Application-Specific Consideration for Next-Generation Battery Technologies

This story is contributed by Asang Mehta, from Xerion Advanced Battery

• Battery materials research, cell design, and business development must all be catered to specific end-use applications right from the onset in order to provide a structured focus to product timelines
• The key to large-scale adoption of battery technologies is to amplify specific State-of-the-Art (SoA) cell specifications based on identified customers while ignoring ancillary ones.
• Electric vehicles (EV) and other stringent end-use applications require product lines catering to customers based on behavioral patterns, location, and other socio-economic factors.

Introduction

Battery development is a dynamic and intricate industry that necessitates cross-linking mineral acquisition, supply chains, research & development (R&D), pilot scale testing, production, and large-scale deployment [1]. Hence, it is extremely important to cater battery materials and cell development towards specific end usage applications. Otherwise, mastering the perfect strategy for new, untested technology that combines all of these vertically integrated aspects of battery business development can take years, if not decades, to come into fruition.

Figure 1: Next-generation solid-state batteries and pure silicon anodes to give a boost in energy and power [2]

There are bound to be tradeoffs in a world with limited resources and an ever-growing human population. Battery development, unfortunately, is no different. While there has been a massive push towards achieving the “Holy Grail” of batteries using solid-state and other emerging technologies (Figure 1), developing the perfect battery at scale has been, and will continue to be a pipe dream. This push for developing the “Holy Grail” for batteries has more downsides than benefits through wastage of materials with a long R&D timeline and thus delaying the energy transition.

Historically, battery materials technologies built from the ground are a result of three characteristics — inventiveness & ingenuity, luck, and sheer brute force [3]. Development that depends heavily on the latter two is more likely to deplete resources if there isn’t a better understanding of the technology. Hence, the focus of R&D and manufacturing for such novel technologies should be to adjust direction so that most, if not all, aspects of the technology can be attributed to ingenuity and innovation. One potential solution to this dilemma is to recognize that no new battery technology is perfect, but that it can be perfected for one or a few end uses and specifications. It provides more direction for R&D efforts while lowering the mountain of risk associated with battery startups.

Understanding End-Use Industries

Being 100% ingenious is obviously easier said than done — after all, we are only human. If we were that scientifically creative, there would be electric rockets scanning the landmass of Saturn right now (yes, Saturn is a gas planet). From a materials science perspective, it is time-consuming and labor-intensive to create a battery that can fast charge, has a long cycle life, unparalleled energy density, and all the other SoA characteristics one can only dream of having at the same time. As capitalistic as it may sound, the focus should rather be on who we are selling the batteries to: this is one strategy that can be implemented to make batteries more ubiquitous in all industries that require some form of energy storage.

Figure 2: Current SoA for energy storage technologies [4]

It is critical to understand which products and industries require which type of energy storage solutions, and how next-generation technologies can be uniquely catered for each of these. With current energy storage trends and available technologies, batteries face an uphill battle in terms of competitiveness and market penetration in industries like stationary storage (Figure 2, 3) in terms of longevity, EVs in terms of cost versus their Internal Combustion Engine (ICE) counterparts, and aviation in terms of energy density. As of 2020, the primary industries that use lithium-ion batteries for energy storage are EVs/transportation (Figure 3), consumer electronics, and defense/space [1]. Ancillary industries where the usage of batteries will grow rapidly are medical devices, tooling/construction, and microelectronics [5].

Figure 3: EVs will dominate demand for Li-ion batteries

One must also recognize that within each industry, there may be unique products with vastly different requirements, and each product may have several variants; another pain point that battery engineers and scientists must keep in mind. For example, defense battery products being sent to space must meet strict requirements for thermal stability and abuse test handling, whereas defense drone products can tolerate high depth of discharge and not be as strict on other requirements [6]. It is important to acknowledge that the battery requirements for next-generation technologies are becoming increasingly stringent in order to meet customer expectations. Customers require upfront cost parity, fast charging, and all other wonderful attributes of an EV so that they view an EV as worthy of replacing an ICE car. Even though it seems like all SoA characteristics must be hit simultaneously, this need not be true and will be discussed in the subsequent sections. The next part will focus on how certain SoA cell parameters and specifications affect demand in these industries, and how they can be tweaked from a business and marketing standpoint to attain mass penetration of secondary cells, specifically in industries that employ the greatest number of batteries in terms of kWh of energy.

Battery End Usage with SoA Cell Specifications and Application

Now that a foundation has been laid for why cell design should be industry-specific, what are these end-usage specifications and why is obtaining all of them such a challenge for every OEM and battery materials startup? On a very general cell and materials level [7], the following are SoA cell specifications (*inserts battery jargon*):

• Cell capacity — the relative size of the battery [expressed in Ah]
• Cycle life — number of times the battery can charge reversibly before decay
• Charge rate — how fast the cell can be recharged without damaging it contents [expressed as a function of cell capacity: e.x [1C means charging a cell’s full capacity in 1 hr]
• Thermal stability — how well the materials can survive huge temperature shifts [several different ways to measure, normally through TGA or DSC]
• Specific energy — energy per unit mass [expressed in Wh/kg]
• Energy density — energy per unit volume [expressed in Wh/L]
• Power density — power per unit volume [expressed in W/L]
• Specific Power — power per unit mass [expressed in W/kg]
• Peak specific discharge power — power burst given in a short period [expressed in W/kg]
• Cost — how cheaply they can be mass-produced [expressed in $/kWh]
• Shelf life — longevity of cells if stored without use [expressed in # of years]

Table 1 below summarizes how sensitive the most important SoA cell performance characteristics are for consumer electronics, EV, and stationary storage energy storage applications.

Table 1: Sensitivity of cell specifications for 3 industries within which product specs do not vary much [8], [9], [10]; Orange (1–2) = very sensitive, Grey (3) = medium sensitivity, Green (4–5) = not sensitive

From the myriad of industries discussed in the previous section, some are very sensitive to certain SoA cell specifications, i.e. those features are an absolute must for products, whereas other specifications that are not so sensitive have greater room to work with [9]. EV cells require a constant interplay between various cell capabilities, making it difficult for new technologies to hit everything simultaneously while being produced in large quantities. Next-generation technologies such as solid-state batteries and silicon anodes are being hailed as potential candidates for breaking the energy density targets required to achieve mass penetration for EVs (Figure 1). Stationary and grid storage systems, on the other hand, are extremely sensitive to requirements such as cost, size, and durability whereas energy and power metrics do not matter much (Table 1). Unfortunately, current SoA battery technologies are not suited to supply adequately to this industry. Finally, only energy density is a rigid need for consumer electronic devices; all other attributes are customizable, which is why batteries are used in almost every mobile or laptop device.

Figure 4: Bottom-Up approach (verticals) of battery materials development. Right before mass production, there is constant feedback to lower verticals for product iteration/improvements

Now the main question still haunts us: what approach should one employ (specifically a new company/ startup) at different horizontals and verticals (Figure 4, 5) of battery development to ensure that the goal of product/industry specific penetration can be met?

Figure 5: Horizontals involved in battery development: ongoing simultaneously with all stages of materials development

Most of us see battery development divided rigidly between these horizontals and verticals, which is why some stages on the vertical timeline may seem more prolonged than required, mainly because something on the horizontal does not align well with the vertical growth. Therefore, in reality, I would like to describe battery development as a “criss-cross” undertaking between these two facets, rather than being viewed as discreet models of progress. This will allow us to see a technology evolution from a scientific and a business standpoint, rather than only focusing on one of them. For example, news headlines could go either as “Company X raised $100M to advance next-generation battery technology” (business) or “Company A cell cycle results could solve the problem of batteries’’ (scientific) rather than “Company XA’s long-lasting battery secures $100M to aid next-generation grid storage”. Changing the mindset to the latter is a stepping stone to application-specific considerations and design.

Next, understanding what type of fundamental materials design is needed to hit each specification individually, and then together is the key to mastering an application space. For example, this industries’ biggest scientific bottleneck has been the inability to simultaneously obtain high power and high energy due to inherent limitations in making both thick and porous (volume space between particles) electrodes [3].

Figure 6: Porous electrodes for high rate (Left); dense electrodes (Right) for high energy[3]; (reproduced with permission from Adrian Yao, CTO of Enpower)

The image above elucidates why it isn’t possible to have high energy and high power with conventional battery architectures. If electrodes are porous and thick, the particles will not adhere to the current collectors (grey and golden ends — Figure 6) and break apart upon repeated usage. This is a huge obstacle to the mass electrification of EVs — A high power and low energy battery pack means a larger pack, smaller car range, and increased cost for a larger pack. A high-energy, low-power battery pack means hours of waiting at a gas station during a lengthy road trip before one’s electric vehicle is ready to leave. Attempts at high energy and high power with different electrode architectures (mainly 3-D) and structures have historically yielded lower cycle life. With these inherent tradeoffs, it is clear that there is a need to cater materials design from the materials level to meet a certain specification.

What about applications such as EVs that have more stringent requirements? The answer lies in giving the customer what they want. Based on region-specific levelized cost of electricity (LCOE), traffic conditions, annual mileage, and other usage trends, it makes the most sense to have a product line that caters to these trends and shifts in response to changes in consumer behavior patterns. “Key Factors Influencing Consumers’ Purchase of Electric Vehicles” [11] describes consumers’ attitudes towards EVs based on various behavioral pattern research models. It concludes that the primary factors influencing their decision to purchase EVs are ease of use and control over resources required for purchase. Since a customer’s definition of “convenient” varies based on the aforementioned trend and behavioral patterns for car usage, having EV specs with some, but not all of these characteristics will benefit on a materials design as well as cost level (Table 2).

Table 2: EV Battery Specifications from a customers’ point of view

For example, a potential EV buyer who lives in the heart of Mumbai and must drive an hour to and from work in traffic would not care much about having an EV that can fast charge — they can plug in the car overnight. They would care more about cost and ease of usage: an average range and charging time would do. On the other hand, an upper middle class American suburban driver who makes long road trips would need a car that both has a decent range and can charge quickly (under an hour) at a station outlet. In this case, a lower cycle life would do if the car manufacturer provides battery leasing and swapping options. Thus, the final EV product will be shaped by a combination of these factors. For low cycle life car batteries that have both high energy and high power, options to lease/swap batteries should also be provided.

Using EVs as an example, similar techniques of customer points and co-relation to materials design can be adopted for all other industries. Given the current SoA grid and stationary storage battery innovations, one can say that it is definitely headed in the right direction with sodium-ion and flow batteries.

Ending Remarks

Now that there is clarity on the why and how of application specific design from a consumer’s point of view, there must be sufficient time and effort spent in determining how to structure most, if not all, aspects of battery development to the end usage application. A combination of figures 4 and 5 is what this would be, rather than as discreet “horizontal” or “vertical” lines. Being mindful of the end customer while deploying resources at various stages will reduce R&D end-goal confusions and give streamlined direction. All this being said, if someone does indeed make a battery that can be used for everything, it will be a win-win for everyone.

About the author:

Asang Mehta is a Battery Engineer at Xerion Advanced Battery Corporation based in Dayton, Ohio. He graduated in May 2020 with a B.S. in Chemical Engineering & Materials Science from the University of Southern California.

Interested in publishing in BatteryBits? Apply at this link to become a contributor.

Join our discussion with other battery professionals at the community slack space Battery Street!

References

[1] Howell, D., 2021. [online] Nrel.gov. Available at: [Accessed 13 July 2021].

[2] Lambert, F., 2021. MIT startup claims a breakthrough for ‘holy grail’ of batteries, doubles energy density. [online] Electrek. [Accessed 10 July 2021].

[3] Yao, A., 2021. The Li(ttle) ion that could. [online] Honestenergy.substack.com. Available at: [Accessed 5 July 2021].

[4] (EESI), E., 2019. Fact Sheet | Energy Storage (2019) | White Papers | EESI. [online] Eesi.org. [Accessed 13 July 2021].

[5] Tyson, Madeline, Charlie Bloch. Breakthrough Batteries: Powering the Era of Clean Electrification. Rocky Mountain Institute, 2019 [Accessed 15 July 2021]

[6] Saft Batteries | We energize the world. 2021. Defense. [online] [Accessed 10 July 2021].

[7] Ghassan Zubi, Rodolfo Dufo-López, Monica Carvalho, Guzay Pasaoglu, The lithium-ion battery: State of the art and future perspectives, Renewable and Sustainable Energy Reviews, Volume 89, 2018, Pages 292–308, ISSN 1364–0321 [Accessed 9 July 2021]

[8] sofeast. 2021. [online] [Accessed 12 July 2021].

[9] IOP Conf. Series: Materials Science and Engineering 252 (2017) 012058 [Accessed July 12 2021]

[10] Congressional Research Service, 2019. Electricity Storage: Applications, Issues, and Technologies. EveryCRSReport.com. [Accessed 13 July 2021]

[11] Sustainability 2019, 11, 3863; doi:10.3390/su11143863