Batteries, Energy Storage

Implications of a Lithium-Ion Storage Transformation

Issue 11 and Volume 121.

AES 30MW Energy Storage Facility. Photo courtesy: SDG&E

The surge of interest in energy storage has propelled Lithium-ion Batteries (LiBs) to a prominent place in the transformation of our power grid into a more flexible, responsive resource. Although energy storage worldwide is still dominated by pumped storage (comprising 96% of all storage capacity,) Li-ion is the fastest growing, estimated to reach 15 GW by 2025, according to Albermarle. Emerging energy storage mandates have been favorably predisposed to LiBs, due to their ability to deploy quickly and participate in smaller-scale development schemes on the “grid-edge” (closer to end-use customers.) In 2016, AES Energy Storage deployed the largest of these farms for San Diego Gas & Electric, connecting 400,000 individual batteries into 24 containers to store 120 MWh of electricity for up to four hours.

These deployments essentially consist of massive banks of batteries, stacked and interconnected, in temperature controlled tanks. The ability to get into the storage business by simply purchasing off-the-shelf batteries from wholesalers, stacking them into tanks, and coupling with a control system is a compelling business proposition for many developers, if the right market conditions and incentives are in place.

Unfortunately, the race to develop storage projects with LiBs has left many questions unanswered. Is there sufficient supply of Lithium, Cobalt, Nickel, etc. to meet the needs of EVs, mobile applications, and stationary grid storage? Are the environmental and human health impacts of the mining and refining processes for these metals worth the transition? How do these supply chain issues impact the sustainability of Li-ion? Investigative reporting has uncovered serious risks in the supply chain for key constituents of the bill of materials, and anecdotal evidence from residents living in regions where these resources are extracted and/or refined has exposed particularly troubling trends. Finally, the cost of transitioning to a Li-ion storage system must be compared with proven storage alternatives available, accounting for both lifetime operating cost and environmental impact. Given the long-term consequences of power grid transformation and the current pace of transition, policymakers and industry must carefully weigh the pros and cons of each energy storage technology, and avoid making hubristic decisions to favor one approach over another.

Is there adequate supply?

Several studies have considered this question, and most have concluded that for Lithium, the answer is “Yes,” at least in the short and medium-term. However, many of these studies are focused purely on whether there is adequate supply for greater levels of EV penetration, without factoring in demand from the stationary grid storage market. The most notable study, published by MIT in 2013, concluded that there is adequate Lithium for the next century if demand for portable electronics (cell phones, laptops, etc.) tracks GDP growth (which it did in the time period studied by MIT of 1994 – 2008.) The study does not consider demand from grid storage applications, and actual growth rates of portable electronics during that time period have been closer to 10 percent Compound Annual Growth Rate (CAGR.) Technavio, a market research firm, estimates that total sector-wide Lithium-ion battery demand will grow at 72 percent CAGR through 2020. Furthermore, EV penetration rates have consistently exceeded the expectations that make it into academic forecasts (although many industry analysts have got them right.) BNEF recently estimated that EVs will reach 54 percent of all new car sales by 2040.

The supply of Lithium and constituent materials (Cobalt, etc.) are often concentrated in only a few countries, exposing their operations to geopolitical risk. Over 90 percent of Lithium is concentrated in four countries – China, Chile, Argentina and Australia. China, home to over 95 percent of current global rare-earth material supply, has demonstrated a willingness to impose export quotas of these materials. Although Bolivia is home to the world’s largest single proven Lithium reserve, relations with the U.S. had deteriorated over the term of President Evo Morales, damaging many relationships with a key trading partner. Over 50 percent of the world’s Cobalt Production comes from the Democratic Republic of Congo, where political instability is a way of life. Cobalt is already experiencing shortages, price run-ups, and hoarding by hedge funds, as analysts forecast that supply will not be able to keep up with demand due to EV growth projections. Macquarie Research estimates a deficit of 885 tons of Cobalt for 2018, 3,205 tons in 2019, and 5,340 tons in 2020. This tightening of the supply chain for key LiB materials must be taken into account to provide a more realistic picture of LiB penetrations in the grid storage sector.

What are the environmental consequences of large-scale implementation?

Each LiBs includes a cathode, an anode, and a porous separator placed in between. The chemistries that make up a LiBs vary, but typically the bill of materials includes cobalt, graphite, nickel, aluminum and copper (aside from Lithium itself.) The extraction of each of these elements poses significant environmental risks and health impacts. Steel production for the battery pack housing and management system are associated with freshwater cyanide emissions. The production of soda that is used in processing lithium salts leads to photochemical oxidation. Aluminum production for the cooling system, cathodes and other parts of the batteries contributes to ozone depletion. According to a lifecycle analysis of Lithium-ion batteries conducted by the EPA, “upstream materials extraction and processing and battery production…are significant contributors to eutrophication potential, ozone depletion potential, ecological toxicity potential, and the occupational cancer and non-cancer hazard impact categories.”

Cobalt is used for the cathodes in LiBs, and makes up approximately 35 percent of the total composition. Cobalt mining has a track record of severe environmental and human health hazards. A recent Washington Post expose on this issue uncovered common injuries and deaths for miners in the Democratic Republic of Congo, the world’s largest Cobalt producer, where an estimated 100,000 miners dig hundreds of feet underground with hand tools, according to the Post. Graphite is used for the anodes in LiBs, and represents about 10-15 percent of the total cost. 70 percent of graphite is produced in China, where pollution from mines and refineries have resulted in stunted and damaged agriculture, living conditions covered in soot, and polluted water. The effects of graphite mining and refining reported by residents in the towns impacted by these operations is consistent – graphite dust on all food (tastes and feels similar to sand,) laundry, and bodies; a chemical smell in the air; oily drinking water due to waste released directly into local water supplies, and poisoned crops. Graphite in its powder form quickly becomes airborne and can cause breathing issues and has been linked to heart attacks.

One ton of Lithium typically requires about 500,000 gallons of water to produce. Closed-basin brines are located almost exclusively in arid regions, where groundwater aquifers are scarce and annual rainfall is limited. Although the water pumped from the brines is undrinkable (10x as salty as seawater,) there are some scientists who posit that the voids left by these operations may be refilled by freshwater. This has the potential to divert valuable water supplies from local communities that may be present in these locations. These communities typically build their livelihoods around water supply, due to the agrarian nature of their subsistence. Anecdotal evidence from local communities suggest that there may indeed be a connection to the pumping of these brines and the source of fresh water – many claim that their sources of water have diminished and previously lush areas have become arid. Other local communities report polluted water and reduced agricultural production. Admittedly, these reports are usually anecdotal – however, the communities that typically inhabit these regions have limited ability to voice their concerns and impact policy on a national scale and are often outmatched in negotiations by the much larger mining operations. Despite widespread mining and investment by these companies, there is a surprising lack of published data on the impact of mining operations on water quality and availability, and this issue requires utmost consideration due to the aridity of these regions. Given the risk and impacts of the full bill of materials for battery production, it remains an open question of whether this form of storage is truly sustainable.

Finally, the end-of-life concerns for Li-ion have not yet been addressed. Given the finite lithium resource worldwide, the recycling of batteries is looked to as a key contributor to future supply. Current efforts to recycle are still in their infancy, as the availability of Lithium supply and relative new application in electric vehicles has not yet forced a strong demand for recycling. However, as prices continue to climb and demand is constrained, new recycling programs and technologies, such as hydrometallurgical separation must come to the fore to meet demand. Lead acid battery recycling efforts benefit from their relative simple chemistry and construction, and regulations prohibiting their disposal (in the U.S.)

These regulations are not yet mature and consistent for Li-ion, and the current recycling infrastructure is not prepared to handle the influx of these more complicated batteries. Even if this infrastructure was rapidly developed, the question remains whether this would relieve the supply constraints for new Lithium. One study on this issue from the University of Adelaide in Australia found that even if 100 percent of all LiBs were recycled, this recycling effort will only reduce demand for Lithium by 26 percent by 2030. An operation to extract and recycle even close to 100 percent of Lithium (not to mention other materials such as Nickel and Cobalt,) would require a recycling operation of massive scale that is not close to being available today.

Can Li-Ion cost effectively perform as needed?

The cost of installing and maintaining LiBs continues to be an inhibitor to widespread implementation. LiBs remain more expensive than nearly every other alternative stationary grid storage technology, including pumped storage hydro and compressed air. Lazard, a financial advisory, estimates the Levelized Cost of Storage (LCoS) for each storage technology. While the highest cost Compressed Air and PSH are still below $200/MWh stored, LiBs range from $270-$600/MWh. The costs for these systems have come down significantly as deployment as increased, with an observed learning rate (cost reduction per doubling of capacity) of around 22 percent. However, with rising costs for many of the raw materials that constitute these batteries, it is unclear if this rate will be able to be maintained after initial economies of scale and manufacturing improvements are realized.

The principle disadvantage of Lithium-ion in a levelized cost analysis is the 10 year life expectancy (compare with 20 years for CAES and 50+ years for PSH, for example.) Material degradation can be a significant challenge for LiBs exposed to large volume of cycles (lithiation and delithiation processes). During each charging cycle, Lithium is deposited on the electrodes and does not completely dissolve, leaving slowly accumulating deposits and reduced capacity. As cycle times increase, primary particle separation and defects can occur, posing a reliability challenge for these batteries and reducing capacity. This results in a shorter lifetime and corresponding higher lifetime cost when compared with alternative forms of storage.

What alternatives are available?

Given the diverse alternatives available for grid energy storage, a comparison of the relative advantages and disadvantages of these alternative energy storage systems is essential for a well-informed discussion about the future of our power grid.

Pumped Storage Hydropower – PSH is the oldest and most proven grid-scale energy storage source today, comprising over 96 percent of energy storage installations globally1. PSH consists of pumping water and generating electricity between an upper and lower reservoir. Water is pumped from the lower to the upper reservoir to “store” as energy (typically during off-peak hours when power rates are lowest and when wind production peaks,) and then is allowed to run back to the lower reservoir to provide energy on-demand (typically when rates are most expensive and energy demand is at its highest.) It is widely considered the most reliable source of energy storage, offering the full suite of ancillary grid services demanded from an energy storage system with an extremely limited environmental impact, particularly from a life-cycle perspective. With greater demands for flexibility from power grids due to variable renewables integration, PSH installations are increasingly strategic and valuable assets for utilities.

Although PSH is among the lowest energy plants to operate on a per MWh basis, there is a high initial cost to construct these plants. One solution under investigation is smaller, modular pumped storage hydropower systems that can be integrated with existing infrastructure closer to load constraints. The U.S. Department of Energy has taken a keen interest in this approach in recent years, and commissioned several studies examining the feasibility of modular pumped storage hydropower. Nature and People First is one company that is innovating in this space, and has developed a method of storing energy utilizing reservoirs that are integrated with urban infrastructure and provides local energy and non-energy services. These systems can serve the needs of critical urban infrastructure, providing resilient grid services to complement local distributed resources such as community wind and solar farms. The proximity to urban infrastructure also provides “Revenue-Stacking” opportunities, by using pumped storage reservoirs to provide other water-related services to municipalities (heating/cooling load reduction, water supply for urban irrigation and greenscapes, emergency fire suppression, etc.) If successful, this hybrid approach may open up new financing models including public private partnerships (PPP) with local municipalities to construct resilient multi-use assets within their existing urban planning schemes.

Compressed Air Energy Storage – CAES systems are functionally equivalent to pumped storage plants, but utilize compressed air instead of water as the storage medium. Typically this air is compressed and stored in an underground cavern, although additional reservoirs can be used, including underwater membranes. When energy is required, this air is then heated and expands in a combustion turbine that drives a generator. There are two commercial CAES installations in operation today. These plants are essentially on-demand gas turbines with combustion air stored underground. CAES offers an advantage over PSH where geology provides preferable storage characteristics, such as salt caverns, abandoned natural gas fields or natural aquifers with limited elevation available above ground to provide upper PSH reservoirs. Unfortunately, suitable geographic formations are not well-dispersed, limiting the siting of these plants and constraining its growth potential.

One Canadian company, Hydrostor, is aiming to develop CAES systems that they claim are half the cost of grid-scale batteries and reduces siting concerns. Hydrostor is developing and promoting an adiabatic system, where heat is not fully utilized within the combustion and generation process. The Hydrostor system also introduces water into their system from nearby sources for the compressed air to replace, which the company claims reduces operational strain on the underground tanks. Hydrostor currently has one 0.7 MW operational system in Toronto Island, Canada, one 1.75 MW system under construction in Ontario, Canada, and another 1 MW system contracted to build in Aruba. Hydrostor appears to be taking a similar approach to Nature and People First by integrating their reservoirs near urban areas, providing greater siting flexibility.

Both PSH and CAES compare favorably to Li-ion from a sustainability perspective. Batteries have greater impacts throughout the material procurement and extraction phase, due to the long list of materials and chemistries that comprise the hardware. In the operational phase, the environmental impact of PSH is less than both LiB and CAES. CAES systems as currently deployed burn natural gas, releasing greenhouse gas emissions. LiBs require extensive cooling systems and higher maintenance requirements than similarly-sized pumped storage plants. Once constructed, PSH plants have exceptionally low operational requirements, emit virtually zero greenhouse gases, and require only routine maintenance to run effectively. A study by the Global Climate and Energy Project at Stanford University found that PSH and CAES are less energy intensive than LiBs, on a lifecycle basis, by a factor of 100!

PSH and CAES are both significantly more cost-effective than any battery technology, according to Lazard’s Levelized Cost of Storage comparison. The LCOS analysis is a unique build-up of cost based on the application of storage, and assumptions are maintained equal across all storage options for a 100 MW, 8-hour storage time system deployed on a transmission grid. Under these assumptions, PSH and CAES carry significantly lower capital cost and O&M cost. LiBs have higher overall efficiency, but must be replaced every 10 years, leading to the higher overall levelized cost.

The principle advantage of Li-ion over both PSH and CAES is the ability to site relatively quickly and easily. PSH and CAES require elevation differences or specific geographic formations, respectively, confining their applications. Li-ion grid storage can be sited directly in or near urban areas, substations, critical infrastructure assets, etc., provided that adequate safety measures are taken. As noted above, both PSH and CAES developers are exploring ways to reduce and modularize their systems to take advantage of the urban environment, with promising results. If these companies are successful, it removes the primary and singular competitive advantage of Li-ion as a source of stationary grid storage.


In the past decade, Lithium has emerged as a leading source of energy storage for mobile applications, and also now comprises the large majority of new development grid energy storage projects. This paper identified the key supply chain and operational limitations of LiBs as a grid-storage application. These weaknesses include supply chain security for key raw materials, negative environmental and societal impacts of material extraction, performance limitations and material degradation concerns, safety, and high lifetime costs due to replacement every 5-10 years. Given these challenges, it is important to weigh the relative pros and cons of LiBs vs. alternative storage technologies. On a qualitative analysis across multiple variables, both PSH and CAES compare favorably with Li-ion, offering superior cost and performance with reduced operational risk. PSH and CAES primarily are limited by siting constraints, although several companies are innovating in this field to remove those limitations. An honest examination of LiBs as a future energy source must take into account the full life-cycle impacts and limitations in the perspective of alternative options available today.

Peter Drown is President of Cleantech Analytics LLC