Search

The principle of the electrochemical decomposition of water in an electrolysis cell has already been known for more than 230 years. The first generation of hydrogen by electricity was done as early as 1789 by van Troostwijk and Deiman using an electrostatic generator as the direct current source. Shortly after Volta invented the voltaic pile in 1800, Carlisle and Nicholson used such a device to decompose water into hydrogen and oxygen. In the same year, Ritter performed comparable experiments in Jena, Germany. Moreover, Cruickshank used a voltaic pile for the electrochemical decomposition of NaCl to hydrogen and chlorine at the beginning of the 19th century. Nevertheless, it took decades before the processes were used in the first technical applications. Around 1890 Charles Renard constructed a water electrolysis unit to generate hydrogen for French military airships. It is estimated that around 1900 more than 400 industrial alkaline water electrolyzers were in operation worldwide and large-scale deployment of the chlor-alkali process started. Later, different types of commercial alkaline water electrolyzers were developed in the 20th century to generate hydrogen needed to produce ammonia fertilizers based on low-cost hydroelectricity. As the 20th century progressed, the more cost-effective production of hydrogen by steam reforming of methane increasingly replaced water electrolysis, and by the end of the 20th century, the process was only used in niche applications. In the late 1960s the development of proton exchange membrane electrolysis began at General Electric using an acidic fluorinated ionomer as a solid electrolyte. But the technology became established in the following decades only in laboratory, military, and space applications due to the high material costs. At about the same time, General Electric, and the Brookhaven National Laboratory also started to develop high-temperature electrolysis with solid oxide cells. In Germany, Dornier followed in the HOT ELLY project this technology until the mid-1980s. But despite of all technical progress made commercialization of the membrane and solid oxide electrolysis could not be launched in that time. But new interest in water electrolysis was stimulated already in the 1990s as hydrogen was regarded as a green energy carrier for renewable energy sources like wind and solar power and to power fuel cells, e.g., in automotive applications. However, it is only in the last 10 years that we have seen a significant increase in global interest for water electrolysis, with the adoption of ambitious national climate protection programs. Water electrolysis is regarded as the central element for sector coupling and is expected to make an important contribution to reducing greenhouse gas (GHG) emissions close to net-zero by 2050.

A regenerative fuel cell (RFC) is a hydrogen accumulator which is charged via an electrolyzer (electricity conversion into H2) and discharged via the fuel cell (H2 conversion into electricity), where the storage media is pressurized hydrogen. The also generated oxygen is mostly not stored in terrestrial applications. There are discrete RFCs (DRFC) consisting of two separate stacks (electrolyzer and fuel cell) and unitized RFCs (URFC) with one single stack working during charge in electrolysis mode and during discharge in fuel cell mode. URFCs show a high specific energy up to 1500 Wh kg−1. Furthermore, it is possible to optimize the power and energy of the system independently, which is important for seasonal storage of larger amounts of energy. In contrast to conventional electrochemical accumulators the discharge power of RFCs is unaffected by the state-of-charge of the system. But unfortunately, due to the long conversion chain with associated losses, RFCs generally have low levels of efficiency compared to conventional electrochemical accumulators. The high specific energy of URFCs makes the system interesting for niche applications in military and space areas where efficiency and related costs are not primary parameters. For civil terrestrial applications, however, costs and electrical efficiency are mostly primary parameters. DRFCs are preferred since the individual stacks (EL, FC) can be better optimized. However, for the time being, there are existing no commercial applications of RFCs, as they are not really competitive, especially not with electrochemical accumulators. In the future could be electricity storage with RFCs possible in off-grid or island applications in which a high level of autonomy is required but also in grid application with high share of renewables to stabilized power supply. In a broader sense are power-to-gas storage systems identically with DRFCs. Power-to-gas technologies and therefore also electrolyzers currently are developing strongly which will decreases the costs and increases the efficiency of RFC as well. The RFC technology is in the moment proton exchange membranes based, but they are relatively costly caused by use of platin group metal catalysts. Alkaline-RFCs could reduce the cost by using non-platin group metal catalysts. Because Alkaline–RFCs show an unsatisfactory performance and poor cyclability, therefore, the development is still in an initial stage. The efficiency of PEM-RFCs (URFC max. 35%, DRFC max. 40%) is relatively low. Higher efficiencies, however, show SO-RFCs. SO-URFCs already reach today in an early development state ca. 45% efficiency and could be used also in terrestrial applications.

Thermal stability of charged LiCoO 2 cathodes with various surface areas of active material is investigated in order to quantify the effect of LiCoO 2 surface area on thermal stability of cathode. Thermogravimetric analyses and calorimetry have been conducted on charged cathodes with different active material surface areas. Besides reduced thermal stability, high surface area also changes the active material decomposition reaction and induces side reactions with additives. Thermal analyses of LiCoO 2 delithiated chemically without any additives or with a single additive have been conducted to elaborate the effect of particle size on side reactions. Stability of cathode–electrolyte system has been investigated by accelerating rate calorimetry (ARC). Arrhenius activation energy of cathode decomposition has been calculated as function of conversion at different surface area of active material.

The safety and overall operation of lithium metal batteries depends largely on the prevention of lithium dendrite build. Numerous methods have been proposed for dendrite suppression such as use of external pressure, modification of the nucleation sites for dendrite initiation and the use of solid electrolytes. Each of these approaches shows promise, but neither has led so far to a practical solution. A conceptually different design approach involves the use of high-surface area, 3D structures. Electrodes built on the principle of porous or three-dimensional substrate enable low current densities, while still offering high overall current. The low current density in turn enables long Sands time and prolonged lithium deposition without dendrite build. The entire concept relies on the creation of a high-surface area current collector in direct contact with both the electrolyte and the active material. This is the, often forgotten, principle of the three-phase boundary. The principles of creating large three-phase boundary are used to design the pore structure and to balance the distribution of the electrode components. While the effective design of the three-phase boundary is critical for both charge and discharge reactions in the lithium metal battery, it’s principles are more obvious during the discharge reaction. It is demonstrated how the effective three-phase boundary affects the cell performance. A high-surface area porous silicon current collector is used for the metal silicon anode as well as for the air cathode. The study depicts the fundamental behavior and dependence of lithium dendrite build on the current density at the anode. Cycling of a symmetrical cell demonstrates stable voltage and current, with no capacity fade, i.e., no dendrite build, for C rates below a critical value. Figure 1 depicts stable capacity during cycling, while figure 2 shows reduction in impedance as the cycling progresses. Further suppression in dendrite creation is demonstrated through the geometrical effects of the structure as well as by modifying the nature of the current collector. The application of this 3D structure for the air cathode is also evaluated and modeled.

In this paper reactions having strong influence on the lifetime of PEMFCs are described. These lifetime limiting reactions are related to the catalyst (increase of the particle size, catalyst dissolution, oxidation of the carbon catalyst support, catalyst deactivation by contaminants), to the membrane (degradation, loss of membrane humidification, increase of membrane resistance), and corrosion of structural components (e.g. metallic bipolar plates). Examples for such reactions are shown and rate determinining parameter are discussed.

Hybrid systems, based on a lead–acid battery and a proton-exchange membrane fuel cell (PEMFC) give the possibility to combine the advantages of both technologies. The benefits for different applications are discussed and the practical realisation of such systems is shown. Furthermore a numerical model for such a hybrid system is described and results are shown and discussed. The results show that the combination of lead–acid batteries and PEMFC shows advantages in case of applications with high peak power requirements (i.e. electric scooter) and applications where the fuel cell is used as auxiliary power supply to recharge the battery. The high efficiency of fuel cells at partial load operation results in a good fuel economy for recharging of lead–acid batteries with a fuel cell system.

The all-vanadium redox-flow battery is a promising candidate for load leveling and seasonal energy storage in small grids and stand-alone photovoltaic systems. The reversible cell voltage of 1.3 to 1.4 V in the charged state allows the use of inexpensive active and structural materials. In this work, studies on the performance of inexpensive active materials for use in vanadium redox-flow batteries are reported. Additionally, a cost analysis for a load leveling and a seasonal energy storage system is given based on a flow battery technology well established in Zn-flow batteries.

For a continuous energy supply of photovoltaic operated and off-grid loads, the storage of the solar generated electrical energy is necessary. About 60% of all over the world manufactured solar cells are used for such stand alone systems. In case of photovoltaic systems, mainly electrochemical battery storage systems are used. The paper describes the requirements for batteries in solar systems. The most important storage systems, such as lead–acid, NiMH and Li-ion batteries are described in detail and further developing trends are discussed. As it is well known that the operation conditions strongly influence the battery lifetime, this paper reviews photovoltaic operation conditions and experience in performance and lifetime in photovoltaic systems.

The ‘all vanadium redox flow system’ is a promising candidate for the storage of photovoltaic energy. The reversible cell voltage of 1.3–1.4 V in charged state is well established at various electrode materials in particular carbon based substrate. The kinetics and mechanism were studied for the V2+/V3+ and VO++/VO2+ (V4+/V5+) couples and a one-electron transfer identified as the rate-determining step at smooth surface. The use of activation layers (carbon cloth, felt, etc.) decisively reduced the polarization. Catalysts, which are required for an increase of the reaction rate and the elimination of undesired side reactions, e.g. Ru(O)2 improved the behavior of the positive electrode. The influence of the separator material on mass transfer phenomena (diffusion, migration) and the charge–discharge characteristics were investigated. The requirements to be met as stand alone batteries for the energy supply of users in combination with photovoltaic plants considering the solar irradiation conditions in south Portugal were discussed and the future development goals defined.

For more than 100 years the lead–acid battery (LAB) has been a well-established battery system, mostly used in the traditional battery market. However, during this time, advanced battery systems, such as Ni–MH, high-temperature systems (Na–S, Na–NiCl2) or Li-systems were under development. These new types are directed especially towards new applications such as electric vehicles, battery energy storage or the 4C market (computer, camcorder, cordless phones, cordless tools). This paper discusses how the LAB may co-exist on the traditional and the new battery market in competition to the advanced battery systems. This discussion is based on such important battery parameters as safety, energy, power, costs and recycling, as well as on the improvement potentials of the LAB. This paper is focussed in detail on improvements in energy, power, user properties and corrosion. The influence of the average electron path in the active mass, the active mass thickness and the active mass structure on the battery energy are discussed. The influence of active mass resistance and the cell connector resistance on the battery power are discussed in detail. Industrial solutions to overcome power constraints are demonstrated by the TMF (thin metal foil) battery and bipolar batteries. The advantages and disadvantages of both gel type and adsorbed glass mat (AGM) type of valve regulated lead–acid (VRLA) batteries are explained. Problems related to the Pb/Pb-alloy corrosion in terms of open circuit corrosion and in-situ corrosion measurements are discussed. Taking into account its improvement potential, the LAB will continue to have a strong market position in the future, especially in the fields of stationary and automotive applications.

The performance of lithium ion battery cells is influenced by microscopic and macroscopic parameters. The microscopic parameters are generally responsible for the electrochemical, thermal, safety and lifetime performance of a battery. These parameters are cell chemistry, particle size, diffusion coefficient, equilibrium potential, reaction rate, thermodynamic parameters, thermal and electrical conductivity, heat capacity of active materials and different components of a cell etc. The macroscopic parameters are cell form, aspect ratio, thickness of active materials current collector and separator, tab size and its location. They are used in the cell design process and are optimized to enhance the performance and lifetime of a given cell. When considering upscaling the cell size, the macroscopic parameters attain significance in determining the overall performance of large format cells, which may be largely ignored in short format cells. Increase in the cell size has certain advantages such as, increased cell capacity, fewer connections to the battery pack (parallel connection), low assembly cost, high weight/volume ratio and high reliability of the components of a battery pack. But on the other hand it increases the cell cost, increased safety risk and difficult thermal management etc. But the main disadvantages in terms of performance and lifetime are related to the inhomogeneities occurring in the cell due to increase in cell size. The inhomogeneities can be listed as temperature, current density and state of the charge (SOC) distribution. The current distribution in the cell is affected by tab configuration, current collector thickness and cell aspect ratio. Recent studies show that increasing the cell size increases inhomogeneous current density distribution. Despite the optimized design, the current flow near the tabs is constricted and the location near cell tabs experience higher current density compared to surfaces far away from the tabs. The current distribution influences the local heat generation and introduces temperature inhomogeneity, especially near cell tabs. It also introduces SOC inhomogeneity as SOC in the integral of current over time. These inhomogeneities are related to one another and will have a cascading effect on the overall cell performance. A long term exposure to these inhomogeneities leads to localized aging of the cell. The localized aging creates stronger and weaker regions in cell and further accelerating the aging and leads to premature end of life (EOL) of the cell. With the existing methodologies it is difficult to evaluate cell inhomogeneity as it is completely a localized process. However, recently some experimental and modelling techniques are able to evaluate cell inhomogeneity. These methods can be classified into two as, direct measurement such as, spatial temperature measurements both inside and on the cell surface, multiple or segmented electrode measurements, in-situ and ex-situ diffraction or neutron imaging tools and the second classification is modelling and simulation techniques, which is the focus of this study. This paper shows two modelling techniques, first with spatially distributed battery model developed in Matlab/Simulink and second with Multiphysics based modelling in COMSOL. Both models are able to evaluate inhomogeneity in a large format cell. This work also compares these modelling techniques with respect to complexity, computation and accuracy of the simulation results. Figure 1

Lithium-rich layered-layered oxide materials aLi[Li1/3Mn2/3]O2·(1-a)Li[NixCoyMn1-x-y]O2 (LMO-NCM) are seen as promising candidates for cathode material in next generation Li-ion batteries [1]. This is due to their high specific capacity (> 250 mAh g-1) [2] and low price due to lower cobalt content. On the other hand, the major obstacles for implementation of these materials are high irreversible capacity in first cycle, voltage fade upon cycling due to structural changes, and poor rate capability [1]. For cathode material, thermal stability is particularly relevant as it influences safety. In addition to well-known tendency of layered oxides to decompose exothermally in charged state and emit O2 [3], the oxygen in above mentioned group of materials also participates in charge-discharge reaction [4]. This raises additional safety concerns. We have shown for 0.5Li[Li1/3Mn2/3]O2·0.5Li[Ni1/3Co1/3Mn1/3]O2 that presence of oxidized oxygen in lattice after first cycle increases decomposition enthalpy of the cathode [5]. On the other hand, reversible oxygen and manganese participation in reactions of first cycle also stabilize the structure, which results in a higher onset of decomposition reaction [5]. It has been found that during the activation of Li[Li1/3Mn2/3]O2 (LMO) component in the first cycle, oxygen is emitted from the surface of the material, and oxygen in the bulk is oxidized to compensate for the removal of lithium [6]. Furthermore, the crystal structure changes while the material is held at higher voltage for longer time [7]. These phenomena affect not only performance, but also thermal stability. Knowing that the LMO component decisively impacts the thermal stability of the cathode, the effects of cycling regime, cycling history, ratio between LMO/NCM component, and active material particle size/surface area on thermal stability are investigated. In this research, cathodes are made using LMO-NCM as active material. They are built into half-cells vs. lithium metal with 1 M LiPF6 in flouroethylenecarbonate/diethylcarbonate (1:1) as electrolyte. Cells are conditioned electrochemically at certain current. At certain state of charge, the cathodes are taken out, washed and dried. Thermal stability is investigated by thermogravimetric analysis (TGA) and differential scanning calorimetry (DSC). Evolved gasses are characterised by mass spectrometry. Stability of cathode-electrolyte system is addressed by accelerating rate calorimetry (ARC). Results of the investigations will be presented at the meeting and are expected to elucidate the safety behaviour of LMO-NCM cathodes and help define the required synthesis conditions and microstructure of the cathode active material with regards to safety. Furthermore, appropriate charging regime and cut-off voltage can be derived from the findings. [1] H. Yu, H. Zhou, J. Phys. Chem. Lett. 4 (2013) 1268-1280 [2] A. Ito, Y. Sato, T. Sanada, M. Hatano, H. Horie, Y. Ohsawa, J. Power Sources 196 (2001) 6828-6834 [3] D.D. MacNeil, J.R. Dahn, J. Electrochem. Soc. 148 (2001) A1205-A1210 [4] H. Koga, L. Croguenec, M. Ménétrier, K. Douhill, S. Belin, L. Bourgeois, E. Suard, F. Weill, C. Delmas, J. Electrochem. Soc. 160 (2013) A786-A792 [5] J. Geder, J.H. Song, S.H. Kang, D.Y.W. Yu, ICMAT 2013, Symposium A, Presentation no. ICMAT13-A-1712, Proceedings in preparation [6] H. Koga, L. Croguenec, M. Ménétrier, P. Mannessiez, F. Weill, C. Delmas, J. Power Sources 236 (2013) 250-258 [7] D. Mohanty, A. Sefat, S. Kalnaus, J. Li, R. Meisner, E. Payzant, D. Abraham, D. Wood, C. Daniel, J. Mater. Chem. A 1 (2013) 6249-6261

The energy of a Li-ion cell is in average about 3,250 kJ/kg. About ¼ of this energy is related to electrochemical energy (chemical energy convertible into electrical energy via normal use or short circuit) and ¾ to thermal energy (chemical energy convertible only in thermal energy released at suitable stimulation; e.g. short circuit). The main safety related events are overcharge, external heating, external and internal short circuits, and mechanical deformations of the cell/battery case. These unpredictable events may leads to a temperature increase and at T > 120 °C by internal chemical processes (SEI layer breakup, anode-electrolyte reaction) with gas development and internal pressure increase to cell container rupture at predetermined breaking points. Gas and electrolyte will escape through the rupture holes into the environment and, eventually, inflame by an external ignition source. A further temperature increase till about 200 °C triggers high rate chemical reactions (cathode decomposition with oxygen development, oxygen reaction with the electrolyte, electrolyte decomposition) leading to a very rapid temperature rise (thermal runaway) followed by flames caused by reaching the self-ignition temperature of the electrolyte (~ 450 °C) and possibly explosions. The specific occurrence of most of the above described safety related events, that may occur during the Li-ion battery life cycle phases of transportation, storage, and disposal, will be discussed in this presentation. During transportation and storage, external heating, external and internal short circuits, and mechanical deformations, are the events most likely to happen. Measures which can prevent them (e.g. reliable and low flammable packaging, thermal barriers) and transport related standards (e.g. UN 38.3) are described. The safety hazard in the disposal phase of the cell/battery are also associated with external heating, external and internal short circuits, and mechanical deformations. After collections it is to proof whether the battery is defective (not all functions properly) or damaged (loss of physical integrity). Defective batteries with capacity ≤ 80 % of the nominal value (end-of-life by definition) could be still used in lower demanding applications, e.g. stationary storage in PV houses. Damaged batteries and defective batteries with << 80 % capacity and other malfunctions have to be recycled. In order to reduce the safety risk before the recycling process (including transport), it is advisable to de-energize the battery by low current discharge and finally short it permanently.

For more than 100 years the lead–acid battery (LAB) has been a well-established battery system, mostly used in the traditional battery market. However, during this time, advanced battery systems, such as Ni–MH, high-temperature systems (Na–S, Na–NiCl2) or Li-systems were under development. These new types are directed especially towards new applications such as electric vehicles, battery energy storage or the 4C market (computer, camcorder, cordless phones, cordless tools). This paper discusses how the LAB may co-exist on the traditional and the new battery market in competition to the advanced battery systems. This discussion is based on such important battery parameters as safety, energy, power, costs and recycling, as well as on the improvement potentials of the LAB. This paper is focussed in detail on improvements in energy, power, user properties and corrosion. The influence of the average electron path in the active mass, the active mass thickness and the active mass structure on the battery energy are discussed. The influence of active mass resistance and the cell connector resistance on the battery power are discussed in detail. Industrial solutions to overcome power constraints are demonstrated by the TMF (thin metal foil) battery and bipolar batteries. The advantages and disadvantages of both gel type and adsorbed glass mat (AGM) type of valve regulated lead–acid (VRLA) batteries are explained. Problems related to the Pb/Pb-alloy corrosion in terms of open circuit corrosion and in-situ corrosion measurements are discussed. Taking into account its improvement potential, the LAB will continue to have a strong market position in the future, especially in the fields of stationary and automotive applications.

The effects of different factors such as the state of charge (SOC), the temperature, the polarization potential, the H 2 SO 4 concentration, and the saturation level of the negative plate on oxygen recombination in valve-regulated lead-acid batteries were investigated. The increase of the temperature and of the negative polarization as well as the decrease of the saturation level of the negative plate can all promote oxygen recombination, yet the SOC has almost no effect on the oxygen recombination as long as the capacity of the negative plate is high enough. Furthermore, it is found that the oxygen recombination rate increases and then drops when the H 2 SO 4 concentration at the negative plate increases from 0 to 5 M. The experiments also show that considering all factors, the saturation level of the negative plate is the dominant factor affecting the oxygen recombination rate.

Center for Solar Energy and Hydrogen Research, D-89081 Ulm, GermanyIn the oxygen cycle of valve-regulated lead-acid~VRLA! batteries, there are two ways in which oxygen can move from thepositive to the negative plates, namely, either horizontally to penetrate the absorptive glass mat ~AGM! separator, and/or transportvertically via the gas space. It is found that the oxygen transport depends on the passageway with big void space in the AGMseparator and its rate is proportional to the oxygen partial pressure. The rate constant of vertical transport is about three ordershigher than that of horizontal transport because of the large void space between the AGM separator and plates. However, in thehorizontal direction, the area is very large and the transport path is very short. So the way and the rate of oxygen transport actuallydepend on the level of saturation in VRLA batteries. The horizontal transport is dominant when the saturation is less than 93%,while the vertical transport becomes dominant when it is higher than 93%. The experiments also indicate that with decreasingsaturation, the recombination of more oxygen at the negative plate may oxidize more active H