The flue gas that has been scrubbed of carbon dioxide flows through a series of expanders with interheating supplied from waste heat, or the heat of compression. The process is enhanced by employing alkalis that chemically react with acid gases, such as carbon dioxide. These same CO2-capture processes can be integrated into processes that produce shaft and / or electrical power.

Description

BACKGROUND

[0001] Global warming is caused by the accumulation of greenhouse gases in the atmosphere.The dominant greenhouse gas is carbon dioxide, most of which is produced by combusting fossilfuels. The majority of fossil fuel combustion occurs in large central facilities such as power plants,refineries, factories, cement kilns, lime kilns, steel mills, etc. Economic recovery of carbon dioxidefrom flue gas emitted from large central facilities would contribute greatly to the reduction ofglobal warming.

[0002] Methods for recovering carbon dioxide from flue gas include chemical absorption intoliquids (e.g., amines), physical absorption into liquids (e.g., methanol), adsorption onto solids (e.g.,mesoporous silicas, zeolites, and metal-organic frameworks), membranes, reaction with alkalis(e.g., lime), and cryogenics.

[0003] This invention employs two methods: (1) physical absorption into liquid, and (2)reaction with alkalis.SUMMARY

[0004] In an embodiment of the disclosure, a method of performing work on a load involvescombusting a fuel containing carbon and an oxidant in a combustion device to form gaseouscombustion products having a temperature and a pressure. The temperature and / or pressure of thesegaseous combustion products is then adjusted to a level suitable for separation, such as by coolingthe products to a suitable temperature or compressing them to a suitable pressure. Followingadjustment, the gaseous combustion products are separated into a carbon dioxide-rich stream anda carbon dioxide-poor stream, with the carbon dioxide-poor stream having its own temperature andpressure. The carbon dioxide-poor stream is subsequently heated to a temperature suitable forexpansion and expanded against the load to perform work.

[0005] In further aspects of this embodiment, the cooling of the gaseous combustion productsmay occur via heat exchange against the carbon dioxide-rich stream and / or the carbon dioxide-poor stream, potentially within the combustion device itself. Additionally, the oxidant and / or fuelcan be preheated by heat exchange against the expanded gas from the expansion step. Incombustors, the fuel may include coal, biomass, gaseous fuels, or those containing oxides of carbonsuch as carbon monoxide, sourced from effluents like those from blast furnaces, oil refineries,steam methane reformers, chemical production processes, or treatment plants. In the combustors,the oxidant is typically oxygen contained within air; however, the air may be enriched in oxygento reduce diluting components such as nitrogen. Separation of the gaseous combustion productscan be achieved by dissolving carbon dioxide in a solvent, such as water or methanol. In anothervariation, the gaseous combustion products are cooled against a second gas at a pressure suitablefor expansion, heating that second gas for subsequent expansion against the load.

[0006] In another embodiment of the disclosure, a heat engine comprises a compressorconfigured to compress the oxidant inducted into the engine and / or the gaseous combustionproducts; a combustion device to combust the oxidant with a carbon-containing fuel, producinggaseous combustion products containing carbon dioxide; a carbon dioxide separator to divide theseproducts into a carbon dioxide-rich stream and a carbon dioxide-poor stream; and an expander toexpand the carbon dioxide-poor stream. The heat engine may further include a cooling device forthe gaseous combustion products, a heating device for the carbon dioxide-poor and / or carbondioxide-rich streams, and a heat exchanger to transfer heat from the expanded stream to the oxidantand / or fuel.

[0007] In yet another embodiment of the disclosure, a method of performing work on a loadincludes cooling an effluent from a first process where the effluent has a first pressure and a firsttemperature greater than 150°C (or greater than 250°C)—to a second temperature lower than thefirst. The cooled effluent is then compressed to a second pressure greater than the first (such asgreater than 40 bar) to create an expansion gas, which is reheated against the original effluent andexpanded against the load. In particular embodiments, the expansion gas is separated into a carbondioxide-rich gas and a carbon dioxide-poor gas, with the carbon dioxide-poor gas heated againstthe effluent and expanded. This separation may involve absorbing carbon dioxide into a solventlike water, an aqueous solution, Rectisol, Selexol, or potassium carbonate.

[0008] While the embodiments described herein may provide certain advantages, such asimproved efficiency, enhanced carbon capture, or adaptability to various fuels and oxidants, it isto be understood that not all embodiments of the disclosure necessarily achieve all suchadvantages, and the scope of the invention is not limited to embodiments that realize these or anyother benefits. The full scope of the invention is defined by the claims.BRIEF DESCRIPTION OF DRAWINGS

[0009] For a more complete understanding of this disclosure and its features, reference is nowmade to the following description, taken in conjunction with the accompanying drawings in which:

[0010] Figure 1 shows Henry's constant for water and methanol as a function of temperature.

[0011] Figure 2 shows a simple schematic of an absorption column.

[0012] Figures 3A, 3B and 3C show profiles of the partial pressure of carbon dioxide alongthe height of the absorption column.

[0013] Figure 4 shows Option A-1 where high-pressure flue gas is scrubbed of carbondioxide in an absorption column using water.

[0014] Figure 5 shows Option A-2 where high-pressure flue gas is chilled using refrigerationand is scrubbed of carbon dioxide in an absorption column using water.

[0015] Figure 6 shows Option A-3 where high-pressure flue gas is chilled using gasexpansion and is scrubbed of carbon dioxide in an absorption column using water.

[0016] Figure 7 shows multi-stage compression using axial fans and intercooling.

[0017] Figure 8 shows multi-stage expansion using axial turbines and interheating.

[0018] Figures 9A and 9B show the layout for axial fans arranged as compressors andexpanders.

[0019] Figures 10A, 10B, and 10C show three methods for promoting carbon dioxidenucleation.

[0020] Figure 11 shows Option B where high-pressure flue gas is scrubbed of carbon dioxidein a series of agitated vessels using water.

[0021] Figure 12 shows Option C where regeneration of the physical solvent is enhanced byraising its temperature.

[0022] Figure 13 shows Option D-1 where recovery is enhanced by adding divalent alkalis tothe circulating water at ambient pressure.

[0023] Figure 14 shows Option D-2 where recovery is enhanced by adding divalent alkalis tothe circulating water at elevated pressure.

[0024] Figure 15 shows Option E where a second absorption column is added to removeother acids gases (e.g., SOx) from the flue gas.

[0025] Figure 16 shows Option F where recovery is enhanced by incorporating monovalentand divalent alkalis.

[0026] Figure 17 shows Option G where recovery is enhanced by utilizing electrochemicalwater-splitting bipolar electrodialysis.

[0027] Figure 18 shows details of electrochemical water-splitting bipolar electrodialysis.

[0028] Figure 19 shows Option H, which employs a physical solvent other than water.

[0029] Figure 20 shows a biological filter.

[0030] Figure 21 shows Option I-1, which integrates carbon capture into a power-productionprocess that approximates the Ericsson cycle and uses air-rich combustion.

[0031] Figure 22 shows Option I-2, which integrates carbon capture into a power-productionprocess that approximates the Ericsson cycle and uses fuel-rich combustion.

[0032] Figure 23 shows the per-stage enthalpy change in expanders with a pressure ratio of2.

[0033] Figure 24 shows Option I-3, which employs recycle to achieve stoichiometriccombustion.

[0034] Figure 25 shows Option I-4, which employs a single combustor rather than a series ofcombustors.

[0035] Figure 26 shows Option I-5, which replaces the compression train with high-pressureliquid water.

[0036] Figure 27 shows Option I-6, which employs serial ambient-pressure combustion withparallel extraction of heat.

[0037] Figure 28 shows Option I-7, which employs serial high-pressure combustion withparallel extraction of heat.

[0038] Figure 29 shows Option I-8, which employs multiple series of combustors, eachoperating at successively higher pressures.

[0039] Figure 30 shows Option I-9, which employs multiple series of combustors, eachoperating at successively higher pressures. The first combustor operates at ambient pressure.

[0040] Figure 31 shows Option J-1, which integrates carbon capture into a recuperatedBrayton power-production process and uses liquid or gaseous fuels.

[0041] Figure 32 shows Option J-2, which integrates carbon capture into a recuperatedBrayton power-production process and uses a lock hopper to introduce solid fuels into thecombustor.

[0042] Figure 33 shows Option J-3, which integrates carbon capture into a recuperatedBrayton power-production process and uses a lock hopper to introduce solid fuels into a gasifier.

[0043] Figure 34 shows Option K, which integrates carbon capture into a recuperatedBrayton power-production process that employes near-isothermal compression.

[0044] Figure 35 shows Option L-1, which integrates carbon capture into a power-productionprocess that includes a methane reformer to produce hydrogen. The combustor operates atelevated pressure.

[0045] Figure 36 shows Option L-2, which integrates carbon capture into a power-productionprocess that includes a methane reformer to produce hydrogen. The combustor operates at near-ambient pressure.

[0046] Figure 37 shows Option L-3, which integrates carbon capture into a power-productionprocess that includes a methane reformer to produce hydrogen and carbon monoxide. Thecombustor operates at elevated pressure.

[0047] Figure 38 shows Option L-4, which integrates carbon capture into a power-productionprocess that includes a methane reformer to produce hydrogen and carbon monoxide. Thecombustor operates at near-ambient pressure.

[0048] Figure 39 shows how ammonia production integrates with processes that capture CO2and produce electricity.

[0049] Figure 40 shows how liquid fuel production integrates with processes that captureCO2 and produce electricity.

[0050] Figure 41 shows Option M-1, which employs a combined-cycle gas turbine thatemploys exhaust gas recirculation. A portion of the high-pressure combustion gas is removedfrom an intermediate stage of the expander so that carbon dioxide is recovered.

[0051] Figure 42 shows Option M-2, which is identical to Option M-1, except that anadditional compression stage is added prior to the absorber.

[0052] Figure 43 shows Option M-3, which is identical to Option M-2, except that the air fedto the combined-cycle gas turbine is enriched in oxygen.

[0053] Figure 44 shows the concentration of carbon dioxide in the flue gas based on thepercentage of oxygen in the feed air.

[0054] Figure 45 shows the energy requirements for increasing the oxygen concentration infeed air.

[0055] Throughout the drawings, like reference numerals denote corresponding or similarelements, with numerals in different figures incremented to reflect the figure or embodiment(e.g., element 123 in FIG. 1 generally corresponds to element 223 in FIG. 2). However, suchcorresponding elements may differ in specific aspects to accommodate changes in the overallsystem configuration.DETAILED DESCRIPTION

[0056] The figures described below, and the various embodiments used to describe theprinciples of the present disclosure in this patent document, are by way of illustration only andshould not be construed in any way to limit the scope of the disclosure. Those skilled in the art willunderstand that the principles of the present disclosure invention may be implemented in anysuitably arranged device or system. Additionally, the drawings are not necessarily drawn to scale.

[0057] Theory

[0058] Physical absorption of carbon dioxide into a liquid is characterized by Henry's Law,which is commonly expressed by the following two formulas:PA=H'XA(1)CA = H PA(2)wherePA = Partial pressure of Species A (i.e., carbon dioxide)XA = Mole fraction of Species A (i.e., carbon dioxide)CA = Concentration of Species A (i.e., carbon dioxide)H' = Henry's constant for Equation 1H = Henry's constant for Equation 2

[0059] Figure 1 shows the Henry's constant H for carbon dioxide dissolved in water andmethanol as a function of temperature. Generally, the Henry's constant for methanol is about 4times larger than water. Methanol is often a preferred solvent for removing carbon dioxide fromindustrial gases, the Rectisol process being the most famous example. For both solvents, Henry'sconstant decreases as temperature increases.

[0060] Figure 2 shows a simplified description of a column where liquid flows from the topdown and gas flows from the bottom up. At any location along the vertical height, the partialpressure of carbon dioxide in the gas phase is greater than the partial pressure in the liquid phase.This difference in partial pressure (chemical potential) causes carbon dioxide to flow from the gasphase into the liquid phase.

[0061] Figure 3 shows the partial pressure profile of Species A (carbon dioxide) in the liquidand gas phases at each elevation in the column. As described above, at every height, the partialpressure of Species A (carbon dioxide) is greater in the gas phase than the liquid phase, whichdrives carbon dioxide to dissolve in the liquid phase.

[0062] The following processes employ both compressors and expanders. It is understood thatcompressors require an input of shaft power and expanders produce an output of shaft power. Theexpanders can drive the compressors directly. Alternatively, if excess shaft power is available fromthe expanders, it can be used to drive a generator that produces electricity. Because the shaft powerfrom the expander can be employed in two manners, for simplicity, the coupling to a compressoror generator is not shown. It is understood that this shaft power can be employed in eitherapplications.

[0063] Add-on CO2-capture Processes

[0064] Add-on CO2-capture processes remove carbon dioxide from the flue gas emitted byexisting processes, such as power plants, chemical plants, refineries, cement kilns, lime kilns, steelmills, etc.

[0065] Figure 4 shows Option A-1, according to an embodiment of the disclosure. Flue gascontaining water vapor, carbon dioxide, and gases (e.g., nitrogen, oxygen, argon) has been cooledto near-ambient temperature. Then, through a series of compression stages (compressors 481, 482,483) with intercoolers (heat exchangers 471, 472, 473), the pressure rises. The intercoolers keepthe gas at near-ambient temperature, which reduces energy requirements in the compressors.

[0066] As shown in Figure 4, as the flue gas is compressed and intercooled, water condensesand is removed from the system (e.g., as shown by the water outlet from vessel 450). The high-pressure flue gas enters 401 a packed column 402 at the bottom while high-pressure water enters417 at the top. The water entering at the top has a low concentration of dissolved carbon dioxide.As it flows down column 402, it absorbs carbon dioxide from the flue gas, which enters 401 at thebottom.

[0067] The water exiting 418 from the bottom of the column 402 has dissolved carbon dioxide,which is nearly saturated at the partial pressure of carbon dioxide in the raw flue gas. This high-pressure liquid flows through a series of pressure-let down devices that reduce the pressure andthereby allows the dissolved carbon dioxide to enter the gas phase. Non-limiting examples ofpressure-letdown devices include Pelton wheel, Francis turbine, Turgo turbine, centrifugalturbines, radial expanders, axial pistons, radial pistons, gear motors, vane motors, gerotors, andisobaric pressure exchangers. In some cases, these devices are used to recover pressure fromdischarged brine in reverse osmosis systems. In Figure 4, the pressure-letdown devices are shownas turbines 404, 407. While two are shown, more or fewer than two may be utilized.

[0068] To facilitate bubble nucleation in vessels 406 and 408, a porous surface is employed,such as sintered metal, zeolite, sand, etc. At each pressure letdown stage, the evolved carbondioxide is fed to the inlet of a multi-stage compressor train (e.g., compressors 485, 486, 487). Τοmaximize efficiency, the compressor inlet pressure matches the pressure of the evolved carbondioxide. As shown, as the total pressure of CO2 increases, liquid water will condense wheninterstage cooling (e.g., heat exchangers 475, 476, and 477) is employed. This condensed watercontains dissolved CO2, so it is recycled so that the CO2 can be captured as shown by the outletfrom vessel 409. From the vessel 409, the CO2 may be compressed, again, via compressor 488 toyield the high-pressure CO2. Because liquid water and CO2 forms carbonic acid, the intercoolersmust be constructed from expensive acid-resistant materials. Alternatively, to avoid this cost, theCO2 can be dried to remove water prior to compression. Typical drying agents include silica gel,zeolites, activated alumina, and calcium sulfate. Once the drying agent is saturated, it isregenerated by swinging the temperature or pressure. Such processes are common and typicallyemploy two beds, one that is actively drying and the other is being regenerated.

[0069] As the carbon dioxide bubbles emerge from the saturated liquid, there is a slight coolingeffect. Under standard conditions, the enthalpies of formation for gaseous and aqueous CO2follow:CO2 (g)ΔΗ= -393.51 kJ / molCO2 (aq)ΔΗ= -413.26 kJ / mol

[0070] The evolution of CO2 from water into the gas phase is represented by the followingequation:CO2 (aq) → CO2 (g)ΔΗ = 19.75 kJ / mol

[0071] The positive enthalpy change indicates that cooling occurs when carbon dioxideevolves from the liquid phase to the gas phase. Additional cooling results from the evaporation ofwater that occurs simultaneously with the evolution of the gaseous carbon dioxide from the liquidphase.

[0072] After the final letdown stage, the water is pumped (e.g., via pump 411) to high pressuresand is returned to the absorption column 402 through line 417. If necessary, the water exiting thepumps is cooled (e.g., via heat exchanger 478) to maintain near-ambient temperature.

[0073] The gas exiting 403 the absorption column 402 has a reduced content of carbon dioxideand is sent to a series of expanders (491, 492, 491) that reduce the pressure. Between the stages,interheating (e.g., via heat exchangers 461, 462, and 463) is employed. The source of heat couldbe from the heat of compression in the compression train; however, flue gas is typicallyaccompanied by large volumes of low-grade waste heat. To maximize power production, the gasbetween each expansion stage is heated to the highest possible temperatures using waste heat.

[0074] In the compression and expansion trains of Figure 4 (and other embodiments describedherein), a variety of heat exchangers could be employed such as, but not limited to, plate-and-frame, shell-and-tube, finned-tube, plate-finned, direct-contact, etc. The incoming and outgoingflue gas can exchange heat with water, air, flue gas, or other fluids.

[0075] Figure 5 shows Option A-2, according to an embodiment of the disclosure. Option A-2 is similar to Option A-1 except that the flue gas and water entering (501B, 517B) the absorptioncolumn 502 are precooled to lower the temperature and thereby increase Henry's constant (Figure1). The impact of lower temperatures is to reduce the water circulation rate, which decreases thesize of the absorption columns and also reduces pumping power.

[0076] In Figure 5, most of the precooling occurs through countercurrent heat exchangers (551,552, 553, 554); however, because of the required approach temperature in the heat exchangers,complete heat exchange is not possible. To overcome this problem, a heat pump is employed toremove heat from the incoming stream and discharge it to the outgoing stream.

[0077] To lower the temperature at which absorption occurs, agents can be added to the water,which lowers the freezing point. Examples of candidate agents follow:• Calcium chloride• Magnesium chloride• Sodium acetate• Potassium acetate• Ethylene glycol• Propylene glycol

[0078] To reduce corrosion, the following additives are examples:• Sodium nitrite• Sodium molybdate• Sodium phosphates

[0079] They may be used in pure form, or as mixtures.

[0080] It should be noted that adding agents to water typically lowers the solubility of CO2 ata given temperature through the so-called salting-out effect. On the other hand, lowering thetemperature improves the solubility of CO2. Specific combinations of chemistry, concentration,and temperature can yield improved loading of CO2 in aqueous systems. Many of these additivesare nonvolatile and will not be released into the environment. Furthermore, many of them areinexpensive and readily available.

[0081] The water 518 and 517B that circulates through absorption column 502 and vessels 506and 508 has constant addition of energy from liquid pumps 511 and the heat pump compressor584. Typically, in a single pass through the system, the circulating water would increase intemperature by about 0.63°C. If this energy is not removed, upon each circulation, the watertemperature will increase further. To remove this energy, multiple approaches may be taken.

[0082] In Approach 1, the water circulating through vessels 506 and 508 has sufficiently hightemperature that the energy can be rejected directly to cooling water or air through conventionalheat exchangers, such as plate-and-frame, shell-and-tube, air-cooled fins, etc. This approach hasthe advantage of simplicity; however, the operating temperature may be so high that the heat dutyon heat exchanger 554 may be excessive and expensive.

[0083] In Approach 2, the circulating water is cooled by a heat exchanger that rejects heat to acold fluid, such as an evaporating refrigerant or chilled water. In Approach 2, the circulating wateroperates at a lower temperature than Approach 1 and therefore reduces the heat duty on heatexchanger 554, but it has an additional expense of a refrigeration system.

[0084] In Approach 3, the circulating water is cooled by operating vessel 508 at sufficientlylow pressure that a portion of the circulating water evaporates and thereby cools the circulatingwater. For example, to cool the circulating water by 0.63°C, it is necessary to evaporate 1.08 kgof water per m³ of circulating water. At 25°C, the vapor pressure of water is 0.0317 bar. Assumethat vessel 508 operates with a water vapor pressure of 0.0317 bar and that the water enteringvessel 508 is 25.63°C. In this scenario, a portion of the water entering vessel 508 (1.08 kg of waterper m³) will evaporate and thereby cool the liquid water exiting vessel 508 to 25°C. At this lowpressure, dissolved CO2 will bubble out of the water and thereby achieve extremely high CO2recovery. For example, if vessel 508 were to operate at a total pressure of 0.0427 bar, the gasphase would be 74 mol% water and 26 mol% CO2. Because the CO2 partial pressure is so low(0.0111 bar), the dissolved CO2 in returned water 517B is extremely low (0.0165 kg CO2 / m³water). In this example, assume that absorber 502 operates at 0°C, total pressure 18 bar, andincoming gas 501B contains 15 mol% CO2. At equilibrium, the CO2 concentration in water stream518 is 9.12 kg CO2 / m³ water. In this scenario, comparing the CO2 concentrations in water streams518 and 517B, the total CO2 recovery is about 99.8%. Approach 3 achieves both cooling andextraordinarily high CO2 recovery. The low pressures in vessel 508 is achieved by compressor585. Examples of such compressors include turbo compressors, lobe compressors (Roots blowers),gerotor compressors, liquid-ring vacuum pumps, jet ejectors, and other suitable devices.

[0085] Figure 6 shows Option A-3, according to another embodiment of the disclosure. OptionA-3which is similar to Option A-2 except that the flue gas is cooled by an expander 695.

[0086] In many applications, the volume of flue gas is large, so conventional compressors maybe too expensive. To overcome this problem, compression can occur using a series of mining fans,or similar axial fans (Figure 7). This type of fan is characterized by having a tip speed of aboutMach 0.5. Interstage cooling can be achieved using direct contact of the compressed flue gas withwetted packing, such as Munters CELdek packing, structured packing typically employed incooling towers (e.g., Brentwood Industries), or even dump packing. Figure 7 shows typicaltemperatures that result from cooling the circulating water against cooling water.

[0087] As shown in Figure 7, the wetted packing has horizontal channels; however, thisarrangement makes it difficult to countercurrently flow of water and gas. By arranging thechannels vertically, it is more practical to achieve countercurrent flow: water flows downward andgas flows upward.

[0088] In Figure 7, the use of wetted-packing, direct-contact heat exchange is shown in thecontext of cooling the flue gas; however, this same concept can be applied throughout the processwhenever compressed gas must be cooled. Because of direct-contact heat transfer and theinexpensive wetted packing, this approach to heat transfer is substantially less expensive thanconventional gas-to-air or gas-to-liquid heat exchangers.

[0089] Figure 8 shows a method for interheating, according to an embodiment of thedisclosure. After each expansion, the gas cools, so it is reheated by directly contacting circulatinghot water. To facilitate heat transfer, the hot water contacts the cool gas through wetted packing,such as Munters CELdek packing, structured packing typically employed in cooling towers (e.g.,Brentwood Industries), or even dump packing. The circulating water is heated by contacting thesource of waste heat. The indicated temperatures are typical for waste steam at 90°C. So the gasexiting the final expansion stage is at near-ambient temperature, the final expansion stages mightnot employ interheating.

[0090] When compressing the flue gas, a hybrid system can be employed that combines axialfans with conventional compressors. Initially, the volume is large, so axial compressors are mostappropriate. Later, as the pressure increases and the volume decreases, conventional compressorscan be employed. Similarly, when expanding the flue gas, the initial pressure is high and thevolume is low; therefore, conventional expanders are appropriate. As the pressure reduces and thevolume increases, axial turbines can be used.

[0091] Figures 9A and 9B show the layout of the axial compressors and expanders, accordingto an embodiment of the disclosure. In the compression train, initially, the axial fans have a largediameter with a smaller rotation speed. Because the per-stage compression ratio is small, multiplefans can be placed in series. It is understood that each rotating fan includes stators that straightenthe flow between the fans. As the compression occurs, temperature increases. Intercooling reducesthe temperature, making compression more efficient. As shown in Figures 9A and 9B, twocompression stages are employed before intercooling; however, this is just illustrative. More stagescan be employed before intercooling.

[0092] In the expansion train, interwarming occurs between the expansion stages. As shown,two expansion stages are employed prior to interwarming; however more may be employed.

[0093] The expanders produce shaft power whereas the compressors consume shaft power. Ifthe expander cannot supply enough shaft power for the compressors, an electric motor is employedto supply the remaining power. All the compressors and expanders that operate at a common speedare located on a single shaft. As the diameter of the compressors and expanders reduces, it maybe necessary to design the system so that the shaft speed increases. If the interwarming temperatureis high enough, the expanders will produce more power than is required by the compressors. Inthis case, a generator captures the excess shaft power and produces electricity.

[0094] Figure 9A shows the configuration where each wetted pad has its own separate heatexchanger that supplies cooling to the compressor and heat to the expander. If no external sourceof heat is available, then the heat required for the expanders can be supplied from the compressorsby simply circulating liquid water between pads of similar temperature and pressure. This approachreduces the cost of heat exchangers, which greatly reduces capital costs. Because more shaft poweris required by the compressor than is delivered by the expander, the extra power input from theelectric motor must be dissipated against cooling water or into the air via finned tube heatexchangers.

[0095] In Options A-1 (Figure 4), A-2 (Figure 5) and A-3 (Figure 6), the evolved carbondioxide from each pressure letdown stage is compressed in a compressor train with intercooling.The compressor train could be a series of axial fans, such as those typically employed in the miningindustry (Figure 7), or they could be conventional compressors. The final discharge pressure issufficiently high so that the carbon dioxide can be fed to a pipeline and ultimately disposed, eitherby utilization or sequestration. Typical sequestration pressures are about 150 bar.

[0096] Figures 10A, 10B, and 10C illustrate three example methods for promoting nucleationof carbon dioxide gas bubbles when the pressure is reduced, according to an embodiment of thedisclosure. Figure 10A shows a tank 1006 that contains a porous high-surface material 1051 (e.g.,sintered metal). The tank diameter is sufficiently large that the liquid disentrains from the gas.Figure 10B shows a hydroclone 1041A that facilitates the separation of gas from liquid. In thisapproach, the wall 1052 of the pipe entering the hydroclone 1041A is covered with a porous high-surface material (e.g., sintered metal) that promotes nucleation. Figure 10C shows a hydroclone1041B that facilitates the separation of gas from liquid. In this approach, small particles of poroushigh-surface material (e.g., boiling chips) are suspended in the liquid. These particles arerecovered by a suitable method (e.g., filtration 1042) and returned to the entrance to the hydroclone.

[0097] Figure 11 shows Option B, according to an embodiment of the disclosure. Option B isidentical to Option A, except that the absorption column is replaced with a series of stirred vessels1159 through which the gas and liquid flow countercurrently. Stirring increases the rate that carbondioxide dissolves into the liquid. Heat exchangers remove the thermal energy generated fromstirring the liquid. It is possible to combine Options A and B; for example, most of the absorptioncould occur within a column and the final absorption stage could be a stirred vessel to ensure near-complete capture of carbon dioxide from the flue gas. While three vessels are generally shown,more or fewer than three vessels may be used in other configurations.

[0098] Figure 12 shows Option C, according to an embodiment of the disclosure. Option C isidentical to Option A-1, except that the liquid is heated to raise the temperature, which reducesHenry's constant (Figure 1) and thereby lowers the concentration of carbon dioxide in the liquidreturned to the absorption column. Furthermore, by raising the temperature, a greater fraction ofthe gas volume exiting 1218 the vessel 1202 is water vapor, which lowers the partial pressure ofcarbon dioxide and thereby further reduces the concentration of carbon dioxide in the liquidreturned to the absorption column. In Option C, the carbon dioxide concentration entering theabsorption column 1202 is reduced, which increases the recovery of carbon dioxide from the fluegas and / or reduces the required pressure in the absorption column.

[0099] Figure 13 shows Option D-1, according to an embodiment of the disclosure. Option D-1 which is identical to Option A, except that a divalent hydroxide is added to the circulating water.A mixing vessel 1310 is shown receiving the divalent hydroxide; however, other structures maybe used.

[00100] Carbon dioxide is an "acid gas" that forms carbonic acid when dissolved in water.CO2 + H2O → H2CO3

[00101] When adding a divalent hydroxide (e.g., calcium hydroxide), it reacts with carbonicacid to form carbonates.Ca(OH)2 + H2CO3 → CaCO3 + 2 H2OMg(OH)2 + H2CO3 → MgCO3 + 2 H2OBa(OH)2 + H2CO3 → BaCO3 + 2 H2O

[00102] The carbonates have low solubility in water and can be recovered via filtration, settling,centrifugation, or other appropriate means (as generally represented by a filter 1399). Therecovered carbonates can be thermally decomposed to create the corresponding oxide.CaCO3 → CaO + CO2MgCO3 → MgO + CO2BaCO3 → BaO + CO2

[00103] The resulting carbon dioxide is present at high concentrations and therefore is easilycaptured. The oxides can be added to water to make the hydroxides, which are then recycled.CaO + H2O → Ca(OH)2MgO + H2O → Mg(OH)2BaO + H2O → Ba(OH)2

[00104] Table 1 summarizes key properties of each chemical.Table 1. Solubility and decomposition temperature of divalent salts.Solubility at room temperature (g / L)Ca(OH)21.73Mg(OH)20.0064Ba(OH)252.3CaCO30.013MgCO30.139BaCO30.024Decomposition temperature at 1 atm (°C)CaCO3700 to 900MgCO3350BaCO31360

[00105] Of the candidate divalent alkalis presented, magnesium is advantageous because thedecomposition temperature is low, which reduces energy costs.

[00106] Adding divalent alkalis allows for near-zero carbon dioxide concentration in the waterentering the absorption column (Figure 3B). In principle, with enough contact time in theabsorption column, the partial pressure of carbon dioxide in the gas exiting the absorption columnwould approach zero. Adding excess alkali to the liquid in the absorption column allows for"residual alkalinity" in the water, which generates the partial pressure profile shown in Figure 3Cand allows the CO2 concentration in the exiting flue gas to approach zero.

[00107] As shown in Figure 13, the low-solubility carbonates are removed by suitable means,such as filtration, centrifugation, settling, etc. After the carbonates are removed, fresh divalentalkali is added. In Figure 13, the removal of carbonates and addition of fresh divalent alkali occursat near atmospheric pressure, which provides many options when selecting equipment.

[00108] Additional acid gases that are found in flue gas include SOx (e.g., SO2, SO3) and NOx(e.g., NO2). These acid gases may be removed from the flue gas using conventional technologies;however, if these conventional technologies are not employed, these acid gases will react in theabsorption column as follows:SO2 + H2O → H2SO3SO3 + H2O → H2SO42 NO2 + H2O → HNO3 + HNO2

[00109] These acid gases react with divalent alkali as follows:Ca(OH)2 + H2SO3 → 2 H2O + CaSO3Ca(OH)2 + H2SO4 → 2 H2O + CaSO4Ca(OH)2 + 2 HNO3 → 2 H2O + Ca(NO3)2Ca(OH)2 + 2 HNO2 → 2 H2O + Ca(NO2)2

[00110] The calcium sulfate and sulfite are poorly soluble and will precipitate, and they will bemixed with the insoluble carbonates. These precipitated salts may be used to amend clay soils toimprove fertility. In contrast, the calcium nitrate and nitrite are soluble and must be purged fromthe circulating water. These salts may be used as nitrogen fertilizer.

[00111] Figure 14 shows Option D-2, according to an embodiment of the disclosure. Option D-2 is identical to Option D-1 except that the divalent oxide is added, which reacts with water in anexothermic reaction. The heat of reaction is captured to elevate the temperature of the finaldesorption stage, which enhances removal of carbon dioxide from water by lowering its solubility(Figure 1).

[00112] Figure 15 shows Option E, according to an embodiment of the disclosure. Option E islike Option D-1, except that an additional absorption column 1502B has been added to capture SOx(e.g., SO2, SO3) from the flue gas. A suspended carbonate (e.g., calcium carbon, magnesiumcarbonate, dolomite) is added to the slurry in the absorption column 1502B. Using calciumcarbonate as an example, the following reactions occur:CaCO3 + SO2 + H2O → CaSO3 + CO2 + H2OCaCO3 + SO3 + H2O → CaSO4 + CO2 + H2O

[00113] The resulting sulfites and sulfates are insoluble and are recovered by filtration, settling,centrifugation, or other appropriate methods. The sulfite and sulfate can be added to clay soils toimprove fertility. The output 1501D of column 1502B enters column 1502a.

[00114] Figure 16 shows Option F, according to an embodiment of the disclosure. Option F isessentially identical to Option D-1, except that a monovalent alkali (e.g., NaOH, KOH) is addedto the water circulating through the absorption column. This circulating monovalent alkaliprovides "residual alkalinity" in the water, which ensures the partial pressure of carbon dioxide inthe gas exiting the absorption column can approach zero (Figure 3C) even with short contact timesin the absorption column. The resulting monovalent carbonate is highly soluble in water, so it doesnot precipitate in the absorption column. However, when contacted with the divalent hydroxide(e.g., magnesium hydroxide), the carbonate precipitates.Na2CO3 + Mg(OH)2 → 2 NaOH + MgCO3K2CO3 + Mg(OH)2 → 2 KOH + MgCO3

[00115] The precipitated carbonate is readily regenerated by thermal decomposition back to thecorresponding oxide, as previously described.

[00116] Additional acid gases that are found in flue gas include SOx (e.g., SO2, SO3) and NOx(e.g., NO2). These acid gases may be removed from the flue gas using conventional technologies;however, if these conventional technologies are not employed, these acid gases will react in theabsorption column as follows:SO2 + H2O → H2SO3SO3 + H2O → H2SO42 NO2 + H2O → HNO3 + HNO2

[00117] These acid gases react with the monovalent alkali as follows:2 NaOH + H2SO3 → Na2SO3 + 2 H2O2 NaOH + H2SO4 → Na2SO4 + 2 H2O2 NaOH + HNO3 + HNO2 → NaNO3 + NaNO2 + 2 H2O

[00118] To recover the sodium ions, the salts can be reacted with calcium hydroxide, as follows:Ca(OH)2 + Na2SO3 → 2 NaOH + CaSO3Ca(OH)2 + Na2SO4 → 2 NaOH + CaSO4Ca(OH)2 + 2 NaNO3 → 2 NaOH + Ca(NO3)2Ca(OH)2 + 2 NaNO2 → 2 NaOH + Ca(NO2)2

[00119] The calcium sulfate and sulfite are poorly soluble and will precipitate. Theseprecipitated salts may be used to amend clay soils to improve fertility. In contrast, the calciumnitrate and nitrite are soluble and must be purged from the circulating water. These salts may beused as nitrogen fertilizer.

[00120] Figure 17 shows Option G, according to an embodiment of the disclosure. Option G islike Option F, except that the alkali is regenerated electrochemically using water-splitting bipolarelectrodialysis. In Figure 17, the electrodialysis system 1758 is simplified; Figure 18 shows thesystem 1758 in greater detail. In Figure 18, the example salt is sodium sulfate (Na2SO4), althoughmany other salts could be employed as well. The electrodialysis system consists of a series ofalternating cation- and anion-selective membranes. When a voltage is applied by the electrodes,ions migrate as shown in Figure 18, including the splitting of water. The total reaction issummarized below:Na2SO4 + 2 H2O → 2 NaOH + H2SO4

[00121] The NaOH and H2SO4 are produced in separate chambers. The NaOH is sent to theabsorption column 1702 where the following reaction occurs:2 NaOH + CO2 → Na2CO3 + H2O

[00122] The sodium carbonate exiting 1718 the absorption column 1702 enters the acid chamberof the electrodialysis system 1758 where the following reaction occurs:Na2CO3 + H2SO4→ Na2SO4 + H2O + CO2

[00123] The carbon dioxide is released as a gas, which is captured and sequestered or utilized.

[00124] Figure 19 shows Option H, according to an embodiment of the disclosure. Option H islike Option A, except that a solvent other than water is employed. Figure 19 shows the solvent ismethanol; however, other solvents may be employed such as ethanol, glycerol, 1,2 propanediol,poly(ethylene)glycol dimethyl ether (Selexol process), dimethyl carbonate, and propylenecarbonate. Because many of these solvents are volatile, it is necessary to prevent them fromescaping into the gas phase. To accomplish this, a solid adsorbent (e.g., activated carbon) isemployed. Two beds 1934 are employed, one that is actively adsorbing and the other that is beingregenerated. Typically, regeneration occurs by heating the bed (temperature swing) or applying avacuum (pressure swing). Two beds 1935 are also shown just before the high-pressure CO2 outlet.

[00125] If a solvent other than water is employed, it is not possible to deploy the alkalinechemistry described previously. In this scenario, a second absorption column would be employedthat contains water with alkali.

[00126] Figure 20 shows a "biological filter," according to an embodiment of the disclosure.The biological filter removes trace amounts of biodegradable contaminants from the vented fluegas. For example, if methanol were used as the physical solvent, trace amounts that escape fromthe adsorbent can be removed in the biological filter. Typically, a biological filter consists of a bedpacked filled with wetted bark chips. The vented gas and air flow through the bark chips. Naturallyoccurring microorganisms establish themselves on the surface of the bark chips and metabolizetrace chemicals in the effluent gas. To ensure that the microorganisms are healthy, small amountsof nutrients (e.g., urea, phosphate) can be added to the water.

[00127] Integrated CO2-capture Processes

[00128] In the previously described add-on processes, compressors are required to pressurizeflue gas, which elevates the partial pressure of CO2 and thereby facilitates its dissolution into thephysical solvent. In addition, expanders are required to recover energy from the remaining high-pressure gas. Unfortunately, the capital and energy costs of the compressors and expanders aresubstantial. These costs can be avoided by integrating CO2 capture into power-productionprocesses that combust fuels. For example, a gas turbine pressurizes air, combusts fuel, andexpands the combustion gases to drive an electric generator. Typically, the combustion pressureis 20 to 60 bar; thus, the resulting CO2 is already at high partial pressure. Of course, it is necessaryto cool the combustion gas before the CO2 can be recovered. Various processes that accomplishthese steps are described as follows:

[00129] Figure 21 shows Option I-1, according to an embodiment of the disclosure. Option I-1is a power-production process that approximates the Ericsson cycle. Option I-1 is very similar toOption A-1. The primary difference is that Option A-1 uses low-temperature waste heat and OptionI-1 combusts fuel (shown generally as 2195) to purposely create high-temperature heat. In OptionI-1, the added air (shown generally as 2197) flows in series through the combustors (one of thecombustors labeled 2188). Some or all the combustors have excess air, which promotes completecombustion.

[00130] Optionally, small amounts of water can be added to the combustors 2188, which coolsthe flame temperature and reduces NOx formation. The water can be added either as a liquid orvapor (steam). Optionally, larger amounts of water can be added to the combustor 2188 such thatstoichiometric, or sub-stoichiometric, amounts of fuel can be combusted without causingexcessively high combustion temperatures.

[00131] In Option I-1, combustion occurs at near-ambient pressure. The hot combustion gasexchanges heat with high-pressure gas that flows through the expanders (2191, 2192, 2193). Inthe final heat exchanger, the combustion gas is cooled sufficiently so that water in the combustiongas condenses as liquid, and it is thereby removed. Ideally, this condensed water is recycled to theoutlet of the heat exchanger, as explained below.

[00132] Water recycle is important because it allows the enthalpy-temperature profiles on eachside of the heat exchanger to nearly match, which reduces irreversibilities and thereby improvessystem efficiency. As shown in Figure 21, the recycled water is supplied directly from the watercondensed from the combustion gas; however, it could come from any source. If sourced from thecondenser, this water will be saturated with carbon dioxide. To reduce carbon dioxide emissionsfrom this source, the pressure of this CO2-saturated water could be dropped, which allows CO2 tobubble out and be recovered. After the CO2 is recovered, the water would be pressurized by apump 2187 and returned to the heat exchanger.

[00133] The cooled semi-dry gas is compressed using a series of compressors (2181, 2182,2183) with intercoolers (2171, 2172, 2173) that maintain near-ambient temperatures in thecompression. Because of the elevated pressure and cool temperature, most of the remaining waterin the gas will condense. The remaining gases enter the absorption column 2102 where CO2 isremoved. The gas exiting 2103 the absorption column 2102 is substantially free of CO2 andcontains primarily nitrogen, argon, and possibly oxygen if excess oxygen is employed in thecombustors. Because most of the water and CO2 have been removed, the mass of this gas mixtureis less than that of the combustion gas. To ensure there is enough thermal mass in the heatexchangers, liquid water and additional high-pressure gas are added to the stream. This additionalhigh-pressure gas could be air; however, as shown in Figure 21, it is recirculated combustion gas,which reduces the oxygen concentration in the heat exchangers.

[00134] The gas exiting the final expansion stage is at near-ambient pressure. It is still hot, soincoming combustion air is preheated (see heat exchanger 2154) to recover thermal energy. Thethermal mass of the incoming air is insufficient to capture all the thermal energy in the outgoinggas, so an additional heat exchanger 2153 can be employed. To optimize energy efficiency, thecaptured waste heat can be used in a "bottoming cycle" that makes additional power.

[00135] As shown, Figure 21 includes a recuperator 2154; however, it could be removed, whichallows more energy to enter the bottoming cycle and thereby make more shaft power.

[00136] As shown in Figure 21, some of the effluent gas is compressed and recycled (e.g., boxlabeled 2155 with compressors and intercoolers), as described previously. The remaining gas isvented to the atmosphere. If the gas contains unreacted fuel or other gases such as carbonmonoxide, optionally it may be sent to a reactor 2156 that completes the oxidation process beforethe gases are vented to the atmosphere. The reactor could be a biological filter (Figure 19) in whichthe oxidant is air. However, other reactors could be employed such as those that use catalysts tofacilitate the reaction. In addition to air or oxygen, other oxidants (e.g., ozone, hydrogen peroxide)could be deployed.

[00137] Figure 22 shows Option I-2, according to an embodiment of the disclosure. Option I-2is similar to Option I-1 except that the added air flows in parallel through the combustors 2288. Iffuel is added in slight excess to each combustor 2288, reducing conditions are maintained in allheat exchangers, which increases the maximum operating temperature of many metals that areused in high-temperature heat exchangers.

[00138] In the multi-step combustion processes shown in Options I-1 and I-2, a key issue is todetermine the number of combustion steps required to achieve stoichiometric combustion of fuel,which will maximize the concentration of carbon dioxide in the gas entering the absorber. Thestandard-state heat of combustion for each fuel is listed below. The inlet air is assumed to contain21 mol% oxygen.Methane CH4 + 2O2 + 2100-21.21N₂ → CO2 + 2 H2O + 2100-2121N₂CH4 + 2 O2 + 7.523 N₂ → CO2 + 2 H2O + 7.523 N₂ΔΗ lower802.5 kJmol CH4X802.5 kJ1 mol CO2 + 7.523 mol N₂94.15 kJmol of 88.3% N2CoalCH0001 + 1.175 O2 + 1.175100-2121N₂ → CO2 +0.45 H2O + 1.175100-2121N₂CH09001 + 1.175 O2 + 4.420 N₂ → CO2 +0.45 H2O + 4.420 N2ΔΗhigher32.962 kJg raw coalX14.5 g CH0.900.10.938 g CH 0.90 0.1509.5 kJmol CH0.900.1ΔΗlower509.5 kJmol CH 0.900.1X0.45 mol H2Omol CH0.900.1X44.02 kJmol H2O489.7 kJmol CH000.1ΔΗ, lower489.7 kJmol CH09001X489.7 kJ1 mol CO2 + 4.420 mol N290.36 kJmol of 81.5% N2Biomass CH15006 + 1.075 0₂ + 1.075100 - xXN₂ → CO₂ +0.75 H2O + 1.075100 - xXN₂CH1.500.6 + 1.075 O₂ + 4.044 N₂ → CO2 +0.75 H2O + 4.044 N2ΔΗ = higher20.75 kJg raw biomassX23.1 g CH1.500.60.994 g CH1.500.6482.2 kJmol CH1.500.6ΔΗ lower482.2 kJmol CH1.500.6X0.75 mol H2Omol CH1.500.6X44.02 kJmol H2O449.2 kJmol CH1.500.6ΔΗ lower449.2 kJmol CH 0.900.1X449.2 kJ1 mol CO2 + 4.044 mol N289.06 kJmol of 80.2% N2

[00139] In the analysis, the following assumptions are employed:• Per-stage pressure ratio = 2 (typical of radial turbomachines)• Combustor = 1200°C• Expander inlet temperature = 1100°C• N2 composition = 81 mol%• CO2 composition = 19 mol%• Isentropic expansion

[00140] Using these assumptions, Figure 23 shows the per-stage enthalpy change, which is theisentropic (theoretical) work that can be produced by each expansion stage. Most of the expansionswill occur at low pressure, so a weighted-average per-stage enthalpy change is about 8 kJ / mol. Ifthe expansions are isentropic, approximately 11 expansion stages are required. Assuming theexpanders are 85% efficient, approximately 13 to 14 expansion stages are required to achievestoichiometric combustion. With a per-stage pressure ratio of 2 and a final expansion pressure of1 bar, the inlet pressure to the expander train must be 214 = 16,384 bar. Clearly, this is not practical.Consequently, the combustors in Options I-1 and I-2 must operate sub-stoichiometric, which iswhy gases (primarily nitrogen) are recycled. Because of gas recycle, the CO2 concentration willbe significantly less than stochiometric, which increases the cost of capturing the carbon dioxide.

[00141] Figure 24 shows Option I-3, according to an embodiment of the disclosure. Option I-3employs a recycle loop to the beginning of the combustor train. The recycle rate is specified sothat the carbon dioxide concentration is approximately the stoichiometric concentration. A purgestream is tapped from the recycle loop, which is sent to the system that captures carbon dioxide.In this embodiment, the recuperator has been removed so that the incoming air is as cool aspossible. This allows more fuel to be combusted and achieve near-stoichiometric carbon dioxideconcentrations. When the recuperator is removed, more thermal energy is available to thebottoming cycle.

[00142] Figure 25 shows Option I-4, according to an embodiment of the disclosure. Option I-4employs a single combustor 2598 with multiple heat exchangers removing heat from thecombustor. The single well-mixed combustor is equivalent to the multi-stage combustor withrecycle (Option I-3). The combustor can operate near-stoichiometrically, which increases thecarbon dioxide concentration and thereby reduces the separation cost. As shown, the recuperatoris included to preheat the combustion air; however, it could be removed and thereby allow morepower to be produced in the bottoming cycle.

[00143] Figure 26 shows Option I-5, according to an embodiment of the disclosure. Option I-5replaces the compression train that provides make-up gas with liquid water. The liquid water mixeswith the gas that has been stripped of CO2 and makes steam in the heat exchanger. This approacheliminates the capital cost of the compressor train and replaces it with an inexpensive pump.Because of the inherent irreversibility associated with vaporizing a liquid (latent heat) with a hotgas (sensible heat), less power is made with this approach; however, the lower capital cost maymake it attractive in some situations. Of course, this option of replacing the compressor train witha liquid water pump can be practiced with any of the embodiments shown herein. Here, it is appliedto Option I-4 as an illustrative example.

[00144] Figure 27 shows Option I-6, according to an embodiment of the disclosure. Option I-6employs series of ambient-pressure combustors with parallel heat extraction. Air and fuel areadded to a series of combustors 2788 in a stepwise manner. As shown, all the air is added to theleftmost combustor and fuel is metered into each sequential combustor 2788 such that thetemperature never exceeds material limits. Alternatively, all the fuel could be added to the leftmostcombustor and air is metered into each sequential combustor. Alternatively, both could be meteredinto each combustor.

[00145] An air compressor 2755 pressurizes air that is fed to heat exchangers 2761 and 2762that heat the compressed air prior to expansion. Because the air contains little moisture and thehot combustion gases never cool enough to condense water, only sensible heat must be transferredbetween combustion gases and air.

[00146] After the combustion is completed to near-stoichiometric conditions, the gas is cooledso that carbon dioxide can be recovered. In this case the temperature of the combustion gas isreduced sufficiently so that water does condense; thus, the heat exchanger exchanges both sensibleand latent heat. To match the enthalpy-temperature profile in the heat exchanger, liquid water mustbe recycled to the heat exchanger.

[00147] As shown, air compression 2755 is accomplished with interstage cooling, which allowsfor highly efficient near-isothermal compression. To preheat the air, a recuperator 2754 isemployed. Alternatively, rather than isothermal compression, air could be compressedadiabatically, which would either eliminate the need for the recuperator, or make it smaller.

[00148] Figure 28 shows Option I-7, according to an embodiment of the disclosure. Option I-7is like Option I-6, except the combustors operate at high pressure.

[00149] Figure 29 shows Option I-8, according to an embodiment of the disclosure. Option I-8employs multi-stage combustion to achieve near-stoichiometric CO2 concentrations in the flue gas.Air is compressed (compressor 2943A) and sent to a combustor 2988A where a portion of thestoichiometric fuel is added. A countercurrent heat exchanger 2944A removes a portion of thethermal energy and transfers it to the stream from which CO2 was removed. Additional thermalenergy is removed in heat exchanger 2945A from the combustion gas exiting heat exchanger2944A and is used to create work in a bottoming cycle. Cooled combustion gas from heatexchanger 2945A is compressed further to a higher pressure in compressor 2943B and combustedwith additional fuel in combustor 2988B. Heat is transferred in heat exchanger 2944B from thecombustion gas exiting combustor 2988B to the stream from which CO2 was removed. Additionalthermal energy is again removed from the combustion gas (heat exchanger 2945B) and is used tocreate work in a bottoming cycle. This process is repeated in compressor 2943C, combustor 2988C,and heat exchanger 2945C, whereby the combustion gas is substantially free of oxygen and thecombined air-to-fuel ratio is near-stoichiometric. After all the useful thermal energy has beenextracted from the flue gas, it is cooled to near-ambient temperature, allowing CO2 to be removed,as in other embodiments. Flue gas from which the CO2 has been removed is reheated in heatexchanger 2945C, expanded in expander 2947C, reheated in heat exchanger 2945B, expanded inexpander 2947B, reheated in heat exchanger 2945A, expanded in expander 2947A, reheatedagainst compressed gas from compressor 2943A, and further cooled in heat exchanger 2953 againsta fluid used in a bottoming cycle.

[00150] Figure 30 shows Option I-9, according to an embodiment of the disclosure. Option I-9is like Option I-8, except the first combustor 3088A operates at ambient pressure, which makes itconvenient for using solid fuels. The remaining combustors (e.g., 3088B, 3088C) operate atelevated pressure where liquid and gaseous fuels are more convenient.

[00151] Figure 31 shows Option J-1, according to an embodiment of the disclosure. Option J-1integrates carbon dioxide capture with a recuperated Brayton cycle. The combustor operates atelevated pressure, so it is most convenient to employ liquid or gaseous fuels, which are easilypressurized. To achieve near-stoichiometric carbon dioxide concentrations, gas is recycled.Because the recycled gas is cold but the compressor discharge is hot, there is some loss inefficiency. This could be eliminated by multi-stage compression with intercooling, as wasemployed in Options I. As described previously, an alternative method for achieving near-stoichiometric carbon dioxide concentrations is to inject water (liquid or steam) into the combustor,which lowers the combustion temperature and thereby allows more fuel to be combusted withoxygen. Additionally, atomized liquid water can be added to the compressor inlet, which helpsachieve near-isothermal compression. Yet another option is to add liquid water (preferablyatomized) to the compressor outlet, which lowers the temperature and thereby allows more heat tobe captured by the recuperator. If sufficient water is added, there is no need to recycle gas. Ofcourse, to achieve near-stoichiometric carbon dioxide concentrations, a hybrid system can beemployed with both gas recycle and water addition. As shown, Figure 31 includes a recuperator3154; however, this could be removed, which increases the amount of energy available to thebottoming cycle 3153.

[00152] Figure 32 shows Option J-2, according to an embodiment of the disclosure. Option J-2is similar to Option J-1; however, in this case, solid fuel (e.g., coal, biomass) is introduced througha lock hopper.

[00153] Figure 33 shows Option J-3, according to an embodiment of the disclosure. Option J-3is like Option J-2; however, in this case, solid fuel (e.g., coal, biomass) is introduced through alock hopper into a high-pressure gasifier. The gases and liquids from the gasifier flow into thecombustor, which operates at elevated temperatures. One advantage of using a gasifier is that itoperates at a lower temperature than a combustor, so the ash may not melt and cause problematicslag.

[00154] Figure 34 shows Option K, according to an embodiment of the disclosure. Option Kwhich is similar to Option J-1, except that a single expander is employed; thus, it is not possible toreheat between the expansion stages. As shown, Figure 34 includes a recuperator 3454; however,this could be removed, which increases the amount of energy available to the bottoming cycle.

[00155] Figure 35 shows Option L-1, according to an embodiment of the disclosure. Option L-1 integrates a steam methane reformer (SMR) 3557 with power production. In this example, theSMR 3557 is integrated with power production Option J-1; however, the SMR 3557 can beintegrated with any of the other power production cycles. Reforming methane with steam is anendothermic reaction, so the reactor must be integrated with a combustor that provides thenecessary heat. The gas produced by the SMR 3557 contains carbon monoxide, which is often nota desired product.CH4 + H2O → CO + 3 H2

[00156] Using the water gas shift (WGS) 3558, carbon monoxide is converted to carbondioxide.CO + H2O → CO2 + H2

[00157] The net reaction is:CH4 + 2 H2O → CO2 + 4 H2

[00158] Because the reactions may not be complete, the final product contains not only CO2,but also CO and CH4. By integrating the process with power production, the energy content inthese unreacted gases is recovered in a productive manner. Furthermore, the CO2 is readilyrecovered in the same equipment that recovers CO2 from the power production process. The fuelconsumption by the SMR is envisioned to be small relative to the fuel consumption in the powerproduction process; therefore, the added cost for capturing carbon dioxide is marginal and takesadvantage of economies of scale. Hydrogen is separated from other gases using pressure swingadsorption (PSA) 3559, membranes 3544, or other suitable means.

[00159] As shown, Figure 35 includes a recuperator 3554; however, this could be removed,which increases the amount of energy available to the bottoming cycle 3553.

[00160] In Option L-1, the combustor operates at elevated pressure, making it most suitable forcombusting gaseous or liquid fuels.

[00161] Figure 36 shows Option L-2, according to an embodiment of the disclosure. Option L-2 integrates a steam methane reformer (SMR) 3657 with power production. In this example, theSMR 3657 is integrated with power production Option I-4; however, the SMR 3657 can beintegrated with any of the other power production cycles. In this example, the objective is tointegrate with a combustor that operates at near-ambient pressure, which is convenient for solidfuels. In particular, if biomass is used as the feedstock, then the entire process is carbon negative.

[00162] Figure 37 shows Option L-3, according to an embodiment of the disclosure. Option L-3 is like Option L-1, except that the product gas is HyCO (primarily a mixture of hydrogen andcarbon monoxide) rather than hydrogen alone. Because it can be used to synthesize manychemicals and fuels, HyCO is also called synthesis gas (syngas).

[00163] To recover HyCO, there are numerous separation options (e.g., pressure swingabsorption, membranes).

[00164] Figure 38 shows Option L-4, according to an embodiment of the disclosure. Option L-4 is like Option L-2, except that the product gas is HyCO (primarily a mixture of hydrogen andcarbon monoxide) rather than hydrogen alone. Because it can be used to synthesize manychemicals and fuels, HyCO is also called synthesis gas (syngas). To recover HyCO, there arenumerous separation options (e.g., pressure swing absorption, membranes).

[00165] Figure 39 shows a process that produces ammonia from hydrogen and nitrogenproduced by steam methane reforming + power production (SMRP), according to an embodimentof the disclosure. For example, Options L-1 or L-2 could be used to produce hydrogen. In yetanother option, a conventional SMR could be combined with a conventional power plant thatemploys add-on carbon capture. Regardless of the power production process, carbon dioxide isseparated, which leaves a residual stream of gas that is primarily nitrogen. The nitrogen andhydrogen are compressed and used to make ammonia. Typically, the per-pass conversion in anammonia synthesis plant is low, so the ammonia product is condensed and harvested. Theuncondensed gas (primarily hydrogen and nitrogen) is recycled to the reactor, where it can react.Because the recycled gas contains small amounts of gases other than hydrogen and nitrogen, apurge stream is required. In this example, the purge stream is returned to the SMRP for furtherprocessing.

[00166] Figure 40 shows a process that produces liquid fuels and chemicals from hydrogen andcarbon monoxide produced by steam methane reforming + power production (SMRP), accordingto an embodiment of the disclosure. For example, Options L-3 or L-4 could be used to producehydrogen and carbon monoxide. In yet another option, a conventional SMR could be combinedwith a conventional power plant that employs add-on carbon capture. Directly from the SMR, theratio of H2 and CO may not be correct for the synthesis of fuels and chemicals; normally, it is toorich in hydrogen. To adjust the CO:H2 ratio, there are a number of options, such as the followingexamples:• Employ a separator (e.g., PSA, membrane) that removes excess hydrogen.• Operate a reverse water gas shift (RWGS) that converts CO2 to CO.• Add CO2 directly to the SMR, which produces CO through "dry reforming"CH4 + CO2 → 2CO + 2H2The fact that the processes described herein produce highly concentrated CO2 is synergistic andprovides opportunities for process integration.

[00167] The most common method for converting syngas is the Fischer Tropsch process;however, there are many other options, such as methanol and ethanol synthesis. In addition toliquid products, it is also possible to produce gaseous olefins, such as ethylene.

[00168] Figure 41 shows Option M-1, according to an embodiment of the disclosure. OptionM-1 employs a that employs effluent gas recirculation (EGR) to increase the CO2 concentration inthe combustion gas to near-stoichiometric levels. Without recirculation, the CO2 concentration inthe effluent gas is about 3 to 5 mol%. In contrast, with EGR, the concentration in dry gas increasesto near-stoichiometric (11.7 mol% CO2 for methane). This higher concentration greatly reducesthe cost of carbon capture by increasing the partial pressure of CO2 in the gas.

[00169] In principle, a side stream is pulled from the combustor, from which the CO2 will beremoved. In practice, the operating temperature of most combustors is too high and exceeds thelimits of most metals that could be employed in heat exchangers. Rather than take a side streamof gas from the combustor, Figure 41 shows the side stream being removed from an intermediatestage in the expander. Optionally, additional fuel can be added to the side stream to ensure thatthe effluent from the combustor is fuel-rich and therefore provides a reducing environment thatincreases the maximum operating temperature of many metals that are employed in heatexchangers.

[00170] As shown in Figure 41, the combined-cycle gas turbine does not employ a recuperator.Although this is the favored configuration in industry, a recuperator could be added to preheat theincoming air.

[00171] Figure 42 shows Option M-2, according to an embodiment of the disclosure. OptionM-2 is identical to Option M-1, except that an additional compression step is added prior to theabsorber.

[00172] Figure 43 shows Option M-3 in which the feed to the power plant is enriched withoxygen according to an embodiment of the disclosure. As shown, oxygen enrichment is performedwith pressure-swing adsorption (PSA); however, other enrichment methods can be employed aswell, such as vacuum-swing adsorption, membranes, cryogenics, addition of oxygen from waterelectrolysis, etc. The following are the equations that describe the stoichiometric reaction ofmethane (CH4), Utah coal (CH0.900.1), and poplar biomass (CH1.500.6) with enriched air:Methane CH₄ + 2 O2 + 2100-XN₂ → CO₂ +2 H₂O + 2X100 - xN2XCoal CH0.900.1+ 1.175 O2 + 1.175100 - x-N₂ → CO2 +0.45 H2O + 1.175X100 - xN₂XBiomass CH1.500.6 + 1.075 0₂ + 1.075100 - xN₂ → CO2 +0.75 H2O + 1.075X100 - xN₂Xwhere x is the percentage of oxygen in the feed air. At stoichiometric conditions, the percentageof carbon dioxide in dry flue gas follows:Methane CO₂ Percentage =11+2100 - x×100%XCoal CO₂ Percentage =11+1.175100X-×100%XBiomass CO₂ Percentage =11+1075 100-x-×100%X

[00173] Figure 44 shows the relationship between the percentage of carbon dioxide in dry fluegas as a function of percentage of oxygen in the feed gas.

[00174] The moles of carbon dioxide per mole of oxygen follows:Methane Ratio =1 mol CO2 mol O22= 0.5mol CO2mol O2X44 g CO₂mol CO2Xmol O232 g O₂= 0.688g COgO22CoalRatio1 mol CO21.175 mol O20.851mol CO2 44 g CO₂mol O2mol O232 g O₂=1.170g CO₂g02Biomass Ratio =1 mol CO1.075 mol O220.930mol CO, 44 g CO,mol O,mol CO22xmol O32 g O₂2=1.279 g CO.,gO2

[00175] Figure 45 shows the energy cost of increasing the oxygen concentration in air usingvacuum-pressure swing adsorption. [4] For example, the energy cost of producing 51% oxygen inair is measured to be about 669 MJ / tonne O2 (186 kWh / tonne O2). In the case of biomass fuel, thisenergy consumption is 145 kWh / tonne CO2. Assuming the grid price of electricity is $0.05 / kWh,the energy cost of concentrating the oxygen is $7.27 / tonne CO2. This energy cost increases theCO2 partial pressure by 2.4 times, which benefits the downstream CO2 separation process byreducing both the size of equipment and energy consumption.

[00176] In the integrated processes described above, power production is combined with CO2-recovery Option A-1. It is understood that power production could be combined with any of theother CO2-recovery options.

[00177] Operating Conditions

[00178] The table below is the operating conditions applicable to the embodiments described.While specific broad, medium, and narrow ranges have been supplied, others may alternatively beused in yet other embodiments.BroadMediumNarrowAbsorber temperature (°C)-40 to 50-30 to 40-10 to 30Absorber pressure (bar)5 to 20010 to 10015 to 60Desorber temperature (°C)0 to 1505 to 10010 to 60Desorber pressure (bar)0.03 to 300.03 to 50.03 to 2Combustor temperature (°C)400 to 2000500 to 1900600 to 1800Combustor pressure (bar)1 to 2001 to 1001 to 60Heat recovery temperature (°C)25 to 130025 to 120025 to 1100Heat recovery pressure (bar)1 to 2001 to 1001 to 60Waste heat temperature (°C)25 to 70030 to 70035 to 700

[00179] What has been described and illustrated herein is an example along with some of itsvariations. The terms, descriptions, and figures used herein are set forth by way of illustration onlyand are not meant as limitations. Many variations are possible within the spirit and scope of thesubject matter, which is intended to be defined by the following claims- and their equivalents—in which all terms are meant in their broadest reasonable sense unless otherwise indicated.

Claims

WHAT IS CLAIMED IS:

1. A method of performing work on a load, the method comprising:combusting a fuel containing carbon and an oxidant in a combustion device to form gaseouscombustion products having a temperature and a pressure;adjusting at least one of the temperature and the pressure of the gaseous combustionproducts to a level suitable for separating the gaseous combustion products;separating the gaseous combustion products into a carbon dioxide-rich stream and a carbondioxide-poor stream, the carbon dioxide-poor stream having a temperature and a pressure;heating the carbon dioxide-poor stream to a temperature suitable for expanding the carbondioxide-poor stream against a load; andexpanding the carbon dioxide-poor stream against the load.

2. The method of claim 1, wherein adjusting at least one of the temperature and thepressure comprises cooling the gaseous combustion products to a temperature suitable forseparating the gaseous combustion products into the carbon dioxide-rich stream and the carbondioxide-poor stream.

3. The method of claim 1, wherein adjusting at least one of the temperature and thepressure comprises compressing the gaseous combustion products to a pressure suitable forseparating the gaseous combustion products into the carbon dioxide-rich stream and the carbondioxide-poor stream.

4. The method of claim 1, wherein the gaseous combustion products are cooled byheat exchange against at least one of the carbon dioxide-rich stream and the carbon dioxide-poorstream.

5. The method of claim 4, wherein the heat exchange occurs within the combustiondevice.

6. The method of claim 1, wherein at least one of the oxidant and the fuel is heated byheat exchange against an expanded gas resulting from expanding the carbon dioxide-poor stream.

7. The method of claim 1, wherein the fuel is coal, biomass, or a gaseous fuel.

8. The method of claim 1, wherein the gaseous combustion products are cooled againsta second gas having a pressure suitable for expansion against a load, whereby the second gas isheated to a temperature suitable for expansion against the load and is expanded against the load.

9. The method of claim 1, wherein the oxidant contains carbon.

10. The method of claim 9, wherein the oxidant comprises effluent from a combustion-fired furnace, a heat engine, a cement kiln, an electric arc furnace, a paper mill, pulp production, asteam methane reformer, or a combustion-fired boiler.

11. The method of claim 1, wherein the fuel contains oxides of carbon.

12. The method of claim 11, wherein the fuel contains carbon dioxide.

13. The method of claim 11, wherein the fuel comprises effluent from a blast furnace,an oil refinery, a steam methane reformer, a chemical production process, an anaerobic digester, alandfill, or a treatment plant.

14. The method of claim 1, wherein separating the gaseous combustion productscomprises dissolving carbon dioxide in a solvent.

15. The method of claim 14, wherein the solvent is one of water or methanol.

16. The method of claim 14, wherein the solvent is water.

17. A heat engine comprising:a compressor configured to compress at least one of an oxidant inducted into the heat engineand gaseous combustion products;a combustion device configured to combust the oxidant with a fuel containing carbon tocreate the gaseous combustion products containing carbon dioxide;a carbon dioxide separator configured to separate the gaseous combustion products tocreate a carbon dioxide-rich stream and a carbon dioxide-poor stream; andan expander configured to expand the carbon dioxide-poor stream.

18. The heat engine of claim 17, further comprising a cooling device configured to coolthe gaseous combustion products.

19. The heat engine of claim 17, further comprising a heating device configured to heatat least one of the carbon dioxide-poor stream and the carbon dioxide-rich stream.

20. The heat engine of claim 17, further comprising a heat exchanger configured totransfer heat from an expanded stream resulting from the expander to at least one of the oxidantand the fuel.

21. A method of performing work on a load, the method comprising:cooling an effluent from a first process, the effluent having a first pressure and a firsttemperature greater than 150°C, to a second temperature less than the first temperature;compressing the cooled effluent to a second pressure greater than the first pressure to createan expansion gas;reheating the expansion gas against the effluent; andexpanding the reheated expansion gas against the load.

22. The method of claim 21, wherein the first temperature is greater than 250°C.

23. The method of claim 21, wherein the expansion gas is separated into a carbondioxide-rich gas and a carbon dioxide-poor gas, and the carbon dioxide-poor gas is heated againstthe effluent and expanded against the load.

24. The method of claim 23, wherein the separation is performed by absorbing carbondioxide into a solvent.

25. The method of claim 24, wherein the solvent is at least one of water, an aqueoussolution, Rectisol, Selexol, or potassium carbonate.

26. The method of claim 21, wherein the second pressure is greater than 40 bar.

Images