Experimental Hydrogen Plant with Metal Hydrides to Store and Generate Electrical Power

2.1 Electrolyzers

Electrolyzers are devices used to break the water molecule into its hydrogen and oxygen components. Currently, there are three types of electrolyzers distinguished according to the type of electrolyte used and temperature of operation: alkaline, proton exchange membrane and solid oxide.

1. Alkaline Electrolyzers: a mature technology, characterized by two electrodes immersed into a liquid alkaline electrolyte composed by a solution of 20–30% of potassium hydroxide (KOH). A diaphragm involving the gases produced in the process to ensure high efficiency and safety separates the two electrodes. The diaphragm should also be permeable to hydroxide ions and water [7, 8]. The energy consumption in these cells ranges from 4.2 to 5.9 kWh/Nm³, while in the complete system (stack, converters, pumps and valves) 4.5–7.0 kWh/Nm³. These electrolyzers operate at temperatures between 60 and 80°C and less than 30 bar pressure for a few megawatts [9].

2. PEM Electrolyzers: are composed by solid proton exchange membranes (PEM). A compact design and operation at high pressures are their main advantages. The energy consumption in the cells is about 4.2–5.6 kWh/Nm³, while for the complete system is about 4.5–7.5 kWh/Nm³. These electrolyzers operate at temperatures between 50 and 80°C and pressures less than 30 bar [9].

3. Solid Oxide Electrolyzers (SOEC): in this type of electrolyzer, the hydrogen production is performed at high efficiency, flexible chemistry, high temperatures and pressures, eliminating noble catalysts [9].

2.2 Hydrogen storage

Hydrogen can be stored in various forms, such as gas, liquid, or also as an intermediate compound. In this project were considered the following forms:

1. Hydrogen storage in gas form: in this natural state, hydrogen has low density (large volumes), leading to difficulties in its storage process. One solution is to store the hydrogen gas by compressing it in high-pressure cylinders. The hydrogen mass becomes 6–10% volume of the total storage at pressures of 35–70 MPa. Although commercially available, this form of storage is expensive with high-energy consumption, taking about 30% of the hydrogen energy, troubles with mechanical fracture and insecurity, and the need of an appreciable compressor structure [10].

2. Hydrogen storage in liquid form: is the more compact storage means. The hydrogen is stored in cryogenic tanks, typically in isolated cylinders, with a 20% hydrogen mass at pressures of 0.1 MPa at –253°C. The drawbacks of this method are the problems associated with the loss of energy during the H₂ liquefaction and evaporation stages [10].

3. Hydrogen storage in intermediate compounds: In this method of storage, the hydrogen is adsorbed in solid state of metal hydrides (MH). Under these circumstances, the hydrogen takes a potential maximum of about 7% of mass with 90 kg/m³, at pressures of 0.1–6.0 MPa. The drawbacks of this type of storage are the distinct adsorption and desorption temperatures. The main advantages are the lower working pressure, representing higher security of operation, and small recharging times, besides occupying lower volumes. The formation/dissociation processes in MH storage tanks require a thermal exchanging system. During formation of the hydrides (exothermic reaction) the tank needs cooling, whereas during dissociation (endothermic reaction) the tank requires heat to recover the stored hydrogen [10, 11].

2.3 Fuel cells

The electrochemical device called fuel cell (FC) consists of an electrolyte (membrane) conductor of ions, an anode, and a porous cathode, which convert energy from chemical reactions directly into electricity. For FC PEM (Proton Exchange Membrane), the electrochemical combination of the fuel (hydrogen) with the oxidizer (oxygen) produces continuous electricity, heat and water [12, 13].

This type of FC operates at temperatures around 20–80°C, with an efficiency of around 40–60%, and operates without any damage to the membrane if kept well hydrated. The relative humidity of the membrane has to be kept in a range between 85 and 100%, but without condensation [10].

Due to their characteristics of operation, the PEM FCs are been highly considered candidates for distributed energy generation, due to its relatively low operating temperature, quick start and its production of water as waste and high power density [10].

3.1 Hydrogen bus

The experimental power plant proposed in this paper is made of three hydrogen sources (pressurized tanks, electrolyzer and metal hydrides) and two consumer loads (metal hydrides and FC). A hydrogen circuit with sensors and valves performs measurement and control of the whole process, as illustrated in Figures 2 and 3. Thus, by monitoring some vital points and using a control program, it is possible to establish different conduction gas pathways in the hydrogen circuit.

From pressurized tanks, the hydrogen is injected into FCs or stored in metal hydrides in accordance to an opening and closure combination of valves. The purified hydrogen generated by the electrolyzer is stored in metal hydrides or injected directly into the FC. In a similar way, the hydrogen stored in metal hydrides is used to supply the fuel cell.

In Figure 2, the hydrogen mass sensor is strategically located to measure the hydrogen flow from any of the three sources or consumers. Another sensor measures the pressure in the equipment.

In addition, manual pressure control valves, to increase the plant safety levels, were installed next the equipment and to serve as decontaminant drains to remove any possible unwanted residues when a pipe is disconnected.

3.2 Alkaline electrolyzer

The choice of the alkaline electrolyzer for this project is due to its technological maturity level and easy commercial availability.

The hydrogen production is made by a commercial alkaline electrolyzer, HP model manufactured by Piel-McPhy Energy. Technical specifications of this electrolyzer are listed in Table 1 [14].

The electrolyzer data assure a hydrogen purity of 99.5%. However, according to the FC manufacturer’s specifications, the minimum purity of the input hydrogen must be at least 99.995%. Then to elevate the purity degree and ensure the FC proper functioning, one purifier was installed at the output of the electrolyzer increasing hydrogen purity to 99.999%.

In addition to the above precautions, the water in the electrolyzer had to be free from minerals that could impair its safe operation. Thus, it was installed an inlet demineralizer.

3.3 Metal hydride

Hydrogen storage in metal hydrides (MH) is used in this project because it works at low pressures and do not demand for pressurizers (reductions of cost, maintenance and losses). So, the plant safety is higher since generation, consumption (FC) and hydrogen storage work at low pressures (<15 bar).

The MH cylinder used in this project is the HBOND-7000L model manufactured by LabTech, whose technical specifications are listed in Table 2. The layout is presented in Figure 4 [15].

Figure 4:

Layout of metal hydrides tanks in the cylinder.

The typical hydrogen storage reaction as an intermediate compound in the adsorption and desorption in a metal alloy is shown in (1) [16].

The direct interaction is an exothermic process (heat release Q) in which the hydrogen is adsorbed by the alloy forming the metal hydride compound. In the reverse reaction, the hydrogen is released from the alloy by supplying certain amount of heat energy (exothermic reaction) [16].

where M is a metal alloy (LaNi5, TiFe, etc.), and (s) and (g) refer to the solid and gas states, respectively.

The heat released in the adsorption process must be removed, since the adsorption rate decreases with an increasing temperature [17]. For that, water circulates (at ambient temperature, approximately 25°C in the internal pipes of the cylinder (Figure 4). On the other hand, desorption of hydrogen is an endothermic reaction, which removes heat from the environment. Therefore, the metal hydrides must be heated so that the stored hydrogen is released by circulating heated water through the internal pipes of the cylinder [18–20].

Selection of the water source (hot or cold) in the MH thermal exchanges was made through activation of solenoid valves to configure adequately the circuit as shown in Figure 5. The pump is of a membrane type and has a speed control performed by a monitor and control program also used to control the solenoid valves (V1, V2, V3 and V4), monitors the water flow (F1), the water temperatures in hot (T1) and cold (T2) reservoirs, and water temperature of the cylinder output (T3).

Figure 5:

Layout of the MH cylinder temperature control.

3.4 PEM fuel cell

Selection of this type of FC was due to its technological maturity level, commercial availability, low operating temperature, quick startup time, and cause only of a low environmental impact.

The energy storage plant discussed in this paper has a fuel cell module depicted in Figures 2 and 3. The FC module model is the GreenHub 3000 model manufactured by Horizon Fuel Cells Technologies. Each module is equipped with the FC basic controls, such as temperature, pressure of the inlet reagents, periodic short circuiting system for voltage recovery, and protection devices. The whole module contains also a DC/DC converter and a DC/AC single-phase inverter across its output providing 220 V/60 Hz. The main technical specifications of each FC module are listed in Table 3 [21].

3.5 Monitor and control program

To monitor and control the storage plant was used a supervisory program developed in a LabVIEW platform to provide high interactivity with operational quantities. A dedicated computer and data acquisition system were used to track the main variables.

Data flow between the supervisory program and the plant is shown in Figure 6 with the controlled variables.

Figure 6:

Data flow between the plant and the supervisory program.

4.1 Electrolyzer

The electrical signals captured along the electrolyzer experimental tests were stack output voltage and current (electrolyzer unit itself) and the whole module (stack, power supply, power converter, ventilation system, valves, and circuits of monitoring, actuation and control, etc.).

The stack internal pressure was set at 10 bar to allow measurement of the net consumed and converted energies into hydrogen storage, because the hydrogen volume drained at electrolyzer output is equal to its maximum capacity.

Figure 7 illustrates the electrolyzer operation and the measured powers. The H₂ flow and energy density (3.29 kWh/L) were used to calculate the amount of converted power in H₂ [12].

Figure 7:

Powers in electrolyzer operation.

During the electrolyzer operation, the power consumption has increased as well as the hydrogen production and, consequently, the electrolyzer efficiency, as shown in Figure 8. This efficiency increases from 30 to 40% due to elevation of the internal electrolyzer temperature. During these experimental tests, the electrolyzer temperature changed from 30 to near 50°C.

Figure 8:

Efficiencies and hydrogen pressure in electrolyzer operation.

To verify the temperature dependence, an experiment was carried out with the electrolyzer with an internal operating temperature of 49°C. The average hydrogen production, obtained from the flow sensor signals and the electrolyzer efficiency are shown, respectively, in Figures 9 and 10.

Figure 9:

Hydrogen production by electrolyzer at a constant temperature.

Figure 10:

Electrolyzer efficiency at a constant temperature.

In Figure 10 can be seen that the average electrolyzer efficiency is around 33% for a 49°C operating temperature.

4.2 Fuel cell

The hydrogen used in the FC experimental tests came from the MH, which was previously generated by the electrolyzer.

To verify the fuel cell operation, a load was connected across the module output and was increased gradually. The powers involved during the tests are presented in Figure 11.

Figure 11:

Powers in fuel cell operation.

The difference between the FC stack output power and the load-consumed power is due to module internal losses (power converter, ventilation system, valves, and circuits of monitoring, actuation and control, etc.).

It is observable in Figure 11, that the hydrogen consumed power has numerous peaks. These surges are caused by the FC module internal control of pressurized H₂ to expunge water, which is result of the chemical reactions, to the outside [22–24]. This expunge technique introduces hydrogen losses, and consequently, power losses.

Figure 12 presents the FC stack output voltage with the respective hydrogen consumption at 1 min lasting experiment. In this figure, the periodicity of hydrogen consumption peaks is about 5 s. However, as the rate of hydrogen release is independent of load, this method introduces a module fixed loss.

Figure 12:

FC stack output voltage and hydrogen consumption.

Similarly, short circuits across the commercial FC stack terminals are applied at fix periods every 10 s. After each short circuit there is a slight recovery of the original cell voltage, justifying so the use of this technique to humidify the cell membranes and to reduce the ohmic losses [25]. Similarly, the purge techniques of hydrogen and water are also performed periodically, without taking into account the voltage drop level by ohmic losses.

Figure 13 displays the FC stack and module efficiency variations along the experiment.

Figure 13:

Efficiencies in fuel cell operation.

In Figure 13 is noticeable that the FC stack efficiency rises with an increases load approximately up to 6 min due to fixed low losses compared to the load power magnitude. At about 7.5 min there is a load increase, causing a higher voltage drop across the stack terminals dropping its efficiency to 38%. The FC module efficiency at full load remains at approximately 28%. Note that the efficiency amounts were obtained from the relationship between the energy contained in the volume of hydrogen entering the FC module and the output power in stack and module. Thus, the efficiency calculations include the energy losses associated with hydrogen used to expunge water out of the membrane.

4.3 Metal hydrides

For operation of the storage plant, the metal hydride cylinder was previously charged with the hydrogen produced in the electrolyzer. During the experimental tests, MH supplied hydrogen to the fuel cell. Variations in the main quantities involved in the metal hydride operation are shown in Figure 14.

Figure 14:

Physical quantities in metal hydrides operation.

It is observable in Figure 14 that initially the hydrogen pressure decreases along the desorption process. However, after 4 min, the activation of the water circulation pump through the inner metal hydrides pipes reestablished the pressure line. On the other hand, after stopping the water pump, after 12 min, the line pressure dropped down again. The pressure reestablishment through forced water circulation is due to an increase of the hydride temperature since desorption is a temperature dependent process. The MH thermal exchanges were monitored by the water temperature rise at the cylinder output after starting the water pump.

The cold and hot waters used in the metal hydride heat exchanges are provided by renewable sources. The cold water comes from geothermal exchangers and the hot water, from solar heaters. Thus, the thermal energy exchanged with the metal hydride was neglected since no electricity was used in the hydrogen adsorption and desorption processes [26].