REVIEW pubs.acs.org/IECR
Solar Energy Storage Methods Yu Hou,†,‡,§ Ruxandra Vidu,†,§ and Pieter Stroeve*,†,§ † §
Department of Chemical Engineering and Materials Science, ‡Department of Mechanical and Aerospace Engineering, and California Solar Energy Collaborative, University of CaliforniaDavis, Davis, California 95616, United States ABSTRACT: Solar energy can provide an abundant source of renewable energy (electrical and thermal). However, because of its unsteady nature, the storage of solar energy will become critical when a significant portion of the total energy will be provided by solar energy. In this paper, current solar energy storage technologies are reviewed. Storage methods can be classified into categories according to capacity and discharge time. New developments in solar energy storage require advances in chemical engineering and materials science. Life cycle assessment (LCA) is an important tool to evaluate energy consumption and environmental impact of renewable energy processes. LCAs of some of the storage methods are reviewed. It is important to note that, while using renewable energy sources such as solar power, storage methods based on nonrecyclable materials or methods that consume significant amounts of energy may undermine the effort to reduce energy consumption.
1. INTRODUCTION In order to protect the environment and provide energy security, energy generated from renewable sources has been extensively studied. Among those, solar energy has many advantages such as availability and lower cost. Electricity generated by photovoltaic has grown significantly in the past years. In developed countries, a total installed capacity of 7.8 GW was reached in year 2007, which is a 40% increase from 2006.1 With the advancement of solar technology, a growing trend of solar energy usage can be predicted. There is one major obstacle that exists, not only for solar energy but also for other renewable energy sources such as wind and ocean wave power. Energy generated during the off-peak hours must be stored to meet peak-hour demand. Therefore, a planner only needs to design the generating capacity at the average load level instead of the peak hour load level, as shown in Figure 1. During a typical day, the load is the lowest around 4 a.m., when there is not much activity in both industry and home. The first and highest peak comes slightly after 12:00 p.m. for industry and later, the second load peak comes around 9:00 p.m. for homes. This applies for energy generated from both renewable and nonrenewable sources. For solar energy, electricity generation is unsteady. Generation of electricity fluctuates for many different reasons.2 Seasonal conditions, weather conditions, and locations cause the energy generation process to vary around the year, day, or even within the hour. In 2007, solar power plants produced 0.22% of the electricity for the state of California,3 as shown in Figure 2. For solar energy, 99% are generated within the state of California. By 2020, the state of California expects to have 33% of its utilities generated from various renewable energy sources.4 Without proper electricity storage, this goal will be difficult to reach. Storage methods can be classified in different ways. It is important to review all the different methods and consider their advantages and disadvantages. The decisionmakers then can pick the proper one or combinations of storage methods to achieve their goals. In addition, life cycle assessment (LCA) techniques can be used to assess the total energy cost and environmental impact for each of the storage methods. Information for storage r 2011 American Chemical Society
methods from construction to operation and to decommissioning will be needed to perform the LCA study. In this review, the main methods of storing solar energy are reviewed. An assessment of the most recent developments for different storage methods is given. Many key parameters for each method are also presented in order to make specific comparisons between the methods of storage. LCA is discussed as a tool to determine the cost and environmental impacts.
2. CLASSIFICATION OF STORAGE METHODS There are many different ways to classify various storage methods. Energy harvested from solar energy consists of two forms: electrical energy from photovoltaic (PV) panels and thermal energy from concentrated solar power. Therefore, in this paper, only the storage methods of electrical energy and thermal energy are reviewed. Since electricity supply is of primary concern, the focus of this review is mainly on electricity storage. 2.1. Electricity Storage. The methods to store electricity generated by solar power can be classified by the form of storage2 or usage5 into the categories shown in Table 1. 2.2. Thermal Energy Storage. Thermal energy is typically stored in a thermal reservoir for later usage. Thermal energy storage can also be classified according to usage. Thermal energy harvested from a solar source can be stored via thermal physical reaction (i.e., using the temperature difference of materials (or phase changes) to store energy). It can also be stored via chemical reaction (i.e., creating new chemical species (solar fuels)). These methods are shown in Figure 3 and are described as follows: (1) Thermal Energy Storage: Thermal energy can be stored directly. In sensible heat storage, such as steam or hot water, by changing the temperature of materials (liquid or solid) during peak-hour energy, the energy is stored by temperature difference of the material. In latent heat Received: February 18, 2011 Accepted: June 9, 2011 Revised: June 9, 2011 Published: June 09, 2011 8954
dx.doi.org/10.1021/ie2003413 | Ind. Eng. Chem. Res. 2011, 50, 8954–8964
Industrial & Engineering Chemistry Research storage, such as phase change materials, by changing the phase of materials (liquid or solid) during peak hour energy, the energy is stored in the form of latent heat.2,6 (2) Sorption Storage: In sorption storage, two chemicals, which are bonded together under standard conditions, are separated using peak-hour energy. Energy is released when the two chemicals are mixed and exposed to standard conditions.7 The choice of materials has great impact on the performance of the storage system. (3) Chemical Energy Storage: Heat generated from concentrated solar power is used to carry out endothermic chemical transformation and produce storable and transportable fuel. Examples are solar hydrogen, solar metal, and the solar chemical heat pipe.2
3. ELECTRICAL ENERGY STORAGE METHODS 3.1. Pumped Hydroelectric Storage (PHS). Pumped hydroelectric storage (PHS) has the largest storage capacity that is commercially available. The basic idea is simple: use the excess
Figure 1. Energy load profile within 24 h (adopted from ref 2).
REVIEW
electrical energy generated at off-peak hours to pump water from a lower reservoir to a higher reservoir. The electrical energy is converted to gravitational potential energy. Since the electrical energy will be supplied from the grid, it can be generated from not only photovoltaic (PV), but also from other kinds of renewable energy sources. When the peak hour comes, the water then will be discharged from the higher reservoir to the lower reservoir. The potential energy of water is coverted to electrical energy as normal hydroelectric power plants do. Typically, a turbine will be used to generate electricity. The typical power rating (i.e., the maximum power output) for a PHS system is typically ∼1000 MW.2 There are some small-scale PHS systems as well. These often have a capacity range from 1 MW to 30 MW.8 The storage capacity of the PHS will depend on the size of the reservoirs and the elevation difference between them. Figure 4 shows a diagram of a pumped hydroelectric storage system. The energy-related cost for pumped hydroelectric storage system is ∼10 $/kWh, and the discharge efficiency is ∼87%.9 It should be noted that the 10 $/kWh is the incremental cost, which is the cost needed to generate one additional unit (kWh) of electricity. The total cost of the system will range from 1100 $/kWh to as high as 2000 $/kWh.10 Energy can be stored in the PHS system for a long period of time. However, the system can be charged and discharged multiple times a day and the discharge time requires just a few minutes.10 The drawback of PHS is that suitable sites are limited and many have already been used. Currently, there are ∼150 operational PHS systems in the United States, with a total capacity of over 25 GW.11 Some of the larger facilities are listed in Table 2. For future construction, the sites will likely to be far away from major load center and require new transmission lines.10 The building of the infrastructures may take years and be very costly. The infrastructures for the PHS system are expected to last for ∼60 years.12 Using LCA, the average life cycle energy of nine different PHS facilities in United States for each MWh storage capacity is 373 GJ.12 This includes all the energy required from construction to operation and maintenance, and then decommissioned over the lifetime of the
Figure 2. California’s electricity mix in 2007. 8955
dx.doi.org/10.1021/ie2003413 |Ind. Eng. Chem. Res. 2011, 50, 8954–8964
Industrial & Engineering Chemistry Research
REVIEW
Table 1. Classification of Electrical Energy Storage Classification According to Storage2 electrical energy storage
direct electricity storage in devices such as capacitors or superconducting magnetic devices; those storage methods have the advantage of quickly discharging the energy stored
mechanical energy storage
storage of electrical energy in the form of kinetic energy such as flywheel or
chemical energy storage
storage in chemical energy form as in batteries, fuel cells, and flow batteries;
potential energy such as pumped hydroelectric storage (PHS) or compressed air energy storage (CAES) chemical energy storage usually has small losses during storage Classification According to Usage5 bulk energy storage
bulk energy storage has discharge power range of 101000 MW, discharge times are 18 h, and
distributed generation
the stored energy range of 108000 MWh; the applications of such storage are in load leveling and spinning reserve distributed generation storage has discharge power range of 1002000 kW, discharge time range of 0.54 h, and
power quality
power quality storage has discharge time range of 0.12 MW, the discharge time is 130 s, and
the stored energy range is 508000 kWh; the application of such storage is in peak shaving and transmission the stored energy range is 0.02816.67 kWh. The applications are end-use power quality and reliability
Table 2. Large Pumped Hydroelectric Storage (PHS) Facilities in the United States location
Figure 3. Conceptual diagram of thermal energy storage.
facility
Virginia
Bath County
2772
New Jersey
Mt. Hope
2000
California Michigan
Castaic Dam Ludington
1566 1872
California
Pyramid Lake
1495
becomes a generator that produces electricity. The system is usually kept in a vacuum containment at pressures of ∼106108 atm.2 The energy storage capacity depends on the speed, the mass of the spinning object, and the size of the flywheel: E¼
Figure 4. Diagram of a pumped storage facility.
PHS system. Details of the LCA study are discussed in a later section. PHS is a mature technology; there is little space for conceptual improvement. However, there are many possibilities for improvement in construction and operation to reduce cost and improve efficiency. Commissioned in 1985, with a construction cost of $1.6 billion (US), the PHS system in Bath County, Virginia is the largest PHS system in the United States.13 In the state of California, Pyramid Lake and Castaic Lake acts as the high and low reservoirs for a PHS system, respectively.14 Commissioned in 1973, as part of the California State Water Project, the 1495 MW storage system is the largest PHS in California. The state of Hawaii operates some small-scaled PHS systems. A 160-MW, 8-h storage system at Koko crater and Keau crater on the island of Oahu and a 30-MW, 6-h storage system on the island of Maui are among them. 3.2. Flywheels. A flywheel energy storage system stores energy in the form of angular momentum. During peak time, energy is used to spin a mass via a motor. At discharge, the motor
capacity (MW)
1 mV 2 2
where E is the energy stored, and m and V are the mass and velocity of the spinning object, respectively. Some flywheels can store 25 kWh of energy with a power rating of 100 kW.15 Flywheel energy storage systems can be categorized into highspeed flywheels and low-speed flywheels. The high-speed flywheel is mostly made of a high-strength composite material and the low-speed flywheel is mostly made of metals. The high-speed flywheel has higher tensile strength and is more durable than the low-speed flywheel; however, it is also more expensive to fabricate. A high-speed flywheel can cost $25 000 for each kWh of energy stored, while a low-speed flywheel only costs ∼$300 per kWh.9 This system can also be categorized as mechanical bearings and magnetic bearing. For magnetic bearing systems, spinning speeds of 20 00050 000 rpm can be achieved.16 The efficiency for the system can be at 90%95% in short amounts of time.2 The system can go through 5 000 000 recharge/discharge cycles.17 The major limitation is the storage duration. The system is generally only good for short-time bridging storage, as defined in the classification chapter. 3.3. Compressed Air Energy Storage (CAES). The basic idea of a compressed air energy storage (CAES) system is to use the off-peak excess electricity to compress air. At a later time, the 8956
dx.doi.org/10.1021/ie2003413 |Ind. Eng. Chem. Res. 2011, 50, 8954–8964
Industrial & Engineering Chemistry Research
Figure 5. Compressed air energy storage (CAES) system.
compressed air can be used along with a gas turbine to generate electricity. As shown in Figure 5, electricity from the grid can be used to compress air into an air storage facility. Ideal air storage facilities are typically underground caverns, usually salt, mined hard rock, or limestone caverns. Beside the PHS system, the CAES system is the only other large capacity system that is commercially available. During off peak hours, excess electricity can be used to run an air compressor to compress air into underground caverns. During discharge, the compressed air is released and a conventional gas turbine is used to generate electricity. The pressure for the compressed air is typically 4.08.0 MPa.2 The storage capacity of CAES system depends on the size of the underground cavern. A cavern size of 700 000 m3 corresponds to a capacity of 1500 MWh.18 The power rating of a CAES system ranges from 50 to 300 MW and the system is expected to charge and discharge on a daily basis.2 The storage efficiency of a CAES system is typically ∼70% and the incremental cost is 3 $/kWh.9 The typical lifetime of a CAES system is ∼40 years.19 The life cycle energy per MWh of storage capacity is ∼270 GJ.12 The drawback of a CAES system is similar to the PHS system: a unique geographic location is required for the system. In addition, fossil fuel is often used for the gas turbine in a combined cycle power plant in a heat recovery steam generator (HRSG), because it combines the Brayton cycle of the gas turbine with the Rankine cycle of the HRSG. Thus, during operation, greenhouse gases (GHGs) are emitted. A hybrid energy storage system based on CAES and a supercapacitor has been studied to improve the efficiency of the system.19 Performance analysis and optimal operation strategies have also been investigated.2022 3.4. Batteries, Flow Batteries, and Fuel Cells. A battery is an electrochemical cell that converts stored chemical energy to electrical energy. Rechargeable batteries can have their chemical reaction reversed by supplying electrical energy to the cell; therefore, they can be used to store electricity generated by solar panels. Battery performance is evaluated by the following characteristics: energy and power capacity, efficiency, life span, operating temperature, depth of discharge, self-discharge (loss during storage), and energy density. Operation life cycle depends heavily on the depth of discharge and operating temperature. Discharging completely and operating at higher than ambient temperature will reduce battery life.23 Voltage and current level were not evaluated because desired voltage and current levels can be achieved by connecting cells in series and parallel. Significant developments happened recently in battery technology. Deep cycle batteries with energy capacities in the range of 1740 MWh and efficiencies of ∼70%80% are currently used in power system applications.24 Rechargeable battery types are summarized in Table 3. The lead-acid battery is the oldest and most mature type. Li-ion, NaS, and NiCd batteries are leading
REVIEW
technologies in high-power-density battery applications. The Li-ion battery possesses the greatest potential for future development and optimization, it is most suitable for portable devices, but it has high cost and short lifetime. The NiCd and lead-acid batteries supply excellent pulsed power, which means they are capable of supplying high power capacity in a short duration, but they are large, contain toxic heavy metals, and suffer severe selfdischarge. The NaS battery is smaller and lighter than the NiCd battery, but a constant heat input is required for operation at 300 °C. The metal-air battery has low cost and high energy density, but it is very difficult to recharge.23 A flow battery has two electrolytes that are stored separately and pumped through an electrochemical cell. It is good for longduration storage, because of its negligible self-discharge. However, the efficiency is reduced, because of the energy used for pumping and to the energy loss in chemical reactions. Its power density is limited by the rates of the electrode reactions and, therefore, the main challenge for future development is to increase the power density by finding better electrode materials.24 Based on the time scale, battery energy storage (BES) application in power system can be classified into four types, as follows: (i) instantaneous applications (from zero to a few seconds), (ii) short-term applications (from a few seconds to minutes), (iii) midterm applications (from minutes to a few hours (
