Dynamic behaviour of integrated chemical looping process with pressure swing adsorption in small scale on-site H2 and pure CO2 production
Introduction
The route to full scale green hydrogen production using water electrolysis powered by renewable energy has attracted a lot of attention from research, industry and policy makers [1] and it represents the long term solution to enable clean H2 at different scales. Despite the technology being under development currently, H2 from electrolysers represent only 0.1% of the global production [2].
Today, H2 manufacture mostly comes from steam reforming of natural gas (50% of the global production [3]). The predominant driving force for this production route is process economics at relevant scale. Hydrogen production with integrated carbon capture, utilisation and storage (CCUS) also referred to as blue H2, is a valid mid-long term alternative to avoid greenhouse gas emissions to atmosphere and in the meantime develop the required infrastructures for H2 distribution and utilisation as an energy carrier for chemicals production [4], fuel cells and ultimately heat decarbonisation [5]. Currently, only a few large demonstration projects have been proposed in the UK, Europe [6], [7], USA [8], Australia [9] and Japan [10] to establish the benchmark and assess the feasibility for large scale implementation. However, there is no existing infrastructure for supply of hydrogen in areas of small demand such as small industries, residential use or refuelling stations for transportation. Alternatively, the use of different carbon-based feedstock such as biogas, waste gaseous fuel from chemical processes, flare gases, typically available at different quantities and the possibility to inject H2 in the natural gas pipelines [9] could unlock other opportunities for small scale H2 production with effective waste recovery. Small-scale production with conventional steam reforming techniques is very costly and CO2 capture is very expensive as well making the implementation not economically viable yet [11]. Among different options, inorganic membrane reactors have been proposed as a solution to convert methane or alcohol to H2 for small scale with high technical and economic performance [12], [13], [14]. However marginal CO2 reduction is obtained from improved reforming efficiency, unless entirely fuelled with renewable feedstock or including a CO2 capture unit.
Chemical looping technology is an interesting alternative for power and H2 production combined with CO2 capture because it presents higher efficiency than conventional CO2 capture technologies [15], [16]. Chemical looping combustion (CLC) is a standard chemical looping process to produce high temperature heat for power generation. It is comprised of two reactors, an air reactor (AR) and a fuel reactor (FR), where combustion is happening in two stages in the presence of a metal oxygen carrier (OC) [17], [18], [20]. In the FR, a fuel is fed to reduce the OC and the main products are CO2 and H2O, from which pure CO2 can be easily captured with condensation, avoiding any extra downstream separation process. In the AR, air is used to oxidised the OC and the main products is O2-depleted air (N2, Ar and traces of CO2, H2O). The heat stream generated from the oxidation of the OC can be used as a thermal source for power generation. Based on the same principles as in the CLC, chemical looping can be deployed for hydrogen or syngas production with one of the most common applications being Chemical looping reforming (CLR). In CLR, heat generated from the redox reactions is used for the endothermic steam/dry reforming of CH4 or other light hydrocarbons which are present in the natural gas (e.g. C2H6, C3H8, nC4H10) to produce syngas. A lot of research is currently focused on hydrogen production chemical looping technologies and a list of the existing pilot-scale plants is presented in Table 1.
The development of oxygen carrier materials such as the conventional metals (Ni, Fe, Cu, Mn, Co), perovskites and core-shell structures [21] is a key aspect to enable the technology. Some examples of recent advances in material development are high-Fe3O4 content magnetite with ultrasonic treatment to increase syngas production and selectivity [22], perovskite-based LaNiO3 on montmorillonite support to minimize LaNiO3 sintering and reduce carbon deposition [23], Ca2Fe2O5 for high CH4 conversion and syngas yield in moving-bed – fluidized-bed reactors [24] and titania supported iron (Fe2O3/Y0.20Ti0.15Zr0.65O1.90) for higher reactivity [25].
The concept of performing chemical looping reforming for the production of syngas in packed bed reactors (CLR-PB), was first introduced by Spallina et al. [33]. In CLR-PB, oxidation, reduction and reforming are performed in three different stages, presented in Fig. 1. In this concept, the solid material is inside the reactor and the gases are sequentially fed to the reactor by means of switching valves. Firstly, the bed gets oxidized with air and the temperature rises across the reacting bed, due to the highly exothermic reaction (e.g., ΔHο298Κ = −479.4 kJ mol−1 for Ni oxidation) with the main product being N2. The second step is to reduce the bed with a low-grade fuel (off-gas) obtained from a downstream process such as the tail gas from a PSA unit (usually a low grade mixture of CO, H2 and CO2) which is converted into H2O and CO2. The reduction reactions are mostly heat neutral (e.g., ΔHο298Κ = −43.26 kJ mol−1CO and ΔHο298Κ = −2.13 kJ mol−1H2 for Ni reduction), therefore the solid temperature slightly changes and only part of the bed is cooled due to the heat front moving from the inlet across the reactor. The CO2 product can be sent to storage, or alternatively used as a reactant for other processes such as urea manufacture [34]. At the start of the third stage, the reforming, most of the heat originating from the oxidation stage is still present inside the bed. In fact, in gas-solid reactions (reduction and oxidation) the reaction front velocity is much faster than the heat front velocity and therefore, while the solid is being converted, the heat generated is not vented off, resulting in a heat storage [35]. This heat is used to enable the endothermic (steam and dry) methane reforming. The high temperature conditions along with the frequent valve switching are the primary technical challenges for this process.
Another way of producing hydrogen with gas-solid reactions is sorption enhanced reforming. In this configuration, CaO is used to capture CO2 generated during the reforming/water gas shift and converted into CaCO3 which is later regenerated to CaCO3 [36], [37], [38], [39]. The endothermic reaction of CaCO3 regeneration, also called calcination, occurs at high temperature by means of an oxy-combustion or by adding Cu to CaO and using the reduction of CuO to Cu (exothermic) [40], [41], [42], [43], [44].
To convert the syngas produced by a CLR-PB unit into high-purity H2, one or more water–gas-shift stages are used followed by removal of impurities such as N2, CO, CO2 and CH4 [45], [46]. In terms of gas separation, PSA is known as a highly well-stablished and versatile technology for hydrogen purification due to its low energy consumption and high product purity. Hydrogen purities from 98% to 99.9999% are reported for the PSA process [47], [48], [49]. The performance of a PSA unit is often defined by product purity, recovery and adsorbent productivity (i.e., the amount of H2 produced per unit mass or volume of adsorbent). In order to achieve high recovery and purity, different adsorbents have been proposed and commercialised and process parameters can be varied. Activated carbons and zeolites are commonly used in H2 PSA units [48], [50], [51], [52] and typically adsorption columns are packed with at least two layers of adsorbent to obtain high purity hydrogen. The feed end is commonly packed with activated carbon to selectively adsorb CO2 and hydrocarbons such as CH4. This is complemented with a zeolite layer that has an enhanced capacity for capturing N2 and CO [45], [52], [53]. Zeolites adsorb CO2 and water vapour strongly and these components cannot be readily desorbed by decreasing the pressure. Therefore, CO2 and water vapour must be prevented from reaching the zeolite layer by adequately sizing the activated carbon layer and potentially adding a short desiccant layer at the inlet of silica gel or activated alumina [54].
PSA units can be used to supply the full spectrum of capacities from small plants producing a few hundred Nm3/h to large-scale plants that produce>400,000 Nm3/h. On the biggest plants, PSA units are normally operated with 10 or more columns with three or more pressure equalisations stages resulting in higher recovery [55], [56], [57]. There is though a growing requirement for hydrogen production facilities in small H2 demand industries. PSA units at this scale are normally 4–6 bed systems with one equalisation stage and this makes it challenging to obtain a high H2 recovery. The recovery values are found to be at or below ca. 75%, mainly due to the lower operating pressures reducing adsorbent capacities and increasing the amount of H2 needed for purging. They can be designed with more than one equalisation to improve the recovery loss, but the cost of a PSA unit increases with more pressure equalisation steps due to the additional vessels required [57], [58].
Both CLR-PB and PSA processes are pseudo-continuous meaning the outlet stream conditions vary with time, but the process is overall operated continuously. Since the PSA process is downstream of the syngas generation process, the variation of feed concentration, temperature and flowrate might change the performance of the PSA unit and subsequently the composition of the off-gas. This could subsequently affect the performance of the reduction step and hence the CLR-PB process. Therefore, the effect of temperature, feed composition and flowrate variation need to be addressed during the design of the PSA and CLR-PB units. Little research has been done on the effect of feed condition variations in these systems. Ahn et al. [59] studied the effects of feed composition on a layered bed H2 PSA process, where they looked at three feed compositions based on different N2 contents. Li et al. [60] also investigated the effect of 24 different feed compositions on purity and recovery of H2. However, none of these studies looked at the real case, where the composition of gases in the feed varies dynamically over time. Additionally, these studies did not investigated the effect of the time dependent changed tail gas composition and flowrate as this stream is returned to the chemical looping system and results in further changes to the inlet feed to the H2 PSA system.
In this work, we have simulated the performance of H2 PSAs based on the gas coming out of a chemical looping reactor on the reforming step from a dynamically operated CLR-PB. The reformate syngas from the CLR-PB is fed to a PSA after one-stage of WGS at high temperature and the PSA off-gas is send back to the CLR-PB for the reduction step. Since both are dynamic processes and mutually dependent, the operation strategy and heat management are essential to identify the main challenges. This is more relevant for small scale on-site H2 generation where the overall volume should be limited to reduce the cost and make the system compact. In all the cases, the feed to the H2 PSA has been modelled using the actual dynamics in composition and flow coming out of a simulation of the CLR. In this study, a layered activated carbon/zeolite LiX process is considered for the H2 PSA unit. Both CLR-PB and PSA processes have been modelled and simulated using an advanced mono-dimensional, non-isothermal model with integrated kinetics to account for the gas-solid and heterogeneous reactions (in case of CLR) and adsorption kinetics (in case of PSA).
Section snippets
Description of the considered plant
The small scale H2 generation system refers to a plant capacity of 130 Nm3/h H2 capacity plant. Such an installation can be used to cover the energy needs of a small industry, a residential building using PEM fuel cells, local transportation or heating demands, while having pure CO2 as a by-product which can be temporarily stored and sold to where the demand is. The installation of such a unit can be also placed near an industry needing pure CO2 such as in the chemical industry (urea, inorganic
Description of the model
A mathematical packed bed reactor model has been developed to simulate the complete dynamic process of chemical looping reforming. A one-dimension (1-D) pseudo-homogeneous model has been used, meaning that averaged radial values were used and no significant differences between the solid and gas phase were accounted for. Axial dispersion effects have not been implemented for as the reactor’s dimensions are close to an industrial one. With respect to the pressure drop, a low superficial velocity (
Design of the process
A high temperature WGS reactor following the CLR system has been implemented and simulated, with the model described in the Supplementary Information document. For gas coming from an adiabatic CLR it was found that the appropriate volume of water-gas-shift catalyst bed was 0.04 m3, whilst that for a non-adiabatic CLR was 0.05 m3. The equations of conservation for the WGS reactor are presented in Eqs. (4), (5).
Fig. 5 shows
Integrated processes under dynamic operations
A total of seven (7) different cases have been examined (Table 7) to cover the effect of different PSA configuration and a sensitivity analysis on heat losses effect. A comparison is made to the benchmark case (BEN), concerning different PSA setups, which covers the use of an 8-bed over a 4-bed PSA (4BED case) configuration and the inclusion of a blowdown tank after the PSA (BT case). With respect to the non-adiabatic cases (HL – 1.5, HL – 3.0, HL – 5.0, HL – 5.0alt), the examined values for
Conclusions
In this paper, the integration of two dynamically operated processes (CLR-PB and PSA) has been assessed showing that combined pure H2 production suitable for fuel cell applications with inherent CO2 separation at high purity is possible at kW scale and for application with fuel cells. The study on PSA design has demonstrated that a fast cycle 8-bed unit performs better than a slower cycle 4-bed unit in terms of H2 recovery factor (+4%), smaller bed volume (25–30%) and a smaller buffer tank
Declaration of Competing Interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
The authors acknowledge the UKCCSRC for providing the funding of this research. The UKCCSRC is supported by the EPSRC as part of the UKRI Energy Programme.
References (70)
- et al.
Conventional and advanced exergoenvironmental analysis of a steam methane reforming reactor for hydrogen production
J. Clean. Prod.
(2012) - et al.
Renewable energy storage via CO2 and H2 conversion to methane and methanol: Assessment for small scale applications
Renew. Sustain. Energy Rev.
(2019) - et al.
A review on the role, cost and value of hydrogen energy systems for deep decarbonisation
Renew. Sustain. Energy Rev.
(2019) - et al.
Direct route from ethanol to pure hydrogen through autothermal reforming in a membrane reactor: Experimental demonstration, reactor modelling and design
Energy.
(2018) - et al.
Hydrogen production with integrated CO2 capture in a membrane assisted gas switching reforming reactor: Proof-of-Concept
Int. J. Hydrogen Energy.
(2018) - et al.
Integration of coal gasification and packed bed CLC for high efficiency and near-zero emission power generation
Int. J. Greenh. Gas Control.
(2014) - et al.
Comparison of carbon capture IGCC with pre-combustion decarbonisation and with chemical-looping combustion
Energy.
(2011) - et al.
Kinetics of redox reactions of ilmenite for chemical-looping combustion
Chem. Eng. Sci.
(2011) - et al.
Improved activity of magnetite oxygen carrier for chemical looping steam reforming by ultrasonic treatment
Appl. Energy.
(2020) - et al.
A novel oxygen carrier for chemical looping reforming: LaNiO3 perovskite supported on montmorillonite
Energy.
(2017)
Syngas and a separate nitrogen/argon stream via chemical looping reforming–A 140kW pilot plant study
Fuel.
Continuous high purity hydrogen generation from a syngas chemical looping 25kWth sub-pilot unit with 100% carbon capture
Fuel.
250 kWth high pressure pilot demonstration of the syngas chemical looping system for high purity H2 production with CO2 capture
Appl. Energy.
A 30-kWth moving-bed chemical looping system for hydrogen production
Int. J. Greenh. Gas Control.
Chemical looping reforming in packed-bed reactors: Modelling, experimental validation and large-scale reactor design
Fuel Process. Technol.
A comparative study on clean ammonia production using chemical looping based technology (under revision)
Appl. Energy.
Hydrogen production through sorption enhanced steam reforming of natural gas: Thermodynamic plant assessment
Int. J. Hydrogen Energy.
Modeling of sorption enhanced steam methane reforming in an adiabatic fixed bed reactor
Chem. Eng. Sci.
Modelling of sorption-enhanced steam methane reforming in a fixed bed reactor network integrated with fuel cell
Appl. Energy.
Combined sorbent and catalyst material for sorption enhanced reforming of methane under cyclic regeneration in presence of H2O and CO2
Fuel Process. Technol.
Process design of a hydrogen production plant from natural gas with CO2 capture based on a novel Ca/Cu chemical loop
Appl. Energy.
A parametric study of layered bed PSA for hydrogen purification
Chem. Eng. Sci.
Activated carbon for hydrogen purification by pressure swing adsorption: Multicomponent breakthrough curves and PSA performance
Chem. Eng. Sci.
H2 pressure swing adsorption for high pressure syngas from an integrated gasification combined cycle with a carbon capture process
Appl. Energy.
Dynamic behaviour of layered columns in pressure swing adsorption
Gas Sep. Purif.
PSA purification of waste hydrogen from ammonia plants to fuel cell grade
Sep. Purif. Technol.
Modelling and simulation of two-bed PSA process for separating H2 from methane steam reforming
Chinese J. Chem. Eng.
CO2 utilization in the perspective of industrial ecology, an overview
J. CO2 Util.
Techno-economic assessment of an integrated high pressure chemical-looping process with packed-bed reactors in large scale hydrogen and methanol production
Int. J. Greenh. Gas Control.
Measurement challenges for hydrogen vehicles
Int. J. Hydrogen Energy.
Experimental demonstration of CLC and the pressure effect in packed bed reactors using NiO/CaAl2O4 as oxygen carrier
Fuel.
The Future of Hydrogen
Industrial Innovation: Pathways to deep decarbonisation of Industry. Part 2: Scenario analysis and pathways to deep decarbonisation
Eur. Comm. DG Clim. Action.
Cited by (20)
Highly-efficient hydrogen production through the electrification of OB-SiC nickel structured catalyst: Methane steam reforming and ammonia cracking as case studies
2024, International Journal of Hydrogen EnergyH<inf>2</inf> purification employing pressure swing adsorption process: Parametric and bibliometric review
2024, International Journal of Hydrogen EnergyA computational-fluid-dynamics study on scaling up a single flat membrane reactor for on-site hydrogen production
2023, Chemical Engineering Research and DesignA parametric computational-fluid-dynamics study on improving the performance of a flat membrane reactor for on-site hydrogen production
2023, Chemical Engineering Research and DesignNumerical modeling of chemical looping oxidative dehydrogenation of ethane in parallel packed beds
2023, Chemical Engineering Journal