NL2016294B1 - Intermittent solar energy conversion: flexible electricity-battery-electrolyzer device. - Google Patents

Abstract

The invention provides a system comprising a photovoltaic cell and a hybrid batteryelectrolyzer configured to select one of (i) a battery charging mode, and (ii) an electrolyzing mode, in dependence of solar light received by the photovoltaic cell. The invention further provides method for providing one or more of electrical energy and hydrogen gas and/or for storing electrical energy, the method comprising using a system comprising a photovoltaic cell and a hybrid battery-electrolyzer configured to select one of (i) a battery charging mode, and (ii) an electrolyzing mode, in dependence of solar light received by the photovoltaic cell, allowing during a period of time solar light to reach the photovoltaic cell, and providing one or more of electrical energy and hydrogen gas and/or storing electrical energy.

Description

Intermittent solar energy conversion: flexible electricity-battery-electrolyzer device FIELD OF THE INVENTION

The invention relates to a system for generating energy or storing energy. The invention further relates to an arrangement comprising such system. Yet further, the invention also relates to a method for providing electrical energy or hydrogen gas or for storing electrical energy.

BACKGROUND OF THE INVENTION

The energy sector is nowadays facing a deep transformation. This transformation is related to the shift from a fossil-fuel based society to a so-called renewables based society. The drivers for such a transformation are mainly: a) the expected increase in energy demand especially in the transportation sector, b) the projected further shift of the population from rural areas to metropolitan areas, c) the depletion of easily accessible fossil fuel resources, and d) the socio-economical consequences of air pollution and climate changes. In this context, the energy demand is highly fluctuating and so are the renewable energy sources. To match the fluctuating curves of demand and supply is possibly one of the greatest sustainable energy challenges.

The issue of fluctuating renewable energy generation is known in the art. US2012116603, for instance, describes systems and methods for associating a load demand with a variable power generation. For example, a method of providing power from a renewable resource includes receiving a signal including an output capability value for a renewable variable power generation module. An input power demand value is calculated for a load demand module based on the output capability value of the received signal. A customer load is controlled on a power transmission and distribution grid based on the calculated input power demand value.

SUMMARY OF THE INVENTION

The experience gained with the diffusion of rooftop solar PV (photo voltaic) has shown that the peak of PV production coincides with hours of little household demand. In the evening hours, when the electricity demand peaks, there is limited to no electricity production from the PY. In addition, there is a seasonal mismatch between sustainable energy production and energy demand.

Hence, it is an aspect of the invention to provide an alternative system, which preferably further at least partly obviates one or more of above-described drawbacks, which may be relatively simple and/or which may be relatively easily implementable. Further, it is an objective of the invention to provide a device able to adapt to varying (solar) illumination conditions and convert (solar) light into either chemical or electrical storage.

While batteries can deal with short term fluctuations, it appears that chemical storage in the form of e.g. H2 will be of paramount importance for seasonal storage and as a fuel for mobile applications. At present the storage of solar electricity into batteries or into hydrogen through electrolysis is generally not cost competitive. Thus, it is of importance to extend the efficiency and operational activity of storage devices. A rather unexplored way to do so is by optimizing the energy conversion efficiency of batteries or electrolyzers under variable illumination conditions.

Amongst others, it is herein proposed to directly couple the PV to the battery and/or electrolyzer.

The direct coupling of PV to batteries or electrolyzers presents the additional challenge of matching the voltage generated by the PV to the potential required to ran the electrochemical reactions. It appears that operating at the maximum power point of the PV module may be the optimum condition for both the electrolyzer and the battery module. This observation suggests that, for conventional Li batteries and electrolyzers, a choice has to be made on whether to store the solar energy as electricity or chemically.

Moreover, in a combined PV-storage system, efficiency arguments seems to recommend the direct use of electricity to be prioritized over its storage and reconversion. It often happens during the daily hours that the electricity production from the PV system exceeds the electricity demand. In this case the electricity should first meet the consumption and then the remaining fraction can be stored. However, this remaining power is going to be lower than, and in some cases not enough, to run the electrochemical reactions for the storage in a battery or electrolyzer system designed to handle the power generated by the PV system at full sun. An analogous situation occurs when the light intensity fluctuates significantly. At low light intensity as well as under diffuse irradiance, the voltage generated by the PV drops and it may not be enough to run the reactions. While the PV power output equals the product of current and voltage, which is a monotonically increasing function of the light intensity, to run the water splitting reactions a minimum potential is required. The mismatch described in the two examples is of a major concern as it will reduce the operative hours of the device. One option would be to oversize the PV module in order to supply the required power (voltage) to the electrolyzer. Alternatively a power conversion device must be supplied to the module. This increases the overall cost of the system.

Here, we demonstrate the use of a solar device composed of a PV cell combined with an integrated battery/electrolyzer cell (“battery/electrolyzer”). Since the battery and the electrolyzer operate at different potentials (the battery component loads from a AV of about 1.17 V and the electrolyzer operates with a AV larger than about 1.48 V), we obtain: a) the integration of electrical and chemical storage in one device; b) a self-regulating system where the electrical (daily) storage is prioritized over the chemical (seasonal) storage; c) compared to PV-electriolysis, an increased device efficiency at moderate light intensity due to the fact that the battery components have a lower threshold potential than the water splitting process. Furthermore, we demonstrate the behavior of an integrated PV-battery/electrolysis device under varying light intensity.

Amongst others, we herein chose two PV cells with an OCV (open circuit voltage) of 2.3 V and 1.8 V. The former to test the PV-battery/electrolyzer device operating near the MPP (maximum power point) of the solar cell. The latter to simulate the behavior of the battery/electrolyzer cell under limited applied potential. When the solar cell with larger OCV is used, the performance of the battery/electrolyzer is comparable to that of the electrolyzer at all light intensity. At low light intensity (10 mW/cm2), the open circuit voltage of the PV is reduced from 2.3 V to 2.1 V. This potential is enough to charge the battery and to perform the water splitting reactions and the performance is limited by the current generated by the solar cell. When the solar cell with smaller OCV is used, at low light intensity (10 mW/cm2), the open circuit voltage of the PV is reduced from ~ 1.8 V down to -1.4 V. At this reduced potential neither the charging of the battery nor the water splitting, can proceed. Therefore, electricity generation is the most efficient process. At moderate light intensity (50 mW/cm2) and above, the charging of the battery component is prioritized over the production of H2. When the battery is fully charged, the hydrogen evolution will take place. With fluctuating applied potential, the electricity is stored partly as electrical storage into the battery and partly as chemical storage in the form of H2. Finally, the energy stored as a battery can be released by discharging the electrodes. This closes the cycle, and regenerates the device. In addition, a modular photo-electrochemical (PEC) device where the PV part can be separated from the electrochemical module is advantageous compared to the integrated PEC approaches as it enables the electricity generation.

Hence, the invention provides a system (herein also indicated as “energy system” or “apparatus”) comprising a photovoltaic cell and a hybrid battery-electrolyzer (herein also called “battery-electrolyzer” or “battery/electrolyzer device”, or “battery/electrolyzer cell”) configured to select (at least) one of (i) a battery charging mode, and (ii) an electrolyzing mode, and optionally (iii) a battery discharging mode, in dependence of solar light received by the photovoltaic cell, and optionally load (i.e. here demand for electricity).

The system proposed herein offers the possibility to size the PV module and the battery electrodes precisely to meet the end-user’s need. Calculations for a residential house with a surface availability of 100 m2 for installation the PV-battery/electrolyzer device show that such a system could make the house independent from the grid. Additionally, enough H2 is produced yearly to provide 43 full tank filling and drive 25700 km with a H2 vehicle. This is in principle enough for 1 or 2 vehicles. The herein proposed flexible solar electricity/battery/electrolyzer device (“system”) seems to be a great promise for - amongts others - application in households. Further, the system may be relatively simple, substantially without electronics for controlling the modes. The system may operate autonomously and in a natural way provide and store electrical energy and provide H2 (and 02). Application of the system may especially be of interested between 40-70° north or south latitudes. Alternatively or additionally, the system may especially be of interest as it is especially able to cope with solar light changes, such as the changes from diffuse to direct irradiance or vice versa, such as when a cloudy sky changes to a blue sky or vice versa. The system may (thus) especially be used for providing one or more of electrical energy and hydrogen gas and/or for storing electrical energy.

The system has an electrical energy storage functionality and an electrolysis functionality. Hence, the apparatus is a combination of a battery and an electrolyzer. By charging the battery, the battery gets ready for use and further hydrogen is produced. Even when the battery is filled, hydrogen production can be continued. This provides a charged battery and hydrogen, which production can e.g. take place when no consumption of energy or energy carrier of the apparatus takes place. The term “energy” especially relates to electrical energy. The term “energy carrier” especially relates to hydrogen gas (H2), which can be used as fuel, e.g. for direct propulsion of an engine, but which may also indirectly be used, e.g. in a fuel cell for the generation of electricity.

Hence, the apparatus may especially be used as charging point for vehicles for electricity and/or hydrogen (and/or 02) (see also below). Hence, the battery-electrolyzer is especially a battery that can be charged (charging mode) and at the same time, or above a certain threshold, also can electrolyzes water (electrolyzing mode).

In embodiments, the hybrid battery-electrolyzer comprises a first cell comprising a first electrode, the first electrode in specific embodiments comprising a magnesium-titanium comprising electrode and a second cell, optionally separated from the first cell by a membrane (herein also indicated as separator), the second cell comprising a second electrode, the second electrode in specific embodiments comprising a nickel comprising electrode.

Hence, especially the system, even more especially the hybrid battery-electrolyzer, comprises a first cell, comprising a first cell electrode and one or more first cell openings for a (basic) first cell aqueous liquid (“liquid”) and for a first cell gas (H2), a second cell, comprising a second cell electrode and one or more second cell openings for a (basic) second cell aqueous liquid (“liquid”) and for a second cell gas (02), and (optionally) a separator, wherein the first cell and the second cell share the separator, wherein the separator is configured to block transport of one or more of 02 and H2 from one cell to another while having permeability for at least one or more of monovalent hydroxide (OH ), monovalent sodium (Na+), monovalent lithium (Li+) and monovalent potassium (K+).

During use, the first electrode and the second electrode are in contact with water. Further, during use the first electrode and the second electrode are in contact with an electrolyte. Hence, in specific embodiments during use the first electrode and the second electrode are in contact with an aqueous liquid. The aqueous liquid used is especially a basic aqueous liquid, such as comprising one or more of KOH, LiOH, and NaOH. Especially, the concentration of OH~ is at least 0.5 mol/1, even more especially at least 1 mo 1/1, yet even more especially at least 3 mol/1. Especially, the concentration of the hydroxide (especially one or more of KOH, NaOH and LiOH) in water is in the range of 4.5 - 8.4 mol/L (25-47 wt.% for KOH). The dissolved salts, such as one or more of KOH, LiOH, and NaOH, can be used as electro lyte(s).

As indicated above, in specific embodiments, the first electrode comprises a magnesium-titanium comprising electrode (see also below). Especially, the magnesium-titanium comprising electrode comprises a Mg(i_X)Tix electrode (can also be indicated as “MgTi electrode”), with x selected from the range of 0.05-0.95, especially x being in the range of 0.1-0.6, such as a Mgo.sTio.2 electrode. In yet further embodiments, the magnesium-titanium comprising electrode further comprises a Pd catalyst. The magnesium, titanium and optional palladium, may be provided as electrode via a sputter-deposition process. The Pd catalyst may be a thin top layer, such as in the range of about 2-20 nm (on top of the Mg(i_ x)Tix). Alternative to a magnesium-titanium electrode a MischMetal-based electrode (such as in Ni-MH batteries), or another battery electrode which charges at a potential positive to that of the water reduction reaction.

Further, as indicated above, in specific embodiments the second electrode comprises a nickel comprising electrode (see also below). Alternative to the nickel comprising electrode a Ni-based electrode also including one or more of cobalt and iron might be applied. The nickel comprising electrode may e.g. be a NiOOH electrode, such as e.g. described in US6020088, EP0225000 or US4985318, which are herein incorporated by reference. Such electrodes also comprising cobalt or iron are e.g. described in WO 2006020167, which is herein incorporated by reference.

With such electrodes, a battery-electrolyzer can be configured having a (battery) voltage difference selected from the range of 0.8-2.2 Y at RT, such as in the range of 1.0-1.5 V, and having a current density selected from the range of 0.001-2000 A/cm2.

Especially, the absolute (maximum) potential difference between the electrodes during electrolyzing mode or during battery charging mode should substantially be the same, i.e. both between about 1.0-2.2 V, like in the range of 1.2-1.5 V. The (absolute) difference between the (absolute) potential differences is especially equal to or smaller than 0.5 V, such as equal to or smaller than 0.3 V.

Especially, the minimum potential to operate the electrolyzer is about 1.4 V, especially at least about 1.48 V. In order to reach practical values of current density (which may correspond to a certain rate of FE production) in the order of 8-2000 mA/cm2, the potential is especially in the range of about ~ 1.6-2.0 V. The battery charging may about be the only process occurring at potentials in the range of about 1.0-1.6 Y, especially 1.17-1.48 V. Above which there can be a competing process with the FE evolution.

Hence, in embodiments the herein described method may comprise charging the hybrid battery-electrolyzer in (“during”) a charging mode when there is a potential difference between two electrodes (i.e. the first electrode and the second electrode) of the hybrid battery-electrolyzer of at least 1 V, even more especially at least about 1.1 V, such as at least about 1.17 V, and comprising electrolyzing water in the electrolyzer mode when there is a potential difference between said two electrodes of at least 1.4 V, such as at least about 1.48 V, like at least about 1.5 V.

The battery may discharge in a potential range 1.0-1.2 V or lower.

Hence, in embodiments the electrodes are configured to have an absolute (maximum) potential difference between the electrodes during the electrolyzing mode in the range of 1.2- 2.2 V, like in the range of 1.5-2.0 V, and/or the electrodes are (also) configured to have a potential difference between the electrodes during a battery discharging mode in the range of 0.6-1.2 V, like in the range 0.8-1.2 V, such as lower than 1.2V.

The first cell and the second cell especially share a separator (or membrane), and are separated from each other by this separator. Hence, liquid may not flow form one cell to the other via the separator. Also, hydrogen gas and/or oxygen gas may not flow from one cell to the other via the separator. However, the separator may be permeable for specific ions, such as at least one or more of OH' ions, neutral H2O, monovalent sodium (Na+), monovalent lithium (Li+), and monovalent potassium (K+). Hence, the first cell and the second cell share the separator, wherein the separator is configured to block transport of one or more of O2 and H2 from one cell to another while having permeability for at least one or more of OH' ions, neutral H2O, monovalent sodium (Na+), monovalent lithium (Li+), and monovalent potassium (K+), especially all. Hence, especially the separator may have a relative high ionic conductivity and a relatively low ionic resistance. For instance, the ionic resistance is lower than <0.3 D.cm2 in 30 wt.% KOH solution (at 30°C). The separator may e.g. comprise a membrane, such as electrolysis membranes known in the art. Examples of membranes may e.g. include alkaline resistant polymer membranes and polymer composite mambranes, such as e.g. a Zirfon (from Agfa) membrane. Such membrane may e.g. consist of a polymer matrix in which ceramic micro-particles (zirconium oxide) are embedded. This body is reinforced internally with a mesh fabric made from monofilament polyphenylene sulphide (PPS) or polypropylene (PP) fabric. It has a controlled pore size of about 0.15 pm and bubble point (especially defined as gas pressure against one side of the membrane required to form bubbles at the other side where there is liquid) of about 2 +/- 1 bar (over pressure). Such membrane may be permanently hydrophilic, by incorporated metal oxide particles, perfectly wettable in water and most common electrolytes. Such membrane may be stable in strong alkaline (up to 6M KOH) and up to 110°C. The pore size may e.g. be in the range of about 0.05-0.3 pm, such as about 0.15 pm; the thickness may e.g. be in the range of about 100-1000 pm, such as about 500 pm. Between the separator and each electrode, a respective spacer may be configured. These spacers may include openings for transport of the aqueous liquids and pro vinding acces for these liquids to the respective electrode.

Further, each cell may also comprise a further opening, especially configured for removal of the aqueous liquid and/or for removal of gas. Both may escape from the same opening. The first cell gas especially comprises H2 gas; the second cell gas especially comprises O2. The aqueous liquid in the cell and the cell gas may escape from the same opening. Alternatively or additionally, two or more openings may be used, e.g. one for the removal of aqueous liquid and one for the removal of gas.

As each cell has two openings, the aqueous liquid may be flowed through each cell, where the flow aids in gas removal, cooling (or heating) when necessary and water refilling. Depending on the applied current per cm2 electrode surface area the flow (in volume/area/time) may be for instance in the range of about 0.3 pl/cm2/h - 3.5 ml/cm2/h (with the former value approximately corresponding to the value of 0.001 A/cm2, and the latter value approximately corresponding to the value of 10 A/cm2).

For a good processing, the system may comprise one or more of an aqueous liquid control system, a gas storage system, a pressure system, etc.. Further, additionally the system may comprise a thermal management system and/or thermal insulation. Especially, the energy system comprises all these items.

Hence, in an embodiment the energy system may further comprise an aqueous liquid control system configured to control introduction of one or more of the first cell aqueous liquid and the second cell aqueous liquid. Such aqueous liquid control system may include one or more valves. Further, such aqueous liquid control system may - during operation -functionally be connected with a service pipe for water. In combination with the pressure system (see also below), the aqueous liquid may also be provided under pressure to the cells. Further, the aqueous liquid control system may include a storage for caustics, such as one or more of NaOH, LiOH, and KOH, especially at least KOH. The aqueous liquid control system may independently provide the liquid to the first cell and the second cell. Further, the aqueous liquid control system may include a return system, configured to receive the liquid downstream from the first cell and/or the second cell and reuse at least part of the first liquid and/or second liquid. Hence, especially during use of the system, there may be a constant flow of aqueous liquid, especially water, to the first cell and the second cell. The term “aqueous liquid control system” may also refer to a plurality of aqueous liquid control systems.

Further, in an embodiment the system may further comprise a storage system configured to store one or more of the first cell gas and the second cell gas external from said cells. To this end the system may comprise a storage system configured to store H2 and/or a storage configured to store 02. At least, the system may comprise a storage configured to store H2. In combination with the pressure system (see also below), the storage system may also be configured to store the one or more of the first cell gas and the second cell gas under pressure (see further also below). The term “storage system” may also refer to a plurality of storage systems.

Hence, in an embodiment the system may further comprise a pressure system configured to control one or more of (a) the pressure of the first cell gas in the first cell, (b) the pressure of the first cell gas in the storage system, (c) the pressure of the second cell gas in the second cell, and (d) the pressure of the second cell gas in the storage system. To this end, the pressure system may comprise a pump, a valve, etc.. In an embodiment, the pressurize system essentially comprises one or more valves. The term “pressure system” may also refer to a plurality of pressure systems. Especially when two or more different types of fluids have to be pressurized, two or more independent pressurize systems may be applied. A first pressure in the first cell and a second pressure in the second cell at a pressure may be maintained at atmospheric pressure or above, such as selected from the range of 5-50 bar, like 10-30 bar. The pressures in the first cell and second cell may be maintained independently of each other.

Further, as indicated above, the system comprises a photovoltaic cell (PV) or solar cell. Such cell is especially an electrical device that converts the energy of (solar) light directly into electricity (by the photovoltaic effect). For the present application, especially PV’s are desired that have an open circuit voltage of at maximum in the range of about 1.5-2.5 V. Hence, in embodiments the photovoltaic cell is configured to provide an open circuit voltage (OCV) in the range of 1.5-2.5 V, such as a (maximum) OCV of 1.8 V, at sun light intensities over 75 mW/cm2, especially at sun light intensities in the range of 75-100 mW/cm2. In yet other embodiments, the OCV is even higher, such as in the range of 2.0-2.5 V at sun light intensities over 75 mW/cm2.

The PV may for instance be selected from the group consisting of crystalline silicon PV’s, such as a monocrystalline silicon PV, an epitaxial silicon PV, a polycrystalline silicon PV, a ribbon silicon PV, a mono-like-multi silicon PV, etc.. The PV may for instance also be selected from the group consisting of thin film PVs, such as a cadmium telluride PV, a copper indium gallium selenide PV, a silicon thin film PV and a gallium arsenide thin film PV. Also a perovskite PV may be applied. For instance, the silicon thin film PV may include amorphous silicon (a-Si or a-Si:H), protocrystalline silicon or nanocrystalline silicon (nc-Si or nc-Si:H), also called microcrystalline silicon. In specific embodiments, the photovoltaic cell comprises a multiple junction photovoltaic cell, such as a double junction, but especially a triple junction or a quadruple junction PV. In further specific embodiments, the photovoltaic cell comprises one or more of a triple junction a-Si-nc-Si-nc-Si photovoltaic cell and a quadruple junction a-Si-a-Si-nc-Si-nc-Si photovoltaic cell.

Especially, the battery-electrolyzer is functionally coupled to the PV. An advantage of the system is that the battery-electrolyzer and the PV may be coupled without the need for additional electronics, such as an AC-DC converter, etc., or (other) electronic control elements. Hence, especially in embodiments the photovoltaic cell and a hybrid battery-electrolyzer are electrically coupled without intermediate electronics. Hence, the PV may comprise two electrical contacts, over which under solar radiation a potential difference may be generated. The battery-electrolyzer comprises two electrodes (first electrode and second electrode) over which a potential difference may be applied to charge the battery-electrolyzer and/or to generate hydrogen gas or - when charged - over which the existing potential difference may be used to power an (external) apparatus. Hence, the electrodes of battery-electrolyzer and the respective electrical contacts of the PV may be in electrical contact, with electrical wires or other electrical conductive means, without further (intermediate) electronics.

Herein, the term “PV” may also refer to a plurality of (different) PV’s. Therefore, in embodiments the system comprises a plurality of (different) photovoltaic cells. Hence, in embodiments a plurality of (different) PVs may functionally be coupled with the battery-electrolyzer. Especially, the PV’s may in such embodiments be configured in series. However, in other embodiments the PV’s may also be configured in series. The system may also include a plurality of battery-electrolyzers. In general, each battery-electrolyzer is functionally coupled with at least one PV.

In yet further embodiments, the functional coupling of the PV and battery/electrolyzer may include a connector, which connector also includes the functionality to temporarily disconnect the functional coupling, such as a switch. For instance, when direct electricity of the PV is needed, for instance when the battery is low whereas solar light is e.g. well over 20 mW/cm2, such as over 50 mW/cm2, an energy consuming system or energy consuming apparatus, may directly be coupled to the PV. Then, the battery/electrolyzer may be disconnected to provide optimally electrical energy from the PV. A control system may be configured to control such connector. Hence, amongst others the system may also be configured to select one or more of (i) a battery charging mode, (ii) an electrolyzing mode, and (iii) an electrical energy delivery mode, in dependence of one or more of (a) the solar light flux, (b) a electrical charging level of the hybrid battery-electrolyzer, and (c) electricity demand (by an (external) energy consuming system or energy consuming apparatus) from the hybrid battery-electrolyzer.

The present system allows by its configuration charging, electrolyzing, discharging and H2 generation. The PV, the battery-electrolyzer and the intensity of the solar light, as well as consumption of electrical energy, may define which mode(s) is (are) active. Especially, the system is configured to select one or more of (i) a battery charging mode and (ii) an electrolyzing mode, in dependence of one or more of (a) the solar light flux, (b) a electrical charging level of the hybrid battery-electrolyzer, and optionally (c) electricity demand from the hybrid battery-electrolyzer (by an (external) energy consuming system or energy consuming apparatus). In yet embodiments, the system is configured to select one or more of (i) a battery charging mode and (ii) an electrolyzing mode in dependence of the one or more of the solar light flux and a electrical charging level of the hybrid battery-electrolyzer. Over a certain electrical charging level, electrolysis and charging may occur at the same time, and when the battery-electrolyzer is fully charged, only electrolysis may take place. Of course, during charging also electrical energy may be consumed by an external apparatus. Especially, the system is configured to select the battery charging mode at sun light intensities selected from the range of 10-75 mW/cm2 and configured to select the electrolyzing mode at sun light intensities over 75 mW/cm2. As indicated above, when the system is in an electrical energy delivery mode, the electrical energy may be retrieved from one or more of the PV and the battery-electrolyzer, for instance in dependence of the electrical energy demand and optionally the solar light flux.

In further embodiments, a geometrical area of the battery-electrolyzer is the same as a geometrical area of the photovoltaic cell. In yet other embodiments, the ratio of the geometrical areas may vary between about 10:1 - 1:10, such as 5:1-1:5, like 2:1-1:2. The term “geometrical area” may also refer to a cross-sectional area (parallel to the configuration of the electrodes, which are also especially configured parallel).

In specific embodiments, the battery-electrolyzer and the photovoltaic cell have a ratio of a geometrical area of the battery-electrolyzer and a geometrical area of the photovoltaic cell in the range of 5:1-1:5, and the system is configured to selected the (i) battery charging mode and/or (ii) the electrolyzing mode at sun light intensity in the range of at least 10 mW/cm2.

As during electrolysis H2 and O2 are produced, in embodiments the system may further comprise one or more of an H2 storage and an O2 storage, functionally coupled with the hybrid battery-electrolyzer (see also above, where a storage system is described).

Amongst others, the system may be used to electrically charge an external apparatus or system (herein also indicated as “energy consuming apparatus” or “energy consuming system”). The system may also be used to provide an external apparatus with H2 and/or O2. Of course, the system may also be used to electrically charge an external apparatus and to provide an external apparatus with H2 (and/or O2). Hence, in yet a further aspect the invention also provides a method for providing one or more of electrical energy and hydrogen gas (to an (external) system or apparatus) and/or for storing electrical energy, the method comprising using a system as defined herein, especially comprising a photovoltaic cell and a hybrid battery-electrolyzer configured to select at least one of (i) a battery charging mode, and (ii) an electrolyzing mode, and optionally (iii) a battery discharging mode in dependence of solar light received by the photovoltaic cell, allowing during a period of time solar light to reach the photovoltaic cell, and providing one or more of electrical energy and hydrogen gas (to the (external) system or apparatus) and/or storing electrical energy. Especially, the herein described system is used. The external (energy consuming)_apparatus may e.g. be a car (but see further also below for other external apparatus and system).

Hence, in embodiments the method may comprise providing one or more of electrical energy and hydrogen gas to a vehicle (temporarily) functionally coupled to the system. Yet further, in embodiments the method may comprise providing electrical energy to an energy consuming system (such as a home, etc., see also below). Alternatively or additionally, the method may comprise providing one or more of an H2 and O2 to the energy consuming system.

As indicated above, the system may in embodiments especially configured to select one or more of (i) a battery charging mode and (ii) an electrolyzing mode in dependence of the one or more of a solar light flux and optionally a electrical charging level of the hybrid battery-electrolyzer.

The system may be functionally connected to the electrical grid. Alternatively or additionally, the system may also be functionally coupled to a smaller energy infrastructure, such as of a home.

Hence, in yet a further aspect the invention also provides an arrangement comprising an energy consuming system and the system as defined herein functionally coupled with the energy consuming system, wherein the energy consuming system is selected from the group consisting of a home, an office, a plant, a hospital, a school, an (indoor) sport facility, etc. . The term “home” may especially refer to a house, an apartment, etc.. In specific embodiments, the term “home” may also refer to a mobile home, houseboat, etc.. The term “office” may especially refer to a room or a plurality of rooms or other area where administrative work is done. The term “plant” may especially refer to an industrial plant for manufacturing a tangible product.

In yet a further aspect, the invention provides an arrangement comprising an energy consuming system and the system as defined herein functionally coupled with the energy consuming system, wherein the energy consuming system is selected from the group consisting of infrastructural energy consuming systems, such as selected from a sluice, a moveable bridge, a system comprising one or more of a traffic light, signal light, an information display, a controllable road barrier, etc..

BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments of the invention will now be described, by way of example only, with reference to the accompanying schematic drawings in which corresponding reference symbols indicate corresponding parts, and in which:

Fig. 1: Schematic configuration for the bias-free solar hydrogen/battery setup. The light intensity is controlled by placing optical density (OD) filters between the solar simulator and the solar cell. The solar cell is coupled to the electrochemical cell and a current meter is connected in series. A nitrogen gas purges constantly the electrochemical cell, and the gas evolution is measured every 18 minutes by a gas chromatograph. Referenece 100 indicates a photovoltaic cell. Reference 701 indicates an optical density filter. Reference 703 indicates light of the light source 702, which is here a solar light simulator. Reference 703 is further also used to indicate solar light (ambient light). Reference 704 indicates a gas flow to a gas chromatograph (GC); sampling is e.g. each 18 minute. Reference 705 indicates an N2 purge. Reference 210 indicates a first cell, with first electrode 211; reference 220 indicates a second cell, with second electrode 222. A separator or membrane is in general avialable (but is not depicted). As reference, the first electrode is a Pt/MeH electrode (metalhydride electrode with Pt catalyst)(counter electrode), and as working electrode or second electrode in the reference a NiOOH electrode is applied. The deposition time for the NiOOH electrode in the electrolyzer or battery/electrolyzer mode, are 2 minutes and 30 minutes, respectively.

Fig. 2: Device components and schematic energy diagram with the available redox couples for the PV electrolysis configuration (2a), and for the PY battery/electrolysis configuration (2b). Light (photons) is shining on the PV cell generating electron-holes pair. Charge separation is provided by the internal electric field at each p-i-n junction. Electron hole recombination occurs at each interface. Each p-i-n junction boosts the energy of the electrons to a level higher of the H2O-H2 redox couple. The electrons are collected at the back of the solar cell, which is connected to the cathode in the electrochemical cell. The energy of the holes is lower than that of the OHVO2 redox couple. The holes are collected at the front of the solar cell, which is connected to the anode in the electrochemical cell. The energy difference between the redox couples for the battery reactions is lower than that for the water splitting (including overpotential). Reference 710 indicats a a glass substrate; reference 714 indicates a TCO (transparent conducting oxide) layer, such as indium tin oxide. Reference 713 indicates a charge collector based on e.g. an AgCrAl stack. Reference 711 indicates a glass substrate. Reference 715 indicates a TCO (transparent conducting oxide) layer, such as indium tin oxide. Reference 221, as indicated above, indicates a first electrode. Reference 230 indicates a separator or membrane. Reference 716 indicates a Pd catalyst layer and reference 717 indicates a (porous) PTFE (polytetrafluorethyleen) layer. The first electrode 221 in Fig. 2a is a Pt electrode; the first electrode in Fig. 2b is a MgTi electrode 221a. The second electrode in Fig. 2a is a NiOOH electrode; likewise, in the embodiment of Fig. 2b, the second electrode is e.g. a NiOOH electrode 222a. During testing, the deposition time for the NiOOH electrode in the electrolyzer or battery/electrolyzer configuration, respectively, is held at 2 minutes for the reference embodiment schematically depicted in Fig. 2a and 30 min for the emebodiment schematically depicted in Fig. 2b.Hence, the PV having two electrical contacts, here the TCO electrode 713 and the charge collector 713, is electrically coupled, without additional electronics, to the first electrode 221 (via a wire and the TCO layer 711) and the second electrode 222 (via a wire). As can be seen in Fig. 2b, a geometrical area of the battery-electrolyzer is the same as a geometrical area of the photovoltaic cell. This may provide an efficient system, though other ratios are not excluded.

Fig. 3: GC test. A) Two electrode voltage response to a galvanostatic current (0.1 mA, 0.3 mA and 1 mA) of the electrolyzer, with C indicating the current and V indicating the voltage/potential (P). B) Hydrogen and oxygen evolution during the galvanostitic measurement as detected by the GC. In Fig. 3a, on the Y as, the current I is indicated, and on the x-axis the elapsed time (ET) in minutes is indicated. On the right y-axis, the potential (P) in volts is indicated. In Fig. 3b, O2 indicates the O2 concentration (see also right y-axis) (dots) and H2 indicates the H2 concentration (see also right Y-axis) (squares).

Fig. 4: current-voltage characteristic of the triple junction a:Si-nc:Si-nc:Si. As the light intensity decreases, both the current density and the OCV are reduced (solid lines); the correspondent power curve is also shown (dashed lines). Light intensity is controlled with optical density filters OD 0.3 and OD 0.1. As light source light simulated AM 1.5 solar illumination (100 mWcm2) with a Newport Sol3A Class AAA solar simulator (type 94023 ASR3) is applied. In Fig. 4, on the right y-axis the current (I) in mA is indicated; on the right y-axis the power (E) in mW; on the x-axis the potential (P) in volt is indicated.

Fig. 5: Schematic illustrating the operating point for solar water splitting under AMI.5 illumination for a single absorber PV+electrocatalyst system (a). The power output under reduced illumination (10% in Fig. 5b relative to Fig. 5a) reduces, and the operating point follows until eventually water splitting is no longer possible (b). EL indicates electrolysis and OP indicates operation point.

Fig. 6: Current-voltage characteristic of the triple junction a:Si-nc:Si-nc:Si at different light intensitis, together with the JV characteristic (two electrode measurement) for the overall water splitting via the Pt-NiOOH electrolyzer. The schematic for this design configuration is shown in Fig. 2a. ELZ indicates “electrolyis” (and may also indicate “electrolyzer”).

Fig. 7: A) Current-voltage characteristic of the triple junction a:Si-nc:Si-nc:Si at different light intensity, together with the JV characteristic (two electrode measurement) for the overall water splitting via the Pt-NiOOH electrolyzer. Dashed lines show the JV characteristic before the long term measurement and the solid lines after the measurement. TR indicates room temperature, which applies for the dashed lines. The closed lines relate to a temperature above room temperature (at the beginning of the experiment the solar cell is at room temperature and performs best. After prolonged exposure of solar light the solar cell heats up and its performance reduces (straight lines)). B) Chronoamperometric measurements for the PV-electrolyzer configurationFig. 2a) under different light intensity: 1 sun (top curve), a 50% OD filter (middle curve), and a 90% OD filter (lowest curve). The current with a 90% filter becomes negligible. Experiments displayed in Figs. 7a-7b were performed at increasing light intensities.

Fig. 8: A) Current-voltage characteristic of the quadruple junction a:Si-a:Si-nc:Si-nc:Si (4jn PV, solid lines) at different light intensity, together with the JV characteristic (two electrode measurement) for the overall water splitting via the Pt-NiOOH electrolyzer. Dashed lines show the JV characteristic for the triple junction a:Si-nc:Si-nc:Si (3jn PV). B) Chronoamperometric measurements for the PV-electrolyzer configuration Fig. 2a top) under different light intensity: 1 sun (top), a 50% OD filter (middle), and a 90% OD filter (lowest curve). Experiments performed at increasing light intensity.

Fig. 9: Half-cell’s reactions at the anode (oxidation) and cathode (reduction) as a function of the applied potential. At the cathode side the formation of the metal hydride occurs at more anodic potential than the hydrogen evolution reaction, and at the anode side the formation of the NiOOH compound requires less energy than the oxygen evolution reaction. As the metal hydride (MH) formation starts at 0.0 V vs. RHE (reversible hydrogen electrode), and the NiOOH forms at 1.4 V vs. RHE, the difference between the two reactions indicates that 1.4 V are required to load the battery part. On the other hand, at least 1.6 volts are required to perform water splitting. Measurements performed in 1 Μ KOH. The scan rate was 50 mV/s in the oxidative side, and 5 mV/s in the reductive side (due to different electrode capacity). RS indicates the reductive side and OS indicates the oxidative side.

Fig. 10: Current-voltage characteristic of the quadruple junction a:Si-a:Si-nc:Si-nc:Si (4jn PV, solid lines) at different light intensity. Dashed lines (at AM 1.5, Opposition Division 0.3 en Opposition Division 1) show the JV characteristic for the triple junction a:Si-nc:Si-nc:Si (3jn PV). The two electrode JV characteristic for the overall water splitting via the Pt-NiOOH electrolyzer (ELZ closed line) (the schematic for the PV-battery/electrolyzer configuration is shown in Figure 2a), and the two electrode JV characteristic for the MeH-NiOOH battery/electrolyzer (ELZ dashed line) are also displayed; the start between about 1-1.5 V and steeply rise between 1.5-2 V. The schematic for the PV-battery/electrolyzer configuration is shown in Fig. 2b. Reference 200 indicates the battery - electrolyzer (Fig. 2b).

Fig. 11:A-C) Chronoamperometric measurements for the triple junction PV-battery/electrolyzer as in Fig. 2b (“B+ELZ” or ref. 200), and the PV-electrolyzer configuration as in Fig. 2a (“ELZ”) under different light intensity: 1 sun (AM 1.5), a 50% OD filter (OD0.3), and a 90% OD filter (OD1). The current in the PV-battery/electrolyzer configuration (“B+ELZ”) is always higher than in the PV-electrolyzer configuration (“ELZ”. Remarkably, the current for the PV-battery/electrolyzer with a 50% OD filter has a step-like behavior. It is 20x higher than for the for the PV-electrolyzer for the initial 1500 s, then decreases to ~ 3x the value for the PV-electrolyzer. The initially higher current is attributed to the charging of the MgTi battery electrode, until full charge is achieved. Experiments performed at increasing light intensity. D-F) Zoom in over the first minute of the chronoamperometric measurement. In the drawings, B indicates battery; ELZ indicates (again) electrolyzer and B+ELZ indicates battery + electrolyzer. The elapsed time (in seconds) is indicated with ET.

Fig. 12: Chronoamperometric measurements for the quadruple junction PV-battery/electrolyzer as in Fig. 2b (black lines), and the PV-electrolyzer configuration as in Fig. 2a (grey lines) under different light intensity: 1 sun (top two curves), a 50% OD filter (middle two curves), and a 90% OD filter (lowest two curves). The current for the PV-battery/electrolyzer under 1 sun has a step-like behavior. The initially higher current is tentatively attributed to the charging of the MgTi battery electrode, and the drop to its activity in electrolyzer mode. Experiments are performed at increasing light intensity. PV-B-ELZ indictates the system including the batter/electrolyzer and battery (Fig. 2b) (in the charging mode); PV-ELZ indicates the combination of PV and electrolyzer (Fig. 2a) during electrolyzer mode.

Fig. 13: Chronoamperometric measurements for the PV-battery/electrolyzer with a 50% OD filter (grey squares (B+BLZ)) as inFig. 11. H2 concentration measured by the GC (black squares). No H2 is detected after 420 and 1500 seconds. H2 detection starts from 2580 seconds, and increases at the 3660th second. The GC data support the interpretation of an initial higher current due to the Mgo.sTio.2 charging, until full charge is achieved. Thereafter, the current is going to hydrogen evolution. H2 indicates the H2 evolution. The time is in seconds, the left axis indicates the current (A) and the right axis the H2 concentration (%).

Fig. 14: Linear sweep voltammetry of the Mg0.8Tio.2 battery electrode at different scan rates. At a scan rate of 5 mV/s the battery electrode shows an onset potential at +0.2 V vs. RHE and a peak at -0.15 V vs. RHE, which is attributed to the hydrogenation of the electrode. At V < -0.22 Y vs. RHE the current density increases in value which is attributed to the hydrogen evolution. The two processes appear separate and can be distinguished with ease (blue). At a scan rate of 10 mV/s the battery electrode shows again an onset potential at 0.2 V vs. RHE, however there is no local maximum of current density. To understand whether the current goes to battery charging or to hydrogen evolution, we have calculated the integral of the JV curves in Figure 14 for the hypothetical hydrogen uptake and release for samples of MgTi of two thicknesses, 200 mn and 400 mn. The integral is calculated assuming all the current under the charging/discharging curves goes to the sorption of hydrogen into/out of the MgTi. When comparing the hydrogen uptake (current IN) to the hydrogen release (current OUT), the hydrogen uptake is always higher. Considering that the hydrogen release is on average at least 90% of the hydrogen uptake as shown in the galvanostatic measurement of Figure 16, the results indicate that upon charging, part of the current does not go to MH formation, but to H2 evolution.

Fig. 15: Polarization curves of the MgTi electrodes at different scan rates. Four features are distinguishable: a) hydrogen uptake into the MgTi electrode; b) hydrogen evolution; c) hydrogen release from the Pd catalyst; d) hydrogen release from the MgTi electrode. In fig. 15a, 200 nm refers to the 200 nm thickness of the MgTi layer of the MgTi electrode; in fig. 15b, 400 nm refers to the 200 nm thickness of the MgTi layer of the MgTi electrode.

Fig. 16: Galvanostatic charging and discharging of the MgTi electrode. The capacity of the electrode is indicated in mAh/g. Here, “a” indicates the charging at 5 A/g (ampere per gram material) (effectively 0.2 mA/cm2) and “b” indicates discharging at 1 A/g (effectively 0.04 mA/cm2).

Fig. 17: Chronoamperometric measurement for the discharging of the battery electrodes through the solar cell.

Fig. 18: Three electrodes galvanostatic charge/discharge of the Ni electrode (left), and of the Mgo.8Tio.2 electrode (right). The intersection of the charging/discharging curves is a good approximation of the potential for the given reversible redox couple, either Ni(OH)2 / NiOOH (18a), or Mgo.sTio.2 / Mgo.sTio.2H2 (18b). A steady potential over a certain time indicates the time for which a certain reaction is occurring, and can be interpreted as the capacitance of the electrode for the given redox reaction. An abrupt change in potential over time indicates the shift to a different reaction which enables the given current density to be maintained. The x-axis scale in Fig. 18a and Fig. 18b plots differs by 3 orders of magnitude, and the current density by 1. These information combined illustrate that the discharging capacity is limited by the Ni electrode. C indicates charging and SD indicates slow discharging; D indicates discharging.

Fig. 19: Three electrodes galvanostatic discharge of the Ni electrode and of the Mgo.8Tio.2 electrode at different discharge current. The intersection of the discharging curves represents the moment where there is no potential difference between the electrodes. It is therefore a good indication of the time necessary to discharge the battery. Discharge Ni(OH)2 is at -30 μΑ; dischrage -> MgTi is at + 50 μΑ, and slow discharge -> Ni(OH2) is at -3 μΑ.

Fig. 20: Schematic diagram of the device under operation. (I) When the electrochemical cell is fully discharged the Ni-based anode is in its low-energy transparent state Ni(OH)2 while the hydrogen-absorbing metal cathode is in the metallic state. If optionally electricity from the PV is needed, a switch () disconnects the electrochemical part and directs the electricity to the users. (II) If the electricity is not immediately needed, the electrochemical module is connected to the PV, and the device begins charging its electrodes by storing the energy according to Ni(OH)2 -> NiOOH at the anode, and Me -> MH at the cathode. This process occurs until the battery electrodes are fully charged. (III)-(IV) Electricity not needed. High light intensity (III), fully charged electrodes (IV. The device automatically switches from battery mode to electrolysis mode for the hydrogen production, just like in a conventional electrolyzer or a PEC cell, with OER (oxygen evolution reaction) and HER running over the efficient NiOOH and Pd catalysts, respectively. (V) Electricity needed. No light. The energy stored into the battery can be released. This occurs by discharging the Ni electrode according to Ni(OH)2 NiOOH, and the hydride electrode via the reaction Me MH. When the battery electrodes are fully discharged the cell is back at the initial point (i).

Reference 731 indicates a stage where electricity is needed (I), and wherein the system is fully discharged (reference 730). Reference 724 indicates a charging. Reference 721 indicates a situation where electricity is not needed (and electrical energy can be stored in the system). Reference 722 indciates a low light intensity stage (II); reference 723 indicates a high light intensity stage (III). Reference 726 indicates a stage where electricity is not needed; reference 725 indciates a fully charged stage. Reference 727 indicates a discharging and reference 728 indicates a stage where electricity is needed (and thus the discharge takes place)(V); energy gets out of the system, which is indicated with reference 729.

Fig. 21: Daily load profile of a residential house. The fraction of the electricity provided directly by the PV is taken to be 1/3 of the total daily consumption. The remaining electricity is to be provided by the battery or H2. El indicates “electricity from battery / H2” and E2 indicates “electricity from PV”. E on the y-axis indicates the power in Watt. On the y-axis the power (E) in watt (W) is indicated, and on the x-axis the time in hours (h).

Fig. 22a-22b schematically depict some embodiments and variants.

The schematic drawings are not necessarily on scale.

DETAILED DESCRIPTION OF THE EMBODIMENTS

Here below, first some experimental work is described. PV synthesis: Triple junctions solar cells where the absorber layers are amorphous silicon/nanocrystalline/nanocrystalline. Quadruple junctions where the absorber layers are amorphous silicon/amorphous silicon/nanocrystalline/nanocrystalline. Each of these cells have a p-i-n structure, p and n supporting layers are SiOx doped with boron and phosphorous respectively. They were deposited by PECVD in a textured glass. The front contact is 300 nm A1 (the front lines) and the back contacts are 100 nm Ag 30, nm Cr, and 500 nm A1 (AgCrAl stack).

Cathode synthesis: The synthesis of MgTi thin films has been carried out by sputter depositing a film of controlled composition of Mg and Ti targets (99.95% purity) onto FTO-coated glasses substrates (TEC-15, 15 Ω/sq; Hartford Glass Co.) in 3 pbars partial pressure of Ar (20 ml/min flow) with 100 W DC electric power, and a substrate temperature of 400°C. A 5 nm Ti adhesion layer was deposited first, followed by the MgTi film. The same parameters were used for the deposition of the Pd catalyst. Moreover the MgTi-Pd electrode is coated with 150 nm sputtered PTFE film. The same parameters were used for the deposition of the Pt film of 10 nm thickness.

Anode synthesis: Prior to the deposition process the substrates were cleaned by three successive cycles of ultrasonic rinsing in a 10% aqueous triton solution, acetone and isopropanol, for 15 min each time. All the chemicals were used as received, without any further purification. The water used to prepare all solutions was deionized and ultrafiltrated by a Milipore Milli-Q system (resistivity > 18.2 ΜΩ cm). The electrodeposition was performed in a single compartment electrochemical cell, using an Ag/AgCl (sat. KC1, sat. AgCl) (XR300, Radiometer Analytical) as reference electrode and a coiled Pt wire as a counter electrode. All potentials are reported versus the reversible hydrogen electrode (RHE), calculated according to the equation (1):

(1)

Where Erhe is the potential versus RHE, EAg/AgCi ([sat.KCl) is the potential applied experimentally and EAg/AgCl (sat.KCl) is the standard potential of the Ag/AgCl (sat. KC1) versus the normal hydrogen electrode (0.1976 V), ApH accounts for the difference in pH of the working solution respect to the normal hydrogen electrode (pH zero). The electrodeposition was performed with a potentiostat / galvanostat (EG&amp;G PAR 283). The Ni-based electrocatalysts were deposited galvanostatically. The galvanostatic electrodeposition of NiOOH was carried out at 300 pAcm'2. The time for NiOOH electrodeposition was calculated according to the real surface area of the working electrode and chosen as 2 min for the electrolyzer and 30 min for the battery/electrolyzer. Theoretical estimation of time needed to fabricate NiOOH films of desired thickness is difficult, since water oxidation reaction can readily occur at potentials facilitating NiOOH deposition.

Electrochemical measurements: The same setup as described in the previous section was used for voltammetric measurements. The polarization curves were obtained by cyclic voltammetry in 1 Μ KOH.

Device testing: Photoelectrochemical characterization was carried out in an aqueous 1 M KOH (pH -13.6). The solution was purged with nitrogen prior and during the measurements to remove any dissolved oxygen. The working area of the electrodes exposed to the electrolyte was 28.3 mm2 (6 mm diameter) for all samples. The potential of the working electrode was controlled by a potentiostat (EG&amp;G PAR 283). In three-electrode measurements, a coiled Pt wire and an Ag/AgCl electrode (XR300, saturated KC1 and AgCl solution; Radiometer Analytical) were used as the counter and reference electrodes, respectively. Cyclic voltammetry measurements were performed with a scan rate of 50 mVs“ ', unless otherwise stated. White light photocurrent measurements were performed under simulated AMI.5 solar illumination (lOOmWcm2) with a Newport Sol3A Class AAA solar simulator (type 94023 ASR3). Reduced illumination is achieved by interposing optical density filters in the light path in front of the PV cell. Electrical contact to the sample was made using a silver wire and graphite paste. In the two electrode configuration, the PV cell was connected directly to the electrochemical cell. The potentiostat was used as the ammeter in the current versus time measurements. The electrochemical cell is sealed and made air tight. The cell was continuously purged with a constant N2 flow rate and vented directly into the gas-sampling loop of a gas chromatograph (GC, Interscience) in order to enable periodic quantification (every 18 minutes) of the gas-phase products. A schematic of such experimental setup (for gas chromatography testing) is depicted in Fig. 1. A schematic energy diagram in the PV and electrochemical cell is shown in Fig. 2a.

Gas chromatography testing: The sensitivity of the GC is tested, with the cell in the electrolyzer mode, by consecutive galvanostatic measurements of 0.1, 0.3 and 1 mA for 72 minutes each in order to collect 4 data points at each current. In Fig. 3a we show the voltage response of the cell to the galvanostatic input. At 0.1 mA, the increase in potential over time is attributed to the charging of the electrodes. At 0.3 and 1 mA the potential decreases over time. This is attributed to an increased water oxidation efficiency as the Ni electrode is overcharged, In Fig. 3b we show the O2 and H2 measurements by the GC at each current. At 0.1 mA no H2 is detected, while the O2 concentration increases and saturates to a level of about 12%. At 0.3 and 1 mA H2 is detected and increases to -100% within the 72 minutes of the measurement. More data points would be needed to confirm the reach of saturation. The O2 detection follows a comparable trend as at 0.1 mA, saturating at -12%. This indicates that an adjustment in the calibration is required. Therefore for a quantitative analysis later in this chapter we will rely on the H2 measurement.

Effect of light intensity on the IV curve of a PV cell and of a (P)EC cell: to understand the behavior of such a PV powered electrolysis device, we first discuss the variation of the open circuit potential in relation to the water splitting process that is run by an electrolyzer. The JV curve of a solar cell is the superposition of the IV curve of the solar cell diode in the dark with the light-generated current. The JV relationship for the solar cells is: (1) where J0 is the dark saturation current, the diode leakage current density in the absence of light (a measure of the recombination in a device), q the electron charge, VA is the potential between the terminals, k the Boltzmann constant, T the temperature and Jsc the short circuit current. At Voc -»J = 0, therefore theJV relationship becomes

(2)

And the Voc can be isolated as in

(3)

The approximation holds when ]sc is much larger than J0, which is normally the case. If the light intensity decreases by i.e. a half, we expect the Jsc to also decrease by half. From equation 3 we observe that Voc decreases too, with the logarithmic dependence on the Jsc. The experimental data for the triple junction solar cell are shown in Fig. 4.

The solar cell efficiency as a function of light intensity can be written as

(4)

With the assumption that the fill factor (FF) does not change significantly under varying light intensity at constant temperature, the solar cell efficiency decreases as the light intensity decreases. In real cases like in Fig. 4 also the fill factor decreases, which leads to a further decrease in efficiency. The further away the fill factor is from unity, the more pronounced the lowering of the OCV upon reduced illumination will be.

Now we need to analyze what happens under fluctuating illumination when an electrolyzer is coupled to such PV cell. This is not trivial, since for the electrolyzer not the power but the potential delivered by the PV is critical. The efficiency for solar energy conversion to chemical products such as hydrogen, is called solar-to-hydrogen efficiency and is given by

(5)

Here Vredox is usually taken to be 1.23 V (at room temperature), based on a Gibbs free energy change for water splitting of 237 kJ/mol. Alternatively, the thermo neutral potential of 1.48 V is also used in an isolated system. ηΡΕ is the Faradaic efficiency, the efficiency for the photo-generated electrons and holes that are actually used for the water splitting reaction. In the case of a modular PV-electrolysis approach, the power input Pin is provided by the solar cell at the operating point OP (the intersection between the PV JV curve and the electrolyzer JV curve), defined as

(6)

Therefore η5ΤΗ can be written as

(7)

Equation 7 shows that the efficiency for solar water splitting is related to the efficiency of the PV module. However, to run the water splitting reactions a minimum potential is required. Since at low light intensity as well as under diffuse irradiance, the generated voltage by the PV drops, the voltage matching between the PV and the electrolyzer is of a major concern as the lack of it will reduce the operative hours of the device.

In Fig. 5 the coupling of a solar cell to an electrolysis cell is visualized in (a, c). The solar cell should provide enough voltage to run the electrocatalytic reactions. Considering the overpotential to be added to the Vredox, the solar cell should have a maximum power point around 1.8-2.0 V. In the latter small- and wide-bandgap semiconductors are combined in a direct photoelectrochemical approach to provide enough photovoltage and photocurrent to run the catalytic reactions. In the PV-electrolysis approach, the operating point of the device is defined as the intersection between the J-V curve of the solar cell and that of the reactions over the electrodes.

When the light intensity decreases both the Jsc and Voc of a PV module decrease, and so does the operating point when a load is directly connected to the PV. Translated into practical terms, this means that in the mornings or during cloudy days the water splitting reactions may proceed at very low efficiency, or even not occur. Fig. 5a a shows a schematic image for the operating point for solar water splitting under AMI.5 illumination for a PV+electrocatalyst system. The highest efficiency is achieved when the reaction JV curve intersects the PV JV curve at its maximum power point. Fig. 5b illustrates the solar water splitting under 10% of the AM 1.5 illumination for a PV+electrocatalyst system. The power output under reduced illumination reduces, and the operating point follows until eventually water splitting is no longer possible.

Behaviour of a Si-based PV-electrolysis cell: Now we demonstrate the concept of Fig. 5 using a triple junction a:Si-nc:Si-nc:Si with an OCV under 1 sun of 1.8 V. Such a voltage is enough to run the water splitting reactions in an electrolyzer consisting of a Pt cathode and a thin Ni(OH)2 anode. The IV curve of the PV and that of the electrolyzer intersect at ~1.7 V, as shown in Fig. 6 Applying an ODO.3 filter (50% light transmission in the visible) the PV IV curve intersects the electrolyzer JV curve at -1.6 V. At these conditions the water splitting reactions are still possible, although they proceed at a lower rate than under full illumination. With a OD1 filter (10% light transmission) the OCV of the PV decreases to <1.4 V, which is less than the required voltage to perform the water splitting reactions and hence no hydrogen is produced.

The overlay of the voltammograms obtained from independent experiments carried out on the solar cell and the electrolyzer is shown in Fig. 6. They are a useful indication of how conceptually the coupling of PV and electrolysis would work. Predictions for the operating point based on the intersection between the JV curves is of over 1 mA under 1 sun, and 0.4 mA with a ODO.3 filter. The actual current output from the PV-electrolyzer device at varying light intensity is shown in Fig. 7. The measurement of the device performance in Fig. 7 reveals that the actual values are one order of magnitude lower. After an initial spike, the current relaxes to a roughly steady value for the remaining 4000 s of the measurement. The increasing temperature of the solar cell most likely plays an important role in decreasing the potential output of the solar cell as shown in Fig. 7a. In addition to that, poor bubble removal and mass transport limitations caused by the geometry of the specific cell could also be involved. Overall, the steady value under 1 sun, 60 μΑ, lowers more than linearly under a ODO.3 filter to 20 μΑ. Moreover, negligible current is measured with a OD1 filter. The negligible current at low light intensity is consistent with the schematic representation in Fig. 5, where we indeed proposed that under reduced irradiation there will be no voltage matching between the PV module and the electrolyzer. One possible solution to allow the electrochemical reactions to occur also at low light intensity would be to oversize the PV module or to employ an additional power conversion device. Herewith we show the effect of using a quadruple junction a:Si-a:Si-nc:Si-nc:Si solar cell which has an OCV under 1 sun of 2.3 Y. When this PY cell is coupled to the electrolyzer the overlay of the voltammograms indicates an operating point at 2.0 V, close to the maximum power point of the solar cell (see Fig. 8a). This quadruple junction solar cell has a better fill factor than the triple junction, therefore the OCV under reduced illumination decreases less significantly and goes from 2.3 V under 1 sun to 2.1 V under 10% illumination (OD1 filter). In this case, the potential provided by the PV cell is always enough to drive the electrolyzer and the PV-electrolyzer device performance is limited by the current generated by the PV. The actual current output from this PV-electrolyzer device at varying light intensity is shown in Fig. 8b. At all light intensity, the current matches the value expected from the intersection of the JV curves and scales linearly with the light intensity. Furthermore, the current density shows a slight decrease which is more pronounced at higher light intensity. This observation allows us to attribute it to the temperature effect on the PV potential.

Building a battery/electro lysis cell (to be used under varying solar conditions): An alternative approach consists of implementing additional battery functionalities to the electrolyzer. The idea of our combined battery/electro lysis cell is that the potential for loading the battery is lower than that required to split water. Therefore it will enable the electrical storage of solar energy under low light condition, i.e. with ODO.3 and OD1. The practical challenge is to find materials that can work as efficient oxygen and hydrogen evolution catalysts, while showing storage behavior at potentials more advantageous than the respective water splitting reactions. Moreover they have to be stable in aqueous environment. Here, we propose the use of a hydride-forming Pd-capped Mgo.sTio.2 electrode, and a Ni(OH)2 electrode. Commercial NiMH batteries employ a Ni(OH)2 positive electrode and a MischMetal-based MH as negative electrode. In this proof-of-concept study we replace the MischMetal-based electrode, expensive and of difficult fabrication, with a cheap and environmentally friendly Mg-based electrode. The Mg-based electrode is an excellent hydrogen absorber since it exhibits a very high gravimetric storage capacity of about 2200 mAh/g (7.6 wt% hydrogen). However, it suffers from low sorption kinetics, and instability towards oxidation. The oxidation of the film can be effectively prevented by depositing a thin film of Pd, which also acts as a catalyst for the hydrogen sorption. It has been shown that alloying Mg with rare earth elements increases the hydrogen sorption kinetics. Hydrogen uptake and release kinetics of Mg thin films can be enhanced significantly by alloying Mg with 20% Ti, while keeping an excellent gravimetric capacity (> 6 wt% hydrogen) up to six times that of the MischMetal-based materials. A series of Pd-capped MgxTii_x films was investigated electrochemically and it was found that Mgo.85Tio.15 and Mgo.72Tio.28 performed best in terms of short activation period, resistance to corrosion, cyclic stability and large discharge capacity up to 80% after 150 cycles (see e.g. Niessen, R. A. H.; Notten, P. H. L. Electrochem. Solid-State Lett. 2005, 8, 10, A534-A538, orXin, G.; Yang, J.; Fu, H.; Zheng, J.; Li, X. Int. J. Hydrogen Energy 2013, 38, 10625-10529, which are herein incorporated by reference).

The Ni(OH)2 ^ NiOOH redox couple is known for its electro-chromic properties and its use as an electrode in the Ni-MH batteries. Its discharging is a downhill reaction which releases energy. Moreover, it seems that NiOOH acts as an efficient water oxidation catalyst. Of importance for the purpose of this work, it also operates in alkaline media just as the metal hydride cathode. Those properties set the basis for the use of this material in our system (see e.g. David S. Hall, David J. Lockwood, Christina Bock and Barry R. MacDougall Nickel hydroxides and related materials: a review of their structures, synthesis and properties Proc. R. Soc. A 471:201407 or Subbaraman, R.; Tripkovic, D.; Chang, K.-C; Strmcnik, D.; Paulikas, A. P.; Hirunsit, P.; Chan, M,; Greeley, J.; Stamenkovic, V.;and Markovic N. M. Trends in activity for the water electrolyser reactions on 3d M(Ni,Co,Fe,Mn) hydr(oxy)oxide catalysts, Nature Mater. 2012, 11, 550-557. The Ni double hydroxide catalyst seems to have a layered structure. The sheets are held together by non-covalent interactions allowing for the movement of water and ions between the sheets and throughout the structure when it undergoes redox chemistry. We hypothesize that the layered structure of the hydroxide catalysts is especially active because the catalytic sites are sandwiched in-between redox-active transition-metal-cation sheets that modify the energetics of the reaction intermediates differently than is possible on the surface of a dense oxide. In between such layers, the water oxidation intermediates OH and OOH from one layer can feel the influence of the other intermediates from the opposite layer. In other words there is a second catalytic surface from which viewpoint OH and OOH do not look identical. As a further point of surface modification, the presence of dopants in the catalyst structure, as well as cations and anions in the electrolyte may also influence the bond strength of the intermediates. This approach seems a promising way to design catalyst that enable the breaking of the scaling relations.The goal of increasing the solar energy conversion and storage under low light condition can be achieved if the potential for loading the battery is lower than that required to split water. In the following paragraphs we will analyze the thermodynamics and kinetics for the hydrogen evolution reaction (the electrolyzer), compared to the reaction for metal hydride formation (the battery). Afterwards, the oxidation reaction will be discussed. REDOX reaction potential: in (photo-)electrochemical water splitting the (photogenerated) electrons are used to perform the hydrogen evolution reaction (HER). This is a well-studied electrochemical reaction and it is understood it proceeds by one of two mechanisms. Each mechanism consists of two primary steps, and begins with the Volmer reaction (5.1) according to: (5.1) M + H20 + e ^ MHad + OH“ (5.2) MHad + H20 + e ^ H2 (g) + OH" or the Volmer reaction (5.1) followed by the Tafel step (5.3): (5.3) 2 MHad ^ 2 M + H2 (g) where M is a hydrogen active metal site. With this approach it has been shown that the formation, or reductive desorption of intermediate hydride species presents a large energy barrier to the overall hydrogen evolution reaction.

On the other hand, when hydrogen is electrochemically stored in a metal hydride, the reaction path becomes: (5.1) M + H20 + e“ ^ MHad + OH“ (5.4) MHad - MHabs

From the reaction mechanisms it follows that the hydrogen evolution and the metal hydride formation are initiated in the same way with the adsorption of one H atom from the surrounding water onto one available metal site via reaction (5.1). This step is followed by the absorption of the H via reaction (5.4) for the formation of a metal hydride. For the hydrogen evolution, a second water molecule and a second electron are involved via reaction (5.2) or (5.3).

Experimentally, the polarization curve of the Pd-Mgo.sTio.2 electrode is shown in the reductive side of Fig. 9. In this curve we assign the peak -0.15 V vs. RHE to the metal hydride formation, and the peak at more negative potential to H2 evolution on the Pd surface. We observe that the formation of the metal hydride (MH) phase occurs at less cathodic potential than H2 evolution. In particular, its onset potential is at +0.2 V vs. RHE. This value is consistent with earlier electrochemical observations on MgTi systems. This observation means that the electrochemical hydrogen absorption MHa(j ^ MHabs (reaction(5.4)) is thermodynamically more favorable than both reactions (5.2) and (5.3). Thus, for the cathode, our goal of loading the battery component at potentials lower than that required to split water is feasible.

However, we also have to compare the battery behavior in the anode. Therefore, we will analyze the electrochemistry for the oxygen evolution reaction (OER) (the electrolyzer), compared to the reaction for the formation of a NiOOH species from a Ni(OH)2 electrode (the battery). If the OER is to be performed by a catalyst based on earth-abundant elements, it will require a minimum of ~0.3-0.4 V overvoltage η to obtain currents in excess of 8mAcm"2. The reason for such a high overpotential can be found in the binding energies of the water oxidation intermediates during the four proton-coupled-electron-transfer (PCET) reaction required for the H2O/O2 redox couple (OH7C>2 in alkaline media): (5.5) 40H“ 02 + 2H20 + 4e“

On the other hand, the charging/discharging of the Ni electrode proceeds according to: (5.6) Ni(OH)2 + OH' NiOOH + H20 + e"

While the Ni charging/discharging reaction (5.6) involves the transfer of a single proton and electron, the OER reaction (5.5) is a four PCET reaction.

Experimentally, the polarization curve of the Ni electrode is shown in the oxidative side of Fig. 9. Consistently with other literature reports, we assign the peak at +1.4 V vs. RHE to the oxidation of the Ni(OH)2 electrode, and the current increase at more positive potential to O2 evolution. Therefore the formation of the NiOOH phase occurs at less anodic potential than O2 evolution. Considering the onset potential, the Ni(OH)2 NiOOH reaction has an onset at ~1.35 V vs. RHE, whereas the OER has an onset at 1.5 V vs. RHE. This halfcell measurement proves that also for the anode, the battery component loads at potentials lower than that required to split water.

When the Ni and the MgTi electrodes are used in an electrochemical cell one can measure the overall two electrode JV curve for the battery/electrolyzer concept. The resulting JV curve is shown as the orange curve in Fig. 10, where it is compared to the overall electrolyzer’s JV curves shown as the green curve. Clearly, the onset potential for the battery mode is 200-300 mV smaller than for the electrolysis mode. This reflects the earlier onset potential for the loading of the MgTi-Pd-PTFE electrode compared to the H2 evolution at the Pt electrode. The lower slope of the JV curve for the MgTi-Pd-PTFE reflects the higher resistivity of the electrode compared to Pt. This implies that the electrodes can be charged when the potential provided by the PV cell is too low to induce electrolysis. PV-battery/electrolysis cell: effect of light intensity and comparison with a PV-electrolysis cell: To verify the actual performance of the battery/electrolyzer cell we powered it by the triple junction PV cell and measure the output as a function of increasing light intensity, Fig. 11. At all light intensities, the current generated in the cell is lower than expected from the individual measurements shown in Fig. 10. This is consistent with what was observed in the PV-electrolyzer system described above (Fig. 6 and Fig. 7) With an OD1 filter (10% light transmitted) the PV cell generates an OCV of 1.4 V. This potential is not sufficient to perform the water splitting reactions, it is however just enough to intersect the IV curve for the battery/electrolyzer in the two electrode measurement of Fig. 10. Therefore, in the measurement of Fig. 11 the measured current is limited to few μΑ. While limited, the measured current represents an improvement compared to the PV-electrolyzer system (Fig. 11). With an ODO.3 filter (50% light transmission), after the initial spike, the current decays within 200 seconds to a plateau at around 500 μΑ. Thereafter, the current decays to a second plateau value of about 100 μΑ. This indicates the exhaustion/completion of one redox process and a stabilization on a second process. We assign the two electrochemical processes to firstly the loading of the battery component until its full capacity is reached, and secondly the electrochemical hydrogen production via water splitting. The last measurement is performed under full sun, AM 1.5 illumination. The chronoamperometric measurement for the PV battery/electrolyzer shows a decay similar to that observed in the PV electrolyzer. However the current stabilizes at 150 μΑ, a value twice as high the 75 μΑ measured for the PV electrolyzer under the same conditions. This observation indicates that the hydrogenated electrode is more efficient in electrolysis at those current density. The performance of the battery/electrolyzer device coupled to the quadruple junction a:Si-a:Si-nc:Si-nc:Si solar cell is shown in Fig. 12. The current measured through the device matches the value expected from the intersection of the JV curves of the PV end the battery/electrolyzer as in Fig. 10. The current is initially always higher than for the PV-electrolyzer configuration. Interestingly, under 1 sun the current drops by about 1 mA after 10 seconds of illumination. This behavior has been reproducible and a galvanostatic discharge measurement for the MgTi electrode revealed that the electrode was not fully loaded at the end of the measurement. Therefore we tentatively attribute the drop in current to a shift in the functionality of the not fully loaded MgTi electrode from battery mode to electrolyzer mode. PV-battery/electrolysis cell: High initial current density with OD0.3 is not due to hydrogen evolution: To verify our interpretation of the current obtained when using the OD0.3 filter (Fig. 11) we performed a gas chromatography (GC) measurement of the evolving gases. In Fig. 13 we plot the chronoamperometric measurement together with the hydrogen concentration as detected by the GC. No hydrogen is detected in the first 1500 seconds. Hydrogen is detected only thereafter and its concentration increases with time. Clearly no hydrogen is produced at the high current plateau. Hence, we conclude that at OD0.3 conditions first the battery component is loaded until its full capacity is reached. It is then followed by the electrochemical hydrogen production via water splitting. The higher current measured in the former process is explained by the lower potential required for the battery loading and the consequent higher current at the operating point as shown in Fig. 10. PV-battery/electrolysis cell: energy storage upon rapidly changing potential: In Fig. 11 and Fig. 13 we showed that at moderate light intensity (ODO.3) the battery/electrolyzer device favors the loading of the battery component over the electrolysis of water. Moreover, once fully charged, the Mgo.gTio^Fh-Pd-PTFE battery electrode acts as an efficient hydrogen evolution catalyst. Since in reality the solar irradiance may change quickly, in this section we investigate the storage behavior under a rapidly changing applied potential. In Fig. 14 we show the linear sweep voltammetry of the electrode at different scan rates. At a scan rate of 5 mV/s the battery electrode shows an onset potential at +0.2 V vs. RHE and a peak at -0.15 V vs. RHE, which is attributed to the hydrogenation of the electrode. At V < -0.22 V vs. RHE the current density increases in value which is attributed to the hydrogen evolution. The two processes appear separate and can be distinguished with ease (blue). At a scan rate of 10 mV/s the battery electrode shows again an onset potential at 0.2 V vs. RHE, however there is no local maximum of current density. To understand whether the current goes to battery charging or to hydrogen evolution, we have calculated the integral of the JV curves in Fig. 14 for the hypothetical hydrogen uptake and release for samples of MgTi of two thicknesses, 200 nm and 400 nm. The integral is calculated assuming all the current under the charging/discharging curves goes to the sorption of hydrogen into/out of the MgTi. When comparing the hydrogen uptake (current IN) to the hydrogen release (current OUT), the hydrogen uptake is always higher. Considering that the hydrogen release is on average at least 90% of the hydrogen uptake as shown in the galvanostatic measurement of Fig. 16, the results indicate that upon charging, part of the current does not go to MH formation, but to H2 evolution. PV-battery/electrolysis cell: battery discharge: Thus far we have shown the energy storage properties of the PV-battery/electrolyzer device as a function of the light intensity, and we have compared them to a PV-electrolyzer configuration. In this section we show that when electricity generation is needed, the device can be switched to battery mode to generate electricity and discharge the battery. In Fig. 17 we show the chronoamperometric measurement performed in the dark by discharging the battery electrodes through the solar cell used as a small resistor. A small however not negligible current is flowing through the circuit. The discharge current shown in Fig. 17 is modest, and rapidly decreases to negligible values within few seconds. Fig. 18 shows chronopotentiometric measurements for the charge and discharge of the Ni-based and the metal hydride electrodes. Fig. 19 compares the time course of the potential of the Ni electrode during discharge, relative to that of the MgTi electrode. Depending of the chosen discharge current for the Ni electrode, there will be no potential difference between the Ni and the MgTi electrodes after 4 s at -30 μΑ, or after 26 s at -3 μΑ. The discharge current of the MgTi electrode is chosen as +50 μΑ. This current density gives a lower limit for the time of potential equilibration between the two electrodes. Moreover we observe that the time necessary to fully discharge the Ni electrode is in the order of tens of seconds. This time is three orders of magnitude shorter than the time required to fully discharge the metal hydride electrode, despite been performed at ten times smaller current density. The timescale for the discharge of the Ni electrode is comparable to the time at which the discharge of the full battery occurs, as shown in Fig. 17. Therefore, we conclude that the battery capacity is limited by the capacity of the Ni electrode. The low capacity of the Ni electrode, despite its thickness, suggests that the Ni film deposited directly onto FTO is not porous enough to be fully charged. As a result only the surface is used in the Ni(OH)2 / NiOOH redox cycling. PV-battery/electrolysis cell: schematic overview under operation: The combination of the results shown in Figures 10-13, gives the full of the flexible PV-battery/electrolyzer device. Fig. 20 shows a schematic diagram of the proposed device in various stages of operation. The electro-chromic properties of both the anode and the cathode are used to visualize the charging/discharging processes. Fig. 20 also shows an embodiment wherein the battery-electrolyzer and the photovoltaic cell have a ratio of a geometrical area of the battery-electrolyzer and a geometrical area of the photovoltaic cell in the range of 5:1-1:5, here in this schematically drawing about 1:1.

When the electrochemical cell is fully discharged (i) the Ni-based anode is in its low-energy state Ni(OH)2, and the hydrogen-absorbing metal cathode is in the metallic state. If the electricity from the PV component is needed, a switch disconnects the electrochemical part and directs the electricity to the users. If electricity is not needed the electrochemical cell is connected to the PV, and the device begins charging its electrodes (ii) and storing energy according to Ni(OH)2 NiOOH at the anode, and Me MH at the cathode. When the solar irradiance increases and the solar cell provides enough power, the device gradually switches from battery mode to electrolysis mode for the hydrogen production (iii) just like in a conventional electrolyzer, with OER and HER running over the efficient NiOOH and Pd catalysts, respectively (iv). When electricity is needed and no power comes from the PV, the energy stored into the battery is released by closing the circuit with the load. The energy release occurs by discharging the Ni-based electrode according to Ni(OH)2 NiOOH, and the MeH electrode via the reaction Me ¢- MH (v). When the battery electrodes are fully discharged the cell is back at the initial point (i).

The concept device proposed in this work, after optimization, offers the possibility to size the PV module and the battery electrodes precisely to meet the end-user’s need. The optimal size will provide enough energy during peak hours (of solar irradiance) to charge the battery component to such an extent that covers the daily storage also during winter times. The excess output from the PV in the summer can be converted into chemical storage as H2. This can be used as a fuel to power a vehicle. We consider as an example a residential house with a surface availability of 100 m2 for installation the PV-battery/electrolyzer device. The electricity daily loading curve is as show in Fig. 21. The solar irradiance is the one available in Delft, The Netherlands, estimated in 1050 kWh/m2'yr. The solar panels efficiency is taken as 15% and generate 11805 kWh/yr. The household consumption is considered to be the average value for The Netherlands, 3000 kWh/yr. The fraction of the electricity provided directly by the PV is taken to be 1/3 of the total daily consumption, thus 1000 kWh/yr. The remaining 2000 kWh/yr is to be provided by the battery or Ffr. This equals to 5.5 kWh/day. To get to the thickness of the MgTi electrode with 5%wt hydrogen storage capacity needed to store (and release) such an amount of current we can consider as a maximum thickness the one used in commercial NiMH batteries. From the literature one can find that 290 pm is the normal thickness for the metal hydride electrode in Ni-MH batteries. Calculations on a per gram basis show that a MgTi film has a 26.8 mAh of capacity at 1 A/g. Using the same surface area as for the solar PV, 100 m2, assuming a 90% efficiency, 100 m2 of290 pm thick MgTi electrode can store up to 24.1 kWh, which is greater than the 5.5 kWh/day required. The next question one may wonder is whether the power output from the PV in the winter is enough to charge the battery to such an extent to cover the daily electricity household consumption. If not, the hydrogen stored in the summertime can be used via a fuel cell to back-up the electricity shortage. Taking Delft (The Netherlands) as an example, the PVGIS website tells us that on average the daily power output in December is 10.40 kWh/day, which is 31% that of the yearly average of 32.3 kWh/day. Given the average electricity consumption of 8.2 kWh/day, a battery with 90% efficiency will be able to meet the consumption. Therefore, the hydrogen produced can be used e.g. to power an electrical/fuel cell vehicle.

Of the 2000 kWh/yr provided by the battery we consider a conversion efficiency for the battery component of 90%, thus 2222 kWh/yr from the PV are needed to cover the 2000 kWh/yr consumption from the daily storage. All the above considered, of the 11800 kWh/yr produced yearly by the 100 m2 PY panels, 3222 kWh/yr are used to cover the 3000 kWh/yr consumption. This leaves 8578 kWh/yr available for e.g. charging a fuel cell or an electric vehicle. We consider the case of a fuel cell vehicle. With an electrolyzer conversion efficiency of 60%, 5147 kWh/yr are stored into 1¾. We consider moreover 200 kWh of H2 is stored in one full tank which guarantees 600 km driving range. Therefore, 5147 kWh/yr are enough to provide 43 full tank filling and drive 25700 km.

In conclusion, 100 m2 of 15% efficient solar panels coupled with an optimized battery/electrolyzer device with 90% efficiency as a battery and 60% efficiency as electrolyzer, are enough to power a typical residential house in The Netherlands throughout the year. Additionally, enough H2 is produced to provide 43 full tank filling and drive 25700 km with a H2 vehicle.

Since the initial cost of batteries (let alone that of the electrolyzers) is an important factor limiting their market penetration, the use of easy-to-fabricate and abundant elements such as MgTi and Ni based electrodes holds promise for the scaling up of the technology. However an alternative material for the Pd catalytic layer should be found. The materials proposed in this proof-of-concept study should not be considered as the final choice. In fact, the materials used in commercial PV and Ni-Fe, Ni-MH and Ni-hydrogen batteries could be employed for pilot plant studies. Another advantage of the proposed system is the limited electronics and balance-of-plant needed to operate.

Herein, we present a proof-of-principle of a device that is able to convert solar energy either in electricity or to hydrogen, depending on the intensity of the irradiance. The device consists of a PV component integrated with a combined battery-electrolyzer cell. The battery/electrolyzer consists of a Pd-capped hydride-forming Mgo.8Tio.2 cathode and a thick Ni(OH)2 anode, which is powered by either of a triple junction a-Si-nc-Si-nc-Si with an OCV of 1.8 V under 1 sun, or a quadruple junction a-Si-a-Si-nc-Si-nc-Si with an OCV of 2.3 V under 1 sun. The device performance under varying solar irradiance is compared to the one delivered by the same PY component coupled to an electrolyzer composed of a Pt cathode and a thin Ni(OH)2 anode. With the quadruple junction PV, the OCV allows for energy storage at all light intensities between 0.1 and 1 sun. With the triple junction PV we show that at low light intensity (10 mWcm"2), the most efficient way to use the solar irradiation is by delivering electricity. In the presence of moderate light intensity (50 mWcrn 2), the device is in battery mode and the renewable energy is stored by charging the cathode battery until full loading. Thereafter, or under an irradiance of one sun (100 mWcm"2), the device automatically switches to electrolyzer mode, and the renewable energy is stored as a mixture of oxygen/hydrogen gas. Finally, when electric power is needed, the device may be switched back to battery mode to discharge the battery. This last step closes the cycle, and regenerates the device.

Since the local storage of solar electricity into batteries or hydrogen is currently not cost competitive with the electricity prices from the grid, it is of paramount importance to extend the storage capacity and the operational activity of the solar energy storage devices. An unexplored way to do so is by maximizing their energy conversion efficiency under the variable illumination conditions typically experienced daily.

Fig. 22a schematically depicts a system 10 comprising a photovoltaic cell 100 and a hybrid battery-electrolyzer 200. The hybrid battery-electrolyzer 200 is especially configured to select one of (i) a battery charging mode, (ii) an electrolyzing mode, and (iii) a battery discharging mode in dependence of solar light received by the photovoltaic cell 100. Further, this drawing schematically indicates the hybrid battery-electrolyzer 200 comprising a first cell 210 comprising a first electrode 211, and a second cell 220, separated from the first cell by a membrane 230, the second cell 220 comprising a second electrode 222. As indicated above, the first electrode 211 may comprise a magnesium-titanium comprising electrode and the second electrode 222 may comprise a nickel comprising electrode. Further, by way of example, the drawing also schematically depicts the system 10 further comprising one or more of an H2 storage 21 and an 02 storage 22, functionally coupled with the hybrid battery-electrolyzer 200. The system 10 may be used to provide electrical energy and/or H2 and/or 02 to an application. Hence, such application, which is not necessarily part of the system is drawn dashed and herein indicated as an energy consuming system 1010, which may (temporarily) be functionally coupled with the system 10. The system 10 can thus be used for providing one or more of electrical energy (“energy delivery mode”) and hydrogen gas (to an energy consuming system 10) and/or for storing electrical energy. The system may also be applied for generation of 02. Fig. 22b schematically depicts a larger system 10, comprising a plurality of PV-battery-electrolyzer units. By way of example, a first unit comprises a plurality of PVs 100 (left). The system may be used to charge electricity to an energy consuming system 1010 and/or H2 (and/or 02). The battery-electrolyzers 200 may be configured parallel or in series. Here, the battery-electrolyzers 200 are configured parallel, which might allow a fast charging. Reference 1010 may refer to an energy consuming system and/or an energy consuming apparatus. Note that optionally the energy consuming apparatus or system may directly be charged by the PV 100. To this end, the system 10 may further include a disconnector (not shown) to temporarily disconnect the battery-electrolyzer 200 from the PY 100. Of course, electrical energy may (during an energy delivery mode) also be provided by the battery-electrolyzer 200 only or by both the PV 100 and the battery-electrolyzer 200.

The term “substantially” herein, such as in “substantially consists”, will be understood by the person skilled in the art. The term “substantially” may also include embodiments with “entirely”, “completely”, “all”, etc. Hence, in embodiments the adjective substantially may also be removed. Where applicable, the term “substantially” may also relate to 90% or higher, such as 95% or higher, especially 99% or higher, even more especially 99.5% or higher, including 100%. The term “comprise” includes also embodiments wherein the term “comprises” means “consists of’. The term “and/or” especially relates to one or more of the items mentioned before and after “and/or”. For instance, a phrase “item 1 and/or item 2” and similar phrases may relate to one or more of item 1 and item 2. The term "comprising" may in an embodiment refer to "consisting of' but may in another embodiment also refer to "containing at least the defined species and optionally one or more other species".

Furthermore, the terms first, second, third and the like in the description and in the claims, are used for distinguishing between similar elements and not necessarily for describing a sequential or chronological order. It is to be understood that the terms so used are interchangeable under appropriate circumstances and that the embodiments of the invention described herein are capable of operation in other sequences than described or illustrated herein.

The devices herein are amongst others described during operation. As will be clear to the person skilled in the art, the invention is not limited to methods of operation or devices in operation.

It should be noted that the above-mentioned embodiments illustrate rather than limit the invention, and that those skilled in the art will be able to design many alternative embodiments without departing from the scope of the appended claims. In the claims, any reference signs placed between parentheses shall not be construed as limiting the claim. Use of the verb "to comprise" and its conjugations does not exclude the presence of elements or steps other than those stated in a claim. The article "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. The invention may be implemented by means of hardware comprising several distinct elements, and by means of a suitably programmed computer. In the device claim enumerating several means, several of these means may be embodied by one and the same item of hardware. The mere fact that certain measures are recited in mutually different dependent claims does not indicate that a combination of these measures cannot be used to advantage.

The invention further applies to a device comprising one or more of the characterizing features described in the description and/or shown in the attached drawings. The invention further pertains to a method or process comprising one or more of the characterizing features described in the description and/or shown in the attached drawings.

The various aspects discussed in this patent can be combined in order to provide additional advantages. Further, the person skilled in the art will understand that embodiments can be combined, and that also more than two embodiments can be combined. Furthermore, some of the features can form the basis for one or more divisional applications.

Claims (21)

A system (10) comprising a photovoltaic cell (100) and a hybrid battery electrolysis cell (200) configured to select one of (i) a battery charging mode and (ii) an electrolysis mode depending on by the photovoltaic cell (100) receives sunlight. The system (10) of claim 1, configured to select one or more of (i) a battery charging mode and (ii) an electrolysis mode depending on one or more of (a) the sunlight flux, (b) a electric charge level of the hybrid battery electrolysis cell (200), and (c) an electricity demand for the hybrid battery electrolysis cell (200). The system (10) according to one or more of the preceding claims, wherein the battery electrolysis cell (200) and the photovoltaic cell (100) have a ratio between a geometric surface of the battery electrolysis cell (200) and a geometric surface of the photovoltaic cell (100) in the range of 5: 1-1: 5, and wherein the system is configured to (i) select the battery charging mode and / or (ii) the electrolysis mode at sunlight intensity in the range of at least 10 mW / cm2. The system (10) of any one of the preceding claims, wherein a geometric surface of the battery electrolysis cell (200) is the same as a geometric surface of the photovoltaic cell (100). The system (10) according to one or more of the preceding claims, configured to select the battery charging mode at sunlight intensities in the range of 10-75 mW / cm 2 and to select the electrolysis mode at sunlight intensities above 75 mW / cm 2. The system (10) according to any of the preceding claims, wherein the photovoltaic cell (100) and the hybrid battery electrolysis cell (200) are electrically connected without intermediate electronics. The system (10) of any one of the preceding claims, wherein the hybrid battery electrolysis cell (200) comprises a first cell (210) comprising a first electrode (211), the first electrode (211) comprising a electrode (211a) comprising magnesium-titanium, and comprising a second cell (220) separated from the first cell by a membrane (230), the second cell (220) comprising a second electrode (222), the second electrode (222) ) comprising a nickel-comprising electrode (222a), in particular wherein the magnesium-titanium-containing electrode (221a) further comprises a Pd catalyst. The system (10) of claim 7, wherein the nickel-comprising electrode (222a) comprises one or more electrode comprising a nickel-cobalt and an electrode comprising a nickel-iron. The system (10) according to one or more of the preceding claims 7-8, wherein the electrodes (211,222) are configured to have an absolute (maximum) potential difference in the range of 1 between the electrodes during the electrolysis mode, 2-2.2 V, such as in the range of 1.5-2.0 V, and wherein the electrodes are configured to have a potential difference between the electrodes in the range of 0.6-1.2 V, such as in the range of 0.8-1.2 V, in a battery discharge mode. The system (10) of any one of the preceding claims, wherein the photovoltaic cell (100) is configured to provide an open circuit voltage (OCV) in the range of 1.5-2.5 V at sunlight intensities of more than 75 mW / cm 2. The system (10) according to one or more of the preceding claims, wherein the photovoltaic cell (100) comprises a multi-junction photovoltaic cell, in particular wherein the photovoltaic cell (100) has one or more of a triple junction a -Si-nc-Si-nc-Si photovoltaic cell and a quadruple junction a-Si-a-Si-nc-Si-nc-Si photovoltaic cell. The system (10) according to one or more of the preceding claims, comprising a plurality of photovoltaic cells (100). The system (10) according to one or more of the preceding claims, further comprising one or more of an H 2 storage (21) and one (V storage (22), operatively coupled to the hybrid battery electrolysis cell (200). An assembly (1000) comprising an energy-consuming system (1010) and the system according to one or more of the preceding claims 1-13 functionally connected to the energy-consuming system (1010), the energy-consuming system (1010) ) has been selected from the group consisting of a house, an office, a factory, a hospital, a school and an (indoor) sports facility. An assembly (1000) comprising an energy-consuming system (1010) and a system according to one or more of the preceding claims 1-13, functionally connected to the energy-consuming system (1010), wherein the energy-consuming system ( 1010) is selected from the group consisting of infrastructure energy-consuming systems, such as selected from (a) a lock, (b) a movable bridge, (c) a system comprising one or more of (i) a traffic light, (ii) a signaling light, (iii) an information screen, and (d) a controllable road closure. A method for (a) providing one or more of electrical energy and hydrogen gas and / or (b) for storing electrical energy, the method comprising using a system (10) comprising a photovoltaic cell (100) and a hybrid battery electrolysis cell (200) configured to, depending on sunlight received by the photovoltaic cell (100), select one of (i) a battery charging mode and (ii) an electrolysis mode to allow the photovoltaic cell (100), and providing one or more of electrical energy and hydrogen gas and / or storing of electrical energy. The method of claim 16, wherein the system (10) is configured to select one or more of (i) a battery charging mode and (ii) an electrolysis mode depending on the one or more of a sunlight flux and an electric charge level of the hybrid battery electrolysis cell (200). The method of any one of the preceding claims 16-17, comprising providing one or more of electrical energy and hydrogen gas to a vehicle operatively coupled to the system (10). The method of any one of the preceding claims 16-17, comprising providing energy to an energy-consuming system (1010). The method of any one of the preceding claims 16-19, comprising charging the hybrid battery electrolysis cell in a charging mode when there is a potential difference between two electrodes of the hydride battery electrolysis cell (200) of at least 1 Y, and comprising electrolyzing water in the electrolysis mode when there is a potential difference between said two electrodes of at least 1.4 V. Use of the system (10) according to one or more of the preceding claims 1-13, for one or more of providing electrical energy and hydrogen gas and / or for storing electrical energy.