Target gas capture process and system using supersonic flow

EP4658392A4Pending Publication Date: 2026-06-24SCOPRA SCI & GENIE SEC

Patent Information

Authority / Receiving Office
EP · EP
Patent Type
Applications
Current Assignee / Owner
SCOPRA SCI & GENIE SEC
Filing Date
2024-02-23
Publication Date
2026-06-24

Smart Images

  • Figure CA2024050222_29082024_PF_FP_ABST
    Figure CA2024050222_29082024_PF_FP_ABST
Patent Text Reader

Abstract

A system for capturing a target condensable gas may include: a supersonic condensation nozzle configured to receive a gas flow having a target condensable gas content, and for accelerating said gas into a supersonic gas flow, at least one particle injector in the supersonic nozzle for injecting particles for the particles to be in the supersonic gas flow, whereby target gas condensate forms onto some of the particles, an expansion device defining a conduit having a convex corner causing the flow of the particles with target gas condensate to generate expansion waves to separate the flow into a first flow path having an increased concentration of particles with target gas condensate thereon and into a second flow path of gas with lowered target condensable gas content, and a divider having at least a first conduit, the divider configured to collect the first flow path.
Need to check novelty before this filing date? Find Prior Art

Description

TARGET GAS CAPTURE PROCESS ANDSYSTEM USING SUPERSONIC FLOWCROSS-REFERENCE TO RELATED APPLICATION

[0001] The present application claims the priority of United States Patent Application No. 63 / 447,743, filed on February 23, 2023, the contents of which are incorporated herein by reference.TECHNICAL FIELD

[0002] The disclosure relates to a capture process for a target gas, such as CO2, using a supersonic flow, and to associated equipment.BACKGROUND

[0003] Capture of carbon dioxide (CO2), such as from industrial processes, remains a global challenge for environmental reasons. For example, a typical 500 MW coal-fired power plant entails a mass flow of released CO2 (CO2 concentration of 12 %voi) of about 50’000 kg / h (=440’000 tonne / yr). For an industrial CO2 capture system, the volume of gas to be processed is therefore substantial. Various processes, technologies and equipment have been developed to capture CO2, for example for its storage, utilization, or sequestration. Some technologies use solid sorbents, liquid sorbents or cryogenic CO2 separation.

[0004] In another application, for capturing CO2 directly from the atmospheric air (CO2 concentration of 410 ppm), a device to capture one metric ton of CO2 per hour, with a capture percentage of 50%, would require the air flow rate to be treated to be about 2’600’000 m3 / h.

[0005] Some gas separation systems use supersonic flow where gas temperature drops significantly, allowing one vapour component to liquefy to be subsequently inertially separated. Such systems typically use a vortex flow for inertial separation and are commonly used to separate water vapour and heavy hydrocarbons from natural gas during extraction. Noticeably, such processes and systems are well suited to achieve separation with hydrocarbons due to their relatively high molecular weight, but such processes and systems may not be appropriate to perform the separation of molecules, such as CO2, which condensates at a lower temperature than water vapor.

[0006] Two phenomena exist for CO2 condensation onset: homogeneous and heterogeneous nucleation. Homogeneous nucleation spontaneously results in nano-sized CO2 particulates, of insufficient mass and size for inertial separation. Heterogeneous nucleation occurs at a higher local temperature and pressure than homogeneous nucleation due to the presence of foreign nuclei. For example, water vapour already present in a gas mixture may solidify, via homogeneous nucleation, at a higher temperature than CO2 and form nano-sized particles onto which heterogeneous CO2 nucleation may occur. The nucleation of CO2 may subsequently be followed by particle growth due to CO2 accumulation.

[0007] Therefore, there exists a need for improving carbon dioxide capture methods by exploiting condensation and inertial separation of particles.SUMMARY

[0008] In a first aspect, there is provided a method for capturing a target condensable gas comprising: accelerating a gas flow having a target condensable gas content in a nozzle into a supersonic gas flow, injecting particles for the particles to be in the supersonic gas flow, whereby target condensable gas condensates onto some of particles, causing the flow of the particles with target gas condensate to generate expansion waves to separate the flow into a first flow path having an increased concentration of particles with target gas condensate thereon and into a second flow path of gas with lowered target condensable gas content, and physically separating the first flow path from the second flow path.

[0009] Further in accordance with the first aspect, for example, the first flow path may be heated to separate the target gas condensate from the particles.

[0010] Still further in accordance with the first aspect, for example, the second flow path is exhausted to the environment.

[0011] Still further in accordance with the first aspect, for example, the gas flow having a target condensable gas is compressed prior to said accelerating.

[0012] Still further in accordance with the first aspect, for example, the gas having CO2 content is collected prior to said compressing.

[0013] Still further in accordance with the first aspect, for example, injecting particles in the supersonic gas flow includes injecting the particles in a converging segment of the nozzle.

[0014] Still further in accordance with the first aspect, for example, injecting particles in the supersonic gas flow includes injecting the particles in a throat segment of the nozzle.

[0015] Still further in accordance with the first aspect, for example, injecting particles in the supersonic gas flow includes injecting the particles in a diverging segment of the nozzle.

[0016] Still further in accordance with the first aspect, for example, heat may be recuperated from the second flow path of gas.

[0017] Still further in accordance with the first aspect, for example, causing the flow of the particles with target gas condensate to generate expansion waves and physically separating the first flow path from the second flow path is performed in a first separation stage, and wherein the method further includes at least a second stage using the first flow path with the increased concentration of particles with target gas condensate, the second stage including: causing the first flow path to generate expansion waves to separate the flow into a third flow path having an increased concentration of particles with target gas condensate and into a fourth flow path of gas with lowered target condensable gas content, and physically separating the third flow path from the fourth flow path.

[0018] Still further in accordance with the first aspect, for example, heat may be absorbed from the gas flow having a target condensable gas content upstream of the nozzle.

[0019] In accordance with a second aspect, there is provided a system for capturing a target condensable gas comprising: a supersonic condensation nozzle configured to receive a gas flow having a target condensable gas content, and for accelerating said gas into a supersonic gas flow, at least one particle injector in the supersonic nozzle for injecting particles for the particles to be in the supersonic gas flow, whereby target gas condensate forms onto some of the particles, an expansion device defining a conduit having a convex corner causing the flow of the particles with target gas condensate to generate expansion waves to separate the flow into a first flow path having an increased concentration of particles with target gas condensate thereon and into a second flow path of gas with lowered target condensable gas content, and a divider having at least a first conduit, the divider configured to collect the first flow path.

[0020] Further in accordance with the second aspect, for example, the divider includes a second conduit, the second conduit configured to collect the second flow path.

[0021] Still further in accordance with the second aspect, for example, a heating unit may be configured to heat the first flow path to separate the target gas condensate from the particles.

[0022] Still further in accordance with the second aspect, for example, the divider has an exhaust to direct the second flow path to the environment.

[0023] Still further in accordance with the second aspect, for example, the second conduit includes a heat exchanger to reclaim heat from the second flow path.

[0024] Still further in accordance with the second aspect, for example, at least one compressor may compress the gas flow having a target condensable gas content upstream of the supersonic nozzle.

[0025] Still further in accordance with the second aspect, for example, a conduit may collect the gas flow having a target condensable gas content and direct same to the at least one compressor.

[0026] Still further in accordance with the second aspect, for example, a heat exchanger may be in the conduit to absorb heat from the gas flow having a target condensable gas content.

[0027] Still further in accordance with the second aspect, for example, the expansion device and the divider form a first stage, the system including at least a second stage receiving the first flow path having the increased concentration of particles with target gas condensate, the second stage including: a second expansion device defining a conduit having a convex corner causing the flow of the particles with target gas condensate thereon to generate expansion waves to separate the flow into a third flow path having an increased concentration of particles with target gas condensate thereon and into a fourth flow path of gas with loweredtarget condensable gas content, and a second divider having at least a first conduit, the divider configured to collect the third flow path.

[0028] In accordance with a third aspect, there is provided a supersonic condensation nozzle comprising: an inlet end and an outlet end; a conduit between the inlet end and the outlet end configured for receiving a fluid flow having a target condensable gas content, a geometry of an interior of the conduit defining, sequentially, from the inlet end to the outlet end: a converging segment in which cross-sectional dimensions decrease in a direction of the fluid flow for subsonic flow, a diverging segment in which cross-sectional dimensions increase in a direction of the fluid flow for supersonic flow, and a throat segment between the converging segment and the diverging segment; wherein, based on fluid flow parameters as a function of the geometry, a condensation onset region is located in the diverging segment, in which condensation onset of the target condensable gas occurs; and wherein a diverging gradient of the diverging segment is greater at the condensation onset region than downstream of the condensation onset region, relative to the fluid flow.

[0029] In accordance with a fourth aspect, there is provided a device for separation of particles from a supersonic particle-laden gas flow comprising: an inlet end and an outlet end; a conduit between the inlet end and the outlet end for gas flow, the conduit having a convex corner on a conduit surface of the conduit causing the particle-laden gas flow to generate expansion waves originating from the convex corner to separate the fluid flow at the outlet into a first flow path having an increased concentration of the particles near a conduit surface opposite the convex corner, and a second flow path of gas with lowered content of the particles near the conduit surface featuring the convex corner.

[0030] In a fifth aspect, there is provided a system that may include the supersonic condensation nozzle of the third aspect and / or the device of the fourth aspect, for capturing a target condensable gas comprising: a supersonic condensation nozzle configured to receive a gas flow having a target condensable gas content, and for accelerating said gas into a supersonic gas flow, at least one particle injector in the supersonic nozzle for injecting particles for the particles to be in the supersonic gas flow, whereby target gas condensate forms onto some of the particles, an expansion device defining a conduit having a convex corner causing the flow of the particles with target gas condensate to generate expansion waves to separate the flow into a first flow path having an increased concentration of particles with target gas condensate thereon and into a second flow path of gas with lowered target condensable gas content, and a divider having at least a first conduit, the divider configured to collect the first flow path.DESCRIPTION OF THE DRAWINGS

[0031] Reference is now made to the accompanying figures in which:

[0032] Fig. 1 is a schematic view of a process and system for the capture of target gas CO2 using a supersonic flow in accordance with an embodiment of the present disclosure.

[0033] Fig. 2 is a graph showing a CO2 phase diagram with isentropic expansion curves.

[0034] Fig. 3 is a schematic view of exemplary dimensions of a basic conical convergent- divergent nozzle.

[0035] Fig. 4 is a graph showing the effect of the supersonic condensation nozzle design Mach number on the final achieved CO2 concentration.

[0036] Fig. 5 is an exemplary graph showing the effect of the supersonic condensation nozzle design Mach number on the gas-particle heat transfer coefficient.

[0037] Fig. 6 is a schematic view showing an exemplary optimized supersonic condensation nozzle in accordance with the present disclosure prior to CO2 condensation onset versus a basic conical supersonic condensation nozzle.

[0038] Fig. 7 is an exemplary graph showing the flow characteristics of the optimized supersonic condensation nozzle prior to CO2 condensation onset versus the basic conical supersonic condensation nozzle of the schematic view of Fig. 6.

[0039] Fig. 8 is an exemplary graph showing the phase diagram of the optimized supersonic condensation nozzle prior to CO2 condensation onset based on dTg / dx = 3°C / m versus the basic conical supersonic condensation nozzle of the schematic view of Fig. 6.

[0040] Fig. 9 is an exemplary graph showing the mass rate of deposited CO2 with respect to the reduction in flow total pressure.

[0041] Fig. 10 is an exemplary graph showing the phase diagram of the optimized supersonic condensation nozzle prior to CO2 condensation onset based on dTg / dx = 0°C / m versus the basic conical supersonic condensation nozzle of the schematic view of Fig. 6.

[0042] Fig. 11 is an exemplary graph showing the phase diagram of the optimized supersonic condensation nozzle prior to CO2 condensation onset based on dTg / dx = -0.3°C / m versus the basic conical supersonic condensation nozzle of the schematic view of Fig. 6.

[0043] Fig. 12 is an exemplary graph showing the flow characteristics of the basic conical supersonic condensation nozzle with particle injection at nozzle inlet versus controlled injection within the divergent section.

[0044] Fig. 13 is an exemplary graph showing the flow total pressure of the basic conical supersonic condensation nozzle with particle injection at nozzle inlet versus controlled injection within the divergent section.

[0045] Fig. 14 is an exemplary graph showing the phase diagram of a 100000 tonne / yr CO2 design case.

[0046] Fig. 15 is an exemplary graph showing the axial CO2 concentration of the 100000 tonne / yr CO2 design case.

[0047] Fig. 16 is an exemplary graph showing the axial particles diameter evolution of the 100000 tonne / yr CO2 design case.

[0048] Fig. 17 is an exemplary graph showing the axial total pressure drop of the 100000 tonne / yr CO2 design case.

[0049] Fig. 18 is an exemplary schematic view of the basic form of a Prandtl-Meyer expansion device.

[0050] Fig. 19 is an exemplary schematic view of the use of a Prandtl-Meyer expansion device combined with a flow divider to remove particles from a supersonic flow.

[0051] Fig. 20 is an exemplary schematic view illustrating a possible arrangement of multiple Prandtl-Meyer expansion devices.

[0052] Fig. 21 is an exemplary full-scale overview of a CO2 capture plant based on the technology of the present disclosure.

[0053] Fig. 22 is cross-section of an exemplary axisymmetric (3D) design of the supersonic condensation nozzle, Prandtl-Meyer expansion device and flow separation device assembly.

[0054] Fig. 23 illustrates the presence of a particle-free zone within a Prandtl-Meyer expansion device.DETAILED DESCRIPTION

[0055] For a process and system designed to capture a target gas such as CO2 using condensation of the target gas and subsequent inertial separation of particles, the optimization challenge is related to the following aspects. The process and related system and devices taught herein apply to other target gases as well, whereby similar principles as taught herein for CO2 apply to other target gases. The most common industrial gaseous contaminants include CO2, SOx, H2S, NOx and others. For simplicity, reference is made herein to CO2, but this includes other target gases. The expressions “target gas” and “target condensable gas” are used herein to describe the fact that the process and system target such gas as being the one to be isolated and separated / removed from a gas flow.

[0056] According to the process described herein, rirst, particles may be injected into the flow to promote heterogeneous nucleation. The selected particle diameter and density should be sufficiently large so as to permit efficient inertial separation, namely over short distances while minimizing total pressure losses. On the other hand, for the same mass flow rate of injected particles, the smaller and lighter the particles, the more surface area is available for CO2 deposition, and the smaller the drop in total pressure of the flow due to particle entrainment. In addition, the particle heat capacity should be small enough to minimize differences with the gas temperature along the flow and accelerate the condensation onset.However, if the particles have a larger heat capacity, they retain their low temperature while accumulating the latent heat of fusion from the solidified CO2, which promotes further CO2 deposition. The injected particles may be in a solid or in a liquid phase, the latter of which may solidify or not before the CO2 condensation onset.

[0057] In addition, the supersonic condensation nozzle geometry (i.e., its length, cross- sectional area variation along its length, and / or shape) into which the flow is accelerated is to be designed optimally to minimize the total pressure drop, and therefore energy and cost required to operate the system, due to friction and particle drag as well as achieving the optimal pressure-temperature conditions for CO2 condensation.

[0058] The location and conditions for particle injection may be either: within the main gas stream before the supersonic condensation nozzle; or at any position within the supersonic condensation nozzle, such as before the CO2 condensation onset, but at a temperature and velocity that may be generally equal to those of the gas at the injection point. After a sufficient quantity of CO2 has been condensed on injected particles, then some form of inertial separation must be performed to isolate the condensed CO2 from the CC>2-depleted gas flow.

[0059] The present disclosure will describe herein a process and system for which the captured gas is CO2. However, the process 10 and related system apply to other gases as well, whereby similar principles as taught herein for CO2 apply to other gases. The most common industrial gaseous contaminants include CO2, SOx, H2S, NOx and others. For simplicity, the present disclosure uses CO2 as the captured gas for the rest of the present disclosure. The present system includes a convergent-divergent nozzle, or other type of supersonic condensation nozzle, where the CCh-laden gases expand to the CO2 condensation conditions, followed by an inertial separation of solid condensed CO2 particles using a Prandtl-Meyer expansion, the collection of solid condensed CO2 particles, then a diffuser to decelerate the gas streams to ambient conditions.

[0060] Injecting micrometric sized particles represents a main principle of the present disclosure. This creates the appropriate well-defined surface (i.e., condensation nuclei) for heterogeneous nucleation of CO2 as well as permitting inertial separation. Referring to the figures and more particularly to Fig. 1 , there is illustrated a CO2 capture process 10 using a supersonic flow with particle injection. The particles provide the surface on which CO2 condenses. In a variant, the particles may be solid materials, i.e., in a solid phase in the operating conditions of the process. The particles may be spherical, tetrahedric or irregularly shaped, as examples among others. The particle diameters may range from, but are not limited to, 0.1 to 50 microns. Exemplary solid materials may be, but are not limited to, silica, activated silica gel, graphite, super activated carbon, titanium dioxide, aluminum oxide, silicon carbide, barium carbonate, magnesium oxide, sundry mining waste, ice, dry ice, or liquids such as water, hydrocarbons, liquid nitrogen.

[0061] The process 10 may collect gas from an industrial process, as shown at A. The collected gas, having a CO2 content, may be compressed at B, using any appropriate compressor(s). The compressor may be adiabatic, or isothermal, or any other polytropic process where compressor cooling is performed. In the isothermal case the compression work may be reduced by around 40% compared to the adiabatic case. The recuperated heat from cooling the compressor may be reclaimed as a waste heat source for various purposes allowing for approaching a net-zero operating energy demand for the CO2 capture system. For example, the reclaimed heat may be used in the industrial process A, or in any other process or equipment requiring heat, e.g., space heating, water heating, process heat, etc.

[0062] Recalling Fig. 1, the process 10 is illustrated schematically and includes at least three sequential steps, such as a step 20 of generating a supersonic flow, a step 30 of performing an inertial separation using a convex corner that generates expansion waves in a Prandtl-Meyer expansion, and a step 40 of collecting CO2-covered particles and decelerating the flow to ambient conditions, which may also be regarded as two steps. Step 40 may also include dividing the output from the Prandtl-Meyer expansion device 31 into solid condensed CO2 particles (e.g., in duct 41 A) and a gas with low CO2 content (e.g., in duct 41 B). While the expression “steps” is used herein, a step may be a substep, and / or a step may include various steps or substeps. For example, as described below, the step 20 of generating a supersonic flow may include a step or substep of performing particle injection, and / or a step or substep of compressing the gas containing the CO2. Moreover, process 10 may optionally include the step of collecting the industrial gas in the industrial process, as shown as A, and / or compressing the collected gas using the compressor(s) at B.

[0063] In step 20, CO2 contained in the gas, such as atmospheric air or combustion products, for example, is condensed so as to agglomerate in a solid state on micrometric particles. This may be achieved by accelerating a pressurized gas flow containing CO2 to a supersonic velocity. The step 20 may also include injecting solid particles in the flow, such as micrometric particles. These particles may be injected at any point prior to conditions that permit condensation being reached, including but not limited to: an upstream plenum, a converging section of the supersonic condensation nozzle, a throat of the supersonic condensation nozzle, and / or a diverging section of the supersonic condensation nozzle upstream of the condensation. It should be noted that exhaust gases from industrial process A may also already contain micrometric particles of the desired size range and in sufficient quantity. When the gas is accelerated, it expands and cools down. Under such conditions, the CO2 contained in the flow changes phase to condense to solid if it attains a suitable temperature; for example, the CO2 sublimation temperature is about -80°Celsius (193 Kelvin) at atmospheric pressure. As an example, as shown in the phase diagram of Fig. 2, the sublimation temperature of CO2 is a function of pressure; at a pressure of 101 kPa, the sublimation curve is at about -80° Celsius (193 Kelvin). This pressure is specifically the partial pressure of CO2 in the main gas stream which is a function of the CO2 concentration level. The condensing of the CO2 to solid stateoccurs at the surface of the micrometric particles that have been previously injected in the flow in step 20. As a result, at the exit of this step 20, there is obtained a supersonic flow of gas depleted from gaseous CO2 which carries the particles covered by solid state CO2.

[0064] In step 30, an inertial separation of the particles occurs. The inertial separation occurs according to the Prandtl-Meyer expansion wave, in which the supersonic flow is exposed to a convex corner to generate essentially stationary and steady two-dimensional or three-dimensional expansion waves. The supersonic gas flow is deviated by the expansion waves, while the micrometric particles preserve their trajectory due to their inertia. Particles can be efficiently separated by this technique if the product of their density times the square of their diameter is sufficiently large. Accordingly, the particles covered by CO2 gather in the exterior path of the convex corner. A magnetic field may be applied, along with particles with appropriate magnetic properties, in order to enhance the separation efficiency. Alternatively, the particles may be electrically charged, either prior to injection or by their interaction with the flow, and an electric field could apply an electrostatic force, or a transversal magnetic field may apply a Lorenz force, to aid in the separation of the particles from the gas flow. The number of Prandtl- Meyer expansion devices either in series or in parallel may be optimized for attempting to reduce total system losses and increase separation efficiency to aim to increase final CO2 purity.

[0065] As part of step 30, the inertial separation uses Prandtl-Meyer expansion waves instead of shock waves or vortices, in contrast to known processes. The use of Prandtl-Meyer expansion results in a cooling of the gas, while shock waves and vortices typically increase gas temperature. By cooling the gas, there is a reduced risk of vaporizing CO2 that is on the surface of the micrometric particles.

[0066] In step 40, some of the particles covered by CO2 are removed from the gas flow. One approach taken in step 40 to perform the separation is to divide the flow in two or more trajectories, such as via two conduits as schematically shown in Fig. 1, divider 41. The divider 41 may also be referred to as divider device, separation device, and may be a pair of duct diverging from a single duct. In a variant, the divider 41 is a pair of separate ducts, conduits, etc, toward which the output from one or more Prandtl-Meyer expansion devices 31 is directed. A first conduit may be for the gas mostly depleted of CO2, and a second conduit is for the gas flow having a high content of particles covered with solid CO2. The two conduits would have the form of a supersonic diffuser in order to decelerate the supersonic flow gradually until stagnation conditions are achieved, and to recover the total pressure of the flow. This deceleration process may be accompanied by shock waves. In a variant, the supersonic diffuser(s) may have the form of a convergent-divergent nozzle but with an adjustable throat to allow for efficient supersonic flow start-up. The supersonic diffuser is to be designed carefully in order to minimize total pressure losses, such as through minimizing the amplitude of the aforementioned shock waves. Accordingly, at the exit of the process 10, there may result a gas with a high content of particles covered with solid CO2. Gas depleted from CO2 may be outputto the environment or may be treated in other ways. In another embodiment of the present disclosure, a turbine may be used, for example in duct 41 B, so as to recuperate work that can be used to drive the compressors at B, as one example of energy reclaim among others.

[0067] The process 10 may include other steps to store, use, or sequestrate the CO2 in solid state on particles. In a variant, the process 10 may perform a step of separation, by which the solid CO2 is separated from the particles by sublimating CO2 from its solid state to a gas state. One considered approach to cause the sublimation is to increase the temperature of the particles above the sublimation temperature of CO2 at about -80° Celsius (193 Kelvin) at atmospheric pressure. This is performed basically during the pressure recovery where the particle-rich gas is decelerated and its temperature augments. The heat of sublimation can be withdrawn from the compressor resulting in a cooling of the compressor and reduction in the compressor work (i.e., approaching isothermal compression). The same can be done, for example, for cooling the compressed gases exiting from the compressor and prior to step 20, cooling the flue gases from the industrial process or directly cooling the particles before reinjecting them in the process. The particles can be otherwise exposed to ambient conditions to allow the CCh to sublimate. The particles may then be recuperated to be reinjected in the step 20 of the process 10. This is one possible outcome for the CO2 in solid state on particles exiting step 40.

[0068] Still referring to Fig. 1 , as part of the step 20, a supersonic condensation nozzle 21 is used, with one or more injection devices 22. The supersonic condensation nozzle 21 has a particular geometry and dimensions as a function of the entering total pressure, temperature, gas flow rate, CO2 concentration, injected micrometric particles diameter, density, mass fraction and heat capacity. The supersonic condensation nozzle geometry here means the axial profile, cross-sectional variation and length. This is optimized as presented hereinafter in order to maximize the amount of captured CO2 versus the total pressure drop across the system which represents the amount of work of the system. The optimization of the supersonic condensation nozzle geometry and the injection parameters are based on mathematical models that integrate simultaneously nucleation and growth models, cross-section variations, wall friction and heat transfer, micrometric particle injection, particle drag and heat transfer. The supersonic condensation nozzle also has a geometry aiming to avoid expansion wave reflection and / or reflection of compression waves and / or the formation of shock waves.

[0069] The total pressure drop within the supersonic condensation nozzle may be due to: particle entrainment, wall friction and heat transfer, and latent heat liberated from CO2 condensation. For a targeted amount of captured CO2, the latent heat would be a fixed value. The wall frictional effects are reduced by increasing the supersonic condensation nozzle scale, i.e., passing the whole gas flow through one large supersonic condensation nozzle instead of dividing the flow into several smaller supersonic condensation nozzles. The remaining loss factor to be minimized is particle entrainment. For a certain particle size and mass fraction, this can be performed either by modifying the supersonic condensation nozzle geometry just afterthe condensation onset to decrease the difference between the particle and gas velocities and / or temperatures, or by injecting the particles just before the condensation onset but at a temperature and velocity generally equal to those of the gaseous phase.

[0070] As illustrated in Fig. 2, an exemplary phase diagram for CO2 is provided, including the flow pressure and temperature, and illustrates zones in which CO2 is solid, liquid and gaseous. At ambient temperature and pressure, CO2 is in a gaseous state. When accelerated to supersonic flow, as in step 20 of process 10, pressure and temperature drop. As presented from Fig. 2, the gas is subject to an almost isentropic expansion based on pressure and temperature conditions, as a function of the flow Mach number. CO2 condensation occurs when the gas is expanded and attains the solid phase zone of the phase diagram. Practically speaking, the condensation onset of CO2 to solid occurs as the properties of CO2 cross the extrapolation 505 of the liquid / vapor curve (V-L line) in Fig. 2.

[0071] The step 20 of process 10 relies on solid condensation to assist in capturing the CO2. In designing the components of the system used for process 10, the nucleation rate and growth rate must be considered. Heterogeneous nucleation rate may be affected by temperature, free energy of droplet formation, particle size, surface tension and / or saturation level. Growth rate may be affected by the size of the particles, the partial pressure of CO2, the flow and particles temperature, the molar volume of CO2, as well as the sticking coefficient. CO2 condensation is an exothermic process, therefore, the gaseous CO2 at the outset must release its latent heat to the surrounding gas to change phase and condense to a solid. This thermal transfer results in a temperature increase for both the gas and the particles, and this temperature increase may slow down or prevent condensation.

[0072] To achieve suitable CO2 condensation, the condensation rate must be above a given threshold. In contrast to prior systems that do not take into consideration condensation control, process 10 and associated system must control condensation to have an efficient capture of CO2. It may be considered to control condensation and to promote nucleation onset by changing the sectional dimensions and / or the cross-sectional area along the length of the supersonic condensation nozzle. The process 10 and associated system may allow CO2 condensation control that is more suited for the capture of CO2, namely by the injection of the micrometric particles into the gas flow. Indeed, the particles injected in the flow act as nucleation sites and increase the useful surface for condensation. In order to significantly increase the useful size, the injected particles must be small, and hence are preferably sized in the micrometric scale. The injection of particles may increase drag because the gas must entrain these particles that are injected at a slower velocity than that of the gas. Due to the use of a supersonic condensation nozzle in step 20, the gas is continuously accelerated within the supersonic condensation nozzle. The drag forces associated with particle injection result in a pressure drop and lowering of the Mach number. Moreover, friction results in temperature increase and this has a negative effect on condensation. Hence, particle injection must take into consideration these factors so as to maintain suitable condensation in step 20.

[0073] The injection of particles also enhances the inertial separation as the particles are denser than gas, and thus have greater inertia, leading to the separation of the particles using the expansion waves in step 30. However, optimal particle size must be used, as larger particles that could facilitate inertial separation may require more energy to be accelerated and may have less of a useful surface for condensation. Accordingly, injection parameters such as the number of particles, injection velocity, uniformity of the injection, must be controlled and the characteristics of the particles as well, such as particle size, density, temperature and surface tension.

[0074] The first aspect of the supersonic condensation nozzle is its design Mach number. Considering a simple convergent-divergent nozzle having a certain length 301 as shown in Fig. 3, the design Mach number will indicate the ratio between the exit area 302 and the throat area 304 (i.e., nozzle area ratio). The supersonic condensation nozzle may be axisymmetric or in 2- D rectangular form or may combine a transition from axisymmetric to 2-D rectangular form or the inverse. From the gas-dynamical point of view, as the design Mach number increases, the area ratio increases and the exit temperature, pressure and density decrease. The decrease in the gas temperature should promote CO2 condensation at a first glance. However, the decrease in the gaseous phase density reduces the heat transfer coefficient between the particles and the gas. Because of this, as the gas becomes cooler, the particle temperature, where CO2 condensation occurs, would no longer match the gas temperature.

[0075] The effect of design Mach number on exit CO2 concentration, therefore capture efficiency, is presented in Fig. 4. For a certain supersonic condensation nozzle length and particle properties, starting with a 15% CO2 concentration in this example, the exit CO2 concentration decreases firstly at 401, to a minimum value 402 at a design Mach number of 6. Increasing the design Mach number further would no longer improve the performance 403 as the particles would become less influenced by the surrounding gas. At an extreme design Mach number of 15, no condensation of CCh is predicted. It should be noted that at elevated design Mach numbers, the residence time of the particles within the supersonic condensation nozzle decreases due to the increased particle velocities. This contributes also to the reduced quantity of captured CO2. A comparison of the formerly mentioned heat transfer coefficient between the design Mach numbers of 6 (point 402) and 10 (point 404) is presented in Fig. 5. The heat transfer coefficient 410 in the case of Mach 10 drops by 10 to 40 times than that of 406 in the case of Mach 6 along the divergent part of the supersonic condensation nozzle due to the decrease in gas density. It should be noted that after exploring the design space, the optimal design Mach number may range from 2 to 10, preferably from 3 to 8.

[0076] Reducing the particle drag effect can be performed by optimally designing the supersonic condensation nozzle profile. The objective is to reduce the mismatch between the particles and gas velocities. The modification in supersonic condensation nozzle geometry can be applied before the condensation onset, for example from the beginning of the supersonic condensation nozzle or starting from the throat. This can, however, delay the condensationonset while minimizing drag effects. The other scenario is performing the supersonic condensation nozzle geometry optimization after the condensation onset. For this case, the supersonic condensation nozzle geometry is kept unchanged from a base, non-optimized, configuration until reaching the condensation onset as soon as possible. Afterwards, the supersonic condensation nozzle geometry is modified gradually to decrease the rate at which the gas is accelerated downstream of the supersonic condensation nozzle, thus to reduce the mismatch between the particles and gas velocities. This is performed, however, without excessively influencing the particle temperature as the particles are cooled by the surrounding gas. If the particle temperature is immoderately increased, the condensation rate of CC>2will be negatively affected. A first example is previewed in Fig. 6: a simple base conical, 4.5 m long, convergent-divergent nozzle 501 is compared to an optimized 4.64 m supersonic condensation nozzle 502. In this case, the supersonic condensation nozzle throat position 550 is at 0.5 m and the condensation onset 503 is observed at around 1 .75 m downstream from the supersonic condensation nozzle inlet as previewed in the flow characteristics of Fig. 7. The supersonic condensation nozzle profile optimization 502 may be performed starting from the 2.2 m position 504 from the supersonic condensation nozzle inlet (i.e. , 0.45 m after condensation onset). The condition to optimize the supersonic condensation nozzle geometry in this case is to set the gas axial temperature gradient to be dTg / dx = 3°C / m, where Tg is the gaseous phase temperature. The gas axial temperature gradient defines the rate at which the gas temperature is allowed (i.e., controlled) to increase incrementally along the length of the supersonic condensation nozzle, with dTg / dx = 3°C / m of the example above representing an increase in gas temperature at a rate of 3°C each meter. The phase diagram for this case is illustrated in Fig. 8. The onset conditions are located at 520 while the conditions at the start of nozzle geometry modification position are located at 525. The increase in the gas temperature 512 for an optimized supersonic condensation nozzle (e.g., 502 in Fig. 6) can be observed if compared with the basic case 511 of a simple conical supersonic condensation nozzle (e.g., 501 in Fig. 6). When the gas state approaches the extrapolation of the V-L line 515, this could result in stopping the CO2 condensation. The rate of CO2 condensation with respect to the increase in the total pressure (dm / dPo) change is represented in Fig. 9. The rate values 900 start at the onset position of condensation 503 (Fig. 6) at 1.75 m. Identical results exist just before the supersonic condensation nozzle profile optimization takes place at the 2.2 m position 504 (Fig. 6). Afterwards, a gain is observed 901 due to the beneficial reduction in particle drag losses. It should be noted that the optimized supersonic condensation nozzle 502 (Fig. 6) according to this value of gas axial temperature gradient is slightly longer in order to produce the same final amount of captured CO2. For this case, the recovery pressure ratio across the optimized supersonic condensation nozzle 502 (Fig. 6) is 18.3% better than that of the simple conical supersonic condensation nozzle 501 (Fig. 6). The recovery pressure ratio here is defined as the ratio of the final total pressure to the initial total pressure. This leads to a 7.3% reduction in the associated compression cost for the optimized supersonic condensation nozzle 502 (Fig. 6) when compared to the supersonic condensation nozzle 501 (Fig. 6).

[0077] The supersonic condensation nozzle may therefore be described as having an inlet end and an outlet end. A conduit is between the inlet end and the outlet end. The conduit configured for receiving a fluid flow having a target condensable gas content (e.g., CO2). A geometry of an interior of the conduit defines, sequentially, from the inlet end to the outlet end: a converging segment in which cross-sectional dimensions decrease in a direction of the fluid flow for subsonic flow, a diverging segment in which cross-sectional dimensions increase in a direction of the fluid flow for supersonic flow, and a throat segment between the converging segment and the diverging segment. Based on fluid flow parameters as a function of the geometry, a condensation onset region (e.g., starting at 503) is located in the diverging segment, in which condensation onset of the target condensable gas occurs. A diverging gradient of the diverging segment is greater at the condensation onset region than downstream of the condensation onset region, relative to the fluid flow, such as at shown at 504. The diverging gradient at the condensation onset region may be said to differ from the diverging gradient downstream of the condensation onset region, by a step in the value of diverging gradient, i.e. , it is not gradual, but it may be gradual in some variants. The diverging gradients described above may be two frustocones arranged sequentially. In an embodiment, there is an edge at the transition. The injection of particles may be in any of the segments, to facilitate heterogeneous nucleation, i.e., converging segment, throat segment, diverging segment. In a variant, the injection is performed downstream of the condensation onset region.

[0078] Exemplary dimensions are provided for the supersonic condensation nozzle, the supersonic condensation nozzle may be axisymmetric or 2-D nozzle, for example, with Fig. 21 showing an example at 806 of an axisymmetric nozzle. In a variant, the minimum size of the throat segment (e.g., diameter) may vary between 0.01 m and 0.50 m, depending on the initial conditions, but other dimensions are considered. In such an embodiment, the Mach number at the outlet end may be between 2 and 10, inclusively, though other Mach numbers are possible. In a variant, for a 2D nozzle, the diverging gradient may be a half-cone angle that is between 3 and 70 degrees, inclusively. In a variant, for an axisymmetric nozzle, the diverging gradient may be a half-cone angle that is between 1 and 48 degrees, inclusively. The modified optimized supersonic condensation nozzle half-cone angle may vary from zero (i.e., constant area duct) to an angle less than or equal to the precedingly specified cone or divergence angles for the basic unmodified supersonic condensation nozzle. The half-cone angle may be defined as being from a central axis of the cone (i.e., a right-angle cone) to a side wall. In a variant, an outlet area (e.g., cross-section dimensions) is from 1.7 to 536 times the smallest throat area (e.g., cross-section dimensions). The inlet area (e.g., cross-section dimensions) may depend on the system piping and may range from 2 to 10 times the smallest throat area (e.g., crosssection dimensions). The convergent section length of the supersonic condensation nozzle may vary be between 1% and 10% of the total nozzle length, inclusively. For a modified optimized supersonic condensation nozzle geometry at the onset point or after the onset point, i.e., the condensation onset region, the distance between the throat section and the start ofmodification of the nozzle profile may be between 10% and 60% of the divergent segment length, inclusively. The modified supersonic condensation nozzle divergent part may be as a straight cone or with a dA / dx which varies from one position to another (i.e., gradual non-linear increase) depending on the required design conditions (i.e., dTg / dx, du / dx, etc.) The transition from the basic supersonic condensation nozzle profile to the modified one at the onset point or after the onset point may be sudden or gradual.

[0079] Another possibility is shown at 506 in Fig. 10 for a 10 m long supersonic condensation nozzle. In this case 506, the design condition after condensation onset is set to be for a gas axial temperature gradient of dTg / dx = 0°C / m. A 30.1% enhancement in the recovery pressure ratio is realized reducing by 8% the compression cost, compared to a base conical supersonic condensation nozzle. A final case 507 is provided in Fig. 11 for a 6 m long supersonic condensation nozzle. Here, a theoretical risk of reaching the extension of the V-L extrapolation line 505 is observed, despite setting the gas axial temperature gradient at dTg / dx = -0.3°C / m. In this case the gas temperature is to be decreased after supersonic condensation nozzle optimization position at 525 but at a smaller rate than for the basic conical supersonic condensation nozzle 511. In fact, the operation line of the basic supersonic condensation nozzle 511 of this case is closer to the V-L extrapolation line 505 due to the smaller particles employed. This increases the surface area for heat transfer between the particles and the gas and leads to relatively hotter gas. The expected increase in the recovery pressure ratio is 20.6% for this case, with 5.6% reduction in the compression cost.

[0080] A summary of the previous cases parameters is presented in Table 1 below. It should be noted that the previous cases are scaled for a 100,000 tonne CCh / year 15% CO2 concentration by volume flue gases source. Therefore, it is shown that the efficiency of the supersonic condensation nozzle for promoting CO2 condensation while minimizing pressure losses can be improved by properly matching the gaseous temperature axial gradient dTg / dx to the other parameters of the process such as: supersonic condensation nozzle geometry (profile and length), pre-existing suspension of liquid-solid-gas composition, gas total pressure and temperature, humidity, particle size, density, specific heat, mass ratio and CO2 concentration within the entering gas. Optimizing the supersonic condensation nozzle geometry to achieve a specific target for the gaseous temperature axial gradient can lead to important improvements in total pressure losses for the same quantity of condensed CO2.

[0081] Similarly, in another embodiment, supersonic condensation nozzle geometry can further be optimized by adjusting the gaseous axial temperature gradient dTg / dx as a function of axial position instead of having a constant value. This is changing dTg / dx from one axial position to another.

[0082] In another embodiment, supersonic condensation nozzle optimization can be achieving by controlling other parameters, such as the gas velocity u axial variation, the gases axial velocity gradient du / dx, the gases axial pressure gradient dP / dx or the gases axial densitygradient dp / dx, or a combination thereof, along the length of the supersonic condensation nozzle.T able 1 : Summary of cases 1 , 2 and 3 operating parameters using graphite particles with a mass fraction of 0.3.

[0083] Reducing the losses due to having to accelerate the injected micrometric particles can be achieved by injecting the particles somewhere within the supersonic condensation nozzle instead of injecting the particles before the nozzle. The particle temperature and velocity should be, however, as close as possible to that of the gases. The particles can be injected within the convergent part of the supersonic condensation nozzle (i.e., before the throat), or within the divergent part either before or after the theoretical condensation onset.

[0084] Recalling Case 1 in Table 1 , a comparison of the flow characteristics is presented in Fig. 12, showing the case 521 of ordinary particles injection within the entering gas suspension versus particle injection 601 within the divergent part of the supersonic condensation nozzle. In this case the injection position 600 is set at 3 m downstream from supersonic condensation nozzle inlet (i.e., around 1.5m downstream of the theoretical condensation onset location 513) to produce the same final CO2 concentration as the ordinary scenario 521 for comparison purposes. As seen in Fig. 12, the gas flow expands at 601 and cools down firstly without any frictional losses due to particle drag, however frictional losses with the walls remain. Then at location 600 in the divergent part of the supersonic condensation nozzle, particles are injected to provide the surface on which CO2 condenses and solidifies. The injected particles temperature and velocity should be equal to those of the gaseous phase at the injection position 600 in order to avoid sudden gas heating, drag losses and possible local shock wave formation which may delay or prevent CO2 condensation. It is also observed that due to reduced particle drag losses, the theoretical condensation onset location 513 is around 0.5 m earlier in the scenario of controlled particles injection position than the onset position 503 (Fig. 6) of the ordinary scenario. The total pressure profiles for ordinary scenario 602 and controlled particles injection position scenario 603 are shown in Fig. 13. This may result in a 120% enhancement and 35% reduction in the recovery pressure ratio and related cost, respectively. Table 2 illustrates further comparisons between both scenarios of particles injection for cases 2 and 3. It should be noted that the supersonic condensation nozzle profile optimization (based on gaseous temperature axial gradient or other parameter or combination thereof) prior to condensation onset 502 (Fig. 6) along with controlling particles injection position scenario 601 (Fig. 12) can be performed at the same time to benefit from both techniques.Table 2: Comparison between ordinary particles injection at supersonic condensation nozzle entry and controlled particle injection within the divergent part of the nozzle for cases 1, 2 and 3.

[0085] The previously mentioned cases show the enhancement potential of the supersonic CO2 capture by employing the presented approaches. However, optimization is dependent on a particular application or situation to provide the most efficient full-scale CO2 capture-separation system. As an example, a design case is presented in Table 3 using a simple conical 4.5 m long supersonic condensation nozzle with 3-micron graphite particles. The phase diagram, the axial CO2 concentration, particles diameter and flow total pressure are provided in Fig. 14, Fig. 15, Fig. 16 and Fig. 17, respectively. It should be noted from Fig. 16 that the particle diameter increases significantly until it reaches 4.4 microns in this case, due to CO2 deposition, which may be beneficial for the design of the separation section.Table 3: 100’000 tonne / yr design case example.

[0086] Still in Fig. 1, the inertial separation step 30 relies on the expansion of the supersonic gas flow, which contains particles covered by solid CO2, to accelerate the gas flow and deviate it using Prandtl-Meyer expansion waves. The particle streamlines thus diverge from the gaseous ones and converge towards the opposite side of the duct, where a flow divider is inserted at a specific height to separate the particle-rich flow. This is performed in an expansion device 31 featuring a convex corner 31A, the expansion device 31 being referred to as Prandtl- Meyer expansion device 31. The corner of the expansion device can be gradual, smooth or sharp, as opposed to being rounded. With the convex corner 31 A being sharp, Sthere may be generated centered steady expansion waves originating from the sharp convex corner. Thisdevice is shown in more detail in Fig. 18. Opposite to the convex corner 31 A is the upper wall 31 B, and adjacent and downstream of the convex corner 31 A is the lower wall 31 C. The upper wall 31 B may be designed to avoid the reflection of the expansion waves 601 and attempt to reduce the risk of a shock wave flow before, during, and / or after separation. A particle 602, covered with solid CO2, follows another path 603 different from that of the gas 604 due to its inertia, impacting the upper wall 31 B at a position 605. The particle 602 is a representation for at least one particle in the flow suspended within the gas flow. The particles may then be captured using a variety of methods including but not limited to, a porous medium which the particles can pass through, an electric field, a magnetic field, a moving belt, an aspiration wall and / or a liquid film along the upper wall 31 B and / or a combination thereof. It bears mentioning that upper and lower are relative to the orientation of the device 31 in Fig. 18. The upper wall is the wall opposite the convex corner, while the lower wall is adjacent to the convex corner. However, in use, upper and lower may not be aligned with gravity, and upper wall could instead be referred to as distal or opposite wall and lower wall could be proximal or adjacent wall, for example.

[0087] One possible separation method, exhibited in Fig. 19 is a flow divider 606, placed relative to the impact zone 610 so as to divide the duct into two separate ducts 41 A and 41 B, with a differing concentration of particles between the two. Ideally, duct 41A contains a higher concentration of particles than duct 41 B. The separated particles trajectories and the accompaniment of process gas with these particles affect the final obtained CO2 purity. The flow divider 606 position and shape are optimized to maximize the collecting efficiency of the particles and / or the CO2 purity of collecting gas. This depends on the impact zone 610 or the trajectory (i.e. , the impact position) of the last particle 611. The flow divider 606 can be positioned before, at, or after the impact position 605. This is a compromise between the purity of the captured CO2 from the particle-rich gas flow and the total percentage of separated particles (i.e., captured CO2). A shock wave 607 may be present on one side of the flow divider 606 without influencing the particles trajectory before separation. This shock can be principally an oblique shock either attached or detached. Depending on the shape of the flow divider and the flow characteristics, this shock wave may appear on either or both sides of the flow divider, but preferably not in duct 41 A which contains a higher concentration of particles.

[0088] Still referring to Fig. 19, the use of a flow divider may be combined with the implementation of other separation techniques, such as but not limited to, an electric field, a magnetic field, a moving belt, an aspiration wall and / or a liquid film along the upper wall 31 B, or any combination thereof. Injecting a liquid film along the upper wall 31 B may be performed before and / or after the impact position 605 and / or downstream within the separation duct 41 A. This is to decrease the rebound of the particles by forming a surface with damping properties. This technique can also prevent process gas from being aspirated with the particles because the particles may penetrate the liquid film. This may lead to an increased final purity of CO2. Theliquid film may be implemented elsewhere within the whole system such as the supersonic condensation nozzle and / or the impact zone 610 and / or the ducts 41 A and 41 B and / or the downstream diffuser in order to decrease abrasion effects on the walls caused by the impact of particles.

[0089] The Prandtl-Meyer expansion device 31 (Fig. 1 ) allows the formation of a generally particle-free zone 1101 along the lower straight wall of the duct following the convex corner 31 A, as shown in Fig. 23. The height H away from the lower wall 31 C of this particle-free zone 1101 depends on both the properties of the flow (pressure, entry Mach number, exit Mach number) and of the particles (size, shape, density). For example, larger, denser particles will generate a larger particle-free zone, as will a lower pressure flow. The upper wall 31 B of the Prandtl-Meyer expansion device 31 and the flow divider device 41 may then be placed in relation to the particle- free zone so as to optimize CO2 purity and / or capture efficiency. In an embodiment where particle rebound is minimized from the upper wall 31 B, for example with a liquid film, the upper wall 31 B may be placed so that all or part of it may align with the border of the particle-free zone. In another embodiment where particle rebound is non-negligible, the upper wall 31 B may be placed significantly within the particle-free zone to be able to account for the particle rebound.

[0090] The dimensions of the Prandtl-Meyer expansion device 31 (Fig. 1 ) also depend on the Mach number and cross-sectional area at the supersonic condensation nozzle exit. However, the design of the supersonic condensation nozzle is dependent on the size of the particle-free zone. Specifically, the height of the exit of the supersonic condensation nozzle or the height of the entrance to the Prandtl-Meyer expansion device 31 may not be significantly varied for given flow conditions without reducing the performance of the Prandtl-Meyer expansion device 31. Thus, in an embodiment of the system using a rectangular cross-section, the width of the Prandtl-Meyer expansion device 31 (the out-of-plane dimension in Fig. 18) must be adjusted such that the cross-sectional area matches that of the exit of the supersonic condensation nozzle. Since frictional effects increase with surface area, it is thus advantageous to choose flow conditions and particle types that allow for the entrance to the Prandtl-Meyer expansion device 31 to approach a square cross-section. However, given that these parameters have an influence on the performance of the condensing subsystem as well, they must be optimized together.

[0091] Table 4 shows examples of certain calculated examples of Prandtl-Meyer expansion device 31 configurations for different particle types. These are calculated for a flow rate of 14.6 kg / s and a total pressure at the entry of the Prandtl-Meyer expansion device 31 of 1.5 bar. Although these cases are calculated with spherical graphite particles, other particle types can preserve the same aerodynamic properties. In cases 1 , 2, 3, the same particles are employed and enter the Prandtl-Meyer expansion device 31 at the same Mach number. Comparing cases 1 and 2, it becomes clear that increasing the final Mach number allows less process gas to becaptured along with the particles. Cases 2 and 3 have the same final Mach number, but different heights. Case 2 therefore minimizes the percentage of process gas captured with the particles, augmenting the final CO2 purity, whereas Case 3, with a larger height at the entrance of the Prandtl-Meyer expansion device 31 , allows for a more square cross-section, minimizing friction effects at the cost of more captured process gas. Cases 4 and 5 are conducted with 10 micron graphite and thus necessitate smaller heights at the entrance of the Prandtl-Meyer expansion device 31 to match the height of the particle free zone. Case 6 with 50 micron graphite allows for the optimal square cross section at the entrance of the Prandtl-Meyer expansion device 31 , whereas Case 7 with 3 micron graphite is near the limit of the size of particle that can be feasibly separated, with a height at the entrance of the Prandtl-Meyer expansion device 31 of 5 mm.Table 4: Separator parameters for 100’000 tonnes / yr design case.

[0092] For a Prandtl-Meyer expansion device 31 such as described above, particles of a size varying between 1 and 100 microns may be used, with material densities between 500 kg / m3and 19200 kg / m3The particle-gas stream may enter the Prandtl-Meyer expansion device 31 at a Mach number between 2 and 5 and may attain a Mach number of 3 to 6 after passingover the convex corner 31 A. In a Prandtl-Meyer expansion device 31 with a rectangular crosssection, the convex corner 31 A flow deflection angles alpha may vary between 2 and 85 degrees. For a flow entering the Prandtl-Meyer expansion device 31 with a total pressure of 1.5 bar, the height of the entrance to the Prandtl-Meyer expansion device 31 may be between 5 mm and 0.5 m.

[0093] The dimensions of the Prandtl-Meyer expansion device 31 , such as width and convex corner 31 A flow deflexion angle alpha, may depend on the exit Mach number and height 706 (i.e., area) at the supersonic condensation nozzle exit. As discussed above, the height 706 depends on the mass flow rate per nozzle, which should be maximized in order to minimize the frictional effects. However, there exists a relaxation length after which the particles trajectories 603 follow the gas streamlines 604 (i.e., the particles no longer become inertially separated). The relaxation length depends on the difference between the particle and gas velocities as well as the particle size and density.

[0094] The Prandtl-Meyer expansion device 31 may be described as being a device for separation of particles from a supersonic particle-laden gas flow, i.e., not necessarily used with a supersonic condensation nozzle. It may be said to have: an inlet end and an outlet end; a conduit between the inlet end and the outlet end for gas flow, the conduit having a convex corner on a conduit surface of the conduit causing the particle-laden gas flow to generate expansion waves originating from the convex corner to separate the fluid flow at the outlet into a first flow path having an increased concentration of the particles near a conduit surface opposite the convex corner, and a second flow path of gas with lowered content of the particles near the conduit surface featuring the convex corner.

[0095] The particle concentration at the impact zone 610 is mostly affected by the relaxation length of the particles within the flow. If the length of the Prandtl-Meyer expansion device 31 is small with respect to the relaxation length, the particle concentration at the impact zone 610 will be sufficiently high and the separation efficiency will be acceptable. However, for a longer Prandtl-Meyer expansion device 31 , there exists a possibility where the upper wall 31 B and the flow divider 606 may be geometrically positioned after the relaxation length. Thus, the particle concentration at the impact zone 610, and thereby separation efficiency, would be significantly reduced. A possible solution to this problem is dividing the Prandtl-Meyer expansion device 31 into two or more shorter Prandtl-Meyer expansion devices. The greater the number of smaller entrance height Prandtl-Meyer expansion devices 31 (correspondingly larger in transverse width to accommodate the required flow rate), the more capability of separating finer particles they would have, but the more the pressure drop would be inherent.

[0096] In another embodiment of the present disclosure, two or more Prandtl-Meyer expansion devices 31 are placed in a parallel cascading arrangement as illustrated in Fig. 20. For illustrative purposes, a Prandtl-Meyer expansion device 31 similar to the device illustratedin Fig. 19 is shown, although other Prandtl-Meyer expansion devices 31 may be used in a similar parallel cascading arrangement, with associated dividers. In Fig. 20, three Prandtl-Meyer expansion devices 701 , 702 and 703 are shown for illustrative purposes, although any number of Prandtl-Meyer expansion devices may be used according to the overall dimensions of the system. The separating wall between ducts is placed such that the leading edge 704 is on the first Mach wave 705 of each Prandtl Meyer expansion device 701, 702, 703, so that each Prandtl-Meyer expansion device functions independently from each other. Weak shock waves may be attached to the edge 704 depending on its thickness and without influencing the global flow. In such an embodiment, the ducts depleted of particles 701A, 702A, and 703A may be connected together downstream of the separation device, as may the ducts containing concentrated CO2 701 B, 702B, 703B. These ducts may deviate out-of-plane from the 2D plane shown in Fig. 20 to accomplish this. This embodiment permits an optimization of capture efficiency by allowing the dimensions of each Prandtl-Meyer expansion device to be chosen according to the relaxation length of the particles and / or the height of the particle-free zone, independent of the dimensions of the exit duct 706 exiting from the supersonic condensation nozzle.

[0097] In another embodiment of the present disclosure, the final cross-sections of the supersonic condensation nozzle may be flattened to a rectangular section, allowing for matching with the aspect ratio of the cross-section of the entrance of the Prandtl-Meyer expansion device. Thus, the in-plane dimensions of the exit of the supersonic condensation nozzle as presented in Fig. 19 may be adjusted according to the relaxation length and / or the height of particle-free zone of the particles within the flow, while the out-of-plane dimension may be adjusted to allow the necessary flow rate of the system. This embodiment may also be combined with other presented embodiments described herein, such as the embodiments described in the previous paragraph.

[0098] As a result of the above described condensation and separation processes, CO2 is mostly maintained in a solid state and carried by particles, yet it is concentrated in a flow path, while a flow depleted from particles and from CO2 follows another path, for subsequent physical separation in step 40, for instance by flow separation device 41 that may have a first conduit 41 A and a second conduit 41 B. A filter may be installed further downstream the particle-rich separated flow to complete the particle separation process. Another scenario is using a cyclonic separator or similar device to separate the solid particles from the CO2-rich gas.

[0099] While the various devices of the system operating the process 10 are shown as discrete components in Fig. 1, the components may be arranged to form a continuous conduit separating in a pair of conduits or more in the flow divider 41. Accordingly, it may be said that the supersonic condensation nozzle 21 , the Prandtl-Meyer expansion device 31 and the flow divider 41 are segments or sections of a conduit system.

[0100] Therefore, the process 10 can generally be described as being for capturing a gaseous contaminant such as CO2, and may include accelerating a gas having CO2 content in a nozzle into a supersonic gas flow, injecting particles in the supersonic gas flow, whereby solid CO2 condensates onto particles, causing the flow of the particles with solid CO2 to generate expansion waves to separate the flow into a first flow path having an increased concentration of particles with solid CO2 and into a second flow path of gas with lowered CO2 content, and physically separating the first flow path from the second flow path.

[0101] The process 10 may operate using a system that can generally be described as having a supersonic condensation nozzle configured to receive a gas having CO2 content, and for accelerating said gas into a supersonic gas flow, at least one particle injector in the supersonic condensation nozzle for injecting particles in the gas flow, whereby solid CO2 condensates onto particles, a Prandtl-Meyer expansion device defining a conduit having a convex corner causing the flow of the particles with solid CO2 to generate expansion waves to separate the flow into a first flow path having an increased concentration of particles with solid CO2 and into a second flow path of gas with lowered CO2 content, and a flow separation device having a first conduit and a second conduit, the flow separation device configured to receive the first flow path and the second flow path and physically separate same in the first conduit and the second conduit.

[0102] A conceptual full-scale overview of an exemplary CO2 capture plant using the process and the system of the present disclosure is presented in Fig. 21 . The CC>2-laden gases entering at 801 may be at a temperature higher than the ambient temperature. The CC>2-laden gases may be cooled, if necessary, through a first heat exchanger 802 designed to handle the gas flow rate and cooling capacity. The CC>2-laden gases to may be cooled to ambient temperature, or to a temperature between the ambient temperature and the original gas temperature. The first heat exchanger 802 may be water-cooled or air-cooled, notably by having an appropriate heat-exchange coil. Outdoor air may optionally be used, notably in cooler days of cooler months. Heat may be reclaimed from the heat exchanger 802 for any appropriate heat load. A compressor stage 803, having one or more compressors 803 in any appropriate arrangement (parallel, cascaded) pressurizes the CCh-laden gases to the required pressure level before the supersonic expansion. The pressurized gases, which may subsequently have an elevated temperature, may be cooled down once again via a second heat exchanger 804. Again, heat may be reclaimed from the heat exchanger 804 for any appropriate heat load. The compressor 803 and the second heat exchanger 804 may be a series of compressors and heat exchangers in series and / or in parallel to handle and / or control a wide range of flow rates and / or realize an isothermal compression with minimized compression work. The recuperated heat from the heat exchangers 802 and / or 804 and / or others may be redirected towards a waste heat recovery application. The injector 805 may be placed to inject the micrometric particles. Injector 805 may be placed at the intake plenum or within any segment or section of supersoniccondensation nozzle 806. The supersonic condensation nozzle 806 may be a single nozzle, or multiple nozzles in parallel to handle and / or control a wide range of flow rates. Particles with solid CO2 are separated through expansion device 807, that may include a divider. The Prandtl- Meyer expansion device 807 with divider may be constituted of a single and / or a series of Prandtl-Meyer expansion devices. The particle-rich flow exits from the 2D plane through an outlet, shown as being a duct 808. The duct 808 may be designed to decelerate the particle-rich flow to near ambient pressure. The separated particles are collected inside a reservoir 809, as a possibility among others. The reservoir 809 may allow the sublimation of CO2 and / or implement this heat of sublimation for cooling purposes. The reservoir 809 may be preceded and / or replaced by a cyclone device to separate the solid particles and / or possible liquid water from the CCh-rich gas flow 810. The recuperated particles 811 may be redirected back to the particles injector 805. The recuperated particles at 811 may be dried from humidity using a heater. The CCh-rich gas flow 810 may be subjected to further purification processes. The particle-lean gas flow 813 exiting from the Prandtl-Meyer expansion device 807 with divider may be decelerated to ambient pressure level using a diffuser duct 812.

[0103] The final achieved CO2 purity can be controlled and / or enhanced by further processing the exit CO2-rich gas 810. This may be performed by coupling two or more CO2 capture systems as the one shown in Fig. 21 in series. Further purification may be performed using cryogenic CO2 purification and / or other gas separation techniques.

[0104] Still referring to Fig. 21 , the supersonic condensation nozzle 806, the Prandtl-Meyer expansion duct 807, the outlet ducts 808 and 812 may be axisymmetric (3D) and / or in 2-D form and / or a transition from one to another. The supersonic condensation nozzle 806 and / or Prandtl- Meyer expansion device 807 with divider can be repeated consecutively in series with or without intermediate particles injection so separate gaseous constituents other than CO2, such as but not limited to, water vapor, SOx, H2S, etc. In the case of separating gaseous constituents other than CO2 the temperature of the reservoir 809 may be controlled to sublimate sequentially each gaseous constituent, each at its corresponding sublimation temperature.

[0105] One possible axisymmetric (3D) design is presented in Fig. 22 combining a supersonic condensation nozzle 1001 having its throat at 1002, a Prandtl-Meyer expansion device 1003 with a convex corner 1004 and a flow separation device 1005, a diffuser duct 1006 for CO2-lean gas 1007 having its throat at 1008 and a diffuser duct 1009 for CCh-rich gas with particles 1010 having its throat at 1011. Such an axisymmetric design with a center-body may be advantageous as it offers a small height over a large perimeter for the Prandtl-Meyer expansion device 1003.

[0106] The off-design operation of the system can occur when one and / or more of the design conditions are varied, such as but not limited to, gas flow rate, compressed gas pressure, compressed gas temperature, CO2 initial concentration, particle diameter, particle density andparticle mass loading or fraction. Mass flow rate control can be achieved by using an adjustable supersonic condensation nozzle throat and / or varying the inlet gas pressure prior to compression. During the startup of the supersonic flow and / or to control the flow properties, particularly the Mach number, at different stations of the whole system, a variation of the geometric area ratio (i.e., supersonic condensation nozzle local area to throat area) may be necessary. This variation of the geometric area ratio may be performed by adjusting the supersonic condensation nozzle throat area, displacing the supersonic condensation nozzle walls by using actuators and / or by introducing bleed ports at different stations of the system.

[0107] The above description is meant to be exemplary only, and one skilled in the art will recognize that changes may be made to the embodiments described without departing from the scope of the invention disclosed. Still other modifications which fall within the scope of the present invention will be apparent to those skilled in the art, in light of a review of this disclosure, and such modifications are intended to fall within the appended claims.

Claims

CLAIMS1. A method for capturing a target condensable gas comprising: accelerating a gas flow having a target condensable gas content in a nozzle into a supersonic gas flow, injecting particles for the particles to be in the supersonic gas flow, whereby target condensable gas condensates onto some of particles, causing the flow of the particles with target gas condensate to generate expansion waves to separate the flow into a first flow path having an increased concentration of particles with target gas condensate thereon and into a second flow path of gas with lowered target condensable gas content, and physically separating the first flow path from the second flow path.

2. The method according to claim 1 , further including heating the first flow path to separate the target gas condensate from the particles.

3. The method according to any one of claims 1 to 2, further including exhausting the second flow path to the environment.

4. The method according to any one of claims 1 to 3, further including compressing the gas flow having a target condensable gas prior to said accelerating.

5. The method according to claim 4, further including collecting the gas having a target condensable gas content prior to said compressing.

6. The method according to any one of claims 1 to 5, wherein injecting particles in the supersonic gas flow includes injecting the particles in a converging segment of the nozzle.

7. The method according to any one of claims 1 to 5, wherein injecting particles in the supersonic gas flow includes injecting the particles in a throat segment of the nozzle.

8. The method according to any one of claims 1 to 5, wherein injecting particles in the supersonic gas flow includes injecting the particles in a diverging segment of the nozzle.

9. The method according to any one of claims 1 to 8, including recuperating heat from the second flow path of gas.

10. The method according to any one of claims 1 to 9, wherein causing the flow of the particles with target gas condensate to generate expansion waves and physically separating the first flow path from the second flow path is performed in a first separation stage, and whereinthe method further includes at least a second stage using the first flow path with the increased concentration of particles with target gas condensate, the second stage including: causing the first flow path to generate expansion waves to separate the flow into a third flow path having an increased concentration of particles with target gas condensate and into a fourth flow path of gas with lowered target condensable gas content, and physically separating the third flow path from the fourth flow path.11 . The method according to any one of claims 1 to 10, including absorbing heat from the gas flow having a target condensable gas content upstream of the nozzle.

12. A system for capturing a target condensable gas comprising: a supersonic condensation nozzle configured to receive a gas flow having a target condensable gas content, and for accelerating said gas into a supersonic gas flow, at least one particle injector in the supersonic nozzle for injecting particles for the particles to be in the supersonic gas flow, whereby target gas condensate forms onto some of the particles, an expansion device defining a conduit having a convex corner causing the flow of the particles with target gas condensate to generate expansion waves to separate the flow into a first flow path having an increased concentration of particles with target gas condensate thereon and into a second flow path of gas with lowered target condensable gas content, and a divider having at least a first conduit, the divider configured to collect the first flow path.

13. The system according to claim 12, wherein the divider includes a second conduit, the second conduit configured to collect the second flow path.

14. The system according to claim 12 or claim 13, further including a heating unit configured to heat the first flow path to separate the target gas condensate from the particles.

15. The system according to claim 13, wherein the divider has an exhaust to direct the second flow path to the environment.

16. The system according to claim 13, wherein the second conduit includes a heat exchanger to reclaim heat from the second flow path.

17. The system according to any one of claims 12 to 16, further including at least one compressorto compress the gas flow having a target condensable gas content upstream of the supersonic nozzle.

18. The system according to claim 17, further including a conduit to collect the gas flow having a target condensable gas content and direct same to the at least one compressor.

19. The system according to claim 18, further including a heat exchanger in the conduit to absorb heat from the gas flow having a target condensable gas content.

20. The system according to any one of claims 12 to 19, wherein the expansion device and the divider form a first stage, the system including at least a second stage receiving the first flow path having the increased concentration of particles with target gas condensate, the second stage including: a second expansion device defining a conduit having a convex corner causing the flow of the particles with target gas condensate thereon to generate expansion waves to separate the flow into a third flow path having an increased concentration of particles with target gas condensate thereon and into a fourth flow path of gas with lowered target condensable gas content, and a second divider having at least a first conduit, the divider configured to collect the third flow path.

21. A supersonic condensation nozzle comprising: an inlet end and an outlet end; a conduit between the inlet end and the outlet end configured for receiving a fluid flow having a target condensable gas content, a geometry of an interior of the conduit defining, sequentially, from the inlet end to the outlet end: a converging segment in which cross-sectional dimensions decrease in a direction of the fluid flow for subsonic flow, a diverging segment in which cross-sectional dimensions increase in a direction of the fluid flow for supersonic flow, and a throat segment between the converging segment and the diverging segment; wherein, based on fluid flow parameters as a function of the geometry, a condensation onset region is located in the diverging segment, in which condensation onset of the target condensable gas occurs; and wherein a diverging gradient of the diverging segment is greater at the condensation onset region than downstream of the condensation onset region, relative to the fluid flow.

22. A device for separation of particles from a supersonic particle-laden gas flow comprising: an inlet end and an outlet end; a conduit between the inlet end and the outlet end for gas flow, the conduit having a convex corner on a conduit surface of the conduit causing the particle-laden gas flow to generate expansion waves originating from the convex corner to separate the fluid flow at the outlet into a first flow path having an increased concentration of the particles near a conduit surface opposite the convex corner, anda second flow path of gas with lowered content of the particles near the conduit surface featuring the convex corner.