New Design of a Separator Unit with Metal Hydride and an Analysis of Its Potential Application in the Process of Hydrogen Separation from a Binary Mixture of Gases

Abstract

Synthesis gases produced in the processes of the high-temperature gasification of otherwise unusable waste, with the use of technologically efficient and cost-effective separation methods, may represent a reliable source of hydrogen intended for applications in the industry and the transport segments. One of the possible solutions to the separation of hydrogen from a mixture of gases is the use of metal hydride (MH) alloys, which are capable of binding hydrogen into their structure. This is the subject of the present article, in which a new design is presented for a fully functional system and a hydrogen separator unit, and the potential application of a commonly available metal hydride alloy in the separation of hydrogen from a binary mixture containing carbon dioxide and hydrogen is discussed. Load testing of the selected type of metal hydride alloy with a high concentration of carbon dioxide in the mixture, representing 40 vol. % and 4 vol. %, was performed. In addition, testing the alloy’s ability to separate hydrogen from a mixture containing H₂ and CO₂ was conducted using small alloy samples and a newly designed hydrogen separator unit. The resulting higher purity of hydrogen after the separation was confirmed by an experiment, in which the hydrogen concentration in the resulting mixture increased by 2.7 vol. %. The purity of the desorbed hydrogen amounted to 99.4 vol. %. The testing also confirmed a high degree of degradation of the alloy, caused by the poisoning effect of CO₂ on the selected alloy type. There was also a significant decrease in the absorption ability of the alloy—from 1.7 wt. % to 1.2 wt. %—and a significant extension of the absorption time caused by the slower kinetics of the hydrogen storage, which occurred as early as after ten absorption–desorption cycles.

1. Introduction

In present times, characterised by abrupt economic, social and geopolitical changes, new requirements are appearing with regard to the production of fuels. In particular, there is a need to produce fuels that will ensure at least the partial energy independence of countries that primarily use imported energy. One of the green alternatives to fossil fuels which is now, together with the use of batteries, being intensively discussed is the use of hydrogen as an energy carrier.

Hydrogen is regarded as an ideal energy carrier, due to its high energy density relative to a mass unit, as well as its availability in sufficient amounts in the form of various compounds on earth, and the fact that zero emissions are produced in the process of its recovery in fuel cells [1,2,3,4,5]. Since the portfolio of industrial applications for hydrogen is growing, there is a need for increased production of this energy carrier, by applying methods different from those used up until now in the production of hydrogen from fossil fuels. Steam reforming represents a well-established commercial method; however, it comprises the recovery of fossil fuels, so the produced hydrogen is classified as grey, i.e., non-ecological. Ecological procedures for the recovery of this energy carrier are currently becoming increasingly preferred. Energy sources that are suitable for the production of hydrogen, in addition to the solar and wind potential available in a given country, may include, for example, the excess energy from nuclear reactors which may be stored during the periods of lower consumption from an electricity distribution system in the form of chemical energy. Potential sources of hydrogen also include the heterogeneous gas mixtures (syngas) formed in the processes of the high-temperature thermal treatment of waste.

Synthesis gas (syngas) is a gaseous product formed in the process of the high-temperature gasification, or pyrolysis, of various types of waste with a high content of organic matter. The composition of the generated gaseous mixture and the type of the processing technology that is used significantly affect the boundary conditions of the thermal processing. The main syngas components are H₂, CO, CH₄ and CO₂ (or plasma-forming gas, in the case of waste processing with the use of plasma technology (N₂)), which from an energy point of view predetermine the mixture to be classified as a combustible gaseous mixture with a low calorific value, the recovery of which is primarily carried out through combustion in systems for the combined production of electric energy and heat. Another potential option for syngas recovery is the use of syngas in the chemical industry, or hydrogen separation from a mixture, aimed at the recovery of hydrogen with the use of green technologies.

Increasing the amount of hydrogen produced from a mixture of gases and the elimination of the presence of poisonous carbon monoxide in the gas is facilitated by a water–gas shift (WGS) reaction, which runs in the presence of various catalysts in the high-temperature or low-temperature conversion mode. A WGS reaction requires not only a high temperature and the presence of a catalyst, but also the presence of a significant amount of steam, the availability and price of which depend on its production process (steam generators using fossil fuels, electric energy, or technological units that recover the waste heat from industrial processes). The steam/CO ratio in a reaction chamber should be as much as 3:1. The conversion of gas is carried out on the basis of a water–gas shift reaction, while the equilibrium constant is defined by the following equation: [6].

$$\mathrm{CO}+{\mathrm{H}}_2\mathrm{O}\hspace{0.33em}\iff \hspace{0.33em}\mathrm{C}{\mathrm{O}}_2+{\mathrm{H}}_2\hspace{1em}\Delta H=-40.6\mathrm{kJ}\cdot {\mathrm{mol}}^{-1}$$

(1)

The water–gas shift reaction is thermodynamically limited and is accompanied by adding excess steam to the process; this improves the ratio of hydrogen and carbon monoxide content at the outlet. Higher hydrogen recovery rates are achieved by extending the conversion with a second-degree process that uses copper-based catalysers and by lower temperatures in the reaction chamber [6].

Gas purification technologies are becoming more crucial, as the demand for high-grade hydrogen is increasing [7]. Hydrogen separation from a mixture of gases may be carried out by applying a number of various technologies, mostly based on the principle of membrane separation of the individual components, or by applying the methods of absorption, adsorption or cryogenic separation. Membrane technology plays a key role in the downstream processing separation and the purification of several products [8,9]. Due to a high purity of hydrogen and a possibility to capture and store CO₂ during the process, it is predicted that this technology may be used for the elimination of greenhouse gases produced in the processing of various secondary fuels [10,11]. However, some failures were observed in the separation of hydrogen from a mixture containing ammonia, in particular, irreversible damage to the membranes [12]. The potential applicability of a polymeric membrane is highly dependent on the membrane materials and on the mechanism of transport. The facilitated transport membranes have attractive transport performances at high operating temperatures. However, solution-diffusion membranes do not operate well at high temperatures, because of a low selectivity between hydrogen and impurities [12]. A disadvantage of the convection membranes produced from ceramics, polymer or metals is that they fail to tolerate the syngas purification conditions or are too costly. A method that is promising, with regard to the existing syngas purification techniques, is the use of a molecular sieving membrane [7]. Capturing the components that are preferred in the industry, or the ballast such as CO₂, from syngas has further led to the development of various adsorption substances [13].

Currently, adsorption-based technologies are widely utilised for the separation and purification of H₂ on an industrial scale, with pressure swing adsorption (PSA) being the most common choice [14,15,16,17,18,19].

A key disadvantage of the PSA process is the high consumption of electric energy; according to some studies, it amounts approximately 0.8–1.5 kW∙m⁻³ of H₂ [20].

In order to remove carbon dioxide from a binary mixture of gases (H₂ and CO₂), it is possible to apply a process involving the saturation of calcium oxide with carbon dioxide to form calcium carbonate.

Cryogenic separation is based on cooling the individual components of a gas mixture at a low temperature. The required cooling can be provided via the Joule–Thomson effect, with external refrigeration/liquefaction cycles, or through the turbo-expansion of the produced H₂ [9,14]. This type of separation, which is highly efficient, is characterised by a high consumption of a coolant and compression work, which has a significant effect on the energy consumption of a cryogenic separation unit [21]. A high purity of the resulting product is often achieved by combining several separation technologies. One of them is described in the paper by Dragomir et al., which involves the purification of hydrogen using a pressure swing adsorption process [22].

The separation of hydrogen from a binary mixture containing H₂ and inert gas (Ar, N₂) is a technically feasible process in terms of the long-term preservation of the cyclic stability of the metal hydride alloy. Nonetheless, problems begin when the alloy is confronted with reactive gases based on carbon oxides and sulphur. Several studies have confirmed a poisoning effect of such oxides on the MH alloy, leading to gradual degradation of the alloy and slowing the adsorption kinetics. The research in this area was initially focused on the effects of poisonous syngas components on the degradation and lower absorption capacity of an MH alloy. In the majority of cases, the performed experiments focused on the exposure of MH alloys to low concentrations of carbon oxides. The gases with a poisoning effect in the synthesis gases that are commonly used in industrial applications are contained in a mixture, in quantities ranging from several ppm up to several tens of a percent. Chemically aggressive gases are also capable of forming surface compounds (oxides and hydroxides, carbonyls, sulphides) which most probably deactivate the surface centres responsible for H₂ dissociation, resulting in the retardation of this process; this represents the rate-limiting step in hydrogen absorption [23]. However, a relatively low contamination of certain metal hydride alloys, caused by chemically aggressive gases, may be partially removed by the regeneration of the alloy, in particular, by prolonged cyclic exposures of an MH alloy to hydrogen and by deep vacuuming.

The purpose of hydrogen separation from syngas is its further use for green technologies. Unlike the hydrogen produced from fossil fuels, the hydrogen generated from waste saves natural resources. Moreover, in the transitional period on the path to climate neutrality, the recovery of waste represents one of the potential alternatives for hydrogen production for use in industry and transport.

With regard to the aforesaid facts, in the present article, the design of experimental testing equipment is described in the form of a hydrogen separator unit using a metal hydride, as well as research into the potential of hydrogen separation from a mixture of gases, and the loading of a relatively accessible and affordable MH alloy containing a high concentration of CO₂ and the effects on its service life.

2. Description of a System for Hydrogen Production from Waste

A high recovery rate of combustible gaseous products from the thermal treatment of wastes, primarily of hydrogen, may be achieved in the process of high-temperature pyrolysis or the gasification of sorted mixed waste containing a significant portion of plastic materials. In this case, the gasification of pre-treated waste was carried out in a plasma reactor at temperatures ranging from 1200 to 1400 °C. The technology was used in a dependent arrangement, with a plasma arc generated between a hollow graphite electrode (cathode) and a graphite hearth (anode). The maximum power of the reactor was 60 KW, and it was powered via a three-phase transformer. The charge was inserted using a screw feeder in the automated mode. The resulting gaseous product of the thermal treatment was a heterogeneous mixture of combustible and incombustible gaseous components, which also captured fly ash or fine particles of particulate matter (PM). In addition to the PM, the process produced a small portion of slag which materialised as the unusable portion of the charge. Tapped vitreous slag is environmentally non-harmful, and in most cases, it exhibits an inert nature.

However, the separation of hydrogen from a mixture of gases requires a complex process of purification and separation of the individual components, which have a substantial effect on the resulting price of the hydrogen. The primary objective of the purification column located at the outlet from the reaction chamber of the plasma reactor is the capture fly ash from the syngas, which contains mainly H₂, CO, CO₂ and CH₄. In the case of the gasification of refuse-derived fuel (RDF) and municipal solid waste (MSW) using plasma technology, more than 80 vol.% of the syngas represents H₂ and CO. In such a case, a charge is a sample of combined municipal waste (origin: Košice City) and RDF waste, formed by sorting out combustible components from the analysed waste sample. All the samples were then crushed to reach a grain size below 5 mm.

With regard to further recovery, it may be stated that the H₂ content in syngas amounts to 44–51 vol.%, and the content of carbon monoxide represents 32–43 vol.%. More detailed information on the composition of the gaseous mixture is presented in a paper [24], and it was obtained with the use of gas chromatography (methods: GC-TCD and GC-FID), carried out in the Labortest U.S. Steel Košice accredited laboratory. In addition to the syngas recovery for power purposes by direct combustion, another interesting alternative for increasing the competitiveness of the combustion-based gasification technologies available on the market is the separation of H₂ from the mixtures of gases. Increasing the H₂ content in syngas is facilitated particularly by the WGS technology, which produces a gaseous mixture with high H₂ and CO₂ content, as is described by Equation (1). With regard to the fact that the desired product after the separation is hydrogen for use in green technologies, the experiments were carried out using a mixture that was formed by syngas recovery in a WGS reactor. This mixture may be regarded as a binary mixture consisting of H₂ and CO₂.

3. Materials and Methods

The hydrogen separator unit is prototype equipment consisting of two low-pressure metal hydride gas storage tanks, regulating components, distribution and safety fittings, a rotary vacuum pump, measuring equipment and a controlling module (hardware, software) which enables users to operate the equipment in a semi-automatic mode. A tandem arrangement of low-pressure gas storage tanks containing a metal hydride alloy facilitates the continuous operation of the equipment. In the phase of filling one of the tanks with a gaseous mixture containing hydrogen, the regulating and controlling technology ensures the extraction of the atmosphere and subsequent emptying of the second tank. The undesirable gaseous components with a low hydrogen content are safely extracted into an intermediate storage tank for further recovery. The separated hydrogen, released from the structure of the MH alloy, is stored in a conventional pressure vessel intended for gases. At the same time, the process of hydrogen absorption by the MH alloy from the supplied gas mixture remains running in the second tank. Both of the pressure vessels containing an MH alloy were made of stainless steel (316 L material). During the absorption process, heat is released that must be effectively removed from the process [25,26,27]. This thermal energy corresponds to approximately 1.01 MJ∙m⁻³ of the stored hydrogen. The rate of filling the MH tank is therefore affected by the efficiency of its thermal management, which should prevent achieving the level of the equilibrium temperature (for a pre-defined working pressure at which the separation process runs). Based on the recorded data of the working temperatures of the prototype equipment, the thermal management system ensures the removal of heat generated in the hydrogen absorption process, in particular, by using a heat exchanger and an external cooling source—the Peltier cell. By reversing the poles of the Peltier cell, it is possible to heat the tanks while the separated hydrogen is being released from the alloy, which accelerates the desorption process. A wiring diagram of the hydrogen separator unit is shown in Figure 1.

A high rate of filling the MH tank containing the mixture of gases combined with a low effectiveness of the removal of thermal energy generated in the hydrogen absorption process leads to losses of H₂, which should be prevented. Such losses may be avoided by filling the tank up to a certain pressure level or by closing it, as this will restrain the possibility of the flow-through separation and facilitate more detailed monitoring of changes and the whole process of the hydrogen separation using MH alloys. The pressure in the pressure vessels is monitored by the controlling system, and the tank where the absorption process is kept running is gradually replenished with the mixture.

In order to demonstrate the functionality of the separation unit, a commonly available MH alloy was used. Its storage capacity relative to the equilibrium pressure at an ambient temperature of 20 °C is shown in Figure 2.

The amount of heat released in the hydrogen absorption process (formation of metal hydride) at the complete saturation of the alloy with hydrogen may be calculated using the following equation:

$$Q_{\mathrm{stor}}=\frac{m_{\mathrm{MH}}\cdot {\alpha }_{\mathrm{MH}}}{{\rho }_{H2}\cdot 100}\cdot Q_{\mathrm{MH}}\hspace{1em}(\mathrm{J})$$

(2)

where m_MH is the mass of the alloy in the tank (kg); α_MH is the mass storage percentage (%); ρ_H2 is the density of hydrogen in normal conditions (kg∙m⁻³); and Q_MH is the quantity of heat released in the absorption of 1 m³ of hydrogen (J).

The value of the thermal power that must be subsequently removed from the tank is calculated using the following equation:

$$P_{\mathrm{stor}}=\frac{Q_{\mathrm{stor}}}{\tau }\hspace{1em}(\mathrm{W})$$

(3)

where τ is the tank filling time (s).

A sharp increase in the pressure of the mixture of gases, which occurs after the step of purification and drying, increases the energy consumption of the technological unit in which the mixture of gases is recovered. Therefore, it was decided that the initial experiments would be conducted at a pressure of 5.0 MPa, at which the hydrogen storage capacity of the MH alloy is 1.7 wt. %.

4. Results

The process of alloy activation, as well as the other experiments, were carried out using a fully automated thermal management system, which maintained the pressure vessel containing the metal hydride alloy. The used active cooling system combined the accumulation of thermal energy and the flow-through cooling. The heat removal during the absorption process was monitored by the software and was facilitated by a circulating coolant, with the temperature adjusted and maintained at the pre-defined level at a temperature of 20 ± 0.5 °C. The heating of the MH alloy during the desorption process, during which the alloy is cooled, was carried out with the use of the same equipment by reversing the poles of the Peltier cells. The temperature of the water in the water shell of an MH low-pressure storage tank was maintained within the range of 17–20 °C. The heat removal and heating of the coolant were carried out using a copper heat exchanger and an external source of heating and cooling, i.e., the Peltier cells [28,29].

The first test that involved the exposure of the metal hydride to the effects of the binary mixture of gases (H₂, CO₂) was carried out using Hydralloy C5 samples, each weighing 3 g. The alloy was activated at an activation pressure of 5 ± 0.1 MPa, and as soon as the pre-defined pressure was achieved in the metal hydride tank, the supply of hydrogen was closed. Subsequent changes in the pressure of the gas in the system over time were monitored; such changes were proportionally correlated to changes in the rate of hydrogen absorption into the structure of the metal alloy (Figure 3). Following the process of the activation of the MH alloy, the quantity of absorbed hydrogen was identified by applying the volumetric method, and the storage capacity was calculated. The results are listed in

Table 1. Identification of the storage percentage of Hydralloy C5 during the activation.

	MH Alloy
	act1	act2	act3
pp (MPa)	5.050	5.106	5.020
tMH (°C)	19.81	23.06	21.19
t0 (°C)	20.25	20.81	20.75
p0 (Pa)	101,364.9	101,364.9	101,364.9
VvoH2 (m3)	0.0024553	0.0025264	0.0025745
Vstor (m3)	0.0005482	0.0005579	0.0005783
ρH2 (kg∙m−3)	0.0899	0.0899	0.0899
mH2 (kg)	4.9283 × 10−5	5.0154 × 10−5	5.1985 × 10−5
mMH (kg)	0.003	0.003	0.003
αMH (wt. %)	1.64	1.67	1.73

p_p—filling pressure from the reservoir of H₂ (relative pressure); t_MH—average temperature of MH; t₀—ambient temperature; p₀—pressure in the measuring cylinder; V_{v0 H2}—volume of hydrogen occupying the free volume; V_stor—volume of stored hydrogen in MH; ρ_H2—hydrogen density, m_H2—weight of the storage hydrogen; m_MH—weight of MH; α_MH—calculated storage capacity of MH.

.

5. Load Testing of the Effects of the CO₂ Concentration on the Metal Hydride Alloy

The activated metal hydride alloy with a mass of 3 g was exposed to the effects of a binary mixture of hydrogen and CO_2, with the aim of monitoring its effects on the storage ability and the kinetics. A cyclic loading of the MH alloy was carried out using a mixture of hydrogen and carbon dioxide, mixed at a ratio of 60:40 (vol. %). The adverse effects of CO₂ on hydrogen absorption by the MH alloy were significant after the execution of only a small number of absorption–desorption cycles. Figure 4 shows the cyclic loading of the MH alloy, with a mixture of H₂ and CO₂ at a ratio of 60:40 (vol. %).

Based on the results of the initial test conducted with a 40% concentration of CO₂, we reduced the CO₂ concentration in the mixture of carbon dioxide and hydrogen. The subsequent experiments were carried out with a mixture of CO₂ and H₂, where the concentration of CO₂ in the mixture was 4 vol. %.

Load tests of the metal hydride alloy—Hydralloy C5—in the hydrogen separator unit were carried out using an adjusted LPMHHST-60 storage tank (low pressure metal hydride hydrogen storage tank, designed by Faculty of Mechanical Engineering, Technical University of Košice), without an external heat exchanger, with a reduced length and with an added new inlet into the tank on the opposite side, as shown in Figure 5. Inside the tank, the internal heat transfer intensifier was maintained, in order to improve the heat removal from the alloy towards the tank walls.

For the initial measurements of the equipment’s ability to separate hydrogen from the mixture containing carbon dioxide, we prepared a sample of Hydralloy C5 with a total mass of 300 g, which was placed inside the tank. The internal volume of the modified metal-hydride tank was 0.031 m³. The concentration of the powder alloy in the horizontally positioned pressure vessel was predicted to constitute 2/3 of the horizontally positioned storage tank in the interlamellar space of the internal heat exchanger. On one side of the tank, there was a thick internal filter, a shut-off valve and a fine filter sized up to 7 μm. On the opposite side, there was an internal thermometer for recording the temperatures in the internal space of the tank during the process of hydrogen separation. During the process, the carbon dioxide was exhausted from the hydrogen separator unit between the absorption–desorption cycles.

During the testing, 10 fillings of the hydrogen separator unit with a mixture of hydrogen and carbon dioxide were carried out; unlike in the previous test, this time, we used a mixing pressure tank for the gas storage, filled with glass spheres with a diameter of 4 mm. The mixing pressure tank, which was interconnected with the hydrogen separator unit in which the metal hydride alloy was placed, was pressurised with carbon dioxide, and the pressure in the system was then increased using compressed hydrogen to a relative pressure of 5 MPa. The prepared mixture was subsequently released from the mixing pressure tank to the tank containing the metal hydride alloy. The separation process using the hydrogen separator unit was launched 15 s after the binary mixture was prepared in the mixing pressure tank. Interconnecting the mixing pressure tank and the hydrogen separator resulted in balancing the pressure to a value of approximately 2.41 MPa, as is shown in Figure 6. A subsequent pressure drop was caused by the hydrogen absorption into the MH alloy.

Based on an analysis of the composition of the gaseous mixture, the concentration of carbon dioxide at the outlet of the mixing pressure tank ranged in the interval of 3–3.5 vol. %. Figure 6 also presents the gradual deterioration of the kinetics of hydrogen storage in the MH alloy, due to a concentration of CO₂ in the mixture of 3–3.5 vol. %.

6. Discussion

In accordance with the expectations, a 40% content of CO₂ in the mixture resulted in a significant deterioration of the absorption ability of the metal hydride alloy with a mass of 3 g. As early as during the first cycle, due to the alloy degradation, it was possible to observe slower kinetics of hydrogen absorption. The storage kinetics were significant, as was also reflected in a sharp drop in the pressure inside the measuring system, which had to be replenished twice during the test. Unlike the activated sample, which was not contaminated or degraded due to the effects of the poisons, the absorption process required a duration more than 10 times longer for the alloy to be fully saturated with hydrogen. This process is associated with the formation of oxides that prevent the absorption of hydrogen into the intermetallic structure of a metal. Adverse effects of CO₂ on the absorption ability of the metal hydride alloy were detected during the second absorption cycle, and such negative effects of CO₂ were reflected in a significant prolongation of the kinetics of the hydrogen absorption into the alloy’s structure. During the same measurement interval, a drop in the pressure inside the closed system was at the minimum level, while the kinetics of the hydrogen storage and the ability to bind hydrogen into the alloy’s structure deteriorated with each cycle. Based on the results of the initial tests, during which the metal hydride alloy was exposed to the effects of the mixture containing hydrogen and carbon dioxide (the mixing ratio of the components of the binary mixture was 60:40), the concentration of CO₂ in the mixture of hydrogen, and carbon dioxide was reduced.

The concentration of carbon dioxide in the binary mixture was reduced to 4 vol. %. The Hydralloy C5 metal hydride alloy with a total mass of 300 g was inserted into the pressure vessel of the hydrogen separator unit. The potential for hydrogen separation from the binary mixture of H₂ and CO₂ was tested using the real semi-automatic equipment. In the tests, a total of 10 experiments of filling and emptying the MH tank were carried out with the same boundary conditions.

Gradually increasing the number of cycles of storing hydrogen from the mixture resulted in a significant deceleration of the kinetics of the hydrogen storage into the metal hydride alloy. Such deteriorated absorption kinetics is documented in Figure 6, where the alloy degradation is clearly visible as a drop in the pressure in the binary mixture in the tank, throughout the monitored time. A decrease in the quantity of carbon dioxide from the used mixture caused a slowing down of the poisonous effect of CO₂ on the metal hydride alloy. Similarly to the first experiment, after the second absorption–desorption cycle, the kinetics were significantly retarded due to a less intensive transfer of elementary hydrogen through the oxidised surface into the core of the alloy grain, where it would be bound to the metallic structure. Again, after the first, third, fifth and tenth filling, control analyses of the storage capacity of the metal hydride alloy were performed with highly pure hydrogen (99.999 vol. %), the results of which are listed in

Table 2. Identification of the storage percentage of the alloy during the experimental testing.

	cont._test_1	cont._test_2	cont._test_3	cont._test_4
pp (MPa)	4.8	4.8	4.8	4.8
tMH (°C)	21.56	21.51	21.3	21.2
t0 (°C)	19.13	20.2	20.52	20.1
p0 (Pa)	101,364.9	101,364.9	101,364.9	101,364.7
VvoH2 (m3)	0.0010721	0.0009902	0.0010949	0.0006242
Vstor (m3)	0.05578	0.05098	0.04501	0.04158
ρH2 (kg∙m−3)	0.0899	0.0899	0.0899	0.0899
mH2 (kg)	5.01543 × 10−5	4.58298 × 10−5	4.04589 × 10−5	3.73784 × 10−5
mMH (kg)	0.3	0.3	0.3	0.3
αMH (wt. %)	1.67	1.53	1.35	1.25

.

The separation ability of the MH alloy during the process of the purification of the binary mixture of H₂ and CO₂ was examined after the second absorption–desorption test (CO₂ “test 2”, Figure 6). Prior to the separation, a sample was collected from the supply pipe of the mixing pressure tank and analysed in accredited laboratories together with the samples of gaseous mixtures collected after the separation. An analysis of the gaseous mixture was carried out with the use of gas chromatography (methods: GC-TCD, GC-FID) in the Labortest U.S. Steel Košice accredited laboratory. The mixture entering the separation process, unlike the pre-defined theoretical value of the mixing ratio of 96:4 (vol. %), was different due to an imperfect mixing of the mixture with a 96.7:3.3 vol. %. After the completion of the absorption cycle, the reduction in the pressure in the MH tank to the atmospheric pressure, and the use of a rotary vacuum pump, the atmosphere inside the pressure vessel of the hydrogen separator unit was exhausted during the slow opening of the fine supply valve, which lasted for 2 s. Subsequently, the MH tank was closed, and with the use of thermal management, the hydrogen was desorbed from the structure of the MH alloy. The collected samples of the gaseous mixture contained: 99.43% H₂ and a 0.57 vol. % of CO₂; and 99.60% H₂ and 0.40% CO₂. The presence of CO₂ in the gas mixture indicated the need for a prolonged exhaustion of the ballast (CO₂) from the MH tank of the separation unit after the completion of the absorption process. However, with regard to the nature and characteristics of the used type of MH alloy, this leads to an undesirable loss of the separated hydrogen, which is released during the process of vacuuming the MH tank.

As the results of the experiments showed, the used type of MH alloy (Hydralloy C5), exhibited a low resistance to the poisonous effects of the carbon oxides. Therefore, from the point of view of the applicability of the separation unit in practice, it is important to find a more suitable type of MH alloy which will meet the requirement of a high resistance of the MH alloy to the gaseous components with a poisonous effect on the MH alloy, as well as a high number of absorption–desorption cycles, without a significant change in the hydrogen storage kinetics. A potential application is expected primarily for purifying hydrogen with a low content of ballast in the form of, for example, carbon oxides.

The degradation effects of CO₂ on the MH alloy may be seen in Figure 7. Figure 7a shows the surface of the activated Hydroalloy C5, which was exposed to pure hydrogen only. Figure 7b shows the surface of Hydroalloy C5 after ten absorption–desorption cycles, which was exposed to a mixture of H₂ and CO₂.

Figure 7b shows the oxidation effects of the mixture on the MH alloy and the formation of compact structures with significant oxidation. In the zone marked in green, the oxidation began to form manganese dioxide.

Due to the aforesaid facts, in particular, the fast degradation of the type of MH alloy, the significant deceleration of the hydrogen absorption kinetics after the completion of 10 absorption–desorption cycles, as well as the need for exhausting a relatively large amount of the gaseous mixture from the MH tank after the absorption cycle in order to remove the CO₂ present in the tank after the separation of hydrogen from the binary mixture, it is necessary to optimise the design of the separation unit. In order to reduce the alloy degradation and increase the effectiveness of the hydrogen separation from a mixture of gases, the design of the MH tank must be supplemented with a slightly porous partition plate, to be installed inside the tank in order to ensure the MH alloy is maintained in the upper section of the pressure vessel where the hydrogen accumulates. Due to the higher density of CO₂, it will accumulate in the lower section of the MH pressure tank, which will facilitate the ideal exhaustion of CO₂ after the completion of the absorption–desorption cycle. Another alternative is the use of an MH alloy which exhibits a high resistance to the poisonous effects of O₂ and carbon oxides. These issues were also investigated by Modibane et al., who performed the separation of hydrogen from a gaseous mixture using an MH alloy of the AB5 type, in particular, metallic hydride materials that offer more attractive solutions to the related problems—La (Ni, Co, Mn, Al)₅. A higher tolerance to poisoning was achieved by modifying the alloy’s surface by fluorination, followed by an electroless deposition of palladium. Integration of the surface-modified material into a prototype hydrogen separation system has shown the feasibility of its application for hydrogen separation from feed gas containing H₂ at a partial pressure of 2.5 bar, with up to 30% of CO₂ and up to 100 ppm of CO. Despite the low process productivity, caused by the slow H₂ absorption due to mass transfer limitations in the gas phase, the hydrogen separation was characterised by stable performances throughout hundreds operation cycles, with only a minor deterioration effect caused by the presence of CO₂ and CO [23].

A significant difference in the density of the individual components with certain mixtures of gases containing high contents of hydrogen may constitute a benefit and may facilitate, after a certain period of time, the gravitational separation of components in the mixing tank. An empty vertical mixing pressure tank, without the use of glass spheres, was again filled, after the completion of the experiments, with the binary mixture of hydrogen and carbon dioxide at a ratio identical to that used in the tests conducted with the hydrogen separator unit. After a period of 30 min after closing the pressure tank, its contents were slowly discharged through a discharge valve installed in the centre of the top base of the tank. During this process, the concentration of carbon dioxide in the mixture was monitored at the outlet from the pressure tank, where the analyser of the Gas Sensor: XDI/XDIwin—F1 type was installed.

A gravitational separation of the individual gas components in the pressure tank after 30 min was confirmed by the results of the experiment. A high proportion of hydrogen in the mixture, compared to the predicted value, was observed until the elapse of one third of the total time of discharging the tank. A gradual decrease in the pressure in the tank, and a decrease in the volume of hydrogen in the pressure tank, further caused a gradual increase in the concentration of CO₂ in the discharged mixture. The pre-defined value of the CO₂ concentration in the mixture (4%) was observed for a short period lasting a few seconds after the lapsing of approximately one half of the total duration of the experiment (Figure 8). The initial pressure in the tank was 0.8 MPa, while with a higher concentration of CO₂ (40%), this value amounted to 4.8 MPa. After opening the fine supply valve, the pressure gradually decreased until it reached the ambient pressure.

A similar trend was observed after the concentration of carbon dioxide in the binary mixture was increased to 40 vol. %. Based on the performed tests, it is predicted that with a gradual discharging of the tank, it will be possible to remove a mixture from a storage tank with a high concentration of hydrogen, just by applying the principle of gravitational separation. However, such a solution requires a suitable design of the storage tank, with a possibility of removing the mixture from the storage tank at the highest possible spot.

7. Conclusions

The transition of global economies to green energy is a process that is both costly and time consuming. In order to achieve this aim, a transitional period will be required during which the general energy demand should be addressed by using, to a considerable extent, wastes that are otherwise unusable. The process of generating gaseous mixtures that are rich in hydrogen logically makes the issue of hydrogen separation and its purposeful utilisation relevant to various economy sectors. One of the potential solutions for the hydrogen separation from a mixture of gases is to separate it while using metal hydride alloys that are capable of storing hydrogen in their structure.

The possibility of hydrogen separation from a binary mixture of hydrogen and carbon dioxide was tested using a newly designed hydrogen separator unit, which was operated in a semi-automatic mode. Thanks to the modularity of the system, the measuring equipment may be subjected to a simple structural modification that is necessary in order to perform the testing of various samples of MH alloys.

The functionality of the measuring equipment, as well as the possibility of using MH alloys and achieving a higher purity of hydrogen at the outlet from the hydrogen separator unit, was confirmed by the results of the performed experiments. The collected samples of hydrogen desorbed from the MH material after the second absorption–desorption cycle exhibited a hydrogen purity of 99.4 vol. %, i.e., 2.7 vol. % higher than that of the mixture entering the separator. The experiments also confirmed the low poisoning tolerance of Hydralloy C5 when exposed to a mixture of hydrogen and carbon dioxide. Based on a summary of the obtained data, the following conclusions can be made:

➢

The poisoning tolerance of the MH alloy that was not subjected to any specific modification was very low. Its direct application for separating hydrogen from a mixture of gases containing a significant amount of ballast in the form of oxygen and carbon oxides is not profitable. Also, based on the performed tests, it is possible to report a significant degradation of the MH alloy due to the effects of oxygen and carbon oxides, which confirms that MH storage systems may only be applied when high-purity hydrogen is being stored.

➢

The poisonous effects of carbon dioxide on the alloy were so substantial that after the completion of 10 absorption–desorption cycles, the alloy was not suitable for the intended purpose. Its degradation significantly reduced its storage capacity (from 1.7 wt. % to 1.2 wt. %), and even more significantly, it prolonged the kinetics of hydrogen storage, which is a negative factor and excludes the future direct use of this particular alloy without any modification that would increase its poisoning tolerance.

➢

Furthermore, the unmodified Hydralloy C5 is not predicted to be applicable in a hydrogen separator unit for the purification of hydrogen with trace amounts of undesirable components, as the degradation effects might be spread over time.

➢

It is expected that the use of an MH alloy with a high poisoning tolerance will facilitate the continuous storage of hydrogen in an alloy until the alloy becomes fully saturated, while a smooth process may be ensured by using at least two tanks arranged in parallel.

The used type of MH alloy exhibited low resistance to the poisonous effects of the carbon oxides. From the point of view of the applicability of the separation unit in practice, it is important to find an MH alloy with a more suitable composition, as well as a high number of absorption–desorption cycles, without a significant change in the hydrogen storage kinetics. A potential application is expected primarily for purifying hydrogen with a low content of ballast in the form of, for example, carbon oxides.

One of the potential supporting solutions to the issues of separating hydrogen from a mixture of gases and increasing its concentration in the mixture entering the process of future purification is to use specifically modified accumulation tanks, based on the principle of gravitational separation of the individual gas components. A slow discharge of the highly concentrated light gases from the upper section of an accumulation tank will facilitate, to a limited extent, increasing the chances of achieving a significantly higher hydrogen concentration in the mixture.