Chapter 6

Hybrid Energy Systems

Renewable energy sources are poised for growth and expected to be second only to natural gas as the primary energy source in the near future, as described in Chapter 1, Introduction to Energy Systems. However, renewable energy systems (RESs) must overcome several techno-economic challenges for the realization of this anticipated evolution in the global energy system. Some of these challenges include intermittency and operational fluctuations, economics of scale-up and commercialization, materials availability and supply limitations, compatibility and integration into the existing infrastructure, and environmental and ecological concerns [1, 2]. The first of these challenges listed above—intermittency and operational fluctuations—arises from the inherent internal characteristics of the primary renewable energy source and is discussed in detail in this chapter, with the next chapter delving into the rest of the challenges.

Renewable energy sources, specifically solar and wind energy sources—the two sources with the greatest potential for growth—are not capable of producing electricity continuously and consistently. It is clear that solar electricity can only be generated during daylight hours, whereas electricity generation from wind is also possible only during limited hours in a day. In general, solar irradiance and wind speeds exhibit natural patterns that are characteristic of the location on earth; however, these patterns are inevitably subject to unpredictable short- and long-term fluctuations [3]. The randomness of these events translates into intermittent and inconsistent power generation from these sources, creating a mismatch with the consumer demand for electricity that also has its characteristic daily cyclic as well as seasonal variations [4, 5]. Hybrid energy systems (HESs) that feature integration of diverse primary energy sources and/or energy storage capability have the potential to overcome the intermittency/inconsistency/operational fluctuation challenges and facilitate the growth of the RESs.

The intermittency/operational challenges are elaborated upon in the next section, leading to a discussion of various HES configurations and the role of energy storage in HESs. Subsequent discussion will focus on separations and processes for the different energy storage alternatives.

6.1 Intermittency in Renewable Energy Systems: Causes and Impacts

Solar insolation at any location on earth exhibits considerable diurnal and seasonal variation. The highest levels of solar insolation are observed in the months of April to August in the Northern Hemisphere, whereas the months of November to February provide significantly less intense insolation for a shorter period. The reverse is true for a location in the Southern Hemisphere. In addition to the seasonal variation, the daily variation in insolation results in the location having a greater intensity of insolation around late morning/early afternoon. These insolation levels are further impacted by the presence of clouds, with the insolation on cloudy days being only one-third or even lower of that on a clear day. Apart from these daily and annual seasonal variations, the solar insolation is also impacted by periodic climate patterns such as El Niño and Pacific/Atlantic oscillations [6]. Similarly, the wind velocities at any location vary considerably, and electricity generation from wind is particularly sensitive to this variation, as it has a cubic dependence on the velocity [7, 8]. As an example, variation in wind power generation at a Canadian wind farm is shown in Figure 6.1. It can be seen that the peak power generation is nearly two orders-of-magnitude higher than the minimum. Wind energy is even more susceptible than solar energy to climate events, and the variability in wind patterns is likely to become even more unpredictable with climate change that is currently present [7].

A graph represents the data related to power generation from a wind farm.

Figure 6.1 Fluctuations in power generation from a wind farm.

Source: Ibrahim, H., A. Ilinca, and J. Perron, “Energy Storage Systems—Characteristics and Comparisons,” Renewable and Sustainable Energy Reviews, Vol. 12, 2008, pp. 1221–1250.

The intermittency and inconsistent availability of primary energy source—sunlight and wind—results in low capacity factors for solar and wind electricity generation. The capacity factor is defined as the average fraction of the time the power plant operates at its rated or designed capacity and is calculated by taking a ratio of the actual power generated to the maximum power that can be generated, that is, the rated capacity of the plant. Solar plants—thermal or photovoltaic—have typically the lowest capacity factors (between 10% and 20%) of all the power plants, whereas wind energy systems are available nearly one-third of the time on average. It should be noted that the average capacity factor is not an entirely satisfactory indicator in describing the ability of the power generator to meet the consumer demand. It can be seen from Figure 6.1 that the instantaneous capacity factor of the particular wind farm, obtained by dividing the power generated at any time by the installed design capacity of 57 MW, can be as low as 1%–5%. Incidentally, this low capacity factor may also coincide with the period of the highest consumer demand for electricity. In contrast to wind and solar renewables, capacity factors of thermal power stations are generally in the range of 70%–90%, and the percentage availability of nuclear reactors is often into high 90s [2]. One of the consequences of the low capacity factors of the RESs is that the power plants need to be overdesigned for them to meet the consumer demand, resulting in higher capital costs. The overall generation efficiency of the plants is also low, as they rarely operate at the rated design capacity.

Apart from cost considerations, reliability is a key feature preferred by the consumers. Electricity consumers expect their demand to be met by the providers irrespective of the demand pattern or challenges faced by the generators. Figure 6.2 shows the fluctuations in the power demand at a New England location on one autumn day in 2010.

A graph represents the data related to daily power demand.

Figure 6.2 Typical daily power demand fluctuations.

Source: U.S. DOE Energy Information Agency, “Demand for Energy Changes through the Day,” https://www.eia.gov/todayinenergy/detail.php?id=830

The two distinct features of the power demand curve are the morning ramp and the peak demand, which usually occurs later in the day. It should also be noted that the curve depicted in Figure 6.2 shows average hourly demand, which has the effect of smoothening out fluctuations in the demand within any hour. The instantaneous or 5-minute peak demand will have higher peaks and minima, meaning greater stress on the power generator than that is conveyed by the curve in Figure 6.2. Furthermore, the demand curve shown is for an autumn day, with significantly less demand for air conditioning than on a summer day or heating demand on a winter day.

Solar or wind RES is unable to meet the reliability requirement of the power consumers as a single source. HESs enable an RES to meet the market demand of electricity through integration with other RESs, nonrenewable sources, or by incorporating energy storage technologies in its design and operation [9].

6.2 HES: Definition and Architecture

Several different definitions of HESs are available in the literature, and some of these are presented as follows:

“Hybrid energy systems incorporate two or more electricity generation options, based either on the exploitation of renewable energy sources (RES) or on small thermal power units, e.g. diesel-electric generators or even micro-turbines.” [10]

“Hybrid energy systems are combinations of two or more energy conversion devices (e.g. electricity generators or storage devices), or two or more fuels for the same device, that when integrated, overcome the limitations that may be inherent in either.” [11]

“Single facility which takes two or more energy resources as inputs and produces two or more products, with at least one being an energy commodity (e.g., electricity or fuel).” [12]

“HES are energy product production plants that take two or more energy resource inputs (typically includes both carbon and non-carbon based sources) and produce two or more energy products (electricity, liquid transportation fuel, industrial chemicals) in an integrated plant.” [13, 14]

“Hybrid energy system is the composed structure of different energy systems supplied by the respective energy source.” [15]

Each of these definitions is successful in capturing some aspect of HESs, and yet is not comprehensive enough to include all possible HES configurations. It may be beneficial to analyze these definitions in the context of the energy system architecture presented in Chapter 1, wherein a primary energy source is transformed into an energy carrier, which in turn provides the motive for service technologies to satisfy the consumer demand.

The abovementioned first definition limits the HES to those producing electricity from multiple sources. A vast majority of HESs indeed feature combinations of primary energy sources—two or more renewables, a renewable-fossil, or renewable-nuclear—that provide electricity with greater certainty or reliability to the consumers. However, the definition excludes systems that generate transportation fuels, heat, and other forms of energy currencies.

The second definition describes the HESs in terms of the energy conversion devices and energy resources (input to the energy system), and does not mention system outputs, which is primarily electricity. As such, it is an incomplete description of the system, and limitations mentioned in the definition are not specified.

The third and fourth definitions require two or more energy resources that provide two energy currencies. These definitions would exclude an energy system that converts a single resource (nuclear energy, for example) to multiple currencies such as electricity and hydrogen. They would also not include a system that produces a single currency (electricity) from multiple resources. The fourth definition also includes industrial chemicals, which are typically not energy currencies, as an output.

The fifth definition defines the energy system as an object structure in the form {[production (solar, wind, etc.)] [utilization (electricity, heat, etc.)]}. In that sense, it is a general definition that would cover most energy systems. However, it does not discuss the energy system in terms of currencies, and the definition is too broad to convey adequate information and is cumbersome to use.

In light of these limitations, a revised energy system architecture is proposed for HESs as shown in Figure 6.3, followed by a general definition of an HES.

An illustration of the HES architecture.

Figure 6.3 HES architecture.

An HES can now be defined based on this architecture in terms of energy resources and currencies:

An HES is defined as

i. a system for the conversion of a single energy source into electricity and other energy currency/currencies, OR

ii. acombination of systems for conversion of multiple energy sources into electricity and possibly other currencies

for satisfying service demands of consumers or effecting storage of currencies when generated in excess of the consumer demand.

As per this definition, an HES is characterized by:

  • Electrical energy as the essential energy currency output common to all types of HESs.

  • Multiple energy sources if electricity is the only output. As different energy sources require different transformer technologies for conversion to electrical energy, this also requires multiple transformer technologies.

  • Multiple energy currency outputs if only a single energy source is used.

  • Possibly having energy currency storage options.

  • Multiple energy source inputs and multiple energy currency outputs in the most general case.

The energy currencies, apart from electricity, may include heat, synthetic transportation fuels, or hydrogen. The system may include energy storage options; however, although it is certainly a desired feature of any energy system, it is not essential for the system to be classified as an HES. The HES may have multiple primary energy sources, even though only two are shown in Figure 6.3. It should be noted that transmission and distribution of energy currencies or service technologies do not factor in determining whether or not an energy system is characterized as hybrid. Furthermore, in the case of multiple energy sources, multiple transformer technologies must be utilized for obtaining the energy currency, so that co-firing of biomass with coal (or a similar approach to other mixtures) in an existing combustor is not considered a hybrid system.

As seen in the figure, an energy currency can be used for multiple services, for example, electricity for transportation, heating, cooling, and illumination, or natural gas for space heating and cooking. Furthermore, multiple currencies can be used for satisfying a single service demand, for example, both hydrogen and electricity can be used for transportation.

The term renewable hybrid energy system (RHES) is typically used to describe an HES that produces electricity using multiple renewable energy resources such as solar energy, wind, and biomass. For example, a geothermal solar energy system consisting of a geothermal power cycle coupled with a solar thermal power [16]. Nuclear hybrid energy system (NHES), as indicated by its name, is an HES that produces electricity and other energy currency, most likely hydrogen or synthetic hydrocarbon transportation fuel from nuclear resources. Other possible energy currency that an NHES can deliver is thermal energy (heat), typically in the form of steam or coolants, such as helium and molten salt. NHESs are attracting particular attention, as they enable nuclear energy to be used for transportation, potentially reducing the demand for petroleum imports for the oil-importing nations. Nuclear-renewable hybrid energy systems feature combination of nuclear energy with a renewable resource (wind, solar, geothermal, biomass, etc.) producing electricity and heat/hydrogen/biofuels [2]. Wind hybrid energy systems (WHESs) typically involve an energy storage system based on the conversion of electricity to hydrogen using solid polymer electrolyte (SPE) cells during excess electricity supply conditions [17]. This hydrogen can be stored and converted back to electricity during excess demand conditions (absence of wind) using a proton exchange membrane fuel cell (the electrolyte cell in reverse). Alternately, excess electricity may also be stored in secondary batteries (electrochemical storage), such as lead–acid, lithium-ion, or redox flow batteries [10].

It should be noted that energy storage is a desired feature for all energy systems, not just RHESs or HESs. An energy system with energy storage capabilities can operate at steady state continuously, maximizing the generation to store energy at times of low demand and cost, relying upon the stored energy to meet consumer demands at times when they exceed the power generation capacity [4]. However, energy storage is a particularly desirable feature for RESs due to their intermittency characteristics.

6.3 Energy Storage: Fundamentals and Alternatives

The term energy storage is commonly understood to mean electrical energy storage in the context of energy systems. However, energy storage is a broader field that encompasses storage of other forms of energy and energy currencies, such as heat or chemicals, as well. Storage of these other forms of energy or energy currencies has its own challenges; however, none are as severe as those encountered in the storage of electricity. The particular difficulty in the storage of electricity can be understood from its basic nature: electricity has two forms—static and dynamic. Static electricity involves accumulation of electric charges at two spatially distinct locations; dynamic electricity involves discharge of the static electricity through the flow of electric charges, that is, the electric current that performs useful work. Electrical energy storage involves processes and equipment for storing this dynamic electricity in some form during periods of excess generation and recovering it from the stored form when the demand exceeds the supply [4].

It is indeed possible to store electrical energy directly as static electricity by separating the charges in ultracapacitors or superconducting magnets (superconducting magnetic energy storage or SMES) [18]. However, both of these options are technologically challenging, limited in size and applications, and expensive. Electrical energy, therefore, is stored indirectly by converting it into some other form of energy or energy carrier. Figure 6.4 provides an overview of the possible options—direct and indirect—for the storage of electrical energy. Each of these options is discussed briefly below.

A classification diagram of electrical storage alternatives.

Figure 6.4 Electrical energy storage alternatives.

(SMES—superconducting magnetic energy storage; CAES—compressed air energy storage)

6.3.1 Direct Energy Storage

Direct energy storage is perhaps the most desirable mode of storing electrical energy from the perspective of recovering the stored energy. Capacitor storage involves separating the positive and negative charges on two different conducting surfaces separated by an insulator. Amount and characteristics of the capacitor storage depend upon the applied voltage, permittivity of dielectric, surface area, and the separation distance between the surfaces [19]. Electrostatic capacitors are the conventional capacitors whose capabilities are enhanced through advanced manufacturing techniques, giving them a superior nanostructure for energy storage applications, particularly where high power density is desired [3]. Electrochemical capacitors are capacitor devices modified by the incorporation of an electrolyte in the space separating the two charged surfaces. This results in the formation of a double layer of charges in the device, increasing the amount of energy that can be stored within the device [3, 4]. Electrochemical capacitors may also have pseudocapacitance properties, when the Faradaic (redox) reactions also occur, enhancing their storage capacities. The Faradaic reactions in pseudocapacitors differ from those in the batteries, in that the charges in pseudocapacitors remain at the electrode surfaces, unlike batteries where they are in the bulk of the electrolyte. The charge–discharge characteristics of the pseudocapacitors are similar to those of the capacitors rather than batteries. The advantages of capacitor devices are related to the nature of direct storage—no conversion devices or technologies are needed for the storage of electrical energy, and hence both energy charging and discharging are fast. This makes them suited for applications where high power is desired, that is, energy discharge over a very small period. Furthermore, these devices typically have a very high cycle life—a practically unlimited number of charge–discharge cycles [19]. The disadvantages of the devices stem from the low energy density, limiting the amount of energy that can be stored in the device and their susceptibility to energy loss due to self-discharge [4].

SMES involves storing electric current in superconducting coils in a magnetic field. The current increases during the charging step and decreases during the discharge step. Energy densities for direct storage in a magnetic field can be orders of magnitude higher than in the electric field [20]. The system requires superconductors as the conductor resistance translates into energy losses from the system. This, in turn, imposes a low temperature requirement for the system. Although the amount of energy stored can be substantial, the system size is quite large and the system temperature–pressure requirements are severe. SMESs also have stringent safety requirements to ensure that the current does not find alternate paths to discharge, which can have disastrous consequences [21]. All these factors serve to make SMES a high-cost, capital-intensive alternative for energy storage. The advantages of SMES stem from its direct storage characteristics—a fast response that is suited for high-power applications [19].

Both the direct storage options have the advantage of fast dynamics making them suited for high-power applications. However, both of the options are limited in the quantity that can be stored effectively and are susceptible to self-discharge, making them unsuitable for bulk energy storage applications.

6.3.2 Indirect Energy Storage

Indirect energy storage involves the conversion of electrical energy into other forms of energy, essentially reverse of the processes described in Chapter 2, Renewable Energy Sources. As shown in Figure 6.4, electrical energy can be converted into mechanical, chemical, or thermal energies for storage.

6.3.2.1 Storage by Conversion into Mechanical Energy

Electrical energy can be converted into either kinetic or potential energy for storage. Conversion to kinetic energy typically involves a flywheel coupled to an electric machine. The energy storage step involves operating the system in the motor mode, where electrical energy input to the device results in the acceleration of the flywheel mass, thus increasing its kinetic energy. The operation is switched to generator mode when the demand exceeds the power generation capacity, and the deceleration of the flywheel transforms the stored kinetic energy into electrical energy [19]. Flywheel energy storage has been used in vehicle operations for a long time; however, its coupling to stationary power generators for large-scale storage is relatively recent [20]. The amount of energy stored in a flywheel depends upon its mass and the square of the angular velocity [18]. Obviously, heavier flywheels can store greater amounts of energy; however, rotation at high speeds puts the flywheels under significant strain. Flywheels made of fiber-reinforced composite materials have superior tensile strength characteristics than metals (stainless steel). Furthermore, flywheels made of composite materials disintegrate rather than explode catastrophically and are inherently safer than metal flywheels [18]. Any kinetic energy storage system must be designed to minimize frictional losses, and the flywheel energy storage system operates in a vacuum in order to minimize drag losses. Furthermore, the flywheels need to be mounted using low-friction bearings or magnetic suspension to minimize energy loss. It should be noted that these losses can be minimized, but not completely eliminated, meaning that the flywheel energy storage system will be effective over shorter durations but lose its efficiency with time [20].

Frictional losses are not a significant concern for the potential energy storage alternatives—pumped hydro storage (PHS) and compressed air energy storage (CAES) system. PHS systems require the presence of two reservoirs of water located at different elevations. Excess electricity available at the times of low demand is used to pump water from the lower reservoir to the upper reservoir, effectively converting electrical energy into potential energy of water. This potential energy is reconverted into electrical energy whenever needed for satisfying excess demand by returning the water from the upper to lower reservoir through turbines [20]. PHS is characterized by high energy capacity and negligible self-discharge losses and dominates over all other storage alternatives, accounting for over 97% of energy storage capacity (https://www.hydro.org/waterpower/pumped-storage/).

CAES bears some similarities to the PHS system, in that excess electrical energy is used to compress air, increasing its potential to perform work. However, the energy discharge step is conceptually different and is linked to a gas-turbine power generation system. A gas-turbine plant essentially consists of four units: compressor, combustion chamber, turbine, and generator. The compressor provides air at high pressure needed for combustion and is responsible for a large parasitic load that may consume up to 60% of the electricity generated in the turbine [20]. CAES enables the gas-turbine plant to avoid this parasitic load by providing high-pressure compressed air for combustion [22].

PHS and CAES are both constrained by geographical considerations. PHS requires an appropriate terrain to locate two reservoirs at different elevations at the site of the power generation system. CAES requires storage of large volumes of air, which is economically feasible only if natural formations, such as underground salt caverns, rock domes, or porous water reservoirs, are available [19]. Furthermore, as mentioned earlier, CAES is coupled to a gas-turbine power generation system, a non-renewable fossil energy source that is not carbon neutral.

6.3.2.2 Storage by Conversion into Chemical Energy

Converting electrical energy into chemical energy avoids geographical limitations inherent in the mechanical energy storage, while also offering the advantage of long-term storage without significant losses. These storage options can broadly be classified into two distinct modes: electrochemical storage and chemical storage.

Electrochemical energy storage is based on rechargeable or secondary batteries, wherein redox reactions occur at the electrodes in electrochemical cells. The reactions are reversible, and the battery is charged by applying external voltage when excess electricity is available. Energy is stored in the chemicals comprising the electrolytes in the electrochemical cells. The direction of the reactions is reversed at the time of excess demand, changing the nature of the chemicals in the electrolyte that results in the delivery of the electric current from the battery to the external circuit [4]. Electrochemical storage offers several advantages over other storage alternatives, such as flexibility, good response characteristics, low self-discharge losses, and ability to scale up to larger systems. Lead–acid batteries are the most common secondary batteries used for energy storage. Other important battery options include alkaline Ni-based batteries, flow-through batteries, high-temperature batteries, and Li-ion batteries [18].

Chemical storage of energy involves converting electricity (or other energy currency) into chemicals such as hydrogen, methane or other hydrocarbons, or alcohols (methanol, ethanol) [23]. The biggest advantage of the chemical energy storage is the high energy density of the chemicals enabling the storage of vast quantities of energy without any self-discharge losses. Hydrogen is considered to be particularly attractive due to its high gravimetric energy density; however, the volumetric energy density is quite low resulting in large system volumes. Energy storage into chemicals also offers the possibility of utilizing them in other applications such as transportation or industrial processes. This will enable RESs to expand their scope beyond power generation into other sectors of the economy.

Hydrogen and methanol, in particular, can also be used in fuel cells, which are often considered another energy storage option and are shown in Figure 6.4 [4, 18]. In reality, fuel cells are merely conversion devices (transformation technologies) that do not store any energy, but merely transform chemical energy stored in a fuel (hydrogen, methanol, etc.) into electrical energy. Processes occurring in the electrochemical and chemical storage devices are discussed in more detail in Section 6.4.

6.3.2.3 Storage by Conversion into Thermal Energy

Thermal energy storage (TES) systems are a natural option for thermal power stations, such as nuclear- or fossil fuel–driven power plants, where high-temperature heat is already available [20]. Incorporating TES into RESs is not as straightforward except for solar thermal/biomass/geothermal systems where high-temperature heat is available. However, electrical energy can always be converted into heat and stored into some medium during times of excess generation. Stored heat can be converted back into electricity through a heat engine at times of excess consumer demand [24].

TESs function in one of the following three ways: (1) sensible heat storage, (2) latent heat storage, or (3) thermochemical heat storage (TCS) [2527]. Sensible heat storage involves heating a medium (molten salts, rocks, pressurized water, etc.) without any phase change during the energy storage step. The heating can be accomplished directly for high-temperature energy generation processes (solar thermal, for example) or by electrical heating for low/ambient-temperature electricity generators (wind, solar-photovoltaics, etc.). The energy discharge step will typically involve transferring the stored heat to obtain pressurized steam for power generation.

Latent heat systems involve changing the phase of the material (molten salts, sodium hydroxide) in which heat is stored without a change in temperature [8]. Typically, the phase change involves a solid to liquid transition for minimal change in volume. Theoretically liquid-to-vapor transition is also possible for energy storage; however, the dramatic increase in the specific volume in the transition presents significant challenges. The energy discharge step involves reversing the direction of the phase change, and latent heat released during solidification is used to obtain pressurized steam for power generation.

TCS systems involve reversible chemical reactions for energy storage-discharge. The energy storage step involves an electricity- or thermally driven endothermic reaction, such as decomposition of hydroxides or carbonates. The products are typically easily separable because they are in different phases. The two products are recombined in the reverse reaction, which is exothermic. The heat released in this step can be harnessed to generate high-pressure steam and electricity as for the sensible and latent heat systems. Systems based on thermochemical reactions have a potential to achieve much higher energy densities than those relying on sensible- or latent heat storage [28, 29]. TCS systems are based on energy storage in the form of chemical energy; however, they differ from the chemical energy storage described in the previous section. The reactions involved in the TCS are cyclic and there is no net consumption of chemicals, whereas the energy discharge step in the chemical energy storage in the previous section results in the oxidation and mineralization of the chemical species.

Processes occurring in the TES systems are discussed in more detail in Section 6.5.

6.3.3 Characteristics and Applications of Energy Storage Systems

It can be seen that there is a wide range of energy storage options that differ widely in their characteristics, such as response time, energy density and capacity, power density, land and capital requirements, self-discharge losses, technology maturity, efficiency, and so on. Options such as capacitors are ideally suited for high-power applications, but have only limited storage capacity. PHS, on the other hand, has practically unlimited energy storage capacity, but the response is too sluggish to meet instantaneous fluctuations in the power demand. Important characteristics of the energy storage system are the following [4, 8]:

  • Storage capacity: total amount of energy that can be stored in the system.

  • Power availability: the maximum power the system is able to provide and how long it can be sustained.

  • Round-trip efficiency: the ratio of energy output to energy input for the system.

  • Calendar and cycle life: calendar life refers to the time period over which the energy storage system can be expected to perform at its capacity. Cycle life refers to the number of times it can be subjected to charge–discharge cycles without deterioration in performance. Calendar life is particularly important for storage devices (e.g., batteries) that deteriorate over a period of time due to structural or other changes or are susceptible to self-discharge.

  • Reliability: dependability of the storage system.

  • Environmental impact: the measure of environmental impacts, if any, of the storage system.

Apart from these considerations, technological maturity of the storage option has a major influence on the choice of the storage option. Above all, capital and operating costs are always overriding concerns for any practical system.

Not all of these characteristics are equally important for all energy storage applications that vary widely both in terms of duration and total energy requirements. The distinct energy storage applications require different solutions that are appropriate for a particular situation. The energy storage applications can be broadly classified into four categories: power quality management, spinning or standby reserve, load shifting, and addressing intermittency issues [18, 30].

Power quality management or system regulation applications involve negating the effects of short-term fluctuations arising from random demand variations, momentary power outages, voltage and power surges, and other transient phenomena over nanosecond and millisecond timescales to provide uninterrupted, reliable power.

Standby or spinning reserve applications involve longer-term phenomena involving a sudden rise in demand for an extended period or a failure of the main power-generating station. In the absence of energy storage, energy systems ramp up power generation from a reserve power generation plant that is operating at lower capacity and efficiency and is positioned to pick up the excess demand. Energy storage would eliminate the need for such spinning reserves.

Load shifting, peak shaving, and load leveling applications involve storing energy at times of excess generation and drawing upon this stored energy when demand exceeds generation.

As discussed earlier in this chapter, some RESs have inherently intermittent power generation characteristics. Energy storage is critically important for these systems, which can have wild variation in power generation.

The applicability of various storage options for various power and energy demands is shown in Figures 6.5 and 6.6. Figure 6.5 shows the positioning of various energy storage alternatives with respect to the energy and power demands, whereas Figure 6.6 shows similar information with respect to the power and operational time requirements for the system.

A graph presents the relationship between energy stored and power output.

Figure 6.5 Energy–power diagram for energy storage system applications.

Source: Ibrahim, H., A. Ilinca, and J. Perron, “Energy Storage Systems—Characteristics and Comparisons,” Renewable and Sustainable Energy Reviews, Vol. 12, 2008, pp. 1221–1250.

A diagram presents the energy storage alternatives based on the system demand.

Figure 6.6 Applicability of energy storage alternatives with respect to the system demand.

Source: Electric Power Research Institute (EPRI), “Electrical Energy Storage Technology Options,” Report 1020676, EPRI, Palo Alto, California, 2010.

It can be seen from the these figures that direct energy storage devices are suited for short-term power applications addressing the power quality issues, whereas large-scale, extended energy demands require PHS and CAES. Electrochemical storage can be useful over several timescales, and is particularly suited as the solution for longer-term power quality and load shifting issues. The total energy capacity of electrochemical storage appears to be lower than what is required for bulk power management; however, batteries are modular by nature, and their energy storage capacity can be increased simply by increasing their number. Thermal and chemical energy storage systems are not shown in these figures; however, these systems typically have high energy densities and capacities, making them ideally suited for extended operations—for bulk power management and addressing intermittency issues of RESs.

Technologies for PHS and CAES are fully mature, and these are the two most dominant energy storage options currently in large-scale usage. These mechanical energy storage options do not involve chemical separations and processes, except in auxiliary systems. Subsequent discussion in this chapter focuses upon some of the separations in chemical storage technologies (including electrochemical systems) and thermal storage technologies (including thermochemical systems).

6.4 Separations and Processes in Chemical Energy Storage

As described earlier, the chemical storage options comprise electrochemical storage (batteries) and bulk storage of chemicals to be used as fuel. Processes in electrochemical storage are discussed first followed by the processes for chemical storage. Processes in fuel cells, also electrochemical devices, are energy conversion processes wherein chemical energy is converted to electrical energy. Batteries affect the conversion of chemical energy to electrical energy as well; however, the storage of chemicals occurs within the battery, whereas the storage is external to the fuel cells.

6.4.1 Electrochemical Energy Storage

The fundamental process occurring in an electrochemical process comprises oxidation–reduction reactions. The oxidation reaction (loss of electrons) takes place at the anode, whereas a reduction reaction (gain of electrons) occurs at the cathode [32]. Both the electrodes may be located in a single electrochemical cell in an electrolyte or separately in two half-cells filled with either an anolyte or a catholyte. A salt bridge provides an ionic connection between the two half-cells. Completing the external circuit by connecting the two electrodes by a conductor allows the electrons to flow, creating an electric current. The overall process in the electrochemical cell can be galvanic—spontaneous reactions converting the chemical energy to electrical energy, or electrolytic—affecting chemical changes through imposition of external voltage. Primary cells or batteries are galvanic devices that convert chemical energy into electrical energy. Primary cells are single-use devices and have to be discarded once the chemicals stored in the cell are exhausted, that is, when redox reactions go to completion. The reactions in a primary cell are irreversible for all practical purposes. A secondary cell, on the other hand, can operate in both galvanic and electrolytic modes, with reversible reactions. These secondary cells function as energy storage devices while operating in electrolytic mode wherein excess electrical energy available at the times of low demand is used to drive the chemical reactions affecting the conversion to chemical energy. When needed, the operation is switched to the galvanic mode, converting the stored chemical energy into electricity. Secondary cells are thus rechargeable and cycle between the galvanic and electrolytic modes as needed.

Secondary batteries are ideally suited for storage of electrical energy, as they typically have a very high round-trip efficiency, fast response time, low self-discharge, and handle a large number of charge–discharge cycles. Their modularity allows a flexibility of sizing, and it is very convenient to switch the operation from charging to discharging mode. Secondary batteries are already widely used at the individual level by the general public for mobile applications (car batteries, batteries for portable electronic appliances) as well as stationary applications (inverters). The same batteries (lead–acid car batteries) and Li-ion (portable electronics) are also used for grid-level electric energy storage. The processes occurring in these and some other batteries are presented below.

6.4.1.1 Lead–Acid Battery

Lead–acid batteries are the most ubiquitous secondary batteries worldwide due to their use in automobiles. Conceptually, it is one of the simplest electrochemical cells with a single electrolyte and two electrodes that have lead as the principal component. The functioning of lead–acid batteries can be understood clearly from Figure 6.7, which shows both the discharged and charged states of the battery.

Two diagrams depict two different states of a lead-acid battery.

Figure 6.7 Lead–acid battery: (a) discharged state and (b) charged state.

(+ and − signs represent positive and negative electrodes, respectively)

Both the electrodes (anode and cathode) of the lead–acid battery are essentially converted to lead sulfate in the discharged state, with diluted sulfuric acid as the electrolyte as seen from Figure 6.7(a). The battery is charged by applying voltage and passing electric current to the charged state as shown in Figure 6.7(b). The electrolyte is also transformed into concentrated sulfuric acid in the process. The equations representing these transformations for the charging process are shown below.

PbSO4(s)+2H2O(l)PbO2(s)+HSO4(aq)+3H++2e         (R6.1)

PbSO4(s)+H++2ePb(s)+HSO4(aq)         (R6.2)

Equation R6.1 represents the anodic reaction involving the loss of electrons during the charging process, whereas equation R6.2 represents the cathodic reaction signifying the gain of electrons. During the charging process, the anodic reaction occurs at the positive electrode, whereas the cathodic reaction takes place at the negative electrode. During the discharge process, the directions of the reactions are reversed. Loss of electrons—anodic oxidation reaction—occurs at the negative electrode, converting lead to lead sulfate, while the cathodic reaction occurs at the positive electrode, transforming lead oxide to lead sulfate through gain of electrons. The overall equation for the process is:

2PbSO4(s)+2H2O(l)Pb(s)+PbO2(s)+2H2SO4(aq)         (R6.3)

The forward reaction represents the charging process, whereas the reverse reaction is the discharge process. The concentration of sulfuric acid in the charged state is ~6 M and is the store of the energy [4]. The standard reduction potential1 for the half-reaction represented by equation R6.1 is 1.685 V, whereas that for the reaction represented by equation R6.2 is −0.36 V. The overall reversible potential of the cell is ~2.1 V, equivalent to a free energy change of −405 J/mol Pb [32]. A typical car battery consists of six such cells connected in series to provide a potential of 12 V.

1. By convention, all redox half-reactions are expressed as reduction reactions, with the standard reduction potential for hydrogen formation [H+(aq)+e12H2(g)] assigned a value of 0.0 V. Potentials for all other reactions are expressed with respect to this base value.

Considerable advances have taken place in the design of lead–acid batteries. The sulfuric acid electrolyte in earlier batteries used to be liquid, and older car batteries often needed topping up—ensuring that the electrode plates remained submerged in the electrolyte—with water. Developments in the battery technology have resulted in the use of gel electrolyte, sealed (VRLA—valve-regulated lead acid) batteries obviating the need and the ability to top-up, use of glass fiber mats that are soaked with electrolyte, advanced electrode materials, and electrode–electrolyte arrangements [33]. The fundamental electrochemical reactions, however, have remained the same since the time of lead–acid battery invention in 1859.

6.4.1.2 Ni–Cd Battery

The Ni–Cd battery, like the lead–acid battery, has a long history of development and is a mature technology. The discharged and charged states of the Ni–Cd battery are shown in Figure 6.8.

Two diagrams depict two different states of a Ni-Cd battery.

Figure 6.8 Ni–Cd battery: (a) discharged state and (b) charged state.

(+ and − signs represent positive and negative electrodes, respectively)

In the discharged state, the positive electrode of the Ni–Cd battery comprises nickel hydroxide, whereas the negative electrode has the formation of cadmium hydroxide. Applying electric current to the discharged battery transforms it into a charged state as shown in Figure 6.8(b), where the negative electrode is of metallic cadmium, whereas the positive electrode is transformed into nickel oxyhydroxide. The alkaline electrolyte generally consists of potassium hydroxide. The equations representing these transformations for the charging process are shown below.

2Ni(OH)2(s)+2OH2NiO(OH)(s)+2H2O(l)+2e         (R6.4)

Cd(OH)2(s)+2eCd(s)+2OH(aq)         (R6.5)

Equations R6.4 and R6.5 represent the anodic reaction (loss of electrons) and the cathodic reaction (gain of electrons), respectively, during the charging process. As with all secondary batteries, the anodic reaction occurs at the positive electrode, while the cathodic reaction takes place at the negative electrode in the charging process. During the discharge stage, the directions of the reactions are reversed. Loss of electrons—anodic oxidation reaction—occurs at the negative electrode, converting cadmium to cadmium hydroxide, whereas the cathodic reaction occurs at the positive electrode, transforming nickel oxyhydroxide to nickel hydroxide through the gain of electrons. The overall equation for the process is:

2Ni(OH)2(s)+Cd(OH)2(s)Cd(s)+2NiO(OH)(s)+2H2O(l)         (R6.6)

The forward reaction represents the charging process, the reverse reaction being the discharge of the stored energy. The standard reduction potentials for the two half-cell reactions (R6.4 and R6.5) are 0.8 V and −0.403 V, respectively, resulting in a nominal potential of 1.2 V for the Ni–Cd battery, equivalent to a Gibbs energy change of −230 J/mol Cd. Ni–Cd batteries for portable applications are of the sealed type; however, a vented type of batteries is used for energy storage applications. These batteries are the flooded type, with 30% aqueous KOH solution serving as the electrolyte. The electrodes are made of plates of Ni and Cd, separated by silicone rubber.

Ni–Cd batteries have higher energy density than lead–acid batteries and are more robust than many other types of batteries requiring low maintenance. The disadvantages of these batteries include low cycle life, concerns about Cd toxicity, and cost compared to lead–acid batteries. Ni–Cd batteries also suffer from the memory effect—full charge is possible only after full discharge. Ni–Cd batteries have been displaced from portable applications by nickel–metal hydride (Ni–MH) and Li-ion batteries due to their superior characteristics. Ni–MH batteries have higher energy densities and do not have the toxicity concerns associated with Cd in Ni–Cd batteries. Li-ion batteries have even superior energy density characteristics due to the use of lighter Li metal [4, 30]. However, these batteries are more expensive than Ni–Cd batteries, which is a major hindrance for large-scale usage. Fairbanks, Alaska, has one of the largest Ni–Cd battery energy storage installations with a capacity of 27 MW, used mainly for power quality management [34].

6.4.1.3 Li-Ion Battery

Li-ion batteries were developed in the 1970s, earning its developers—John B. Goodenough, M. Stanley Whittingham, and Akira Yoshino—the 2019 Noble Prize in Chemistry. The battery is based on shuttling Li-ions between positive and negative electrodes in charge–discharge cycles and has come to dominate the portable electronics market as well as electric vehicle applications [4, 32]. Figure 6.9 shows the discharged and charged states of the Li-ion battery.

Two diagrams depict two different states of a Li-ion battery.

Figure 6.9 Li-ion battery: (a) discharged state and (b) charged state.

(+ and − signs represent positive and negative electrodes, respectively)

The positive electrode is made of lithiated metal oxide, such as lithium cobalt oxide or lithium manganese oxide, whereas the negative electrode is generally graphite (C6) with a layered structure. Recent development in electrode technology includes a phosphate-based positive electrode such as one made of lithium iron phosphate [35]. Like all alkali metals, Li reacts violently with water, and hence the electrolyte is made of an organic salt of Li, such as LiPF6, dissolved in organic carbonates (ethylene carbonate) [4]. A charged battery will have Li-ions intercalated in the graphite layers, as shown in Figure 6.9(b). Discharging the battery results in transfer of these ions into the metal oxide structure, resulting in the state shown in Figure 6.9(a). The half-cell reactions occurring at the positive and negative electrodes during the charging process are shown in equations R6.7 and R6.8, respectively.

LiCoO2(s)CoO2(s)+Li+(org)+2e         (R6.7)

C6(s)+Li+(org)+2eLiC6(s)         (R6.8)

The overall equation for the process is:

LiCoO2(s)+C6(s)CoO2(s)+LiC6(s)         (R6.9)

The forward reaction represents the charging process, the reverse reaction being the discharge of the stored energy. The potential of the first reaction (reaction at the positive electrode) is ~4.0 V with respect to the Li/Li+ reaction at the negative electrode, resulting in a nominal cell potential of 3.6–3.85 V for lithium metal oxide type batteries [36]. Batteries having the lithium iron phosphate as the positive electrode have a slightly lower potential of ~3.2 V.

The obvious advantage of Li-ion batteries is a very high energy density due to the use of Li. In addition, it suffers from no memory effects and has low self-discharge. However, cost and safety considerations are a barrier to their use for large-scale energy storage. Advances in materials and other technologies, as well as increasing usage in the transportation sector, will drive the costs downward; however, the increased use in other sectors limits the availability of lithium for energy storage applications.

6.4.1.4 Sodium–Sulfur (Na–S) Battery

The Na–S batteries differ from the earlier battery types in two important ways: (1) the electrolyte is solid β-alumina and the electrodes are made of molten elements (sodium and sulfur) and (2) the operational temperature is high (300°C–350°C). The conceptual functioning of the Na–S battery is shown in Figure 6.10.

Two diagrams depict two different states of a Na-S battery.

Figure 6.10 Na–S battery: (a) discharged state and (b) charged state

(+ and − signs represent positive and negative electrodes, respectively)

The ceramic β-alumina electrolyte is a key component of the battery maintaining a separation between the anolyte and the catholyte, allowing only the sodium ions to pass through. The energy charging reactions are [37]:

SX2xS(l)+2e         (R6.10)

2Na+(l)+2e2Na(l)         (R6.11)

The overall equation for the process is:

Na2Sx(s)2Na(l)+xS(l)         (R6.12)

As with the other equations stated earlier, the forward reaction represents the energy storage step and the reverse reaction is the energy discharge step. The need for higher operating temperatures stems from the requirement to maintain both Na and S in the liquid state. The standard reduction potentials for the reactions, represented by equations R6.10 and R6.11, are −0.51 V and −2.71 V, respectively, yielding a potential between 1.8 and 2.2 V at operating conditions for the battery.

As seen from the equations, the energy discharge process results in the formation of sodium polysulfide (Na2Sx, x: 3–5) at the positive electrode (cathode for the galvanic process). Both sulfur and polysulfide are electrical insulators, and electron transfer is facilitated by using carbon felt. During the discharge process, sodium is transferred from the anodic compartment to the cathodic compartment, decreasing its level in the anodic compartment and reducing the electrolyte area through which the transfer occurs. The battery needs to be designed to account for this loss, as well as to ensure that complete drying does not occur [37].

Na–S batteries were developed initially for electric vehicle applications, where Li-ion batteries have practically displaced all other battery types. However, the battery technology is quite mature and has advantages of high energy density, excellent cycle life, and use of abundantly available materials. Na–S batteries have been used at over 50 power generation installations worldwide with a total capacity of 365 MW, primarily for power quality management and load-shifting applications [34]. Larger installations have power ratings of tens of megawatts, and total energy capacity to last up to 6–8 hours [30]. However, they operate at high temperatures and have significant self-discharge losses. Furthermore, failure of the separator membrane can have disastrous consequences due to a direct reaction between the sodium and sulfur.

Sodium–metal halide batteries are similar to Na–S batteries, in that they use the same β-alumina electrolyte and are functionally similar, with the same reactions at the negative electrode. Sodium ions combine with the halide ions, reducing the metal at the positive electrode (cathode) during discharge. Sodium–metal halide batteries have higher voltage, energy density, lesser corrosivity, and greater safety than the Na–S batteries [37]. The Na–NiCl2 battery has received particular attention lately and is the focus of developmental efforts in several countries.

6.4.1.5 Vanadium Redox Battery

Two of the characteristics shared by all of the conventional secondary batteries, certainly the ones described earlier, are that all involve electrodes that participate in the reactions with concomitant cyclic chemical and structural changes and the chemical energy store is contained within the battery. A redox flow battery is a different type of secondary battery, wherein the electrodes themselves are typically inert, and chemical redox couples undergoing transformations remain in the electrolyte solutions. As a result, the storage of chemicals (and chemical energy) can be spatially decoupled from the electrochemical cell. Electrolyte solutions flow through the electrochemical cell, and species undergoing oxidation/reduction are carried out to storage vessels. The two major advantages of such an arrangement are that there is no deterioration in the performance of the electrodes due to structural degradation, and practically any amount of energy storage can be achieved, limited only by the size of storage tanks for the electrolytes [38]. Redox flow batteries typically have high round-trip efficiency and can be scaled to meet any power or energy demand. Inert electrodes also permit a high depth of discharge without any cycle life limitations [39]. The disadvantages of redox flow batteries stem from the need to store chemicals, which lowers the power and energy density considerably when storage volumes are considered. Furthermore, the chemicals comprising the electrolyte may be corrosive, toxic, and environmentally hazardous and require careful control of chemistry to prevent precipitates or undesirable species.

Several redox flow batteries have been considered for energy storage including all vanadium, vanadium–polyhalide, vanadium–polysulphide, iron–chromium, hydrogen–bromine, and so on. The conceptual functioning of the all-vanadium or vanadium redox flow battery is shown in Figure 6.11.

An illustration of vanadium redox flow battery is shown.

Figure 6.11 Conceptual schematic of vanadium redox flow battery.

The charging reactions for the battery are:

V3+(aq)+eV2+(aq)         (R6.13)

and

VO2+(aq)+H2O(l)VO2+(aq)+2H+(aq)+e         (R6.14)

The overall equation for the process is:

V3+(aq)+VO2+(aq)+H2O(l)V2+(aq)+VO2+(aq)+2H+(aq)         (R6.15)

The forward reaction represents the charging process, and the reverse reaction represents the discharge process. As seen from Figure 6.11, the V2+/V3+ (hypovanadous/vanadous ions) cycling occurs at the negative electrode and the V5+/V4+ (vanadic–VO2+/vanadyl–VO2+ ions) cycling at the positive electrode. The standard reduction potentials for the reactions represented by equations R6.13 and R6.14 (negative and positive electrodes) are −0.26 V and 1.00 V, respectively, resulting in a net voltage of 1.26 V when operating in the galvanic mode [40]. As seen from the earlier equations, electrolytes in both compartments are vanadium salts in various oxidation states. The counter ion in both cases is sulfate. The acidic electrolytes are obtained by dissolving the vanadium salts in sulfuric acid to obtain the concentration of vanadium ions around ~2 M, while the concentration of the sulfate ions is ~5 M. Increasing the vanadium concentrations enhances the energy density, however, there is a limited concentration window where vanadium salts remain in the dissolved state, with the preparation of electrolyte for the positive compartment (vanadyl sulfate) particularly challenging [40]. The electrodes used in vanadium redox batteries (VRBs) are typically graphite and carbon felt, while the ion exchange membrane that permits the transfer of the hydrogen ion, but not of vanadium ionic species, is made of materials such as Nafion 117 [41]. The advantage of the all-vanadium battery over other flow batteries mentioned earlier is that electrolytes in both the compartments contain ions of the same element, and the consequences of potential failure of the separator membrane are less severe. Developments in VRB technology are focused on addressing the need for advanced electrodes that have faster redox kinetics while minimizing parasitic oxygen and hydrogen evolution reactions that adversely affect the efficiency.

VRB energy storage systems have been installed at several locations worldwide, with a typical power rating of tens of megawatts for power quality management as well as peak shaving/load shifting applications.

6.4.1.6 Comparison of Electrochemical Storage Alternatives

Important parameters that dictate the choice of an energy storage system include round-trip efficiency, energy density and capacity, calendar and cycle life, response time, technology readiness level, safety, and capital and operating costs. These parameters incorporate the effects of self-discharge (impact reflected in efficiency and operating costs), depth of discharge (energy density and capacity), performance deterioration over time (cycle life), and so on. As expected, batteries based on lighter elements (Li and Na) typically have higher energy densities than those based on heavier elements (lead, Ni–Cd, or even vanadium). The round-trip efficiency of the Li-ion battery is also very high (>90%) compared to lead–acid or Na–S batteries (~75%). The Na–S battery suffers from high self-discharge, which is not a concern for VRBs [35]. The cycle life of VRB is very high compared to that of other batteries [33]. Thermal needs of the batteries also vary widely, with the Na–S battery requiring heating while other systems need cooling [30]. The cost advantage offered by the lead–acid batteries—installed costs per kilowatt-hour are two to ten times lower than those of other batteries—cannot be matched [35].

It should also be noted that not all parameters are equally important for all applications. The energy and power needs differ vastly both in terms of timescale and total energy needed. The energy storage needs of an RES differ from those of fossil fuel or nuclear power plants, in that in addition to the normal needs of power quality management and load shifting, the energy storage system needs to address the intermittency concerns. Rechargeable batteries are able to store tens of megawatt-hours of electrical energy, as seen from Figure 6.5, and also experience such energy storage installations worldwide. Significantly higher energy capacities are needed for longer-term, bulk power management demands, necessitating a consideration of other storage options.

6.4.2 Chemical Energy Storage

The gravimetric energy density of the rechargeable batteries is quite low, ranging from ~50 Wh/kg to ~250 Wh/kg for Li-ion and Na–S batteries [30]. This has the effect of making the size of electrochemical energy storage quite large and uneconomical for bulk power management due to their modular nature and other limitations. In contrast, the gravimetric energy densities of chemicals are nearly two orders of magnitude higher, with hydrogen having the highest energy density of 33.3 kWh/kg (120 MJ/kg). Methane and liquid hydrocarbons (gasoline, diesel) have energy densities that range from ~11 to 14 kWh/kg, while even the less energy dense alcohols (methanol, ethanol) have energy densities of 6–8 kWh/kg [23]. However, unlike electrochemical energy storage, chemical energy storage requires conversion processes of higher complexity for producing chemicals using electrical energy, as discussed below.

6.4.2.1 Hydrogen

Hydrogen is quite possibly the most attractive chemical for energy storage applications, not only because of its high gravimetric energy density (the highest of all chemicals) but also because it does not contain any carbon, can be converted back to electricity with relative ease using fuel cells, and can be used directly in transportation applications. The only raw material requirement is that of water, which can be split using electricity (available from all renewable energy sources) or heat (from solar thermal power plants).

The overall reaction of electrolysis of water can be represented by the following equations:2

2. The actual half-cell reactions differ from those represented by equations R6.16 and R6.17, depending on the type of electrolyzer used. The overall reaction is represented by equation R6.18 in all cases.

2H++2eH2(g)         (R6.16)

H2O(l or v)2H++2e+12O2(g)         (R6.17)

H2O(l or v)H2(g)+12O2(g)         (R6.18)

The hydrogen evolution reaction, represented by equation R6.16, involves gain of electrons (reduction, cathodic process) and takes place at the negative electrode. The oxygen evolution reaction involves the loss of electrons (oxidation, anodic process) and takes place at the positive electrode. The standard reversible potential (at 298.15 K, 1 atm pressure, and unit activities of species) for the overall water-splitting reaction (equation R6.18, water in liquid state) is −1.23 V, indicating that the reaction is not spontaneous, and an externally imposed potential of at least 1.23 V is necessary to effect the decomposition if the water is present in the liquid state [32]. Furthermore, the standard enthalpy and Gibbs energy changes (∆H° and ∆G°, 205respectively) for the splitting of water (equation R6.18) are 286 and 237 kJ/mol, respectively [42]. This imposes an additional energy (heat) input requirement above the electrical energy input (equivalent to the Gibbs energy change ∆G°) to achieve the total energy input equal to the enthalpy change ∆H°. It is possible to provide the additional energy input by increasing the electrical energy input (impose higher voltage). This higher voltage, which obviates the need for any thermal energy input, is the thermoneutral potential and equal to 1.47 V at standard conditions.

Thermodynamically, the reversible potential for electrolytic splitting of water decreases (becomes more negative) with an increase in pressure and increases (becomes less negative) with rising temperature. As a result, a lower externally imposed voltage is needed for effecting the electrolysis at higher temperatures and lower pressures [43]. The total energy required for the decomposition of water (equal to the reaction enthalpy) always increases with temperature; however, the fraction of the electrical energy input for the process decreases with temperature.

The three distinct technologies used for the electrolysis of water are alkaline electrolysis, SPE electrolysis, and solid–oxide electrolysis [44].

Alkaline electrolyzers are the most common type of electrolyzers currently in commercial operation. The electrolyte consists of concentrated (~30%) KOH, and the electrolysis cell operates at low temperatures (70°C–100°C) and pressures ranging from 1 to 30 atm. The anodic and cathodic half reactions are different from those stated earlier and are given below [45]:

Anodic:2OH(aq)12O2(g)+H2O(l)+2e         (R6.19)

Cathodic:2H2O(l)+2eH2(g)+2OH(aq)         (R6.20)

A separator—an asbestos diaphragm or a membrane—that permits the movement of hydroxide ions across it, but prevents the mixing of the evolved gases, divides the electrochemical cell into anodic and cathodic compartments. The electrodes are typically made of cheaper iron- or nickel-based material. Anode in conventional alkaline electrolyzers is made of nickel, whereas the cathode is made of stainless steel or nickel.

The electrochemical cell in an SPE electrolyzer has a solid, proton-conducting polymer electrolyte, which also functions as the gas separator, sandwiched between the two electrodes as part of a membrane electrode assembly (MEA). The cathodic and anodic reactions are those given by equations R6.16 and R6.17, respectively. Water dissociates into molecular oxygen, protons, and electrons at the anode, and the electrons flow through the external circuit to the cathode, where they combine with the protons transported across the electrolyte to form hydrogen gas. MEA is the core of the electrolyzer and consists of the catalysts deposited on a porous carbon support on either side of the electrolyte. The catalyst for the cathodic reaction (hydrogen evolution) is typically Pt, whereas specialty Ru-/Ir-Ru-based catalysts have been found to be the most effective for the anodic reaction of oxygen evolution [46, 47]. The electrolyte is typically a perfluorosulfonic acid polymer—Nafion—made by DuPont.

Solid oxide electrolyzers are characterized by an oxide ion conducting electrolyte made of a ceramic oxide, typically yttria-stabilized zirconia (YSZ). The anodic and cathodic reactions for a solid oxide electrolysis cell (SOEC) are [45]:

Anode:O2O2(g)+2e         (R6.21)

Cathode: H2O(v)+2eH2(g)+O2         (R6.22)

The electrolysis is conducted at high temperatures (850°C –100°C) and pressures up to 30 atm, and referred to as high-temperature electrolysis (HTE) or high-temperature steam electrolysis (HTSE). As mentioned earlier, the thermodynamics of the water-splitting reaction become favorable at higher temperatures, with the decrease in Gibbs energy resulting in lowering of the reversible voltage requirement to 0.95 V at 900°C. Consequently, the electrical energy requirement is lower [48]. A single SOEC is composed of porous electrodes deposited on either side of the electrolyte. The cathode is typically a nickel–zirconia cermet, and the anode is made of strontium-doped lanthanum manganate [49, 50]. The schematics of the three electrolyzers are shown in Figure 6.12.

Three different electrolyzers are shown.

Figure 6.12 Electrolyzers for water splitting: (a) alkaline cell, (b) solid polymer electrolyte cell, and (c) solid oxide electrolysis cell.

It should be noted that the potentials mentioned earlier are the reversible, that is, equilibrium potentials, where no net change in the system conditions occurs. Significantly higher voltages need to be imposed in order to accomplish the splitting of water at reasonable rates to make the process economically feasible. The voltage requirements for alkaline electrolyzers are the highest, ranging from 1.8 to 2.25 V, whereas those for the HTSE range from 0.95 to 1.3 V. The SPE electrolyzers have intermediate voltage requirements (1.4–2.0 V). Both alkaline and SPE electrolyzers operate at temperatures slightly above ambient (50°C–90°C for alkaline, 80°C–150°C for SPE), whereas the SOECs operate at much higher temperatures 550°C–900°C [51]. The overall energy requirements for the alkaline electrolyzers are the highest, ranging from 4.3 to 4.9 kWh/Nm3 H2, whereas those for SOECs are around 3.5 kWh/Nm3 H2 [50]. SPEs have intermediate energy consumption. Alkaline electrolyzers are technologically simple, robust, based on proven technology, and do not have any special materials requirements. On the other hand, they have relatively low efficiency and higher energy requirements per unit of hydrogen as compared to the other two options. SPE electrolyzers are more efficient than alkaline electrolyzers, operate at nearly the same temperatures, and do not need any corrosive chemicals. However, they need specialized membranes and expensive catalysts for operation. SOECs have the lowest energy consumption and a more robust membrane than in the SPE electrolyzer. However, they operate at significantly higher temperatures than those available in RESs, except for a solar thermal system or biomass combustion systems.3 Furthermore, the materials requirements are substantially more severe than the other two types due to higher temperatures.

3. Technically, it is always possible to use electricity from lower-temperature RESs to heat materials and equipment to any temperature required for HTSE. However, using electricity for heating is expensive and a source of additional inefficiency in the water-splitting process.

Electrolysis is currently the only practical way of splitting water to obtain hydrogen. Theoretically, it is also possible to accomplish the splitting of water using only thermal energy, provided sufficiently high temperatures are achieved. Direct thermolysis of water is a conceptually simple, single-step process involving the decomposition of the water molecule through application of thermal energy:

H2OHeatH2(g)+12O2(g)         (R6.23)

The reaction is reversible, and the equilibrium constant for the reaction is expressed by the following equation:

KP=PH2PO21/2PH2O         (6.1)

The equilibrium constant is related to the Gibbs energy change for the reaction:

KP=exp(ΔGRT)         (6.2)

The equilibrium constant at the standard temperature is 4.2 × 10−76 bar0.5, calculated from the standard Gibbs energy change. This low value clearly indicates that it is impossible to split water at such a low temperature, even if energy equal to standard enthalpy change is supplied through heat. The reaction must be conducted at higher temperatures, if decomposition to any appreciable extent is desired. The percentage of water decomposing is only 0.69% at 2000 K, increasing to ~57% at 3500 K, if the reaction is conducted at atmospheric pressure [52]. The ∆G value goes to zero at ~4200 K, and this is the minimum temperature needed for a spontaneous complete thermolysis of water into hydrogen and oxygen at atmospheric pressure [53].

Direct thermolysis of water faces several significant technical challenges including separation of product gases (hydrogen and oxygen) to prevent recombination back into water and materials development for handling high temperatures [54, 55]. The only energy source capable of providing such high temperatures is solar energy; however, harnessing this energy to achieve such high temperatures is a practically insurmountable problem. Indirect thermolysis of water is a potential solution, wherein the hydrogen evolution and oxygen evolution reactions are spatially or temporally separated, avoiding the problem of recombination back to water. Furthermore, the individual evolution reactions occur at much lower temperatures (compared to 4200 K), making the materials of construction requirements less severe [56].

Indirect thermolysis of water involves a thermochemical cycle consisting of a series of reactions in which all the intermediate species are recycled within the process so that the net reaction is the decomposition of water. A simple two-step cycle will be represented by the following reactions:

A+H2OAX1+X2         (R6.24)

AX1A+X1         (R6.25)

where X1 and X2 represent hydrogen or oxygen. Reagent A is converted in the first step to either an oxide (X1 is oxygen, an oxide reaction sequence) liberating hydrogen or a hydride (X1 is hydrogen, hydride reaction sequence) liberating oxygen. Decomposition of AX1 in the second step regenerates A for reuse in the first step, liberating the second gas (oxygen or hydrogen, respectively) in the process. The two reactions are spatially or temporally separated, circumventing the problem of hydrogen-oxygen separation [57]. It should be noted that the process is cyclic and the first step is taken to be the one involving the reaction of water. Following are examples of typical oxide and hydride reaction sequences [53, 58]:

  • Oxide Reaction Sequence:

    Zn(s)+H2O(v)ZnO(s)+H2(g)         (R6.26)

    ZnO(s)Zn(s)+12O2(g)         (R6.27)

  • Hydride Reaction Sequence:

    Reverse Deacon Reaction:Cl2+H2O2HCl+12O2         (R6.28)

    2HClH2+Cl2         (R6.29)

It is not necessary that the evolution of either hydrogen or oxygen take place in the first step, and the first step in cycles containing three or more steps will involve transferring these elements to separate compounds that then undergo decomposition in subsequent steps to yield hydrogen and oxygen. The sulfur-iodine (SI) cycle (also called iodine-sulfur, IS, or ISPRA Mark 16 cycle) is an example of the three-step cycle [59].

SO2(g)+I2(v)+2H2O(v)H2SO4(v)+2HI(v)         (R6.30)

H2SO4(v)H2O(v)+SO2(g)+12O2(g)         (R6.31)

2HI(v)H2(g)+I2(v)         (R6.32)

Considerable effort has gone into the research and development of thermochemical cycles for hydrogen production, since inefficient thermal energy to electrical energy (needed for electrolysis) conversion can be avoided. Most of this effort is focused on harnessing nuclear heat for producing hydrogen to be used as transportation fuel. As far as RESs are concerned, only the solar thermal energy systems can potentially provide such high-temperature heat.

The greater the number of steps, the greater the process complexity of the thermochemical cycle. However, the maximum temperature requirement for the process typically goes down with increasing the number of reaction steps. The maximum temperatures in a two-step cycle are around 2000 K compared to ~1000 K for a three-step cycle. Advanced nuclear reactors, still in the development stages, can deliver thermal energy at temperatures up to ~1000 K and match well with the maximum temperature requirements of thermochemical cycles with three or more steps. On the other hand, solar energy can be harnessed to achieve much higher temperatures, and it is more attractive to couple with it a two-step thermochemical cycle for hydrogen production [60].

The two-step thermochemical cycle, termed a hybrid chlorine cycle (also the Hallett Air Products cycle), represented by equations R6.28 and R6.29, is the most common of the hydride reaction sequence cycles. The cycle is not a purely thermochemical cycle, as the decomposition of hydrogen chloride (equation R6.29) is accomplished electrolytically—hence the descriptor hybrid chlorine cycle. Equilibrium limitations of the first (reverse Deacon) reaction, separation of species, and the need to use electrical energy in addition to thermal energy contribute to its inefficiency, and oxide reaction sequence thermochemical cycles are generally considered more promising for splitting the water molecule [60, 61].

The oxide reaction sequence involves a high-temperature decomposition of oxide in an endothermic reaction and a lower-temperature exothermic hydrogen generation reaction. Transition metal oxides, such as those of Fe, Zn, Co, Mn, and Ti, are considered the most promising candidates for the oxide reaction sequence cycle, with those involving zinc and iron being more attractive than those based on other metals. The reaction sequence for the Fe-based cycle is:

3FeO(s)+H2O(v)Fe3O4(s)+H2(g)         (R6.33)

Fe3O4s3FeOs+12O2g         (R6.34)

The temperatures needed for the decomposition of metal oxides (equations R6.27 and R6.34) are in the range of 2250°C–2500°C and this step is driven by solar energy. It should be noted that the extent of decomposition depends on temperature and the partial pressure of oxygen liberated in the reaction, and it is possible to lower the reaction temperature by lowering the oxygen partial pressure to achieve the same extent of decomposition. In general, the Zn-based cycle is considered to be more attractive due to superior thermodynamics and higher energy density [62].

Finding suitable materials of construction is practically as much of a challenge for the two-step thermochemical cycles as it is for direct thermolysis. The biggest problem, though, is in accomplishing separation of products (metal and oxygen) of the decomposition reaction to prevent recombination. The recombination reverses the conversion and contributes to the inefficiency of the process. The rate of the reverse reaction can be reduced dramatically by quenching of the product streams; however, this represents a significant energy loss, again contributing to process inefficiency. The reformation of oxide in the hydrogen production step is a noncatalytic gas–solid reaction subject to reacting surface limitation, and a transfer of the solid is involved between the oxide decomposition and hydrogen production reactors [63]. Mixed metal oxide systems, perovskites, and oxides undergoing partial decomposition have received increased attention recently due to the possibility of lowering the decomposition temperatures [62]. However, the thermodynamics of such systems is much more complex, the phase stability is an issue, and substantial research and development are needed before these cycles become serious contenders for thermochemical hydrogen production.

Hydrogen’s utility as an energy storage medium as well as a feedstock for a variety of industrial processes has led to the U.S. DOE’s [email protected] initiative aimed at enhancing U.S. energy security, resiliency, and environmental and economic benefits to facilitate large-scale production, storage, transmission and distribution, and utilization across practically all sectors of the economy. Significant challenges related to all these aspects need to be overcome in order to realize its potential [64]. Until this vision of hydrogen having the prime role in energy systems is realized, other chemical energy storage alternatives will continue to play an important role in energy systems.

Hydrogen is a unique substance for chemical energy storage, for it is the only one that does not contain carbon. The other compounds mentioned above require carbon either with oxygen (alcohols) or without it (methane). Using any of these compounds for energy storage in an RES requires a carbon source, which is available only with biomass systems. For all other systems, an external source of carbon is needed, which is generally carbon dioxide. It is possible for a gas stream enriched in carbon dioxide, such as that from cement or steel plants, to be available in close proximity of the RES; however, in the absence of such a source, it can be harvested from air [65]. Carbon dioxide so harvested can be combined with hydrogen (produced from water through processes described above) to synthesize the alcohols or hydrocarbons for energy storage. While this synthesis can be effected catalytically at moderate temperatures (200°C–300°C) and high pressures (5–10 MPa) without the use of electricity, the focus of the discussion is on the use of such chemicals for energy storage coupled to RESs is on power-to-gas (methane) and power-to-liquid (methanol) processes [66, 67].

6.4.2.2 Methanol

Industrial synthesis of methanol is based on catalytic conversion of syngas (CO:H2 ratio of 1:2) using metal oxides (Zn, Cr, Mn, Al, or more recently Cu) at temperatures and pressures mentioned above [68]. The power-to-liquid route to methanol is based on electrochemical reduction of carbon dioxide to carbon monoxide, which is mixed with electrolytic hydrogen to obtain syngas. This syngas can then be fed into the industrial methanol synthesis process.

The electrolyzer for the low-temperature electrolytic reduction of CO2 comprises two half-cells—cathodic (negative) and anodic (positive)—separated by a cation exchange membrane (CEM, such as Nafion) or an anion exchange membrane (AEM, typically perfluorinated sulfonic acid based, e.g., Fumasep). The process is conducted at ambient temperatures and various pressures ranging from atmospheric to pressures as high as 5 MPa [6971]. The electrolyzers based on CEMs can function under acidic/neutral or alkaline conditions as shown in Figures 6.13(a) and (b), respectively.

Two different environments in CEM electrolyzers are shown.

Figure 6.13 Low-temperature electrolytic reduction of CO2 in CEM electrolyzers: (a) acidic environment and (b) alkaline environment.

The relevant half-cell equations for the process shown in Figure 6.13(a) where a proton is transported across the CEM from the anode to the cathode are:

Anode:H2O(l)12O2(g)+2H++2e         (R6.35)

Cathode:CO2(aq)+2H++2eH2O(l)+CO(g)         (R6.36)

2H++2eH2(g)         (R6.37)

Standard half-cell potentials for the reactions represented by equations R6.35 and R6.36 are −1.23 and −0.20 V, respectively [32]. Equation R6.37 represents the hydrogen evolution reaction for which the standard reduction potential is 0.0 V.

The corresponding half-cell reactions for the case shown in Figure 6.13(b) where the ionic transport involves K+ migrating across the CEM are:

Anodic:2OH(aq)12O2(g)+H2O(l)+2e         (R6.19)

Cathode:3CO2(aq)+H2O(l)+2e2HCO3(aq)+CO(g)         (R6.38)

2CO2(aq)+2H2O(l)+2e2HCO3(aq)+H2(g)         (R6.39)

The anolyte consists of an aqueous KOH solution, while the catholyte is a KHCO3 solution. The standard half-cell (reduction) potentials for the three reactions (equations R6.19, R6.38, and R6.39) are 0.4 V, −1.0 V, and −0.8 V, respectively.

It can be seen that the hydrogen evolution reaction has a lower potential requirement, and hence is thermodynamically favored over the carbon dioxide reduction. This problem is particularly severe under acidic conditions [Figure 6.13(a)], which also requires costly noble metal catalysts in electrodes [71]. Alkaline electrolyzers, as shown in Figure 6.13(b), are more attractive, as use of precious metals for catalyzing the electrode reactions can be avoided, in addition to suppression of hydrogen evolution [72]. Considerable research and development work is ongoing to develop catalysts that have favorable kinetics for carbon dioxide reduction. Zinc–copper oxide mixtures have been shown to hold considerable promise in this regard to obtain the desired CO:H2 ratios in the syngas [73].

The processes and reactions occurring in an AEM-based electrolyzer are different from those in the CEM-based electrolyzers, as shown in Figure 6.14.

A figure depicts the setup of an AEM electrolyzer.

Figure 6.14 Low-temperature electrolytic reduction of CO2 in AEM electrolyzers.

The corresponding reactions are:

Anode:2HCO3(aq)12O2(g)+2CO2(aq)+H2O(l)+2e         (R6.40)

Cathode:3CO2(aq)+H2O(l)+2e2HCO3(aq)+CO(g)         (R6.38)

2CO2(aq)+2H2O(l)+2e2HCO3(aq)+H2(g)         (R6.39)

The cathodic reactions are the same as in the alkaline CEM electrolyzer. This system operates under alkaline conditions as well, with anolyte and catholyte both typically being KOH solutions. As with the alkaline CEM electrolyzers, conditions in the AEM electrolyzers offer the benefits of use of nonprecious metal catalysts, slower hydrogen evolution reaction, and manipulation of electrocatalyst formulations for obtaining the desired CO:H2 ratio in the syngas for the synthesis of chemicals.

A bipolar membrane can also be used for the reduction of carbon dioxide, with the benefit that the reduction reaction at the cathode is almost exclusively that of carbon dioxide reduction. The schematic of the operation of the bipolar membrane is shown in Figure 6.15.

A figure depicts the usage of bipolar membrane.

Figure 6.15 Bipolar membrane electrolyzers for electrolytic reduction of CO2.

Bipolar membranes, such as Fumasep FBM, are at the heart of these electrolyzers. The membrane is sandwiched between the negative (cathode) and the positive (anode) electrodes that are typically made of nickel foam and a carbon gas diffusion electrode containing silver catalyst, respectively [74]. The anolyte and catholyte consisting of dilute (0.1–0.5 M) solutions of potassium hydroxide and potassium bicarbonate, respectively, circulate through the respective compartments. The key reaction in the bipolar membrane is the decomposition of water into hydrogen and hydroxide ions as shown below [74].

H2O(l)H+(aq)+OH(aq)         (R6.41)

The bipolar membrane serves to retain the hydroxide and hydrogen ions on the anodic and cathodic sides, respectively. The reaction catalyzed at the anode is of oxygen evolution, the same as in the case of the alkaline CEM electrolyzer (equation R6.19). The sequence of reactions in the cathodic compartment resulting in the reduction of carbon dioxide to carbon monoxide are shown below [69]:

Membrane:

H+(aq)+HCO3(aq)CO2(aq)+H2O(l)         (R6.42)

Cathode:

2HCO3(aq)+CO2(aq)+2eCO(g)+2CO32(aq)+H2O(l)         (R6.43)

2H+(aq)+CO2(aq)+2eCO(g)+H2O(l)         (R6.44)

H2O(l)+2eCO(g)+2OH         (R6.45)

The overall reaction in the bipolar membrane electrolyzers is simply a carbon dioxide reduction reaction (CRR):

CRR:

CO2(g)COg+12O2g         (R6.46)

If bipolar membrane electrolyzers are used for the reduction of carbon dioxide, then another electrolysis process is needed, producing hydrogen in order to synthesize methanol (or any other alcohol/hydrocarbon). Alternatively, carbon dioxide and water vapor can be subjected to co-electrolysis in a SOEC at high temperatures to synthesize various organic compounds directly for use as fuel. The solid oxide electrolyzers are essentially identical to those shown in Figure 6.12(c). The feed to the cathodic compartment of the electrochemical cell consists of a mixture of water vapor and carbon dioxide. The electrochemical reactions at the two electrodes are shown below [75]:

Anode:O2O2(g)+2e         (R6.21)

Cathode: H2O(v)+2eH2(g)+O2         (R6.22)

CO2(g)COg+O2         (R6.47)

Conceptually, the high temperature co-electrolysis (HTCE) is similar to HTSE and uses solid oxide—yttria- or scandia-stabilized zirconia (YSZ or SSZ)—as the electrolyte. Several different materials have been investigated for use as electrodes in the solid oxide electrolyzers, with Ni-YSZ being the most commonly used cathode. Lanthanum containing perovskite-type oxides—lanthanum–strontium–cobalt–manganese or lanthanum–strontium–cobalt–iron oxides, for example, appear to be the most promising materials for the anode [75, 76]. The electrolysis is conducted at high temperatures 700°C–900°C and an applied potential of ~1.5 V.

Carbon monoxide is formed in the HTCE process through two mechanisms: electrolytic reduction of carbon dioxide, as shown above, and through the reverse water gas shift (RWGS) reaction:

CO(g)+H2O(g)CO2(g)+H2(g)         (R6.48)

The forward reaction in equation R6.48 is the water gas shift (WGS) reaction, used in the hydrogen production processes based on steam reforming of methane and other hydrocarbons for maximizing the yield of hydrogen. In the HTCE process, the reverse reaction—the RWGS—is desired in order to obtain syngas. Thermodynamically, WGS is favored at low temperatures, while RWGS is favored at temperatures >~800°C, which is within the operating range of the solid oxide electrolyzers. Depending upon the materials used and operating conditions, the formation of carbon monoxide may be exclusively due to the RWGS.

The temperature requirements of HTCE make it compatible with only the solar thermal systems, and possibly biomass energy, if used for electricity production in thermal plants. Furthermore, the above processes yield syngas, which is the raw material for methanol production. It is possible to convert carbon dioxide directly to methanol electrocatalytically, with the following cathodic reactions in acidic and alkaline environments, respectively:

CO2(g)+6H+(aq)+6eCH3OH(aq)+H2O(l)         (R6.49)

CO2(g)+5H2O(l)+6eCH3OH(aq)+6OH(aq)         (R6.50)

The standard reduction potentials for the two reactions are −0.38 V and −0.81 V, respectively, making both reactions thermodynamically unfavorable to the hydrogen evolution reaction [77]. The overall process is plagued with low yields and selectivity due to thermodynamic and kinetic limitations, and the indirect reduction route through syngas is preferred for methanol synthesis.

6.4.2.3 Methane

Methane is an attractive alternative for energy storage, as it has the highest hydrogen to carbon ratio of any hydrocarbon, resulting in lower CO2 emissions per unit of energy. Its volumetric energy density is also higher than that of hydrogen. Another significant advantage of methane is that extensive infrastructure already exists in large parts of the world for storage, transmission, and distribution of methane directly to the consumers. It is already used as an energy currency for a variety of services including cooking, heating, and transportation applications. Biomass conversion to methane through fermentation and anaerobic digestion is a well-known technology and, if needed, energy storage via methane can be incorporated with relative ease into a biomass RES.

Such integration is not readily feasible with other RESs, where power-to-gas conversions must be employed for synthesizing methane. This necessitates a renewable hydrogen production process as discussed in Section 6.4.2.1. Renewable carbon necessary for the synthesis needs to be harvested from carbon dioxide as discussed in Section 6.4.2.2. This renewable hydrogen and carbon dioxide can be subjected to low-temperature co-electrolysis to generate methane. The relevant cathodic (reduction) reactions in acidic and alkaline systems are given by equations R6.51 and R6.52, respectively.

CO2(g)+8H+(aq)+8eCH4(g)+2H2O(l)         (R6.51)