|Fig. 1: (1) Compressed air energy storage schematic. (b) Pumped hydroelectricity storage schematic.|
While all the focus these days is directed towards the search for viable alternative sources of energy, an equally significant issue is finding a reliable means to store the massive amounts of energy generated by these sources. This nontrivial problem is in fact partly a result of the basic ideas behind some of these alternative sources. For instance, while wind farms are an excellent manifestation of natural, renewable, and non-wasteful energy generators, they are also sporadic in their production: manufacturing a plethora of useable electrical energy on windy days, but becoming motionless and fruitless on other days. Since it is often difficult to predict when the next large wind current will flow through, it would be ideal to possess the ability to efficiently store a prolific day's yield for use on quieter days.
Unfortunately, this same concept applies to many other alternative energy ideas including solar power (night/day, cloudy/clear skies) and tidal power (high/low tide velocities). When these sources inevitably become more prevalent in the future, the combination of production unpredictability and lack of mass storage will result in energy waste, offsetting any potential benefits gained. Therefore it is of the utmost importance to research and develop effective means for large scale energy storage.
Currently, with crude oil as the primary source of energy, the most effective and extensively used method of storage is chemical bonds. This includes the use of well-known petroleum products such as gasoline, natural gas, diesel fuel, and liquefied petroleum gas. The conversion of crude oil into the useable forms listed is typically quite straightforward and is performed at one of hundreds of oil refineries operating around the world. Well-established methods such as distillation and purification can efficiently separate crude oil into a number of petroleum products, storing energy for widely varying purposes.
Accessing the energy stored in petroleum products can simply be accomplished through combustion whether the purpose involves running a motor vehicle, producing electric power, or just powering a household heater. Of course, however, all of these positive points stemming from chemical bond energy storage ignore the many well-documented disadvantages from the method including the production of harmful greenhouse gases and the occasional yet highly damaging oil spill/leak. This is not even to mention the inevitable disappearance of crude oil itself in the near future. With the general understanding, then, that crude oil and its established methods of energy storage can no longer be relied upon, creative ways for mass-storing renewable energy must be formed.
In the relatively nascent process of brainstorming and developing techniques for large scale renewable energy storage, some promising progress has already been made. While some of the ideas described in the following sections are challenging to implement both technologically and ecologically, they are very much necessary since they provide a precedent on which to base future innovation on the topic.
One idea that has experienced some success on a larger scale is the pumping and compression of air underground, which has actually been systematically implemented in some places since 1978.  Here, air is gradually pushed under the ground, sometimes into specially made vessels and other times just into abandoned mines, and then kept there until the energy contained by the compressed air is desired. When the air is to be extracted for energy, it is released and decompressed, forcibly rotating turbines that generate electricity. Figure 1(a) shows a schematic of the described setup. From simple thermodynamics, the energy stored is the work it takes to squeeze the air underground, which is
where VA and VB are the respective initial and final volumes occupied by the air and P is the pressure of the atmosphere. This, of course, assumes no energy losses and negligible changes in pressure from above ground to below ground, which is very much unrealistic.
In practice, the compression of air leads to the generation of heat, which flows away into the walls of the containment vessel, raising the temperature of the environment, possibly even to unsafe conditions. This heat, which would otherwise be stored as energy in the ideal equation above, is instead wasted, making the entire process less efficient. Additionally, when the air is decompressed, heat is required for the expansion, wasting even more energy that could be converted to electricity. Without supplementary processes, then, the overall method is not effective.
The solution to this problem, and the direction in which engineers in the field are headed in, is adiabatic air storage. This means that thermodynamically, the technique yields no net heat transfer to or from the system. Preferably, this would be accomplished by somehow collecting the heat produced in the compression phase and then utilizing it later in the decompression phase. The use of heat exchangers and the search for better materials to create containment vessels are currently being evaluated by scientists and engineers. Though the realization of this would still not result in 100% efficiency, a decent conversion rate would be expected. Even with potential safety concerns such as vessel ruptures or overheating, the storage of compressed air does not appear to be such an unreasonable method for large scale energy storage given the obvious abundance of air and relatively large amount of underground space available for storage.
In contrast to compressed air storage, a fairly mature and widely-used large scale storage method involves pumping water from lower elevations to higher elevations. This practice is currently the most frequently used way of storing electricity, accounting for over 129 GW worldwide.  Fortunately, the physics behind the process can again be, at the most basic level, described using fundamental concepts, this time from mechanics. Here, given ideal conditions, the amount of energy stored is equal to mgΔH where m is the total mass of the water being elevated, g is gravitational acceleration, and ΔH is the change in elevation. Simply put, when excess generated energy is yielded by a plant, it is used to pump water to higher elevations by ΔH. Subsequently, in the event that there is a strong demand for energy, the raised water is let loose and used to turn turbines to create electricity as it falls back to its original height as seen in Figure 1(b). Interfering factors disrupting the above equation, however, include losses due to evaporation from exposed water surfaces and potential leakage. Considering these drawbacks, the method is still quite effective, churning out between 76% to 85% efficiency currently. 
With a wide collection of countries already commonly utilizing pumped hydroelectricity including the United States, China, and Japan, the method appears hopeful as a major component in the movement towards future large scale energy storage.  The natural next step, then, is to make sure pumped hydroelectricity technology is fully compatible with new age sources of energy. Given that water is pumped upward using only excess electricity nowadays, it makes sense to find another means to pump it. Wind and solar power seem to be the natural answers to this as they are generated intermittently just like excess electricity. A direct connection between the output of wind farms and solar panels to the input of hydroelectricity plants does not look overly ambitious, although it certainly places restrictions on apt geographical locations. If engineers can implement this and overcome problems such as the need to find specific locations with water, high elevation changes, and alternative energy source generators along with high construction costs, then pumped hydroelectricity is expected to make great progress and be used more often than it already is.
Another interesting proposal stems from the improvement of battery technology, which has been around for centuries and is commonly known for small scale energy storage. Batteries gain the ability to store energy through chemical reactions that occur inside of them. Depending on the materials that make up a particular battery and subsequently the ions that are involved in the chemical reactions, a voltage difference ΔV is built between cathode and anode terminals. When current is allowed to run through these terminals, from one to the other, ions/charges q flow and the energy stored is simply qΔV as explained from basic electromagnetism. In contemplating the use of batteries for large scale energy storage, then, it is obvious that either more charges need to flow or voltage differences must be larger. The far more controllable of the two options happens to be the latter since various combinations of materials can be tested to create the largest difference in voltage. This search typically involves finding materials that have high energy densities, long lifetimes in terms of the number of eventual charge/discharge cycles, and relatively low cost.
Currently, lithium-ion batteries lead the charge, with very high energy densities and greater than 90% conversion efficiency.  However, a huge drawback lies in their cost of production, which is workable for the manufacturing of small, personal devices, but becomes unaffordable if they have to be mass produced for large scale storage purposes. The challenge, then, becomes finding other equally high energy density materials that are cheaper or conceiving of a way to more efficiently and inexpensively output massive numbers of batteries.
In the materials quest, plenty of innovative battery designs are being explored catering to specific materials, including sodium sulfur, vanadium, iron, and even molten salt (liquid sodium). [1,3] Additionally, with the availability of supercomputers and the relative simplicity of the physics, millions of combinations of materials can be assessed to gauge the voltage differences they might produce.  These simulations could factor in all sorts of dependencies including cost and design.
Perhaps more interesting, though, is the movement away from exhaustive materials searches. For instance, some researchers have discovered that using solid-state lithium-ion batteries, in which the liquid electrolytes in current lithium-ion batteries are replaced by solid versions, can be just as effective, if not more so, as existing lithium-ion options.  Another option, although not by definition a battery, is the use of so-called "digital quantum batteries."  Here, the incorporation of billions of nanoscale capacitors onto small chips store energy in electric fields instead of chemical reactions with much higher energy densities than lithium-ion batteries. In converting this technology to larger scales for mass energy storage, much less physical space would be occupied due to the compactness of the chips. With so many options available, the improvement of battery technology seems like a viable pursuit.
Briefly, two other potential ways to store energy on a large scale are flywheels and a smart grid. The concept behind flywheels is fairly simple in that it is just the conversion of electrical energy to rotational kinetic energy for storage and then conversion back to electrical energy using a generator for extraction. This rotational kinetic energy is described by the basic mechanics equation 1/2 Iω2 where I is the moment of inertia of the flywheel about its center of rotation and ω is the wheel's angular velocity. In reality, however, there are other factors at play when illustrating the physics of a flywheel, namely friction, the biggest counteracting force in the process. Depending on the type of environment the flywheel sits in (vacuum or air), the kind of bearings used (magnetic or mechanical), etc., the impact of friction can be either quite debilitating or of minimal interference. In the best conditions (vacuum and magnetic bearings), flywheels can store energy with up to 85% efficiency.  The chief advantages with this storage technique other than conceptual straightforwardness include long storage lifetimes and ease of construction.  However, a major disadvantage, which is both destructive and dangerous, is the well-known potential for flywheels to mechanically explode.  As a flywheel's angular velocity increases, it becomes more likely for the structure to break apart. And unless the material from which the flywheel is built has a particularly high tensile strength, disaster is imminent in any energy storage situation. Thus, the implementation of flywheels for large scale storage is certainly possible, although to store more energy, either much larger flywheels must be constructed or they must be made to spin faster and consequently be built from stronger materials. These appear to be difficult limits.
A smart grid is a network of storage systems designed for maximum efficiency and featuring automatic, real-time adjustments to how energy is stored depending on demand and usage. In essence, it potentially utilizes all of the methods discussed above, but interweaves them with each other so that the advantages and disadvantages of each technique are balanced out. In this way, excess energy to be stored for the long-term would automatically be distributed to a number of long-term storage facilities such as compressed air and pumped hydroelectricity units while energy to be used immediately would be sent to more readily accessible entities like batteries. A good example of this sort of smart grid implementation and thinking is the use of batteries in electric vehicles for large-scale energy storage in a vehicle-to-grid system.  Here, a smart grid would store excess energy in electric vehicles connected to outlets in times of low demand and extract the energy during peak demand. An idea like this would be highly effective when petroleum fuel based vehicles are in the minority and electric vehicles are more prevalent. Smart grids would seem to be the best technique for large-scale energy store since they would be able to make the best out of all available techniques.
In the end, the need for efficient large-scale energy storage techniques is highly significant. Given the energy usage trends of today, petroleum fuels will deplete at more alarming rates and current alternative energy generation ideas are more sporadic than continuous. Thus, the need for energy storage for times when loads of energy are simply too difficult to produce will be crucial to the sustenance and stability of daily operations in future society. The best solution is continued development of the aforementioned storage techniques along with a smart, efficient way to combine them all together such as a smart grid.
© Mason Jiang. The author grants permission to copy, distribute and display this work in unaltered form, with attribution to the author, for noncommercial purposes only. All other rights, including commercial rights, are reserved to the author.
 D. Lindley, "The Energy Storage Problem," Nature 463, 18 (2010).
 D. Rastler, "Energy Storage Technology Options: A White Paper Primer on Applications, Costs, and Benefits," Electric Power Research Institute, Technical Update 1020676, December 2010.
 T. Bayar, "Batteries for Energy Storage: New Developments Promise Grid Flexibility and Stability," Renewable Energy World Magazine, 30 Aug 11.
 N. Kamaya et al., "A Lithium Superionic Conductor," Nature Mat. 10, 682 (2011).
 A. Hübler and O. Osuagwu, "Digital Quantum Batteries: Energy and Information Storage in Nano Vacuum Tube Arrays," Complexity 15, 48 (2010).
 D. Castelvecchi, "Spinning into Control," Science News 171, 312 (2007).
 J. Motavalli, "Power to the People: Run Your House on a Prius," New York Times, 2 Sep 07.