Hydrogen production via a three-reactor chemical-looping (TRCL) process using perovskite-type oxygen-carrier materials.
Lead Research Organisation:
Newcastle University
Department Name: Sch of Engineering
Abstract
This project will investigate the use of perovskite-type oxygen-carrier materials for hydrogen production via chemical looping from the water-gas shift reaction (CO + H2 <=> CO2 + H2O), specifically focusing on the Three-Reactor Chemical Looping (TRCL) process that involves the addition of an air oxidation step and can use CH4 or syngas as the reducing agent. This process eliminates the need to separate the H2 and CO2 product streams since the reactants do not come into contact with each other; the products and the oxygen carrier material intermediate are in different phases.
Perovskite-type materials are able to demonstrate a wide range of different oxidising potentials due to their non-stoichiometry. In addition, their ability to incorporate a number of different metal cations of differing valance states within their cubic ABO3 structure allows for materials engineering of desired properties with point defect chemistry. In chemical looping from the water-gas shift reaction, these materials have typically been strontium doped lanthanum ferrite oxides of the form La1-xSrxFeO3-o.
CO is an ideal reducing agent to use in laboratory-scale experiments involving chemical looping as it oxidises directly to CO2, thereby avoiding selectivity complications and simplifying the modelling involved. However it is not a realistic feed gas to use on an industrial scale since it does not naturally occur in large concentrations and its main production method is via the reverse Boudouard reaction. CH4 from natural gas is more abundant and has been identified as one of the more likely alternatives as a reducing agent. However the large endothermic heat of reaction when the oxygen-carrier material is reduced via oxidation of CH4 results in an energy deficit in two-step chemical looping systems with the water-gas shift reaction. The TRCL process has been proposed as a solution to this problem, with the addition of the air oxidation step allowing for the process to potentially be operated auto-thermally on an industrial scale, and maintaining full oxidation of the oxygen carrier material. Studies of the system efficiency using a steam compression and combustion turbine with metal-oxide oxygen carrier materials have predicted that significant amounts of heat could be recovered from the gas streams to make the TRCL process near self-sufficient in terms of electrical power consumption.
The TRCL process conventionally involves a fluidised bed system with a separate fuel reactor, steam reactor and air reactor. For this study, an existing single packed bed reactor with counter-current gas flows will be used. The perovskite-type oxygen carrier material, which is fixed in place within the reactor, undergoes a series of cyclic reductions and oxidations as separate feeds of carbon-based fuel (such as CH4) and H2O in a balance of inert followed by air enter the reactor, allowing for the production of a H2 stream. Following each step in a cycle, the reactor is purged with a stream of inert, typically argon, in order to remove any trace amounts of the reactants or products. This is to prevent the uncontrolled oxidation of the reactants or products existing in concentrations above their lower explosive limits, which is an important safety concern.
This work will provide a comparison of the benefits of the TRCL process with two-step chemical looping configurations using perovskite-type oxygen carrier materials by developing a mathematical model that shows how the principle of chemical looping might be applied to an industrial scale. An existing model for the two-step process is based on a defect chemistry model that relates the virtual oxygen partial pressure to the delta parameter and the amount of strontium doping. A laboratory-scale reactor rig will be used in order to verify the model and further explore the potential of the TRCL configuration.
Perovskite-type materials are able to demonstrate a wide range of different oxidising potentials due to their non-stoichiometry. In addition, their ability to incorporate a number of different metal cations of differing valance states within their cubic ABO3 structure allows for materials engineering of desired properties with point defect chemistry. In chemical looping from the water-gas shift reaction, these materials have typically been strontium doped lanthanum ferrite oxides of the form La1-xSrxFeO3-o.
CO is an ideal reducing agent to use in laboratory-scale experiments involving chemical looping as it oxidises directly to CO2, thereby avoiding selectivity complications and simplifying the modelling involved. However it is not a realistic feed gas to use on an industrial scale since it does not naturally occur in large concentrations and its main production method is via the reverse Boudouard reaction. CH4 from natural gas is more abundant and has been identified as one of the more likely alternatives as a reducing agent. However the large endothermic heat of reaction when the oxygen-carrier material is reduced via oxidation of CH4 results in an energy deficit in two-step chemical looping systems with the water-gas shift reaction. The TRCL process has been proposed as a solution to this problem, with the addition of the air oxidation step allowing for the process to potentially be operated auto-thermally on an industrial scale, and maintaining full oxidation of the oxygen carrier material. Studies of the system efficiency using a steam compression and combustion turbine with metal-oxide oxygen carrier materials have predicted that significant amounts of heat could be recovered from the gas streams to make the TRCL process near self-sufficient in terms of electrical power consumption.
The TRCL process conventionally involves a fluidised bed system with a separate fuel reactor, steam reactor and air reactor. For this study, an existing single packed bed reactor with counter-current gas flows will be used. The perovskite-type oxygen carrier material, which is fixed in place within the reactor, undergoes a series of cyclic reductions and oxidations as separate feeds of carbon-based fuel (such as CH4) and H2O in a balance of inert followed by air enter the reactor, allowing for the production of a H2 stream. Following each step in a cycle, the reactor is purged with a stream of inert, typically argon, in order to remove any trace amounts of the reactants or products. This is to prevent the uncontrolled oxidation of the reactants or products existing in concentrations above their lower explosive limits, which is an important safety concern.
This work will provide a comparison of the benefits of the TRCL process with two-step chemical looping configurations using perovskite-type oxygen carrier materials by developing a mathematical model that shows how the principle of chemical looping might be applied to an industrial scale. An existing model for the two-step process is based on a defect chemistry model that relates the virtual oxygen partial pressure to the delta parameter and the amount of strontium doping. A laboratory-scale reactor rig will be used in order to verify the model and further explore the potential of the TRCL configuration.
Organisations
Studentship Projects
| Project Reference | Relationship | Related To | Start | End | Student Name |
|---|---|---|---|---|---|
| EP/N509528/1 | 30/09/2016 | 30/03/2022 | |||
| 1948790 | Studentship | EP/N509528/1 | 30/09/2017 | 29/03/2021 | Mustafa Ungut |
| Description | Chemical looping refers to the concept whereby a gas-phase reaction can be broken down into multiple sub-reactions via the use of a solid intermediate material that undergoes redox cycling. When applied to the water-gas shift reaction this process (known as chemical looping water-gas shift or CLWGS) allows for separate hydrogen and carbon dioxide streams to be produced thereby reducing or eliminating the need for downstream separation associated with current widely used production processes such as steam methane reforming. This work investigates the design and application of the solid intermediate material, which in this case is an oxygen-carrier material (OCM). For OCMs with sufficiently fast kinetics, it was found that the thermodynamic behaviour, particularly the relationship between the oxygen capacity of the OCM and oxygen partial pressure of the gas phase, can be used to predict how successful the material would be in a CLWGS process. A reactor model was developed to identify the effect of changing this relationship. Previous work has found that the perovskite OCM La0.6Sr0.4FeO3-d was able to produce nearly pure hydrogen and carbon dioxide streams. However this material also had a much lower oxygen capacity when compared to simple oxide materials such as iron oxide which are more commonly used in CLWGS (but have thermodynamically limited conversions). An OCM with larger oxygen capacity would mean less material would have to be used in a fixed bed reactor scenario leading to improved scale-up potential. In this work TG studies and XRD show that Mn doped materials from the La0.6Sr0.4Fe1-yMnyO3-d family of perovskites exhibit larger oxygen capacity in the active pO2 partial pressure range for CLWGS (roughly -19 to -15 log pO2 (bar)) than with no Mn doping, as a result of the addition of Mn shifting the oxygen capacity-pO2 relationship towards the active pO2 range. Furthermore, the cycling performance of La0.6Sr0.4Fe0.67Mn0.33O3-d (among other high capacity materials that were identified) was investigated in a laboratory-scale fixed bed reactor. This was done by flowing consecutive counter-current streams of carbon monoxide and steam in a balance of argon over a fixed bed. Steady hydrogen production was achieved for a large number of cycles using feed lengths of 3 minutes, and water and carbon monoxide conversions of >90% were obtained, higher than with iron oxide. In-situ x-ray diffraction was carried out tracking the changes in lattice parameter of the material as a function of position in the bed. This data showed the material to be phase pure, although during extensive cycling additional phases were noticed at the carbon monoxide inlet end. For comparison the material with no Mn doping could only achieve 90% water conversions at the shorter feed length of 1 minute. A reactor model was also used to verify these results. |
| Exploitation Route | The design principles identified will aid other researchers to predict how other materials being developed might perform in such a CLWGS process provided that data on the oxygen capacity vs pO2 relationship of that material is available. The knowledge gained regarding the performance of the higher oxygen capacity materials will help researchers investigating the scale-up of the process. The ability to multiply the duration of the feed length and still achieve comparable conversions allows for less down-time where the bed must be purged with an inert gas resulting in a more continuous and consistent production of desired hydrogen product, as well as fewer additional reactors that would need to operate simultaneously in a large scale hydrogen production plant to maintain a particular output. |
| Sectors | Agriculture Food and Drink Chemicals Energy Environment Manufacturing including Industrial Biotechology Transport |