Energy Storage: A Critical Link In the Renewable Energy Chain

An issue that has always grabbed my attention is the critical role I and others foresee for energy storage in the eventual widespread use of variable (intermittent) renewable energy sources such as wind and solar. In fact it was the focus of my first decision when I assumed responsibility for DOE’s renewable electricity programs in 1994. That decision was to establish a comprehensive storage program to complement the established generation programs – up until that point the only storage program was a small effort on underground hot water storage at a university in South Carolina (no doubt related to the fact that the Chairman of the relevant budget authorization subcommittee was from South Carolina). The new program, in addition to thermal storage, added battery storage and superconducting magnetic energy storage (SMES) – superconductivity was another of the programs I managed.

Energy storage is one of two critical renewable energy issues that I have always said I would ‘fall on my sword for’. The second is the need for a national smart grid that will allow renewable electricity generated in one part of the country to be shared with other parts. I have touched briefly on the energy storage topic in earlier blog posts; this post takes a much more detailed look at various storage options.

The need for storage to steady the output from a variable energy source is not new. In fact, in December 1861 the following words and illustration appeared in an agrarian newsletter:
A Mighty Wind One of the great forces nature furnished to man without any expense, and in limitless abundance, is the power of the wind. Many efforts have been made to obtain a steady power from the wind by storing the surplus from when the wind is strong. One of the latest and simplest of these is illustrated in the accompanying engraving. A windwheel is employed to raise a quantity of iron balls, and then these balls are allowed to fall one by one into buckets upon one side of a wheel, causing the wheel to rotate, and thus to drive the machine.”


If one substitutes water for the iron balls and attaches a generator to the rotating machine you have today’s system of pumped water storage and generation. A modern version of the 1861 system, using gravel instead of iron balls, is shown in the following sketch:


Since the discovery of electricity generation using rotating coils of wire in magnetic fields by the British scientist Michael Faraday in 1820, people have sought ways to store that energy for use on demand. Without such storage, or use in some other way (e.g., to electrolyze water to create and store hydrogen, heat water, bricks or phase change materials that store heat , or refrigerate water to create ice) surplus electricity generation is lost. With modern societies increasingly dependent on energy services provided via use of electricity, the need for electricity storage technologies has become critical. This is especially true as more and more variable renewable energy enters the grid, to avoid grid destabilization. This can occur because electric power supply systems must balance supply and demand, and because demand is highly variable and hard to control the balancing is routinely achieved by controlling the output of power plant generators. If these generators are variable solar and wind, and their grid contribution becomes significant, achieving the balance is that much more difficult, and a means of stabilizing these variable outputs is needed.

There is also strong economic and social incentive for storing electricity in a localized, distributed manner. Today’s 100-year-old centralized utility business model, in which large central power plants deliver electricity to customers via transmission and distribution lines, includes the imposition of peak demand charges that can account for a significant fraction of a business’ or an individuals’ electricity bill. With the use of localized generation (e.g., PV panels on your roof), combined with storage at your site, these demand and peak charges can be reduced if not eliminated, and independence from the utility, to some degree, can be achieved. This reality is taking place in Germany (and coming to the U.S.) and threatening the utility business model in Germany to the extent that German utilities have gone into the solar-energy storage business. They now sell or lease or maintain roof-mounted PV and battery storage systems.

Today’s menu of devices that allow storage of surplus electricity for use at other times includes: solid state batteries and supercapacitors, flow batteries, flywheels, compressed air energy storage (CAES), and pumped hydropower. Hydrogen generated from any electricity source via electrolysis of water, and combusted or used in fuel cells, is, in many ways, the ultimate storage technology for surplus electricity. Flywheels, pumped storage, and fuel cells are discussed in earlier blog posts ; other storage technologies are discussed below.

Historically, electricity has been stored in lead-acid batteries, and this is still the dominant battery storage technology today in cars and elsewhere because of low cost, high power density, and high reliability. Disadvantages are low specific energy storage capacity, large size, high weight, and the need for an acid electrolyte. Lead is also a toxic material. Research to improve batteries has been underway for more than a century, and considerable progress has been made (e.g., improved lead-acid batteries that require no maintenance and recycling of used batteries to recover the lead), with considerable promise for further developments in the future.


Most battery attention today is focused on lithium-ion batteries where cost and safety are prime concerns. Research into post lithium-ion batteries is also being actively pursued.
Lithium-ion batteries are widely used today because! “pound for pound they’re some of the most energetic rechargeable batteries available.” For example, it takes six kilograms of a lead-acid battery to store the same energy as one kilogram of a lithium-ion battery. Lithium-ion batteries (there is a variety of battery chemistries) also hold their charge well (losing about 5% per month), have no memory effect (therefore no need to fully discharge before recharging), can handle many hundreds of charge/discharge cycles, and have good ’round trip efficiency’. The story does have a negative side – lithium-ion batteries are sensitive to heat, can’t be fully discharged (thus requiring a computerized battery management system), are still costly (although costs are coming down), and certain chemical formulations can occasionally burst into flame if damaged or otherwise overstressed. One person making a big bet on lithium-ion batteries is Elon Musk, who has announced plans for a $5 billion battery factory, to provide lithium-ion batteries for his Tesla electric vehicles and other applications. Through such large scale production Musk hopes to reduce the cost of the batteries by 30 percent (to about $10,000 for a 60 kWh battery pack).

Supercapacitors store energy in electric fields and fill a gap between ordinary capacitors and rechargeable batteries. Their claim to fame is that they can be charged/discharged much more rapidly than batteries and can tolerate many more charge/discharge cycles. They are widely used as low current power sources for computer memories and in cars, buses, trains, cranes and elevators, including energy recovery from braking.

Redox (reduction/oxidation) flow batteries are large scale rechargeable energy storage systems that are on the verge of wide application in the electric utility sector. They are particularly well suited to storing large amounts of energy, e.g., the surplus energy created by hours of solar or wind power generation. The energy storage materials are liquids that are stored in separate tanks, and when energy is needed the liquids are pumped through a ‘stack’ where they interact to generate electricity. Many different chemical liquids have been tested for flow battery operation, with most current attention being focused on vanadium compounds which are expensive. Flow batteries also have relatively low round-trip efficiencies and response times. Because of the vanadium cost concern many other chemical possibilities are being evaluated.


CAES (compressed air energy storage) utilizes surplus electricity to compress air to high pressures in large caverns, which can then be heated and released as needed to power expansion turbines that generate electricity. Such a CAES system has been operating successfully in Alabama since 1991, and gases other than air (e.g., carbon dioxide) can be used as well.


SMES stores energy in the magnetic field of a circulating dc electrical current in a superconducting coil. The superconductor has no electrical resistance and the current continues indefinitely unless its energy is tapped by discharging the coil. A typical SMES device has two parts, a cryogenic cooler that cools the superconducting wire below its transition temperature at which it loses its electrical resistance, and power conditioning circuitry that allows for charging and discharging of the coil. Its advantages are ultra fast charge and discharge times, no moving parts, nearly unlimited cycling capability, and an energy recovery rate close to 100 percent. Disadvantages are cost of the wire, the need for continuous cooling, large area coils needed for appreciable energy storage, and the possibility of a sudden, large energy release if the wire’s superconducting state is lost. SMES devices are often used to provide grid stability in distribution systems and for power quality at manufacturing plants requiring ultra-clean power (e.g., microchip production lines). One MWh SMES units are now common and a twenty MWh engineering test model is being evaluated.


To summarize, there are many energy storage options that work and tradeoffs are often required – e.g., among storage capacity, power capacity, round-trip efficiency, and most importantly cost. Lots of research is underway to reduce costs, given the large potential markets and the need to safely integrate variable renewable energy generation from solar and wind into the utility grid system. I have no doubt that cost-effective storage systems will soon be available, facilitating the needed rapid transition to a renewable electricity future.

Peter Varadi

It is an excellent and timely review. In my opinion DOE should not spend any more money on PV research, as private PV companies have more money to spend on research than DOE. DOE should spend all the money on research and subsidies of electrical energy storage. The German government is started to subsidize storage.

Appreciate the kind words. Re your push for more DOE support for energy storage: couldn’t agree more. Some DOE support for innovative PV technologies that the private sector will find hard to support is, in my opinion, still justified.