Search in book...
Toggle Font Controls
Create new playlist

Name your new playlist

Playlist description (optional)
Sign In

Email address

Password

Forgot Password?

or

Continue with Facebook

Continue with Google
Sign Up

Full Name

Email address

Confirm Email Address

Password

or

Continue with Facebook

Continue with Google

Chapter 9

Liquid Air Energy Storage

Yulong Ding^*

Lige Tong^**

Peikuan Zhang^**

Yongliang Li^*

Jonathan Radcliffe^*

Li Wang^**
^* Birmingham Centre for Cryogenic Energy Storage, School of Chemical Engineering, University of Birmingham, Edgbaston, Birmingham, UK
^** School of Mechanical Engineering, University of Science & Technology Beijing, Beijing, China

Abstract

Liquid air energy storage refers to a technology that uses liquefied air or nitrogen as a storage medium. The chapter first introduces the concept and development history of the technology and then follows it up with thermodynamic analyses. Applications of the technology are then discussed through integration under different scenarios particularly with gas turbine-based peaking plants, nuclear power plants, solar thermal power generation, and liquefied natural gas regasification. Finally, comparisons are made between liquid air energy storage technology and a number of other energy storage technologies both technically and economically.

Keywords

liquid air energy storage

cryogenic energy storage

integration

thermodynamic analyses

economical and technical comparison

1. Introduction

Liquid air energy storage (LAES) refers to a technology that uses liquefied air or nitrogen as a storage medium [1]. LAES belongs to the technological category of cryogenic energy storage. The principle of the technology is illustrated schematically in Fig. 9.1. A typical LAES system operates in three steps. Step 1 is the charging process whereby excess (off-peak and cheap) electrical energy is used to clean, compress, and liquefy air. Step 2 is the storing process through which the liquefied air in Step 1 is stored in an insulated tank at ∼196 °C and approximately ambient pressure. Step 3 is the discharging process that recovers the energy through pumping, reheating, and expanding to regenerate electricity during peak hours when electrical energy is in high demand and expensive. Step 2 also includes the storage of heat from the air compression process in Step 1 and high-grade cold energy during the reheating process in Step 3. The stored heat and cold energy can be used, respectively, in Step 3 and Step 1 to increase the power output and reduce the energy consumption of the liquefaction process.

Figure 9.1 Schematic illustration of liquid air energy storage technology.

The concept of LAES technology was first proposed by researchers at the University of Newcastle-upon-Tyne (United Kingdom) in 1977 for peak shaving of electricity grids [2]. Although the work involved mainly theoretical analyses, it led to subsequent development of the technology, particularly by Mitsubishi Heavy Industries [3] and Hitachi [4–5] in Japan, and by Highview Power Storage in collaboration with the University of Leeds (United Kingdom) [1,6–9]. The work by Mitsubishi Heavy Industries led to a 2.6 MW pilot plant with air liquefaction and power recovery processes operated independently, leading to low roundtrip efficiency [3]. The work by Hitachi looked at integration of the air liquefaction and power recovery processes through a regenerator [4–5]. Such a regenerator stores cold energy released during the power recovery process and reuses the stored cold energy to reduce energy consumption of the air liquefaction process. Both simulation and experiments were carried out on the regenerator using solid materials and fluids as cold carriers. Based on the results, Hitachi claimed that the roundtrip efficiency of the cryogenic energy storage system could exceed 70% so long as the regenerator was efficient. However, no fully integrated system demonstration was done. Working with the University of Leeds, Highview Power Storage started to design and build the world’s first integrated LAES pilot plant (350 kW, 2.5 MW h) in 2009. The pilot plant was colocated with a Scottish and Southern Energy (SSE) biomass power plant in Slough (United Kingdom) and the whole plant became operational from 2011. The pilot plant has now been relocated to the University of Birmingham (United Kingdom) for further research and development. In collaboration with Virido and with the support of the UK Department of Energy and Climate Change, Highview Power Storage is currently building a 5 MW/15 MW h demonstration plant in Virido’s Manchester (United Kingdom) landfill gas power plant.

2. Energy and exergy densities of liquid air

Fig. 9.2 shows the exergy density of liquid air as a function of pressure. For comparison the results for compressed air are also included. In the calculation the ambient pressure and temperature are assumed to be 100 kPa (1.0 bar) and 25 °C, respectively. The exergy density of liquid air is independent of the storage pressure because compressibility of liquid is extremely small. Fig. 9.2 indicates that, although the mass exergy density of liquid air is only 1.5–3 times higher than that of compressed air, the volumetric exergy density of liquid air is at least 10 times that of compressed air if the storage pressure of compressed air is lower than 10 MPa (100 bar). Such a high volumetric exergy density of liquid air makes it highly competitive even compared with current battery technologies [10].

Figure 9.2 Exergy density of liquid air and compression with compressed air.

Assuming the specific heat C_p of a sensible thermal storage material is a constant an increase or a decrease in its temperature by ∆T from the ambient temperature T_a will lead to an amount of thermal energy δQ being charged into or discharged from the material:

$δ Q = C_{p} ∆ T$ $δ Q = C_{p} ∆ T$

(9.1)

Considering a reversibly infinitesimal heat transfer process the exergy change dE of the material can be calculated by:

$d E = d H - T_{a} \cdot d S = d H - T_{a} \cdot \frac{δ Q}{T}$ $d E = d H - T_{a} \cdot d S = d H - T_{a} \cdot \frac{δ Q}{T}$

(9.2)

where dH is enthalpy change; and dS is entropy change. The exergy ∆E stored in the material is therefore obtained by integrating Eq. (9.2) from T_a to (T_a + ∆T):

$∆ E = C_{p} [∆ T - T_{a} \cdot ln (\frac{T_{a} + ∆ T}{T_{a}})]$ $∆ E = C_{p} [∆ T - T_{a} \cdot ln (\frac{T_{a} + ∆ T}{T_{a}})]$

(9.3)

Combining Eqs. (9.1) and (9.3) gives the percentage of the available energy stored in the material η as:

$η = \frac{∆ E}{|∆ Q|} = \frac{∆ T - T_{a} \cdot ln (\frac{T_{a} + ∆ T}{T_{a}})}{|∆ T|}$ $η = \frac{∆ E}{|∆ Q|} = \frac{∆ T - T_{a} \cdot ln (\frac{T_{a} + ∆ T}{T_{a}})}{|∆ T|}$

(9.4)

Eq. (9.4) is illustrated in Fig. 9.3 where the ambient temperature is assumed to be 25 °C. It can be seen from Fig. 9.3 that, for heat storage, only a significant temperature gap can give a reasonable percentage of available energy. For cold storage, however, the available energy grows far quicker with increasing temperature gap, suggesting that cold storage would be a very attractive option. The physics behind this conclusion rests mainly with the role played by entropy, which increases with increasing temperature.

Figure 9.3 Exergy percentage as a function of temperature difference for heat and cold storage.

3. Liquid air as both a storage medium and an efficient working fluid

Currently, low- to medium-grade heat is often recovered by steam cycles with water/steam as the working fluid [11,12]. However, water/steam is not an ideal working fluid for efficient use of low-grade heat due to its high critical temperature of 374 °C compared with the ambient temperature and its extremely high critical pressure of 22.1 MPa (221 bar). It is because of this that a large proportion of heat is consumed to vaporize water during phase change in subcritical or even transcritical cycles. This leads to the loss of a large portion of exergy in heat transfer processes due to temperature glide mismatch between the heat source and the working fluid—the so-called “pinch limitation” [13,14]. The use of liquid air as a storage medium as well as a working fluid in the power recovery step of LAES technology is thermodynamically more efficient than water in terms of recovering low-grade heat, as demonstrated in the next paragraph.

Consider a heat transfer process between a heat source and a working fluid, with the working fluid heated from ambient temperature t_a to t_H = 400 °C, {T_a to T_H = 673 K}, and define normalized heat

\bar{Q}

$\bar{Q}$

as the ratio of heat load at a certain temperature T to total heat exchange amount during the whole process. This gives:

$\bar{Q} (T) = \frac{H (T) - H (T_{a})}{H (T_{H}) - H (T_{a})}$ $\bar{Q} (T) = \frac{H (T) - H (T_{a})}{H (T_{H}) - H (T_{a})}$

(9.5)

where H is enthalpy. Fig. 9.4 shows the results of Eq. (5) for liquid nitrogen (the main component of air) and water. For comparison the results for liquid methane and hydrogen are also included. One can see that, given a working pressure, the specific heat (the slope of the lines) is approximately the same for liquid nitrogen, hydrogen, and methane. However, water exhibits very different behavior. If the working pressure is lower than its critical value the specific heat of water changes greatly due to phase change, leading to ineffective use of the heat source considering that the heat sources are mostly provided by fluids with a constant specific heat (e.g., flue gases or hot air). Although water behaves in a similar manner to methane under supercritical conditions (e.g., the case with pressure of 30 MPa (300 bar) in Fig. 9.4), the high working pressure increases technical difficulties in bringing the process about.

4. Applications of LAES through integration

Like other mechanical-based energy storage technologies, issues such as capital cost, roundtrip efficiency, and annual operating hours remain key challenges in the industrial takeup of LAES technology. Integration of LAES with other processes/systems provides a way to address the challenges. Examples are given in the following subsections.

Figure 9.4 Normalized heat as a function of cold-side working fluid temperature.

4.1. Integration of LAES with Gas Turbine-Based Peaking Plants

Integration of LAES with a gas turbine-based peaking plant provides an opportunity to use the waste heat from the power generation process leading to high peak-shaving capacity and increased overall efficiency [6]. Integration can also capture CO₂ to give dry ice at no additional efficiency penalty. Fig. 9.5 shows the process diagram of the cycle, which works in the following manner: during off-peak hours, excess electricity generated by the base load is used to power an air separation and liquefaction unit (ASU) to produce oxygen and liquid nitrogen while the rest of the system is powered off. The oxygen and liquid nitrogen produced are stored in a pressurized vessel and a cryogenic tank, respectively, for generating power via the high-pressure turbine (HT) and low-pressure turbine (LT), and assisting combustion in the combustor (B) at peak hours. The liquid nitrogen produced also serves as an energy storage medium.

Figure 9.5 Process diagram of an integrated LAES and gas turbine-based peaking power system[6].
ASU, air separation unit; B, combustor; CS, CO₂ separator; G, generator; GT, gas turbine; HE, heat exchanger; HT, high-pressure turbine; LT, low-pressure turbine; P, cryogenic pump; WS, water separator.

At peak hours, natural gas is compressed in compressor (C1) to the working pressure. The working fluid then mixes with the oxygen in B where combustion takes place to give high-temperature and high-pressure flue gas consisting of CO₂ and H₂O. Combustion of the natural gas in an oxygen environment can produce a temperature that is too high for the gas turbine (GT). To control such a temperature, an appropriate amount of helium gas is mixed with the flue gas before entering the GT for power generation through a generator (G). Note that the helium gas is not consumed but circulates in the system; see later. The flue gas containing helium from the GT then goes through a series of heat exchange processes via heat exchanger 1 (HE1), 2 (HE2), and 3 (HE3) to recover the waste heat by passing the heat to a nitrogen stream from the liquid nitrogen storage tank; see later. During the heat recovery processes, steam in the flue gas is removed via a condenser (WS), whereas CO₂ is removed in the form of dry ice through a solidification process in the CO₂ separator (CS)—the triple point of CO₂ is 571.8 kPa (5.718 bar) and 56.6 °C. As a result the flue gas stream after CO₂ removal contains only helium. The helium stream is then cooled down further in HE3 and compressed in compressor (C2) to the working pressure, and finally goes through further heat exchange in HE2 and HE1 before flowing back to the combustor.

The nitrogen stream starts from the cryogenic storage tank where liquid nitrogen is pumped to the working pressure by a cryogenic pump (P). High-pressure nitrogen is then heated in HE3, HE2, and HE1 in series and expands in two stages via, respectively, a high-pressure turbine (HT) and a low-pressure turbine (LT) to generate electricity. HE1 serves as an interheater between the two-stage expansion. After expansion the pure nitrogen can be used to purge the sorbent bed of the ASU dryer.

From the above, it can be seen that the integrated system consists of a closed-loop topping Brayton cycle [6] with He/CO₂/H₂O as the working fluid and an open-loop bottoming nitrogen direct expansion cycle. The topping Brayton cycle can be identified as 4 → 5 → 6 → 8 → 9 → 11 → 12 → 13 → 14 → 15 → 16 → 4, whereas the bottoming cycle is 18 → 20 → 21 → 22 → 23 → 24 → 25 → 26. It is the combination of the two cycles that produces electricity at peak hours. The Brayton cycle uses natural gas, which is burned in the pure oxygen produced by the ASU during off-peak hours. Helium is only used to control the turbine inlet temperature (TIT) and is recirculated. The working fluid of the open cycle, nitrogen, is the actual energy carrier of off-peak electricity. As CO₂ is captured, only water and nitrogen are given off by the process.

The optimal energy storage efficiency of such an integrated system is nearly 70% whereas the CO₂ in the flue gas is fully captured. Economic analyses also show that if the integrated system is used for energy arbitrage and peak power generation both the capital and peak electricity costs are comparable with the natural gas combined cycle (NGCC), which are much lower than the oxy-NGCC if the operation period is relatively short [6]. Note that not only helium but also oxygen could be used as the recirculating fluid and similar conclusions could be obtained.

4.2. Integration of LAES with Concentrated Solar Power Plants

Additional heat sources can enhance the roundtrip efficiency of the LAES system. Such heat sources can come from industrial processes and renewable solar radiation. This subsection explores the use of solar heat in large-scale concentrated solar power (CSP) plants. Fig. 9.6 shows an integrated LAES and CSP hybrid power system [15]. As can be seen, no liquefaction process is included in the system so an external supply of liquid air/nitrogen is needed. This is practicable if there is a large-scale centralized liquefaction plant within a reasonable distance. The system shown in Fig. 9.6 consists of a direct expansion (open cycle) of liquid air/nitrogen from an elevated pressure and a closed-loop Brayton cycle operated at medium to low pressure. The use of the Brayton cycle in place of the conventional Rankine cycle gives a more efficient heat transfer process and a much lower working pressure. The expansion occurs sequentially in three stages (high-, medium-, and low-pressure turbines), and solar heat is used to superheat the working fluid. Simulation results show that such a system provides over 30% more power than the summation of a solar thermal power–only system and an LAES-only powered system.

Figure 9.6 Integration of LAES with a solar power system.
AC, adiabatic compressor; CT, cryogen tank; CP, cryogenic pump; HC, high-temperature carrier tank; HP, high-pressure turbine; HX, heat exchanger; IP, intermediate-pressure turbine; LC, low-temperature carrier tank; LP, low-pressure turbine; MP, intermediate-temperature carrier tank; P, pump; SC, solar collector.

4.3. Integration of LAES with Nuclear Power Plants

To balance demand and supply at off-peak hours, nuclear power plants often have to be downregulated, particularly when the installations exceed base load requirements. Part load operations not only increase the electricity cost but also impose a detrimental effect on the safety and lifetime of nuclear power plants. Integration of nuclear power generation with LAES provides a promising solution to effective time shift of electrical power output. Fig. 9.7 shows a flow sheet of such integration [16]. The integrated system consists of a nuclear power plant subsystem and an LAES subsystem. The nuclear power plant subsystem in the integrated system is similar to the conventional pressurized water power station. The only difference lies in that there are two three-way valves in the secondary loop, which enables the working fluid to feed into either the steam turbine to produce electricity or heat exchanger 4 to superheat high-pressure air in the LAES subsystem. The LAES subsystem consists of an air liquefaction unit in the left part and an energy extraction unit in the right bottom part of Fig. 9.7. The integrated system could have three operational modes depending on end-user demands as described briefly in the following:

• Energy storage mode: during off-peak hours when demand is much lower than the rated power of the power plant, the power plant operates in a traditional way to drive the steam turbine to produce electricity and excess power is used to drive the air liquefaction unit to produce liquid air.

• Energy release mode: at peak hours when demand is higher than the rated power of the plant, the energy extraction unit is turned on to produce additional power.

• Conventional mode: when supply is approximately balanced by demand, both the air liquefaction unit and energy extraction unit are switched off so that nuclear power operates in a conventional way to drive the steam turbine to produce electricity.

Figure 9.7 Integration of LAES with a nuclear power plant.

The air liquefaction subsystem works in a similar way to the simplest Linde–Hampson liquefier except for the use of external cold energy in heat exchanger 6 (Fig. 9.7). It should be noted that in the air liquefaction unit a cryoturbine is used to generate a liquid product instead of a throttling device in a conventional setup. The working fluid expands in a near-isentropic manner in the cryoturbine with both temperature and enthalpy decreasing and hence generates more liquid product while producing additional shaft power.

The cold storage and recovery steps act as a bridge between the air liquefaction unit and the cryogenic energy extraction unit. Such an arrangement enables recovery of the cold energy released in the liquid air preheating process. In this process, air is in a supercritical state, and as a result cold energy is produced in the form of sensible thermal energy. Fig. 9.8 shows the heat capacity of air as a function of temperature at different pressures. One can see that the heat capacity of air changes only slightly in the heating process, particularly at very high pressures. This is similar to the use of liquids as sensible heat storage materials. In fact, cold energy can also be stored in thermal fluids and such fluids can give a good temperature gradient match during heat exchange and, hence, efficient cold recovery. In this process, thermal fluids are used not only as a working fluid but also as a cold storage medium. Fig. 9.9 shows the heat capacities of some commonly used fluids that may be used as storage media. Clearly, no single fluid can fully cover the working temperature region of the liquid air preheating process. However, the combination of propane and methanol could work both as cold storage liquids and working fluids for heat transfer. Such a combination covers the required temperature range and has high heat capacity. For each of the two fluids a two-tank configuration is proposed for cold recovery and storage (Fig. 9.7). The two thermal fluids are pumped from the hot tanks to the cold tanks during the cold storage process (energy storage mode) and flow back during the cold release process (energy release mode). The use of thermal fluids for both transferring and storing thermal energy can greatly simplify the design of the system in that no additional heat exchangers will be needed. Moreover, the operating strategy can be much more straightforward—the amount of cold energy and the objective temperature can easily be adjusted by controlling the flow rate of the fluids. This is extremely difficult to achieve using the conventional way of storing cold in a packed pebble bed.

Figure 9.8 Heat capacity of air at different pressures.

Figure 9.9 Heat capacity of different cold storage fluids.

The cryogenic energy extraction unit is coupled with the nuclear power plant through the thermal energy utilization process via heat exchanger 4. One can see that hardly any thermal energy is wasted in the cooling process and hence the power output is expected to increase significantly.

By integrating with LAES technology the reactor core and the primary loop of nuclear power plants can operate steadily at full load at all times while net output power is adjusted only by the LAES unit. As the energy extraction process in the LAES subsystem is similar to power generation using a gas turbine a much faster rate of power change could be achieved in comparison with the conventional downregulation of nuclear power plants.

The combination of nuclear power generation and LAES enables the use of cryogen instead of steam as the working fluid in power generation process. As a result this provides an efficient way to use the thermal energy of nuclear power plants, delivering around three times the rated electrical power of the nuclear power plant at peak hours, effectively shaving the peak. Simulations have been carried out on this process, which show that the roundtrip efficiency of LAES is higher than 70% due to the elevated topping temperature in the superheating process.

4.4. Integration of LAES with Liquefied Natural Gas Regasification Process

LAES can also be integrated with liquefied natural gas (LNG) regasification plants to make use of waste cold in the air liquefaction process [16]. The waste cold in LNG import terminals is significant due to large-volume LNG storage. LNG is normally regasified by heating with seawater and burning some natural gas. This leads to wasting of cold contained in the LNG and the burned natural gas. If LAES were colocated at the LNG terminal, and air rather than seawater was used to provide heat for the LNG regasification process, the resulting cold air could then be fed into the air liquefier, potentially reducing its electricity consumption by as much as two-thirds. Currently, there are a number of nitrogen liquefiers in operation at LNG import terminals in Japan and South Korea, which take advantage of this refrigeration to reduce power consumption. The only challenge to be overcome is to reduce the capital cost of such an integrated system.

5. Technical and economic comparison of LAES with other energy storage technologies

In this section a brief comparison is made between LAES and other energy storage technologies. The comparison will be from both technical and economic aspects as detailed in the following two subsections.

5.1. Technical Comparison

Only pumped hydro storage can be currently regarded as a mature technology—it has been practiced for over 100 years. Although compressed air energy storage technology has been developed and is commercially available, actual applications are not widespread. LAES—together with flow batteries, hydrogen storage, and a number of other energy storage technologies [10]—is still under development.

Like pumped hydro and compressed air energy storage technologies, LAES offer a long discharge time (hours) compared with coupled energy storage technologies. However, the power discharge rate depends on the scalability of the power-regenerating unit of energy storage technologies. Pumped hydro storage uses hydraulic turbines for power regeneration and as a result offers the largest power discharge rate (up to several gigawatts). Compressed air energy storage uses traditional gas turbines or steam turbines for power regeneration so the power rate is of the order of hundreds of megawatts. An LAES turbine is somewhat like a gas turbine but with a lower expansion temperature so the power rate is expected to be a little lower than compressed air energy storage, but can still reach hundreds of megawatts. However, the scalability of flow batteries and hydrogen storage is a big challenge and hence their power rates are expected to be less than a megawatt.

Flow batteries and pumped hydro storage have a high (system-level) roundtrip efficiency of (65–85)%. The roundtrip efficiency of compressed air energy storage ranges from about 40 (commercialized and realized) to about 70% (still at the theoretical stage). LAES has a low roundtrip efficiency of about (50–60)% mainly due to the low efficiency of the air liquefaction process. However, it should be noted that the roundtrip efficiency of LAES can be significantly enhanced if waste heat is available.

In terms of energy density, hydrogen storage has the highest volumetric energy density of (500–3000) W h L^–1 depending on the storage methods (e.g., compressed gas, liquid, physical/chemical adsorption etc.). As an extremely flammable gas, however, the technical requirements for hydrogen storage are high. The energy storage density of LAES is an order of magnitude lower at (60–120) W h L^–1, but the energy carrier can be stored at ambient pressure. Pumped hydro storage has the lowest energy density of (0.5–1.5) W h L^–1 while compressed air energy storage and flow batteries are at (3–6) W h L^–1.

5.2. Economic Comparison

Economic comparison can be based on the costs per unit amount of power that storage can deliver (dollars per kilowatt) and costs per unit amount of energy (dollars per kilowatt-hour) that is stored in the system. It is difficult to evaluate a specific technology since the costs are influenced by many factors such as system size, location, local labor rate, market variability, local climate, environmental considerations, and transport/access issues. The capital costs provided in this section are intended to provide a high-level understanding of the issues and are not intended as cost inputs into a design.

The capital costs per unit amount of power relate mainly and directly to the cost of charging/discharging devices. Hydrogen storage is characterized by high capital costs for power (>$10 000 kW^–1). Pumped hydro storage, compressed air energy storage and flow batteries, and LAES have more or less the same capital cost for power (about $400–2000 kW⁻¹). The capital costs per unit amount of energy cannot be used accurately to assess the economic performance of energy storage technologies mainly because of the effect of discharging durations. An alternative measure is the capital cost of storage devices such as a dam for pumped hydro storage and a storage tank for LAES. It is expected that hydrogen storage and compressed air energy storage have the highest storage costs, as the energy carriers are either combustible or at a high pressure. Pumped hydro storage has a low cost due to low energy density. LAES and flow batteries have the lowest cost even though insulation is required.

In terms of lifecycles, mechanical-based technologies including pumped hydro storage, compressed air energy storage, and LAES should last (20–60) years, as these technologies are based on conventional mechanical engineering and the lifecycle is mainly determined by the lifetime of mechanical components. By contrast, the lifetimes of hydrogen storage and flow batteries are currently expected to be about 5–15 years.

..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.

Table of Contents for Chapter 9: Liquid Air Energy Storage

Create new playlist

Sign In

Sign Up