Part II — Storage: The Missing Link

Dawn of new renewable energy technologies. Photo by Consumer Energy Alliance

My first blog on The Future of Energy made references galore to the importance of nuclear energy, hydrogen technology, and energy storage for the deep decarbonization of the global economy, which entails the adoption of cleaner forms of energy for heavy industries like cement and steel manufacturing and long-distance transportation like shipping, aviation, and trucking. This second blog will focus on energy storage, specifically lithium-ion batteries and their transportation and electricity grid applications. I will briefly introduce the alternative forms of electrochemical energy storage and also comment on other solutions for storing energy, including but not limited to mechanical, thermal, chemical, electrical, and gravity-based storage. (Figure 1)

Figure 1: There are numerous energy storage technologies, each with different characteristics and the electricity grid makes use of all of them.

I am focused on lithium-ion batteries because of their pre-eminence over other energy storage solutions. The precipitous price decline of lithium-ion batteries resulted in their diffusion, first across the electric vehicles, and then grid storage, becoming a preferred solution for electric utilities. Other storage solutions are nowhere near as close to scalable deployment as lithium-ion; the majority stuck in the research and development and demonstration stages.

1. Recent Developments

The cost of renewable energy has declined considerably over the last decade, especially for solar. Yet solar and wind constitute single-digit percentage points of energy source used for electricity generation in the U.S. A significant reason is the electricity generated from renewables is not dispatchable and cannot meet demand at a moment’s notice. We need to pair renewables with storage to supplant fossil fuels, which are the preferred energy source. Lithium-ion batteries are certainly one of the remedies for the intermittency of solar and wind energy. The recent large-scale projects have incontrovertibly established that solar and wind, when coupled with lithium-ion batteries, are cost-competitive to coal and natural gas for short-duration storage applications.

How did lithium-ion batteries scale so quickly? The answer is Tesla. Elon Musk’s obsession with helping the world transition to more sustainable transportation led to the development of multiple Gigafactories — each of the size of 100 football fields — that induced competitors to follow suit, leading to a ten-fold decrease in cost from $1,183 in 2010 to $156 in 2019. (Figure 2)

Figure 2: The actual data derived primarily from BloombergNEF. The pricing data is based on the industry volume-weighted average and is not intended to be representative of cost leaders such as Tesla and CATL. According to me, Tesla was already below $100 per kilowatt-hours at the cell level, and below $140 per kilowatt-hours at the pack level in 2019. © Abhishek Kumar

The price/performance improvements in lithium-ion are expected to level off when the Levelized cost of storage (LCOS) reaches $62 per kilowatt-hour. At this point, EVs will become cost-competitive with internal combustion engines (ICE). Since the most expensive component of EV is its battery pack (Figure 3), the retail price of a Tesla Model 3, will decline from $35,000 to $25,000 in 5 years, if lithium-ion battery technology advances as per its present learning rate of 18 percent.

Figure 3: Cost walk of ICE to electric vehicle (EV) in 2019. There is still a cost gap of more than $10,000 between electric vehicles and internal combustion engine vehicles, which will soon fade due to the on-going lithiom-ion battery technology revolution. (Source: McKinsey, March 2019)

1.1. Global Scale

Remember when between 2010 and 2015, the Chinese started dumping solar panels in global markets driving their European and American competitors — both startups and mature companies — into bankruptcy. The history is repeating itself, as the dragon nation is building lithium-ion battery manufacturing facilities with the same ferocity! Of the seven largest lithium-ion battery manufacturing facilities in the world, five are in China. (Figure 4) While foreign entities (Samsung SDI and LG Chem) own two of these, locals own three (BYD, CATL, and Lishen) serving international clients like BMW, Volkswagen, Daimler, Volvo, Toyota, Honda, and now Tesla. Lithium-ion battery capacity is expected to pentuple from 221 gigawatt-hour in 2018 to 1100 gigawatt-hour in 2028, according to Benchmark Mineral Intelligence, and will drive the electrification of things — from drones to robots to more electric vehicles. China will dominate battery manufacturing by considerable margins, followed by the U.S., Germany, Japan, and South Korea. In my opinion, India will plausibly miss a chance to play a significant role in battery manufacturing unless the country sheds its reliance on China and builds a thriving lithium chemical, battery cathode, battery cell, and EV supply chain for itself.

Figure 4: World lithium-ion mega-factory capacity, 2010–2028. (Source: Benchmark Mineral Intelligence)

1.2. Technology Improvements

As a material scientist, I can appreciate battery innovations and their implications for EV performance, reliability, and safety more than a layman. Take the example of batteries used in Tesla Model S and Model X, which hold 7,104 cylindrical battery cells manufactured by Panasonic across 16 modules (444 per module) and are capable of storing up to 85 kilowatt-hours of energy. Through iterations in anode design and by increasing the battery cell capacity, Panasonic was able to endow these vehicles with 90 kilowatt-hours of energy density, a 6 percent increase. Further improvement in battery performance came from the reconfiguration of battery packs where Tesla engineers equipped each module to hold 516 cells instead of 444, increasing energy density further to 100 kilowatt-hours and letting the cars enjoy a range of over 300 miles.

I am also privy to other strategies to increase the EV battery performance. For instance, LG Chem tested new materials and separators, where they changed battery cathode mixture from NMC111 to NMM442 to NMC622 and finally to NMC811, where NMC stands for nickel manganese cobalt. The latest composition of elements with a high nickel content and a low cobalt content improved energy density besides reducing cost. Secondly, LG Chem employed nanotechnological principles to enhance its proprietary separator’s safety, which translated into better thermomechanical, shock resistance, and precluded short-circuiting compared to conventional polyolefin separators.

I am keeping close tabs on products like ‘million-mile batteries’ being advertised by Tesla, which will serve both transportation and renewables grid where for the former sector, it will render business models like battery swapping more viable. Read section 3 for more details.

There is no dearth of examples of lithium-ion battery improvements, yet I addressed the ones which I consider have resulted in substantial benefits to businesses and customers.

1.3. Filip for Clean Energy businesses

Notwithstanding lithium-ion battery weaknesses, it does make previously untenable enterprises tenable. Solar and wind energy variability previously discouraged entrepreneurs and investors from participating and fostering renewable businesses, but now solar panels and wind turbines, when bundled with batteries, offer investors lucrative return profiles. Solar panel manufacturers like REC, Jinko Solar, and Trina Solar — businesses with notoriously low margins, wind farm developers like Vestas, Siemens, Gamesa, and Vattenfall, and solution providers like SolarEdge, Sunrun, and EnPhase have all become more valuable by adopting energy storage in their portfolios.

Lithium-ion batteries’ acceptance by disparate industries also has monetary implications for lithium producers — Albemarle, SQM, and Livent. Prominent ride-hailing, ride-sharing, and automotive OEMs — Lyft, Ola, Didi Chuxing, Grab, General Motors, Nissan, Volkswagen, Volvo, and Zipcar have all pledged to completely transition their fleets from internal combustion engines (ICE) to battery-powered vehicles. And lithium producers will be there to reap the rewards of this shift towards sustainable transportation. Lithium pure-play Livent could see the most upside out of these three notable producers as lithium only accounts for 40 percent Albemarle’s and 43 percent of SQM’s total revenue.

The reason that many of our energy storage needs are already satisfied by the leading battery technology of our time led the Nobel Committee to award the Nobel Prize for Chemistry in 2019 to John B. Goodenough, M. Stanley Whittingham, and Akira Yoshino for their work towards the development of practical lithium-ion batteries.

1.4. Obstacles

My enthusiasm for lithium-ion batteries is, however, not unfettered. Allow me to broach the supply of lithium carbonate and lithium hydroxide, which are critical ingredients required to manufacture Lithium-ion batteries whose demand for which rose 20 percent in 2019. Most of the world’s lithium is in South America’s Lithium Triangle, covering three countries — Chile, Argentina, and Bolivia. While these deposits are enough to electrify the current EV fleet 2.5x over, the current annual lithium production is insufficient to make enough batteries for even 10 percent of annual vehicle sales if called upon in a pinch.

Other symptoms of supply chain bottlenecks became manifest in Audi’s decisions to halt the production of the e-Tron, Jaguar of I-pace, and Mercedes of its EQC, all EVs due to the unavailability of cobalt, another crucial battery component. Audi and Jaguar source their batteries from a Polish factory owned by LG Chem, which was ailing from structural undersupply of cobalt and caused the disruptions mentioned above.

Cobalt is present in 75 percent of lithium-ion batteries and concentrated in the “Democratic” Republic of Congo (DRC) — which controls 65 percent of global supply and an atrocious human rights record. Multiple reports have chastised the Congolese government and private sector for its treatment for miners, including coercing children into the activity and hazardous working conditions. Other Lithium-ion battery supply chain issues include its use of specific graphite forms whose extraction is smothering Chinese peasants and their families. As we witness the burgeoning demand of cobalt — expected to be 430,000 tons in the next decade — and graphite, we must keep the ethics of battery supply chain stakeholders in mind.

And then there are product shortcomings. The range anxiety is a real issue and not imagined as some have contended. I can attest to the feeling of having driven Tesla Model 3 in the U.S., Nissan Leaf in Italy, Mahindra e2oPlus in India, and BYD e6 in China. Secondly, while battery engineers have made considerable progress towards making lithium-ion batteries failure-proof but their degradation that stems from undesirable side reactions during battery recharging is yet to be appreciably mitigated.

Notwithstanding effective remedies of the above supply chain, even if lithium-ion batteries continue to decrease in price precipitously, we will still not have achieved battery supremacy. They can only take us so far down the road to transport electrification and decarbonization. To get to the end of that road, the following two breakthroughs in energy storage are imperative among others:

• Fast charging
• Long-duration storage for weekly, monthly, and seasonal fluctuations in energy supply and demand

2. Fast Charging

The key indicators that determine the role any battery will play in the electrification of transportation are energy density, charging speed, the pervasiveness of charging infrastructure, and durability.

While in section 1.2, I discussed R&D strides industry has made towards improving energy density and safety, the attributes I question most are related to charging. An iPhone takes several hours depending on the iteration, whereas an EV takes 45 minutes to recharge using a fast charger. Contrast this with internal combustion engine cars whose gas tanks you can fill in less than 5 minutes. Slow charging speeds have business implications as more time a vehicle takes to charge, the less is its utilization. Consider the time an entire fleet owned by Uber or Tesla takes to charge. Longer these EVs are plugged-in, the punier profit these companies make. This very lithium-ion battery weakness was an important reason behind the non-success of the Ola Cabs’ EV pilot project in the Indian state of Maharashtra. Secondly, we are also not witnessing any discernible sprawl of charging infrastructure, at least in the U.S. But hopefully, we will soon. Finally, EVs undergo more wear and tear on African and Asian roads. The batteries due to the undeveloped road and grid infrastructure are subject to more electrical, mechanical, and thermal abuse, which reduces their life expectancy.

In my opinion, the charging is a critical constraint to mass market adoption and is not on track to make EVs competitive functionally. And for all the reasons listed above, we will need technologies superior to the current generation of batteries and fast chargers fast enough to get us near 50 percent electrification of vehicles.

There are necessarily two possible solutions to this core problem, and both are beginning to progress from laboratories to pilot projects.

The first is ultra-fast charging. Ultra-fast chargers designed to deliver 300–400 kilowatts can charge EVs ten times the rate of fast chargers. But of course, there is a catch. Due to the inconvenient laws of physics, these chargers are more expensive to install and draw as much power from the grid as an entire neighborhood. Another irksome trade-off is the impact ultra-fast charging will have on battery longevity as larger zaps of voltage delivered by ultra-fast chargers will lead to significantly more rapid battery degradation. Consequently, my foresight is that any transition to ultra-fast chargers will be accompanied by supplantation of lithium-ion batteries with solid-state technology that can endure high voltages during recharging.

A second resolution to charging woes is less technological and more operational. It entails swapping batteries, which has been attempted unsuccessfully by now-defunct start-ups like Israel’s Better Place. While it turned investors off the strategy to redress the nuisance of recharging an EV, I remain intrigued by its possibilities to eliminate reliance on the electric grid and stymying battery dilapidation. An EV running low on the charge could pull up at swapping station and exchange the spent battery for a fresh one. The swapping time is comparable to when the same user would take to refuel ICE vehicles. SunMobility in India helmed by serial entrepreneur Chetan Maini, who invented the country’s first EV (Reva), is implementing the exact business model. (Figure 5) To keep the business software-focused and less capital intensive, Chetan forged partnerships with Microsoft and several auto OEMs.

Figure 5: The concept of battery swapping. The owner of the EV can drive anywhere and replace the battery like we refuel petrol or diesel. © Abhishek Kumar

The challenges for battery swapping are threefold. Firstly, swapping works if enough vehicles have standardized on a single battery form factor and swapping-ready design. Secondly, it means purchasing extra batteries that will, at times, be sitting idle outside of a vehicle. Thirdly, there is the apparent technical challenge of swapping a bulky battery in and out of a car. For two-wheelers and three-wheelers, the nations can hope to adopt a swappable model of lithium-ion batteries, which can be easily carried within the vehicles or charged at home.

Battery swapping is undoubtedly a daunting combination of the business model and technological hurdles. It is one that I hope capable entrepreneurs like Chetan Maini can overcome.

In summary, we will need batteries with twice the density of today’s lithium-ion batteries, resilience to ultra-fast charging, and lower cost of parts and assembly for the complete electrification of transportation. Figure 6 showcases a tiny piece of my recent work on the technology categories and companies pursuing such innovations.

Figure 6: Advancements in lithium-ion battery technology © Abhishek Kumar

3. Long-duration Storage for Weekly, Monthly, and Seasonal Variations in Energy Supply and Demand

I want to turn to lithium-ion batteries used for electricity now briefly. Tesla’s Hornsdale Power Reserve in Australia, which is the largest lithium-ion battery in the world at 100-megawatt, 129-megawatt hour, is commercial proof of lithium-ion batteries’ ability as a grid storage solution. Renewable energy (wind in this case) is stored during periods of oversupply and discharged during periods of excess demand, increasing the penetration of renewables in Australia’s energy mix.

Upcoming projects by Tesla such as the Megapack, with three times the energy capacity of Hornsdale Power Reserve at 100-megawatt, 400-megawatt hour, will further corroborate lithium-ion batteries as a grid storage solution and assert its superiority over the natural gas turbines used for the same purpose of balancing supply and demand of electricity second by second, preventing load-shedding blackouts, and managing the load to store energy when prices are low and sell it when demand is high. Figure 7 explains the multiple applications of batteries in a simple way.

Figure 7: Use of energy storage during the day and night. Energy storage is needed on a grid-scale for three main reasons. The first reason is to “bridge” power — in other words, to ensure there is no break in service during the seconds-to-minutes required to switch from one power generation source to another. The second is “power quality management” — the control of voltage and frequency to avoid damaging sensitive equipment — is an increasing concern that storage can alleviate whenever needed, for a few seconds or less, many times each day. Finally, to “balance load” — to shift energy consumption into the future, often by several hours — so that more existing generating capacity is used efficiently. © Abhishek Kumar

But again, for all its virtues, there are limitations to using lithium-ion batteries as a grid storage solution. These batteries can only provide the energy for short durations (<4 hours) and have a high self-discharge and a low relative round trip efficiency compared to other grid-scale storage solutions. Lithium-ion batteries’ LCOS rises longer the duration electricity has to be stored, reaching $800–1,000 per kilowatt-hour for 8–12 hours of storage. The batteries are incapable of helping renewables cope with seasonal variability of demand. For example, in colder regions like Alaska, heating consumes five times more electricity in the winter than in the summer. I will save a detailed technical explanation for another blog. Still, this problem is because one type of lithium cell is used for power applications, and a different type of lithium cell is needed for energy services, and a single storage system cannot accommodate both. The power and capacity of a lithium-ion battery are inherently coupled, which means that increasing the battery capacity through the addition of cells, intrinsically increases power, given constant chemistry, temperature, and other specifics. The costs escalate the longer you want to store energy in lithium-ion batteries, and this coupling restrains us to the 4-hour duration for economical energy storage. So what are the most promising solutions for long-duration storage? There are three, in my opinion.

3.1. Excess solar and wind capacity

Without an energy storage solution, one way to continually match any given kilowatt-hour of electricity demand with supply is by laying out an excess of solar panels and wind turbines over vast tracts of land. In my simple analysis for promulgating excess capacity, if half of the energy from any given wind or solar farm is wasted, that means the cost of the useful energy it produces will need to be doubled, and as long as that double-priced solar or wind energy is less expensive than different energy storage solutions, it will make sense to build more of it. The approach to over-build solar and wind capacity is not a universal solution, but a serious one, if we accept that long-duration energy storage is a decade away.

3.2. Vehicle-to-grid (V2G)

According to the IRENA report, the aggregate number of EV batteries globally will be able to store a combined 40 terawatt-hour of energy by 2050. The world’s electricity consumption will be a fraction of that, so while EV is patent to most as means of transportation, very few people have grasped its importance as distributed storage for the electricity grid given, especially since these cars sit idle 90 percent of the time. EVs can easily plug into a smart outlet. As bidirectional electron flows become possible between EV owner and grid, the former can provide the latter with benefits like greater renewable penetration, preclude renewable curtailment, peak load shifting, and other demand response services.

3.3. Other diverse energy storage technologies

The future of energy storage is diverse (refer to Figure 1). The appropriate solution for any storage problem depends on cost, location, duration requirements of the customer, and the pervasion of supporting infrastructure. There are many energy storage systems besides electrochemical batteries — pumped hydro, thermal energy storage (steel at 650°C/1200°F), liquid air and advanced compressed air, and gravity-based energy storage, among other technologies that might be appropriate for long-duration storage over days, weeks, months, even seasons.

Even with electrochemical storage, many battery chemistries are battling for supremacy, making bold claims about efficiency and cost, not all of which stand up to scrutiny. Suggestions made by entrepreneurs include Silicon-based batteries (Si), Sodium-sulfur batteries (NaS), Proton batteries (H), Graphite dual-ion batteries (C), Aluminum-ion batteries (Al), Nickel-zinc batteries (NiZn), Zinc-air batteries (Zn), Salt-water batteries (H2O), Paper-Polymer batteries (Paper), and Magnesium batteries (Mg). Flow batteries and pumped hydro, for example, allow for the decoupling of power and energy, and these technologies utilize different materials, which are stable, non-explosive and non-flammable, and non-toxic.

For many technologies, it will be a question of access to manufacturing facilities of appropriate size that will drive scale and therefore cost reductions. For others, it is a question of getting proven through successful deployment beyond the pilot phase, and for others still, it remains a matter of technology development. And while they cannot all be the ‘cheapest’ as they claim, there is probably room for diverse long-duration technologies in tomorrow’s world. For sure, the race is on, and the window to deliver is getting shorter, even if the hours of energy stored and discharged are going to get longer.

Final Thoughts

As an entrepreneur and investor, I am focused on batteries more than exotic energy storage solutions. I want to reiterate that lithium-ion batteries are not the end game for decarbonizing transport and the electric grid. We need batteries that pack more power and energy at a lower cost and greater amenability to ultra-fast charging. There is research currently underway to build storage systems for cars that do not necessarily accommodate it as it will be called upon to cycle once or twice a year entirely. Let us see where these efforts lead.

Share this article with your friends, especially if they are interested in any of the areas outlined above. You can also connect with me on Medium, Linkedin and Twitter.

Coming up next is Part III — Breakthroughs in Energy and Transportation (2020–2030)

Disclaimer: The article expresses my own opinions. The assessment of storage trends is based on the latest available data and announcements by governments and companies, as of June 25, 2020, tracking progress with individual projects, interviews with leading industry figures, and incorporates also the latest insights and analysis from across McKinsey & Company, Benchmark Mineral Intelligence, Wood Mackenzie, and BloombergNEF work.