Introduction

The relentless advancement of modern technology is made possible by an electrical infrastructure that supplies the essential power and energy needed for future progress. Macro trends in computing, AI, mobile devices, and the electrification of transportation have combined to bring power generation and delivery to the forefront of public consciousness. 

The electrification of everything is now spurring innovation across all segments of electrical infrastructure, emphasizing the importance of energy storage. What was once in the background is now at the forefront. New requirements and technologies are building on each other, resulting in an explosion of new products and applications.

Examples abound:

• Rooftop solar installations are now large enough to cause load defection.
• In the opposing direction, significant increases in electricity demand are jolting the global electric utility industry out of traditional growth rates.
• Energy storage technologies—mechanical, thermal, electromagnetic (capacitors), or electrochemical (batteries/fuel cells) —are rapidly advancing.

Among these, energy storage, batteries in particular, have swiftly transitioned from being overlooked for nearly a century to becoming the focus of major technological advancements, global investments, massive deployments, and even Nobel prizes. Batteries for electric vehicles (EVs), long-duration grid storage, and short-duration applications are now a central concern for technologists, general users, and governments worldwide.

This paper focuses on the expansion in segmentation and applications beyond traditional long duration energy storage (hours to days) to include short duration energy storage (minutes to microseconds) utilizing mechanical, thermal, electromagnetic, and electrochemical technologies. Note that all energy storage and short duration power systems share the characteristics of being charged and discharged when desired. This is distinct from power conversion systems which have no storage: examples of converters are wind turbines, solar panels, or diesel generators that convert wind, solar, and chemical power into electricity. Long duration energy storage applications have traditionally been labeled ESS (Energy Storage Systems) and long duration electrochemical (battery) technologies logically evolved to be known as BESS (Battery Energy Storage Systems). In this electrified everywhere age, short duration energy storage segments and applications have expanded rapidly based on the surge in use cases and new enabling technologies.

A fundamental difference between long and short duration energy storage is that storage capacity (i.e., energy density) is the highest value characteristic in long duration applications and discharge capacity (i.e., C rate or power density) is the highest value characteristic in short duration applications. Short duration energy storage examples here include mission critical power backup systems, commercial and military pulse power applications, and short duration industrial and grid power stabilization support. The widening gap between long and short-duration energy storage segments and applications, along with the emergence of new technologies tailored to each, necessitates the creation of a new category aptly named Immediate Power Solutions (IPS).

Under the umbrella of IPS, the short duration energy storage stakeholder community can now effectively identify and differentiate their requirements from the long duration labels used incorrectly due to lack of an alternative. An example of the effect of this overlap is the ever-growing family of lithium-ion rechargeable battery technologies, which are inherently high energy density chemistries. These products work well in long duration energy storage applications such as renewables grid storage and EVs. Due to familiarity or cost, or both, they have also been applied to short duration applications where high energy density is a poor fit or even a disadvantage. In this new era, one-size-fits-all application of battery technologies has become obsolete in favor of optimizing products and technologies for the specific applications.

The IPS category helps short duration users, technology providers and services suppliers self-identify and come together to optimize short duration energy storage implementations.

The widening gap between long and short-duration energy storage segments and applications, along with the emergence of new technologies tailored to each, necessitates the creation of a new category aptly named Immediate Power Solutions (IPS).

Immediate Power Solutions Defined

A requirement of short duration energy storage applications is the availability of instantaneous, high-rate power for a range of minutes to microseconds.

Applications for this type of short duration power delivery exist across multiple verticals including industrial and manufacturing, data center, electric vehicle charging infrastructure and even support of long duration energy storage and generation products as they ramp up to peak power. The IPS group of technologies and applications are vital to the growth of all power infrastructure with many of these applications being categorized as critical, mission critical, or life safety. The presence of instantaneous, high-rate power in a mission critical application requires the additional characteristic of safety for IPS technology to remain viable. Further, mission critical applications demand high reliability along with sustainability, which are now prime factors in all short duration energy storage use cases.

Given the above characterization, Immediate Power Solutions (IPS) can be concisely defined as: safe, reliable and sustainable, short duration, high-rate power technologies for critical applications.

Immediate Power Solutions (IPS) address the needs of electrical applications that are defined by the power levels and typically short duration response time required to maintain desired system operation. For example, a 480 VAC three phase UPS system in a datacenter that must provide 1 Megawatt of power within 10 milliseconds and for up to 2 minutes to assure the successful start and synchronization of backup generators. This is an Immediate Power Solution that requires AC and DC power, real and reactive power considerations, and energy typically discharged by mechanical, electromagnetic, or electro-chemical means. IPS is not a new idea, but frequency and severity of problems caused by power outages in computer systems, electric vehicles, medical monitoring equipment, etc. brought forth the importance of the category. As is often the case, new IPS challenges spur the development of new, innovative IPS solutions. Moreover, each new IPS solution will not only address its target application but will also be applicable to a range of similar challenges and applications.

Immediate Power Solutions (IPS): Applications and a Look Ahead

Now that we have defined Immediate Power Solutions (IPS), we can explore their current state and future directions. As illustrated in the previous section, there are multiple technological approaches to IPS, each with its own set of performance characteristics. This allows users to identify applications that fall into the IPS category and compare their requirements to the available technologies.

Understanding the IPS distinction is critical for matching the right technology to the use case. While the world demands more power and energy, not every application requires a long-duration Energy Storage System. Using ESS technology where it’s not the best fit is inefficient. Conversely, employing an IPS product for a 4-hour grid support application is impractical and costly.

Technologies within the IPS space are designed to serve immediate high-power needs for applications such as EV charging infrastructure, mechanical temporary overloads, critical power for data centers, and artificial intelligence (AI)/machine learning (ML) power influxes.

These use cases would be inadequately served by ESS products. For instance, ZincFive nickel-zinc (NiZn) batteries can bridge short duration power gaps (1 minute to 5 minutes) in multi-megawatt data center installations with a small footprint, and supercapacitors can fill millisecond voids for electronics to ensure proper function. There are IPS products tailored to each of these needs.

All examples discussed in this paper so far are existing applications with new challenges associated with immediacy, power, and duration. Consider electronic power supplies used for many applications. Power supplies typically have surge power specifications supported on a limited basis by internal capacitance. When placed in an IPS application, traditional power supplies don’t have the ability to respond properly to the new high variability loads. The additional challenges of meeting reliability requirements while safely supporting high variability loads create the opportunity to look to IPS technology as a solution.

New IPS technologies like batteries that exhibit superior power and energy density compared to capacitors, similar cycle life, and simple battery management characteristics are gaining traction in applications like power supply surge capacity.

Another common future IPS application example is microgrid power stabilization when exposed to unpredictable loads and availability of power sources. Microgrids have the same problems as utility scale grids, but on a small enough scale that introduction of a single stabilizing microgrid component can make a significant overall performance difference. IPS technology enables easy management of hybridized fuel cells, diesel, and gas turbine generators by allowing any type of generator to provide immediate power at “time zero.”

A final unique category of the issues IPS can resolve includes unrecognized problems associated with the status quo. Consider a factory that sequentially starts all its process motors to avoid power surges that exceed utility-supplied power limits. An IPS solution can help this factory achieve the operational efficiency benefits of simultaneous motor starting. This type of IPS application example highlights the direct benefits of an IPS solution to both known problems and yet-to-be discovered future IPS application opportunities.

Conclusion

The IPS technologies and applications are essential for the expansion of power infrastructure, with many being classified as critical or mission critical. Clearly differentiating between ESS and IPS is just as significant for addressing power infrastructure challenges effectively. Consequently, this significance necessitates its own distinct category in the market to address the unique demands and ensure robust solutions for these vital applications. Selecting the right IPS solution for each application is essential as we continue to face and solve the power issues created by and resolved by the developing IPS infrastructure.

Executive Summary

The world’s business happens in data centers. While data centers are typically associated with technology companies, every industry in every corner of the globe needs space to store and process valuable data.

To meet the demands of the modern, AI-driven era, data center capacity will have to dramatically expand. Data center operators need to pursue greenfield projects, building new facilities from the ground up, as well as brownfield projects that modernize older data centers or retrofit other kinds of existing structures.

Creating more capacity is a straightforward goal. However, greenfield and brownfield data center buildouts come with their own respective sets of challenges. Greenfield and brownfield projects also offer different advantages, and an organization has to be strategic in their approach.

This paper aims to assess both strategies. After tracing the evolution of the data center market to the current capacity crunch, this paper identifies the key criteria for any project: meeting cost, sustainability and safety objectives. It details why these are a data center operator’s main priorities and how to meet them.

Next, this paper dives into why an organization may want to embark on the costly, ambitious endeavor of a greenfield data center buildout and what challenges may arise. Then it explores the advantages of brownfield buildouts, which are especially appealing for any organization with substantial investments in existing infrastructure that still hold value.

Lastly, in both greenfield and brownfield buildouts, immediate power delivery is an essential consideration. This paper demonstrates how nickel-zinc (NiZn) battery chemistry, an innovation in battery technology led by ZincFive, delivers immediate power while helping to address many of the obstacles that slow down both greenfield and brownfield projects.

Given the major investments needed to build a modern data center — whether from the ground up or otherwise — decision-makers are sure to consider the most reliable, future-proof technologies available. NiZn battery technology is becoming more common in the data center world as infrastructure operators look for safe, sustainable, cost-efficient ways to grow.

Background and Current Challenges

The data center market has its origins in the mid-1900s, when IBM housed mainframes in designated computer rooms. Since then, the concept of a data center has come to refer to any facility that offers dedicated space for centralized computer and telecommunications systems. In addition to the requisite compute, storage and networking hardware, data centers comprise supporting infrastructure such as power sources, backup power systems and cooling systems.

An enterprise that needs to leverage computing or telecom capabilities has a few options. First, it can build an on-premise data center for its own use. Alternatively, it can use a colocation facility, where computing resources like servers and storage devices are available for rent. The colocation host covers expenses like power, cooling infrastructure and physical security, benefitting from the economies of scale that come from hosting multiple customers. An enterprise can also choose to run its digital operations via cloud-based applications housed in hyperscale data centers — massive facilities run by technology giants like Amazon, Google, Meta and Microsoft.

The amount of data that flows through data centers has skyrocketed in the digital age. One way to measure data center growth and activity is by their power consumption; this helps us gauge the power draw of the actual servers housed within a data center, which may not directly correlate with a facility’s physical footprint.

In 2014, data center power consumption in the US amounted to approximately 7 gigawatts. By 2022, it reached 17 GW, and it is expected to reach 35 GW by 2030, according to McKinsey analysis¹ — more than doubling in less than a decade. Meanwhile, most of the growing demand for data center power is coming from hyperscalers and colocation facilities. In 2014, enterprises were responsible for nearly 60% of data center power consumption; they now account for around 20%.

Data center demand is growing because commerce, communication and other elements of everyday life are increasingly happening in the digital sphere. The beginning of the AI era is sending demand even higher; enterprises and organizations are taking a growing interest in generative AI, which requires more processing power and more storage.

To keep up with this demand, data center operators will need to create more space. As of mid-2023, data center vacancy rates were declining across the globe². In Q1 2023, Northern Virginia — by far and away the largest data center market — had a vacancy rate of just 1.8%, down from 2.6% a year earlier.

Given market realities — such as data center operators’ budget constraints and the sunk costs of legacy infrastructure — meeting demands for capacity will require a two-fold approach. First, many organizations will find it worthwhile to build entirely new, modern data centers. Building from the ground up on a previously-undeveloped site is known as greenfield construction. Alternatively, it may be more logical for an organization to retrofit existing facilities. Modernization projects that leverage existing facilities are referred to as brownfield projects.

Both brownfield and greenfield data center projects face a number of challenges, including supply chain disruptions, talent shortages, construction delays, a changing regulatory environment and real estate constraints.

Additionally, the industry will have to overcome major power supply constraints, a challenge that’s slowing down data center development around the globe. Supplying sufficient power to data centers will require adequate power generation, as well as improved transmission and distribution. Data center developers can also tackle the challenge with infrastructure that uses power more efficiently.

Given the myriad challenges facing the IT industry, there are key considerations for any data center buildout.

Key Criteria for Greenfield and Brownfield Projects

Any brownfield or greenfield data center buildout must be cost effective, meet increasing sustainability mandates and meet safety requirements.

Data center operators are making major investments to expand and upgrade their facilities. The global data construction market is forecasted to grow from $50.34 billion (as of 2022) to $73.43 billion by 2028. A decade ago, data center spending accounted for roughly 36% of overall IT spending. Now, it’s projected to account for nearly 53%.

Hyperscalers continue to make eye-popping investments in new facilities. Amazon Web Services in January 2024 announced it would spend 2.26 trillion yen³ (15.3 billion USD) on cloud infrastructure in Tokyo and Osaka between now and 2027. At the same time, Google announced a new $1 billion investment⁴ on just one new data center on a 33-acre site in the UK.

Given the sizable investments at stake, data center operators want every dollar well spent. Furthermore, organizations face serious macroeconomic headwinds such as persistent inflation combined with high interest rates, global political uncertainty, and supply chain disruptions that have increased construction and procurement costs. Limited real estate availability also drives up costs in regions where data center operators want to expand their facilities. Meanwhile, data center operators must look past their initial construction costs and consider the evolution of operating expenditures as well as capital expenditures.

Data center buildouts can keep costs in check with wise investments in different components. Cooling systems, for instance, can account for 15% to 20% of the overall cost of a data center project⁵, making the system design a financially critical decision.

Nickel-zinc batteries, meanwhile, can reduce equipment costs in multiple ways. ZincFive’s battery cabinets offer the smallest footprint-per-watt on the market, minimizing the amount of costly real estate that must be dedicated to backup power systems.

NiZn batteries are also low maintenance and have long life; ZincFive battery cabinets come with a 10-year warranty, as do some lithium-ion battery cabinets. By comparison, traditional lead acid battery cabinets come with a warranty lasting around 3 to 7 years. Additionally, since nickel-zinc battery technology has no thermal runaway at a cell level, it can operate at a wider temperature range than competing battery technologies, saving money on cooling systems.

A decade ago, data center spending accounted for roughly 36% of overall IT spending. Now, it’s projected to account for nearly 53%.

BC 2 Battery Cabinets Powered by Nickel-Zinc Batteries

Increasingly, every data center buildout must consider its environmental impact. Data center architects and operators are expected to rely on renewable energy when possible, minimize energy consumption and use energy as efficiently as possible. These goals are tracked in part by measuring a data center’s power usage effectiveness (PUE), water usage effectiveness (WUE), and carbon footprint.

As governments grapple with climate change, they’re imposing new sustainability requirements on the data center market and the broader economy. The European Union, for instance, updated the Energy Efficiency Directive (EED)⁶ in 2023, requiring data centers to monitor and report their energy consumption and emissions. In the United States, states have largely driven the regulatory environment. For instance, California last year adopted two new laws⁷ requiring companies to disclose more information related to their greenhouse gas emissions and climate-related financial risks. Corporate entities are also feeling the pressure to achieve certain sustainability goals from their shareholders and the general public. According to Gartner⁸, 2022 was the first year surveyed CEOs called environmental sustainability a top 10 strategic business priority.

There are a variety of tactics that data center architects and operators can leverage to meet these sustainability requirements. Hyperscalers, for instance, are adopting innovative air-cooling solutions to drastically cut back on data center water usage.

NiZn batteries are more environmentally friendly than other battery chemistries. A Climate Impact Profile by Boundless Impact Research and Analytics⁹ found that, compared with lead-acid and lithium batteries, NiZn has advantages with lower GHG emissions, water footprint, energy footprint, and other sustainability metrics. Specifically, NiZn batteries’ lifetime greenhouse gas emissions are 4x lower than lead-acid and 6x lower than lithium-ion emissions. Meanwhile, a NiZn battery demands 96% less water from cradle to gate than the average lithium-ion battery. The energy footprint, manufacturing to gate, for NiZn is 23%-33% less than that of lithium-ion batteries and lead-acid pure-lead batteries.

Nickel-zinc batteries use common, widely available, conflict-free materials. Nickel and zinc are abundant in the Earth’s crust, respectively, 4x and 5x more than lithium and lead. They are also fully recyclable, environmentally friendly with a low carbon footprint, and not hazardous.

Because NiZn batteries are highly recyclable, they are a logical component of a “circular economy.” The concept of a “circular economy,” which is gaining traction in corporate conversations, refers to a holistic system that minimizes waste and pollution, keeps products in use, and regenerates natural resources.

As more data centers are built around the globe, delivering mission-critical services for countless businesses and entities, safety is paramount.

Fires are not only an obvious safety risk within the data center but also a real business risk. The Uptime Institute found that 7% of data center outages in 2023 were caused by fires¹⁰. While not all data center fires are the result of batteries, components such as lithium can certainly accelerate fires, potentially destroying millions of dollars of servers.

Consequently, data center operators have to invest in proper fire suppression, deflagration venting and other safety features. Alternatively, they can invest in safer technology.

NiZn battery chemistry is inherently safer than both lead-acid and lithium-ion, with no thermal runaway at the cell level. The aqueous electrolyte lacks the flammability and reactivity of organic solvent-based electrolytes used in lithium batteries, which, when failed, can emit toxic fumes.

Compared to lead-acid batteries, ZincFive’s nickel-zinc batteries do not out-gas during normal operation.

UL 9540A testing shows that ZincFive’s nickel-zinc batteries do not exhibit thermal runaway, making them non-flammable and non-reactive to air and water.

The result is reduced safety-related infrastructure, lower costs, and peace of mind.

Greenfield & Brownfield Data Center Projects

Greenfield Data Center Projects

As the demand for data center capacity continues to grow, hyperscalers are leading the way in greenfield deployments. Hyperscalers have the capital to start from scratch. It takes significant resources to scout out available real estate in strategically sound locations, and then build an entirely new facility that can accommodate leading-edge innovation. A new hyperscale data center typically costs upwards of $200 million, with the potential to reach into the billions.

In 2023, hyperscalers operated approximately 900 data centers worldwide, according to Synergy Research Group, accounting for about 37% of worldwide capacity. Overall data center capacity is expected to double by 2028, with hyperscalers accounting for more than half of it.

Building a greenfield data center is about more than simply adding capacity. As hyperscalers extend their reach to new markets and geographic regions, they need to bring their infrastructure closer to their customers. Furthermore, cloud customers are increasingly reliant on distributed workloads, leveraging data centers in different regions of the world to better serve their own customers. More distributed infrastructure also allows cloud customers to better utilize edge-based technologies.

Building from the ground up also gives an organization the opportunity to take different design approaches. Traditionally, data center buildouts take a stick-built approach: materials are delivered to the site of construction, and the facility is built from the ground up. However, data center operators are increasingly turning to modular designs, relying on pre-built components to quickly establish a larger data center footprint.

Modular data centers represent a small but growing portion of the market: in 2022, they accounted for 3.6% of overall data center revenue, according to research firm Omdia. That figure should reach 5% by the end of 2026. The modular data center market in 2026 is forecast to be worth $5.25 billion, up from $3.25 billion in 2023.

In some cases, an organization may want to take a hybrid approach, building a custom-designed data center that includes pre-built power or cooling modules. With current mainstream battery technology, this can be a challenging approach. With some lithium-ion batteries, for instance, a standalone power module would need to include an HVAC system to ensure a maximum temperature within the module of 28 degrees Celsius. It would require a fire suppression system and a deflagration vent. The entirety of the internal structure would have to meet higher burn ratings. On top of all that, these changes to the structure of the container in some areas require different permitting, taxation, and can hamper the data center’s overall footprint.

By comparison, with NiZn battery technology, the option of installing a power module outside of a main server building becomes significantly simpler, safer and more cost effective. Given that NiZn batteries have no thermal runaway at the cell level, the modular unit does not require a fire suppression system. Additionally, container operations can reach up to 35 degrees Celsius; thus, it reduces cooling costs and reduces the footprint occupied by the HVAC system. NiZn batteries also offer higher power density, discharging high levels of power quickly. Compared to conventional lead-acid batteries, ZincFive’s NiZn batteries offer up to 3x the power density while being half the size and one third of the weight.

All told, using NiZn batteries instead of lithium-ion batteries can reduce the length of a modular power container by several feet. This is noteworthy, given that the average container costs approximately $10,000 per foot. Additionally, ZincFive’s modular battery cabinets are significantly more cost effective because the batteries ship from the ZincFive factory to their destination inside the cabinet. Due to the volatile nature of lithium-ion batteries, some brands must be shipped separately and installed upon arrival.

Meanwhile, hyperscalers intent on optimizing the footprint of new data centers have resources they can leverage, such as the Open Compute Project’s data center design principles.

For instance, the collaborative industry organization — helmed by major companies including Meta, IBM, Intel, Nokia, Google and Microsoft — opted to include a distributed offline backup power strategy in its design principles, as opposed to a standard inline centralized UPS for backup power. ZincFive’s in-rack battery backup units allow data center organizers to take a distributed approach.

As with any major capital investment, a number of external factors can impact the construction of a greenfield data center. Macroeconomic headwinds, global political uncertainty and supply chain challenges all impact the data center market. In key markets, the limited availability of land — whether due to high costs, political interference or both— is a major challenge. 

Northern Virginia, for instance, is the largest data center market in the world and remains a desirable location for new infrastructure. However, data center buildouts have experienced delays in recent years, in part because of local opposition.

Limited power capacity is also a serious constraint on the greenfield data center market. In Singapore, for instance, the government in 2019 enacted a three-year moratorium on data center construction in response to the energy consumption they require. Meanwhile, in London — the second-largest data center market — “a key electricity substation upgrade in the western corridor has been delayed,” according to CBRE. “As a result, securing power from the grid operator is almost impossible for the next few years. This is a problem for the hyperscalers, given their desire to expand their availability zones in the western corridor.”

Brownfield Data Center Projects

While greenfield buildouts will significantly contribute to the growth of the overall data center market, breaking ground on an entirely new facility isn’t always the logical way to build more capacity.

Organizations of all kinds — including hyperscalers, enterprises, medical facilities and educational institutions, to name a few — have existing data centers or other sites that they can retrofit with modern data center technology. Brownfield buildouts allow organizations to leverage the real estate already in their possession, as well as legacy IT equipment. By pursuing a brownfield buildout, an organization can potentially avoid the regulatory hurdles that can slow down greenfield projects. All told, brownfield projects can be completed relatively quickly and with a smaller budget.

Brownfield projects are sure to look more appealing to enterprises ready to integrate AI into their workflows. While hyperscalers are already making significant investments in AI-powered products, the opportunities to leverage AI extend far beyond the walls of Microsoft or Google. Nearly half of organizations surveyed last year said they were evaluating the business use cases for AI. However, to keep up with the demand generated by AI, data center capacity will need to grow by nearly 300%, according to one estimate.

Modular data center expansions provide ways for organizations to quickly add capacity to existing operations. For instance, an enterprise could add a power module unit outside of its existing data center, delivering more power capacity to support increasing compute power within a facility. Alternatively, adding a power module outside of the data center would allow an enterprise to strip out old power and cooling infrastructure that sits within the facility, occupying valuable floor space. As is the case for greenfield projects, brownfield projects can add power modules that leverage NiZn battery technology for simple, safe and cost effective power expansions.

NiZn batteries also provide an option for data center operators who want to replace existing lead-acid batteries with a safe, longer-lasting and more sustainable option. ZincFive offers a NiZn drop-in replacement for lead-acid UPS batteries. It is adapted to use the same charging system as lead-acid batteries, making the replacement process seamless. NiZn batteries have an operating life up to 3x that of lead-acid batteries, thanks to the stable, non-corroding positive nickel current collector in nickel-zinc batteries.

NiZn batteries also give data center operators a way to make existing facilities more sustainable. Generally speaking, sustainability goals are easier to reach with greenfield projects, which can use the most sustainable components for every part of the facility. Brownfield projects may include building materials that are less energy efficient, for instance. Older buildings may also have different layouts that simply make them harder to cool efficiently. However, the cradle to grave carbon footprint of a nickel-zinc battery is significantly less than lead-acid or lithium batteries.

Additionally, NiZn batteries’ lifetime greenhouse gas emissions are 4x lower than lead-acid and 6x lower than lithium-ion emissions. Furthermore, nickel-zinc batteries use common, widely available, conflict-free materials.

Brownfield projects face many of the same external challenges as greenfield projects, including economic headwinds and supply chain disruptions. While limited real estate can be a challenge for greenfield projects, it may counterintuitively serve as an advantage for brownfield buildouts — there may be economic incentives available for organizations that can modernize older buildings.

Data center operators may also run up against resistance within their own organization, if leaders are inclined to use existing infrastructure as long as possible before undertaking modernization projects. However, technologies like NiZn can help make a compelling case for facility upgrades. For one thing, NiZn batteries are more reliable than lead-acid or lithium-ion batteries — the cells remain conductive even when weak or depleted. Most leaders would agree that the risk of letting business-critical systems fail is not worth taking simply to squeeze a few more years out of existing battery backup systems. Furthermore, NiZn batteries can help reduce operating costs by reducing the need for safety infrastructure.

Conclusion

Data center operators have reached a critical point where they must decide how to add capacity to their operations. If they do not, they risk being left behind while competitors host new, in-demand, power-hungry AI tools.

There are a few key considerations for every data center buildout: cost, safety and sustainability. Along with other bleeding-edge technologies, ZincFive’s field-proven nickel-zinc battery technology can help data center architects and operators meet requirements in each of these areas. Entities that need more data center capacity can pursue greenfield projects or brownfield projects. One route may make more sense than the other, given an organization’s size, existing infrastructure, available capital and capacity needs.

In each of these scenarios, incorporating nickel-zinc battery technology is a wise choice.

Citations

1 https://www.mckinsey.com/industries/technology-media-and-telecommunications/our-insights/investing-in-the-rising-data-center-economy

2 https://www.cbre.com/insights/reports/global-data-center-trends-2023

3 https://press.aboutamazon.com/aws/2024/1/aws-plans-to-invest-2-26-trillion-yen-into-its-japanese-cloud-infrastructure-by-2027

4 https://www.prnewswire.com/news-releases/google-to-invest-1-billion-in-united-kingdom-data-centre-302038632.html

5 https://www.businesswire.com/news/home/20230703133703/en/Worldwide-Data-Center-Construction-Market-Report-2023-A-73.43-Billion-Market-by-2028—Market-Growth-Enablers-Restraints-and-Trends—ResearchAndMarkets.com

6 https://energy.ec.europa.eu/topics/energy-e³ciency/energy-efficiency-targets-directive-and-rules/energy-effciency-directive_en

7 https://watershed.com/blog/california-sb-253-and-sb-261-a-guide-for-companies

8 https://www.gartner.com/en/newsroom/press-releases/2022-05-18-gartner-survey-reveals-signifcant-shifts-in-ceo-thinking-on-sustainability-workforce-issues-and-infation-in-2022

9 ZincFive Climate Impact Profile

10 https://uptimeinstitute.com/resources/research-and-reports/annual-outage-analysis-202

Executive Summary

Advanced energy storage solutions are increasingly needed to transition the electricity grid, transportation, building and industrial sectors towards renewable energy sources. ZincFive’s nickel-zinc battery is a high-capacity battery with environmental and safety advantages. The materials comprising ZincFive’s battery are non-flammable and environmentally benign compared to lithium-ion and lead acid batteries. ZincFive’s ecological advantages and performance specifications drive its demand in a wide range of high-growth applications, such as data centers, intelligent (communicating) transportation, motive and start-stop applications.

Alignment with SDGs

Climate Impact Score: 9.4/10

Boundless performed an environmental assessment of ZincFive’s nickel-zinc battery, to quantitatively measure environmental outcomes and illustrate the relative benefits of its technology compared to traditional battery chemistries such as lithium-ion, lead-acid and sodium sulfur. Using its customized life cycle assessment (LCA) methodology, Boundless measured the impact and retained an independent battery industry expert to review the findings.

The assessment identified and analyzed the Carbon Return on Purchase (CROP), GHG Footprint, Energy Footprint, Water Footprint, Volatile Organic Compounds (VOC) Footprint, Carbon Payback Time (CPT) and the Levelized Cost of Storage (LCOS) of ZincFive’s battery. The life-cycle inputs and impacts were evaluated considering raw material production, procurement and battery cell fabrication. Results were normalized relative to one kWh of stored energy and compared against lead-acid variants including pure lead (PbA Pure Lead), absorbent glass mat (PbA AGM), and acid–gel (PbA Gel), and lithium-ion variants including nickel-manganese-cobalt (Li-Ion NMC), nickel-cobalt-aluminum (Li-Ion NCA), and iron-phosphate (Li-Ion LFP), as well as sodium sulfur (NaS).

Using a formulaic comparison to measure relative performance across all metrics, ZincFive’s technology scored a 9.4 out of 10 impact score on its overall performance. This Boundless Climate Impact Score is based on per kWh impact for above mentioned performance indicators and shows that ZincFive has significant environmental advantages over its competitors. The Climate Impact Score includes an analysis of the Levelized Cost of Storage of ZincFive’s nickel-zinc battery. however this metric is excluded from this report for confidentiality reasons. Details to the score rationale can be found in appendix F.

Boundless assessed the Carbon Return on Purchase (CROP) metric using technology specific assumptions for energy storage and standardized depth of discharge rates to compare alternatives on an energy basis. The CROP metric measures the greenhouse gases avoided by ZincFive’s customers per kWh of customer energy storage. Analysis showed that ZincFive’s customers can realize significant GHG savings by investing in ZincFive’s nickel-zinc battery, compared to lithium-ion, lead-acid and sodium sulfur batteries. Customers purchasing ZincFive’s battery can save up to six times more GHG emissions compared to lithium-ion NMC and NCA batteries, and even higher compared to lithium-ion LFP batteries due to their relatively high GHG footprint. Analysis showed that up to four times more GHG emissions can be saved compared to lead-acid AGM and gel batteries. Per one-million dollar ZincFive batteries purchased, 148,255 tonnes of CO2e savings can be realized by enabling renewable energy to enter the grid. This can be compared against 23,243 tonnes of CO2e savings per one-million dollar lithium-ion batteries purchased (averaged of all lithium-ion chemistries in this analysis), and 8,251 tonnes per one-million dollar lead-acid batteries purchases (averages of all lead-acid chemistries in this analysis). The results of the complete analysis have been summarized in the spider-chart on the next page.

The ZincFive battery GHG Footprint was estimated to be 59 kgCO2e per kWh of stored energy, which is significantly lower than the GHG Footprint of lithium-ion (Li-Ion), lead-acid (PbA), and sodium sulfur (NaS) batteries. 38% of the GHG Footprint for the ZincFive battery is derived from the two main materials that compose the battery, zinc oxide (13%) and nickel(II) hydroxide (15%). ZincFive batteries only use safe and abundant materials that mitigate battery hazards, health risks and scarcity concerns. ZincFive’s battery uses only a small amount of cobalt compared to lithium-ion batteries, so the sustainability concerns around this material are minimized. Both zinc and nickel are relatively abundant materials, four-times and five-times more abundant than lithium in the earth’s crust, respectively.

In addition to the CROP and the GHG Footprint of ZincFive’s battery, Boundless also analyzed the Carbon Payback Time (CPT), which measures the time it takes for a battery to offset its GHG Footprint by supporting more renewable resources to supply the electricity grid. The CPT was estimated to be between 0.16 and 0.21 years, four times faster than lithium-ion and lead-acid batteries and up to six times faster than sodium sulfur batteries. All other batteries reported a potential CPT exceeding one year.

The Energy, Water and VOC Footprint of ZincFive’s battery were also analyzed. Volatile Organic Compounds (VOCs) are emitted as gases from solids and are known for creating short- and long-term adverse health effects. Unlike lithium-ion and lead-acid batteries, the ZincFive battery does not use VOCs in production. The Water Footprint of the ZincFive battery, including water requirements for raw material extraction, was estimated to be 96% lower than the average Water Footprint of lithium-ion batteries. For comparison, the Water Footprint of lead-acid batteries is 99% lower than the average lithium-ion battery, and roughly three times lower than ZincFive’s battery. Lastly, the Energy Footprint of ZincFive’s battery was estimated to be between 20 and 35 percent less than lithium-ion batteries, sodium sulfur batteries and lead-acid pure lead batteries. It was found to be slightly higher than lead-acid AGM and lead-acid gel batteries, mainly because these batteries are highly recyclable, and a strong recycling infrastructure is in place. 

Boundless engaged Dr. Kent. J. Griffith, a battery chemistry expert and postdoctoral researcher at Northwestern University, to review the ZincFive assessment and validate the assumptions made in calculating the environmental metrics. Dr. Griffith concluded that the inputs to the study are detailed and correct, and that the ZincFive battery was compared to a diverse range of relevant energy storage technologies. A summary of Dr. Griffith’s review is included in Appendix D.

Environmental Key Performance Indicators (EKPIs)

We evaluated the life-cycle inputs and impacts per kWh for the ZincFive battery, considering raw material production, procurement and battery cell fabrication. Results are normalized relative to one kWh of stored energy and compared against lithium-ion, lead-acid, and sodium sulfur technologies, more specifically for: lead-acid variants including pure lead (PbA Pure Lead), absorbent glass mat (PbA AGM), and acid–gel (PbA Gel), and lithium-ion variants including nickel-manganese-cobalt (Li-Ion NMC), nickel-cobalt-aluminum (Li-Ion NCA), and iron-phosphate (Li-Ion LFP), and sodium sulfur (NaS).

Calculations of environmental metrics used to determine climate impact benefit.

NOTES: Consistent with conventions within the financial sector, we use the Roman numeral “M” to denote “thousand” and “MM” for “millions.”

Carbon Return on Customer Purchase

Measures the greenhouse gases avoided by ZincFive’s customers per kWh of customer energy storage.

• The Carbon Return on Purchase (CROP) shows that ZincFive’s customers can realize significant GHG savings by investing in ZincFive’s nickel-zinc battery, compared to lithium-ion, lead-acid and sodium sulfur batteries.
• ZincFive’s CROP ranges from 21 to 26 kgCO2e per kWh energy storage. Customers purchasing ZincFive’s nickel-zinc batteries can save up to 6 times more GHG emissions compared to lithium-ion NMC and NCA batteries, and higher compared to lithium-ion LFP batteries. Up to 4 times more GHG emissions can be saved compared to lead-acid AGM and Gel batteries.
• Technology specific assumptions for energy storage and depth of discharge were used to compare alternatives on an energy basis, each with standardized 500 cycles per year for every year of operating life.
• Each kWh of stored energy (and associated losses) are assumed to be supplied by non-emitting electricity and displaces marginal U.S. grid electricity.

Carbon Payback Time

Time required for emissions savings from the product’s use to offset the GHG of its production. All scenarios assume 500 cycles per year for every year of operating life and that each kWh of stored energy (and associated losses) are supplied by non-emitting electricity and displace marginal U.S. grid electricity.

• The ZincFive battery takes between 0.16 and 0.21 years to offset the embedded carbon due to its production, which is significantly lower than the lithium-ion, lead-acid and sodium sulfur batteries.
• The Carbon Payback Time of lithium-ion batteries is on average 4.1 times longer than the Carbon Payback Time of the ZincFive battery. The carbon payback time for lead-acid and sodium sulfur batteries is on average 3.9 and 6.2 times longer.
• All other batteries report a potential Carbon Payback Time exceeding one year.

Product GHG Intensity

GHG emissions were measured as CO2 equivalent per kWh of stored energy.

• GHG emissions for the ZincFive battery ranged from 51.0 to 66.5 kgCO2e/kWh
• The GHG Footprint for PbA Gel and Pba Pure Lead was calculated starting with the GHG intensity of PbA AGM and replacing glass by fumed silica and recycled lead by virgin lead respectively.
• ZincFive’s estimated GHG Footprint is roughly half of its competitors. The GHG Footprint of ZincFive’s battery is 63% lower compared to lithium-ion batteries, 37% lower than lead-acid batteries and 54% lower than sodium sulfur batteries.

Energy Footprint

A measure of the energy input per kWh of stored energy.

• The estimated Energy Footprint of the ZincFive battery ranged from 581 MJ to 872 MJ per kWh.
• The Energy Footprint for PbA Gel and PbA Pure Lead was calculated starting with the energy intensity of PbA AGM and replacing glass by fumed silica and recycled lead by virgin lead, respectively.
• Energy Footprint estimates for the ZincFive battery were 33% less than lithium-ion batteries, 23% less than lead-acid Pure Lead batteries and 32% less than sodium-sulfur batteries.
• The Energy Footprint of PbA AGM and PbA Gel batteries is 12% lower than the ZincFive battery, this is mainly because these batteries can be efficiently recycled.

Solvent / VOC Footprint

Measures the Volatile Organic Compounds (VOC) required for manufacturing per kWh of stored energy

• The ZincFive battery uses no solvents or toxic chemicals. This is also the case for sodium-sulfur batteries.
• Average VOC Footprint of lithium-ion batteries is 5.5 grams / kWh.
• Average VOC Footprint of lead-acid batteries is 12 grams / kWh.
• Our independent expert review suggests that the values reported appear low for lithium-ion battery cathodes deposited via NMP solvent, because the typical solvent volume fraction may be approximately 30% and the cathode comprises a significant fraction (>1/3) of the cell.¹

Water Footprint

Water use for raw material extraction and the manufacturing process per kWh of stored energy.

• The Water Footprint of the ZincFive battery ranges from 7.2 to 10.8 gallons per kWh, and averages 9.0 gallons per kWh.
• The ZincFive battery requires 96% less water for material extraction and production than the average lithium-ion battery, whereas the Water Footprint of lead-acid batteries is 99%  lower than the average lithium-ion battery.
• Lead-acid batteries have a low Water Footprint because of their high degree of recyclability and recycling infrastructure.

Environmental Highlights

Summarized below are most relevant impact categories and codes that refer to the United Nation’s Sustainable Development Goals (SDGs). The associated metrics highlight the most important factors that explain how this technology is impacting the environment.

Appendix A: Methodology

Key Goals

Key goals of this analysis were to:

1. Examine environmental performance in conjunction with financial data to arrive at environmental and hybrid environmental-financial metrics for ZincFive’s storage technology versus existing technologies.
2. Provide equitable comparisons among relevant alternative technologies.
3. Incorporate a variety of methodological considerations that are relevant to the energy storage industry and which were expected to bear upon the results.

To ensure that these key goals were reached, an independent industry expert reviewed the study and assumptions to ensure that the methodology was coherent with industry standards. The expert review and commentary notes are provided in Appendix D.

Methodology 

To address the first goal, Boundless researched the material, energy, and performance characteristics for ZincFive’s energy storage technology, based on detailed information provided by ZincFive, describing the material components and energy inputs. At the core of the methodology is a life-cycle assessment (LCA) model for a kWh of stored energy on the ZincFive battery. The functional unit (FU) of this LCA was a kWh of stored energy, such that embodied energy and emissions are estimated for the battery production. We used SimaPro v9.0.0.41 and employed the IPCC 2013 methodology when calculating life-cycle impacts of material and energy systems not described elsewhere in the literature. The complete set of detailed calculations, impact assessment factors, assumptions, and references are available as Supporting Information (SI) upon request.

Each metric compares ZincFive’s technology against alternative technologies. Metric construction for industry alternatives relies on comparisons, for which we relied on scientific literature, industry reports, white papers, as well as assumptions provided by the industry expert. The impact metrics are reported graphically using bar charts to illustrate a baseline result value, along with sensitivity bars reflecting a range of possible result values around deployment scenarios and key variables.

Research Approach

• Followed a life-cycle analysis approach and leveraged professional LCA software/data and scientific literature.
• Investigated non-GHG metrics, including water footprint and minerals use.
• Accounted for emissions offsets occurring from hypothetical marginal electricity system impact assuming energy storage facilitated renewable generation on a 1:1 basis.
• Identified sources of uncertainty and quantified their impact on results.
• Included important financial and operational variables to estimate the cost of production.

Appendix B: List of Metrics

EKPI	Unit of Measure	Description
Energy Intensity	MJ / kWh	A measure of the energy input per kWh of stored energy.
GHG Intensity	kgCO2e / kWh	A measure of the greenhouse gas impact per kWh of stored energy.
Water Footprint	Gallons / kWh	A measure of the water use per kWh of stored energy.
Solvent / VOC Footprint	mg / kWh	A measure of the VOC avoided by using water-based manufacturing, measured per kWh of stored energy.
Carbon Payback Time	Years	A Measure of the time that it takes for a product’s use to offset the GHG of its production.
Carbon Return on Purchase	kgCO2e / kWh Installed	Measures the greenhouse gases avoided by customers per kWh of customer energy storage.

Appendix C: Summary of Life Cycle Product Inventory

Appendix D: Independent Expert Review

Independent Industry Expert

Kent J. Griffith holds a PhD in battery materials from the University of Cambridge, United Kingdom. He has ten years of experience in electrochemical and battery research and development. Kent is also the founder and CTO of a start-up company commercializing efficient, fast charging and high-power lithium-ion batteries based on new, patent-protected electrode materials. His experience in technical subfields includes cathode and anode chemistry, solid electrolytes for all solid-state batteries, nickel-rich NMC degradation and protection, fast charging battery applications, high power chemistries and electrode formulation, characterization and specification of batteries for individual applications (e.g. energy density, safety, power, variable temperature operation), mineralogy, materials synthesis and recycling.

Summary of Expert Review

Boundless Impact Investing analyzed the environmental impacts of the ZincFive nickel–zinc battery technology. The comprehensive life cycle analysis explicitly considered factors including air, water, carbon, levelized cost, and energy. The outputs of the report – e.g. environmental footprints and carbon payback time – are the result of evaluating individual cell components with data-supported environmental impact measures. The inputs are detailed and thus the assumptions are minimal. A diverse range of relevant energy storage technologies are included for comparison, including sub-categories of the major competing technologies: lithium-ion and lead-acid.

New battery technologies are emerging on the market now and in the next five years, and the trend is toward environmentally-friendly products. The strong trend toward high energy-density batteries is dependent on nickel-rich cathode chemistries, which have the simultaneous benefit of eliminating toxic cobalt. This report accounts for the trend toward nickel-rich LIBs as well as the variability in the nickel content of cells on the market. Water-based processing of cathodes is targeted but challenging, particularly for nickel-rich materials. For non-NMC/NCA chemistries, LFP is making movement for EVs in China, and lithium–sulfur will be coming online ca. 2023 from a large plant in Brazil. Silicon is partially-replacing graphite for lithium-ion anodes but the cost, energy, and environmental impacts are minimal at the present levels. Several other technologies one hears about, such as solid-state batteries and lithium metal batteries, are not mature and thus appropriately left out of this report. Cost may be driving the move away from cobalt and organic solvent processing/waste, but the environmental advantages will be there too. The analysis here focuses on the present state-of-the-art technologies with realistic considerations of materials, cost, and energy density.

The cost of energy storage depends strongly on the application. ZincFive batteries are suited for large-scale applications requiring a high degree of safety and reliability with a long shelf-life. Their energy density is intermediate between lead-acid and lithium-ion energy cells. However, the ZincFive stated power density is higher than conventional lithium-ion batteries, even when the latter are optimized in power cells. Thus the ZincFive technology may be well-suited for certain high-power applications. The Boundless report accounts for the variation in application of lithium-ion batteries with a range of energy densities that covers energy and power cells. For completeness, the metrics are evaluated on both a mass and energy density basis.

Battery collection and recycling practices are mature for lead-acid batteries but at a juvenile stage for lithium-ion batteries. The variation in lithium-ion battery chemistry, the minor contribution of each of many components, and the low resale value of most materials hinder lithium-ion battery recycling. The recyclability of the nickel–zinc battery is likely to be considerably better than lithium-ion because the electrodes are simpler composites than lithium-ion, and closer in a recyclability-sense to lead-acid, which has an excellent recycling record.

Appendix E: Global Warming Potentials

How Global Warming Potential Scenarios highlight the importance of investing in Emission Reduction Technologies 

The methane impact from emissions depends on which Global Warming Potential (GWP) is used. GWP is a metric measuring how much heat a greenhouse gas traps in the atmosphere up to a specific time horizon, relative to carbon dioxide. The larger methane molecule provides a warming potential that is 28-36X that of CO2 in a 100-year timeframe. (That is, over 100 years, methane traps 28 times more heat per mass unit than carbon dioxide). The lifespan of methane in the atmosphere was estimated at 9.6 years, and CO2 is much longer (estimated from 20-200 years). In the shorter 20-year timeframe, methane’s impact would, therefore, be 84-87X that of CO2, and the GHG savings for all landfill technologies would be greater. Investment in methane reduction using this shorter timeframe increases the return for investment by a factor of 2.2-3X. The 20-year timeframe is especially important when considering critical climate change mitigation efforts needed over the next two decades.

Appendix F: Score Rationale

Climate Impact Score

The climate impact value is a number (1=worse to 10=best). This number represents an overall indicator of a company’s climate impact performance against its most relevant industry competitors. The value is obtained by comparing the average of each resulting EKPIs for the company against its competitors. The score for each metric can be read from the summary Spider Chart of the profile for each product. The EKPIs are developed and displayed in the detailed graphs for both the target company and the competing companies.

ZincFive has a generally advantageous, performance when compared to its competitors. For example, ZincFive’s technology has a lower GHG Footprint than its competitors, but a higher Energy Footprint than its lead-acid competitors. Using a formulaic comparison to measure relative performance across all EKPIs, ZincFive’s technology scored a 9.4 out of 10 on its climate performance.

Appendix G: Report Development Team

Paul Meier, Director of Climate Impact

Paul has worked with industry, government and public interest groups on energy and environmental issues since 1995. His efforts have focused extensively on the use of energy systems modeling to support decision-making. Paul has led multi-disciplinary research efforts to evaluate energy alternatives at the national, regional, and state levels and spanning electricity, transportation, and building energy sectors. From 2006–2016, Paul served as a Scientist and Energy Institute Director at the University of Wisconsin-Madison. From 2016–2018 he served as Director of Engineering for Blumont Engineering Solutions. Paul has environmental engineering degrees from Purdue University and Clemson University and earned his Doctorate from the Nelson Institute for Environmental Studies at the UW – Madison. He is a licensed professional engineer.

Fernanda Avila Swinburn, Research Analyst

Fernanda graduated from Columbia University in 2018 with a Master’s degree in Sustainability Management and a focus on renewable energy, sustainability strategies, data analysis, and life cycle assessment. Prior to Columbia, she graduated from Universidad de Chile with a master’s degree in Electrical Engineering. Fernanda has experience modeling demand side management systems for micro-grids and renewable resources forecasting. Her work on these topics was recognized with the first place of the Eco-Logicas Monograph competition, given by the “Instituto para o Desenvolvimento de Energias Alternativas na América Latina”. She has worked as a consultant performing energy price projection and the modeling of power purchase agreements for developers and financial institutions. She also has experience developing sustainability strategies and life cycle assessment for organizations in different sectors, such as a music festival, a foundry plant, and a coffee roasting company.

Andreas van Giezen, M.S., Research Analyst

Andreas graduated from Delft University of Technology (TU Delft) in The Netherlands in 2018 with a Master’s degree in Management of Technology, focusing on Infrastructure & Environmental Governance. He received a special annotation with his degree for his thesis work focusing on sustainable development of technologies. Prior to TU Delft, he graduated from Inholland University of Applied Sciences with a Bachelor’s degree in Aeronautical Engineering. Andreas interned for research & development projects at universities in both the Netherlands and China and won a nationwide contest for engineering students active in the energy industry in the Netherlands. He was previously employed at an international engineering consultancy firm, researching the social and technical impacts of ultra-deep geothermal energy projects. Andreas also has experience with academic research on ocean plastic collection logistics.

Michele Demers, Founder, CEO

Boundless Founder and CEO Michele Demers has 20 years of experience as a philanthropy executive, strategist, and social entrepreneur. She is Founder and CEO of Boundless Impact Investing, a market intelligence platform that provides high-quality, objective, and actionable research and tools to family offices and private investors interested in maximizing the social and environmental impact of their investments. From 2010–2013, she was Vice President at Foundation Source where she built a knowledge platform on best practices in philanthropy that was used by a network of 1200 family foundations. From 2007–2008, Michele was the Director of Communications for Humanity United. She has been involved in the successful development of more than two-dozen philanthropic and nonprofit start-ups, including her own, Tattersall Consulting, from 2002 to 2007. Michele is regularly called upon for her innovative thinking about impact investing and social enterprise. She is a graduate of Pennsylvania State University and has a Master’s in International Relations and Communications from Boston University.

Jack Cederroth, Director of Operations

Jack is an accomplished global operations and platform management leader with more than thirty years’ experience in the Financial Services and Financial Technology arena. During that time, he has worked extensively creating, implementing, and supporting enterprise-wide data and analytics platforms and services for the investment community. He is versed and trained in Lean Six Sigma techniques and principles. Prior to joining Boundless, he was the Global Head of Operations at S&P’s Securities Evaluation business unit where he established and led a twenty-four-by-seven follow the sun customer support model for their enterprise reference data and evaluation feed products. As a member of the leadership team, he helped define the strategy for the sale and subsequent integration of the entity to ICE Data Services. Jack believes that effectively creating strategic alliances with partners, leaders internally and at client organizations is the key to successful business initiatives. He is a graduate of Fordham University with a B.S. in Finance.

About Boundless Impact Investing

Driven by the latest research by independent industry and academic experts, Boundless Impact Investing offers analysis, market trends, and evidence of best practices in a growing number of emerging sectors that address major social and environmental challenges. We are an advanced consulting firm that enables investors to connect with industry leaders and peers for expert analysis, diverse perspectives, and real-time collaboration. Our investor education and expert advisory services offer proprietary access to both subject-matter experts and other impact investors.

The information provided in this report by Boundless Impact Investing and accompanying material is for informational purposes only. The information in this report should not be considered legal or financial advice, nor an offer to buy or sell or a solicitation of an offer to buy or sell any security, product, service, or investment. Boundless Impact Investing does not make any guarantee or other promise, representation, or warranty as to the accuracy or completeness of the statements of fact contained within, or any results that may be obtained from using our content. Neither this content, nor the investment examples cited, should be used to make any investment decision without first consulting one’s own financial advisor and conducting one’s own research and due diligence. To the maximum extent permitted by law, Boundless Impact Investing disclaims any and all liability in the event any information, commentary, analysis, opinions, advice, and/or recommendations prove to be inaccurate, incomplete, or unreliable, or result in any investment or other losses.

Contact Us

Boundless Impact Investing

www.boundlessimpact.net

Michele Demers, CEO and Founder

mdemers@boundlessimpact.net

Citations

¹ Zakeri B and Syri S (2014) Electrical energy storage systems: A comparative life cycle cost analysis. Renewable and Sustainable
Energy Reviews

² https://periodictable.com/Properties/A/CrustAbundance.v.html

Introduction

From process control to retail, businesses everywhere are using data-driven strategies to improve their product and service offerings. They are building Internet of Things (IoT) infrastructures that collect large volumes of data from edge devices for delivery to cloud-based servers for processing. However, with its population of 50 billion IoT edge devices,¹ the huge volumes of data the IoT generates creates considerable challenges as well as opportunities.

The most effective way of extracting actionable information from these high data volumes is to use data analytics technologies including Artificial Intelligence (AI), Machine Learning (ML), and Deep Learning (DL). These are no longer futuristic concepts, but rather practices that are here today and being integrated with and deployed into a variety of business.

As a new generation of AI-driven compute intensity is accelerating rack-level power density, backup battery solutions must keep up with rack-level power density expectations.

Compute Density Drives Battery Backup Requirements

This rapid adoption of analytics has created an explosion in workloads which are both compute and power intensive. For example, OpenAI—an artificial intelligence research laboratory—has released an analysis showing that since 2012, the amount of compute used in the largest AI training runs has grown by more than 300,000 times.² In other words, the computer resources consumed by AI has doubled every 100 days.

This rise of AI/ML/DL is driving future rack power densities that will far exceed those of today. Rack-based servers already contain multiple CPUs/GPUs (with hundreds of cores) that will exceed 300W each in the near future.³ They are integrated with terabytes of memory and multiple high-speed communication channels. There will also be DDR5 RAM memory power and channel number increases, PCIe Gen4/5 bus power and lane increases, 100G+ Ethernet, and increasing NVMe protocol adoption, which will only be moderately offset by efficiency gains. All this will lead to accelerating rack power densities.

The appetite for compute solutions associated with the global digital transformation, coupled with the rise of AI/ML/DL, suggests that future rack densities will far exceed those of today.”

Dell 2020 Server Trends and Observations Brie

Together, this AI compute usage and compute density have created a turning point in power density requirements, which has implications for data centers, the server hardware within them, and, in turn, the batteries utilized for backup.

The massively increasing data center workload is gravitating to companies focused on vast cloud businesses, driven by industry leaders such as Amazon, Google and Microsoft. There are already over 500 hyperscale data centers operated by such firms.⁴ While this is a minority of the total data center population, hyperscale data centers consumed 47 percent of servers in 2020.

Some of these data centers’ operators are opting for power infrastructures with battery backup distributed out to individual racks, in architectures such as those defined by the Open Compute Project (OCP). However, within this scenario of sharply increasing rack power density, backup batteries must deliver more power while occupying less space; it’s the computer hardware rather than the battery that earns the revenue for the data center.

With twice the power density, NiZn batteries exhibit ½ the size and weight of comparable lead-acid batteries.

Battery Power Density is the Key Factor

With these continuing space constraints on in-rack power backup, battery power density will be the key competitive factor. While lead-acid battery technology has been the workhorse for decades, newer technologies are introducing fresh opportunities to meet the challenges of increasing power density in server racks. Nickel-zinc (NiZn) technology, in particular, has specific advantages over lead-acid solutions – and lithium-ion chemistry as well – in terms of performance, reliability, safety, cost, and eco-friendliness.

In particular, ZincFive NiZn battery backup solutions offer dramatically higher power density than lead-acid batteries when measured by either weight (Watt hours per kilogram) or by volume (Watt hours per liter). The size of the NiZn battery is reduced to half that of a comparable lead-acid type. This means that NiZn batteries have two times the power density and half the weight of lead-acid batteries.

NiZn batteries are also simpler to use in a rack format. They do not require trickle charging to maintain capacity performance, which simplifies system design and is more energy efficient. Additionally, unlike lead-acid batteries, NiZn’s alkaline chemistry does not sulfate over time and has a higher operating temperature range; another contributing factor to a significantly longer life with low maintenance. In fact, NiZn batteries have three times the product life compared to lead-acid batteries.

The Safer, More Sustainable Choice

Placing battery backup in the rack, instead of a separate UPS facility, heightens the need for entirely safe operation to protect employees and equipment. Data center operators concerned about the possibility of thermal runaway in lithium-ion batteries will be interested to note that NiZn batteries have been rigorously tested to the UL 9540A test method at cell level, and they did not exhibit thermal runaway in any of the five arduous and destructive test types in that test method.

Comparable lithium-ion battery systems require a Battery Management Systems (BMS) to manage safe battery operation during UL 9540A testing—a clear disadvantage.

NiZn batteries are also more reliable, in part due to their battery string behavior. When a lead-acid or lithium-ion battery cell fails, it creates a high impedance or an open circuit that halts string operation. By contrast, a weak or depleted cell in a NiZn battery remains conductive, allowing the string to continue operating. In addition, NiZn strings tolerate string imbalances to a greater degree than either lead-acid or lithium-ion systems. With rack-based battery storage being distributed by nature, this behavior serves to lessen maintenance activities and costs.

At the same time, increasing density and power levels of batteries in the data center can have implications for data center sustainability. In a recent Climate Impact Report, ZincFive’s NiZn batteries achieved the best overall score of all surveyed technologies.⁵

ZincFive NiZn batteries have been rigorously tested to the UL 9540A test method at cell level and they did not exhibit thermal runaway.

The High-density Batteries for High Density Compute

As rack-level compute density drives power density levels continuously upward, backup battery requirements rise accordingly. The high-power density of NiZn batteries makes them the ideal choice to keep up with the increased density requirements, with their smaller size and weight allowing easier integration into server racks. They also can reduce operating expenses by virtue of their superior safety, high reliability, wider operating temperature range, long life and simpler maintenance requirements. By simplifying the challenge of high density battery backup, NiZn batteries are the better choice for the data center operators chasing ever greater compute density.

Citations

¹ Dell EMC’s 2020 Server Trends and Observations – ‘Data is King’, p2

² Programmer Info – ‘An Exponential Law For AI Compute’

³ Dell EMC’s 2020 Server Trends and Observations – ‘Data is King’, p13

⁴ ZincFive White Paper – Optimizing Data Center Operations with NiZn Backup Technology

⁵ Boundless Climate Impact Profile August 2020

Data Center Sustainability

Greenhouse gas (GHG) emissions are the primary contributor to climate change, and companies, governments and the general public have all taken an active interest in aggressively minimizing these emissions. Investors, consumers and other stakeholders are increasingly expecting companies to report their GHG emissions and reduction goals, in order to assess their performance compared to competitors, gauge their preparedness for existing and anticipated regulation and ensure that the company’s values are aligned with their own.

Accurate, transparent reporting of emissions data is becoming a standard expectation across industries, including data centers. In addition to a data center operator’s internal emissions tracking and reporting, customers and partners that rely on these data centers as part of their business operations require access to this GHG emission data as well, incorporating them into their own reporting. Data centers are feeling pressure to not only operate sustainably, but also conduct comprehensive reporting to relay this information to their stakeholders.

However, despite the increased demand for data center emissions data, many reports have gaps in their coverage. Among these gaps is the reporting of scope 3 emissions (as defined in the GHG Protocol), which are the emissions resulting from activities from assets not owned or controlled by the reporting data center company itself.¹ For data centers, scope 3 emissions can be tied to activities such as facility construction, energy sources, cooling services, and uninterruptible power supplies (UPS) including energy storage systems.²

As supply chain sustainability draws increased attention from stakeholders, scope 3 emissions data need to be included in carbon accounting to provide an accurate assessment of a data center’s climate impact. Once this data is widely available, companies can ensure actions to reduce these emissions, including improving the sustainability of their energy storage practices using battery chemistries with low climate impact such as nickel-zinc (NiZn) used in UPS Battery Cabinets.

The Importance of Scope 3 Emissions Reporting

Oftentimes, companies will focus their efforts on tracking and reporting emissions directly tied to their own operations and electricity consumption. These include scope 1 emissions, which are direct GHG emissions that occur from sources that are controlled or owned by an organization, as well as scope 2 emissions, which are indirect GHG emissions associated with the purchase of electricity, steam, heat, or cooling.

Scope 1 and 2 emissions, however, fail to capture the emissions that a company is responsible for that occur away from its own facilities. Scope 3 emissions, which represent emissions associated with a company’s value chain, often represent the majority of an organization’s total GHG emissions.⁴

For most companies, 65-95% of their carbon emissions can be attributed to their supply chain and end use, but may ultimately not be reported if scope 3 emissions are excluded from carbon accounting.⁵

Scope 3 emissions often represent the majority of an organization’s total GHG emissions.

Thus, reporting and action focused solely on scope 1 and 2 emissions are incomplete, and don’t paint an accurate picture of a company’s carbon footprint.

Despite the significance of scope 3 emissions in relation to a company’s overall carbon footprint, scope 3 emissions quantification is currently not required by the GHG Corporate Protocol.⁶ Due in part to its optional nature and complexity, a low number of companies are reporting scope 3 emissions. According to MSCI, only 18% of the constituents of their investable market index were reporting scope 3 emissions in 2020, with even lower rates of reporting for the specific categories within the scope 3 accounting methodology.⁷ This lack of information on emissions associated with companies’ supply chains results in a significant gap in stakeholders’ ability to understand and compare carbon footprints.

Although the level of transparency is currently low, scope 3 emissions represent a significant opportunity for GHG emission reduction. Leaders are emerging, as over 3,000 companies have reported scope 3 emissions under the Carbon Disclosure Project, according to investor sustainability advocate Ceres.⁸ Companies that report on their scope 3 emissions stand out from their competitors who lag behind in their disclosure, and are able to identify and act upon areas of their supply chain that offer room for improvement. The measurement of these emissions is necessary for goal setting and action, and companies taking the lead can take advantage of the economic, reputational, and environmental benefits before others.

Leaders in reporting and acting on scope 3 emissions are also better equipped to manage their preparedness for existing or potential regulations. Starting in 2023, German companies will be responsible for social and environmental issues tied to their global supply chain networks.⁹ Investors and other stakeholders are paying attention to these global developments, and are seeking reassurance that companies are managing any risks they face should such regulations be enacted in the U.S. To improve performance and preparedness, the initial step that needs to be taken is to gather these data and improve transparency.

The Right Batteries Can Reduce Data Center Scope 3 Emissions

As data centers and their customers look to meet the growing expectation to disclose scope 3 emissions, they can turn to their energy storage systems that are part of the UPS as an opportunity to establish a lower carbon footprint compared to their competitors. Energy storage is an increasingly significant part of a data center’s business operations, and the battery chemistry tied to a data center’s UPS offers significant trade-offs in terms of the sustainability of their supply chain and environmental footprint. As data centers and their customers look to assess their scope 3 emissions, the battery chemistry used in their UPS systems should be measured, incorporated into goals and seen as an opportunity to stand out.

NiZn batteries used in the ZincFive BC 2 UPS Battery Cabinet represent a more sustainably sourced and environmentally friendly alternative to other batteries used in data centers such as lead-acid and lithium-ion. Boundless Impact Research and Analytics performed a life cycle analysis of lead-acid, lithium and nickel-zinc batteries to prepare a scope 3 emissions Climate Impact Profile of these battery types. For the first time, energy storage users can compare and utilize scope 3 level environmental data comparing lead-acid, lithium-ion and ZincFive’s NiZn batteries along key performance indicators including GHG emissions, water footprint, energy footprint, and hazardous material requirements.¹⁰ In multiple ways, ZincFive’s NiZn batteries proved a more climate-friendly option:

Material Use

Nickel and zinc are four and five times more abundant in the earth’s crust, respectively, than lithium and lead. In addition, while lead exposure is a global health concern and lithium’s reactivity to air and water makes it a fire hazard, nickel and zinc are safer and non-flammable.

Greenhouse Gas Intensity

Since nickel and zinc sourcing require fewer emissions, ZincFive’s estimated GHG Footprint is roughly half of its competitors. The GHG Footprint of ZincFive’s battery is 63% lower compared to lithium-ion batteries, 37% lower than lead-acid batteries.

Carbon Payback Time

Carbon Payback Time (CPT) measures the time it takes for a battery to offset its GHG Footprint by supporting more renewable resources to supply the electricity grid. NiZn chemistry’s CPT is between 0.16 and 0.21 years – four times faster than lithium-ion and lead-acid batteries.

Volatile Organic Compounds (VOCs)

Emitted as gases from solids – including those used to produce lithium-ion and lead-acid batteries – VOCs are infamous for causing short-and long-term adverse health effects. A healthier alternative, the ZincFive battery does not use VOCs in production.

Water Footprint

Even including water requirements for raw material extraction, the ZincFive battery still demands 96% less water than the average lithium-ion battery.

Energy Footprint

The energy footprint for ZincFive’s battery is 23-33% less than lithium-ion and lead-acid pure lead batteries.

The results of a formulaic comparison measuring relative performance across all metrics showed that ZincFive’s technology scores a 9.4 out of 10 impact score on its overall performance, with higher scores representing higher positive environmental impact.

Citations

1 https://www.epa.gov/climateleadership/scope-3-inventory-guidance/

2 https://www.datacenterdynamics.com/en/broadcasts/london/2020-virtual/case-study-scope-3-emissions-and-next-generation-sustainable-data-centers/

3 https://www.epa.gov/climateleadership/scope-1-and-scope-2-inventory-guidance

4 https://www.msci.com/www/blog-posts/scope-3-carbon-emissions-seeing/02092372761

5 https://www.cnbc.com/2021/08/18/apple-amazon-exxon-and-the-toughest-carbon-emissions-to-capture.html

6 https://www.nature.com/articles/s41558-020-0837-6#ref-CR14

7 https://www.msci.com/www/blog-posts/scope-3-carbon-emissions-seeing/02092372761

8 https://www.cnbc.com/2021/08/18/apple-amazon-exxon-and-the-toughest-carbon-emissions-to-capture.html

9 https://www.cnbc.com/2021/08/18/apple-amazon-exxon-and-the-toughest-carbon-emissions-to-capture.html

10 ZincFive Climate Impact Profile

Introduction

As cloud-based services and storage become central to business operations, data centers are under increasing pressure to ensure the resilience of their mission critical facilities. Data centers rely on uninterruptible power supply (UPS) systemsto secure continued operations during power outages and other disturbances, most often powered by lead-acid batteries. However, UPS systems that utilize nickel-zinc (NiZn) battery technology have specific advantages over lead-acid in terms of performance, reliability, safety, lifetime cost and climate impact.

Despite these advantages of alternative battery chemistries, most existing UPS systems were already designed to work with lead-acid batteries. Since each battery chemistry has its own charge/discharge profile, these UPS systems often would require additional hardware and re-engineering to accommodate non-lead-acid batteries. The time and cost needed for such modifications prevent the switch from lead-acid to alternative chemistries.

Thus, the best solution to this dilemma is a “drop-in replacement” battery storage system that not only enables the incorporation of alternative battery chemistries, but also functions in an existing UPS system without the need for additional modifications. This characteristic enables backward and forward compatibility that saves UPS Original Equipment Manufacturers (OEMs) and data centers time and money by integrating new battery types into traditional UPS systems without expensive modifications.

As the world’s first NiZn BESS (Battery Energy Storage Solution) product featuring backward and forward compatibility with megawatt class UPS inverters designed for lead-acid batteries, ZincFive’s BC Series UPS Battery Cabinet offers a drop-in replacement for battery storage systems in both new and existing UPS installations utilizing lead-acid batteries. Through its use, data centers can efficiently upgrade their UPS systems to harness the benefits of ZincFive’s NiZn batteries.

Why Switch from Lead-Acid to Nickel-Zinc?

Historically, most data centers depend on lead-acid batteries to power their UPS systems. The lead-acid battery was the first chemistry used and remains popular today, but alternative battery chemistries such as lithium-ion and nickel-zinc offer compelling value propositions driving many data center end users and OEMs to switch to a new battery chemistry.

Backward and forward compatibility saves UPS Original Equipment Manufacturers (OEMs) and data centers time and money by integrating new battery types into traditional UPS systems without expensive modifications.

Modern data centers bear a larger compute and power burden each year. Digital transformation, 5G, edge computing, the Internet of Things, artificial intelligence and machine learning all require a continually increasing level of computing power and density. For example, Dell expects that future rack densities will far exceed those of today,¹ which have already vastly grown. Since 2012, the amount of computing used for the largest AI training runs has expanded by more than 300,000 times,² and the computer resources consumed by AI double every 100 days.³

The rising power density per rack and the need to provide space for as much compute power as possible in the data center also applies to infrastructure UPS and energy storage densities need to increase to reduce the infrastructure footprint. Additionally, data centers are being designed and built both faster and larger, emphasizing these trends further.

The shortcoming of lead-acid batteries is particularly evident when compared to the capabilities of ZincFive’s NiZn batteries:

Power Density

NiZn batteries have higher power density than lead-acid batteries, so by choosing NiZn battery backup, data center designers can reduce the footprint needed for energy storage. In fact, NiZn batteries have twice the power density, and so can reduce the footprint by as much as 50% compared to lead-acid batteries. Similarly, NiZn batteries can deliver the same amount of power with half the weight of lead-acid batteries. This reduces the structural requirements of the UPS room. Smaller footprint and less weight will reduce capital expenditure (CapEx) costs associated with energy storage.

Safety

Data center operators are responsible for the safety of their facilities and personnel. NiZn technology is inherently safer than lead-acid or lithium batteries. NiZn batteries do not exhibit thermal runaway and are non-flammable, as proven through testing at the cell level using the Underwriters Laboratories UL 9540A test method.⁴ Their superior weight and safety makes the batteries easier to handle.

Reliability

Battery string reliability is key to avoiding unplanned maintenance: when a lead-acid or lithium-ion battery cell fails, it creates an open circuit that halts string operation. A weak or depleted NiZn cell, on the other hand, remains conductive, allowing the string to continue operating. This turns an emergency situation with other battery chemistries into a simple battery replacement at the next planned maintenance cycle with little cost and no operational impact.

Lifespan & Maintenance Costs

For lead-acid batteries, operating expenditures (OpEx) over the life of the UPS system can be more expensive than up-front CapEx. NiZn solutions dramatically lower OpEx through their much longer useful life and simplified maintenance. Additionally, the size and weight advantages of NiZn batteries compared to lead-acid batteries can lower the investment needed to house energy storage within data centers. NiZn batteries also have a wider operating temperature range than other technologies, which reduces demands on data center cooling systems. By lowering OpEx, ZincFive solutions reduce ownership costs over lead-acid based UPS products when looking across the total UPS useful life.

Sustainability

NiZn batteries are mainly composed of common, highly available, non-hazardous materials as opposed to lead-acid batteries containing significant amounts of lead and other hazardous materials. Even after a long operational life, NiZn is one of the most recyclable battery chemistries on the market. Unlike many other materials, both nickel and zinc can be recycled while maintaining their physical and chemical properties. ZincFive’s zinc electrode contains no hazardous lead, cadmium or mercury, and so is environmentally benign. In fact, NiZn batteries have the best climate impact score – 9.4 out of 10 – of backup battery chemistries typically used in data centers, according to a recent 3rd party Climate Impact Report.⁵

As rack-level compute density rises, so do the requirements for backup power in smaller, safer footprints. With their high power density, longer lifespan, sustainability and greater reliability, NiZn batteries are the ideal choice for backing up the increasing data center loads.

Why Alternative Battery Chemistries Need Backward/Forward Compatibility

While NiZn batteries’ characteristics give data centers more than enough reason to make the switch from lead-acid, one challenge often stands in the way: the average data center UPS system was not built with them in mind.

Rather, most data centers were designed specifically to charge lead-acid batteries. Since every battery chemistry has a different charge/discharge profile, alternative batteries can’t charge from the typical UPS system without a costly and time-consuming reengineering process.

Fortunately, one workaround exists to avoid that extra time and cost. A NiZn battery chemistry can, in fact, charge from a UPS made only for lead-acid battery profiles without additional hardware – as long as it has backward/forward compatibility, the ability to seamlessly operate with a UPS designed for a different battery chemistry than its own.

ZincFive’s BC Series UPS Battery Cabinet: the Key to an Easy Upgrade

ZincFive‘s BC Series Battery Cabinets, while housing NiZn batteries with all of their associated benefits, adapt the lead-acid charging system to the needs of the nickel-zinc battery. The battery cabinets’ intelligent charging & monitoring system emulates the profile of lead-acid batteries to the UPS. Since the UPS doesn’t “know” that the NiZn battery string has a different chemistry, it operates the string just as it would for its previous lead-acid inhabitants.

This way, ZincFive’s UPS Battery Cabinet’s specialized profile allows operators to use NiZn battery cabinets as “drop-in” replacements. Able to support 1MW of UPS output power with only four battery cabinets and the industry’s smallest linear footprint, this NiZn Battery Cabinet offers a lower TCO, lower maintenance and higher performance alternative for data center UPS energy storage.

With ZincFive’s BC Series Battery Cabinets, data center operators can easily integrate NiZn’s benefits into existing UPS systems with no need for hardware modifications.

Conclusion

By utilizing ZincFive’s BC Series Battery Cabinets, data center operators can easily integrate NiZn’s benefits into existing UPS systems with no need for hardware modifications. The cabinets’ backward and forward compatibility with megawatt class inverters allows them to function as a straightforward drop-in replacement for existing lead-acid battery cabinets. This allows data centers to take advantage of NiZn’s higher power density, superior reliability, lower costs, longer lifespan, and multiple environmental advantages that modern data centers need: the perfect choice to safely back up their power and maintain critical operations.

Citations

¹ https://www.delltechnologies.com/asset/en-us/products/servers/industry-market/2020-server-trends-and-observations-brief.pdf

² https://www.technologyreview.com/2019/11/11/132004/the-computing-power-needed-to-train-ai-is-now-rising-seven-times-faster-than-ever-before/

³ https://openai.com/blog/ai-and-compute/

⁴ https://www.ul.com/services/ul-9540a-test-method

⁵ https://www.boundlessimpact.net/climate-impact-assessment

Executive Summary

When utility power goes down, data center Uninterruptible Power Supply (UPS) systems immediately switch over to backup batteries, which provide power to the facility until the generators come online. This transitional period is known as “battery runtime” or “ride-through time.” In the past decade, data center designs have continued to reduce battery runtime.

The reduction in battery runtime provides an opportunity for data center owners to re-examine the batteries they use for backup power in their facilities. Due to limits in their chemistries, traditional lead acid and lithium-ion batteries cannot be optimized for runtimes of less than 5 minutes. In either case, data centers must oversize their battery bank, buying more batteries than they actually need in order to provide adequate short-term power.

As data center owners adopt data center designs with faster automatic failover from UPS to generator, they should invest in a battery that:

1. Can be optimally sized for runtimes of 5 minutes or less;
2. Provides a small footprint with reliable and safe short-term power burst;
3. Offers total cost of ownership (TCO) savings, by requiring a smaller number of batteries to support UPS designs with shorter runtimes.

Nickel-zinc (NiZn) battery technology offers all these advantages and is a smart choice for the future of data center backup power as UPS technologies continue to improve.

The Industry Moves Towards Shorter Runtimes

An Uninterruptible Power Supply (UPS) system serves as part of the fail-safe power mechanism for a data center. Should the utility power experience an outage, the backup power system seamlessly transitions from UPS battery backup to longer duration backup generators. This ensures business continuity for servers housed at that facility, and prevents costly downtime, loss of revenue, and potential damage to the company’s reputation.

The Trend in Reduction of Battery Runtime is Due to Several Factors:

Improvements in generator technology – Today’s generators are able to start up faster. When a power outage hits, the generators can usually start up and come online in under 50 seconds. When the UPS sees that the main input has been transferred to the generators, they slowly transition the facilities load over from the batteries to the generator typically over a 10 second ramping period known as the walk-in period. In some facilities, this total startup and transition to generator happens within 30 to 45 seconds.

Improvements in power delivery architecture – Generators may not be the site’s only backup power source and may not even be the first option during automatic failover. For example, if a facility has dual power feeds and one utility feed goes down, the UPS will automatically switch the facility over to the second utility feed. The backup batteries may still provide power to the facility during this switchover which only typically requires around 5 seconds to make the transfer.

Some corporations have achieved multiple-site redundancy by setting up data centers in several cities, with servers running synchronized applications. For example, if a Denver data center which is hosting eCommerce applications goes down, the company’s IT systems will automatically switch customers over to those same applications hosted on servers in their Atlanta and Phoenix data centers.

Multiple-site redundancy doesn’t directly affect UPS failover times. The switchover of applications usually happens in less than 1 minute – but in this case, the switchover is software-driven. Still, having data centers in several cities means that a business will not be interrupted if one facility goes down. This gives a company peace of mind since it lowers the risk of losing business due to a data center outage.

Even with faster automatic failover and reduced battery runtime, UPS systems are as reliable as ever in providing continuous uptime to data center facilities with reliability reaching 99.99999%. In fact, some data center owners are now adopting new and more robust automatic failover plans to accommodate faster turnover. They now expect their generators to come online faster, and count on the UPS to reliably transfer the data center’s load to generator power in less than 1 minute.

With data center design and construction teams adopting these new and more robust failover plans and reducing their battery runtimes, it creates a highly competitive landscape and is causing other providers to reevaluate their runtime requirements to better optimize their UPS and battery systems. They have an opportunity to choose a battery solution that supports this inevitable move to shorter runtimes.

A few key qualities they should consider include:

1. Reliability and safety during short-term, high-power discharge.
2. High-density power delivery that enables right-sizing of battery loads for data center UPS systems with runtimes moving below 5 minutes.
3. Total Cost of Ownership (TCO) savings.

Choosing a Reliable Battery for Shorter UPS Runtimes

There are batteries available today that offer reliable power delivery for UPS systems for shorter runtimes. Due to output current limitations, a majority of these battery solutions have to be oversized to meet the UPS power requirements. Nickel-zinc batteries are uniquely capable of safely supporting extremely high-power discharges in very small footprints, consequently reducing the UPS room’s overall space requirements and the quantity of cabinets necessary for installation.

The right choice of a high-power battery solution for short runtimes can simplify the site installation and reduce the site risk, provided that it is a safe and reliable solution.

As seen in Figure 1 in a lead acid or lithium-ion battery string, failed cells are typically accompanied by a steep rise in internal resistance which will block the current flow from surrounding cells when the battery string is expected to provide runtime for an outage. This impedance block will create an open circuit, and subsequent failure to discharge. A single bad battery cell can bring down an entire string. Data center owners usually try to compensate for this by adding additional redundant battery strings – which increases the TCO of the battery system.

Contrary to the other chemistries, a weak or depleted nickel-zinc cell will see a reduction in internal impedance and, instead of creating an open circuit, will remain conductive and allow the battery string to discharge. This means that even with a weak or depleted cell the nickel-zinc battery system will continue to deliver runtime, reducing the need for redundant battery strings required by other chemistries. That is inherent reliability at the electrochemical level.

Choosing a Safe Battery for Short-Duration Discharge

It is possible to choose a battery that not only delivers a reliable electric charge for a short-duration runtime, but also delivers that charge safely.

Nickel-zinc batteries are alkaline batteries. Unlike lead acid and lithium-ion batteries, the nickel-zinc battery has zero risk of thermal runaway. This gives the battery several advantages when it comes to safety:

1. A nickel-zinc battery can provide a high-power, short-term discharge without the risk of thermal runaway. This makes the battery a safe option for UPS systems with shorter runtimes.
2. In the case of a bad battery cell, the nickel-zinc battery can still provide a safe, sustained discharge, thanks to the battery’s electrochemistry.
3. The nickel-zinc battery can provide high levels of current without posing a safety risk.

Lithium chemistries specifically rely on Battery Management Systems (BMS) to ensure the cells are functioning and to prevent any safety risk that could occur. The risk this poses to the reliability of the lithium battery system is that if a single battery cell is determined to be out of tolerance, the BMS must protect the battery string and disconnect itself from the system as not to cause any adverse conditions or potential safety concerns. If this happens and the battery system is disconnected by the BMS, that means no runtime for the data center when it is needed most.

Unlike with other types of batteries, the BMS for nickel-zinc batteries does not automatically shut down a battery string or prevent the battery from discharging for any reason.

Nickel-zinc batteries do not require these controls as they are safe at the cell level. These systems do not have discharge parameters that are required to be managed by a BMS as there is no safety risk or thermal runaway concerns with this chemistry. This means the battery system can provide runtime for the data center without any concern of the BMS preventing a discharge.

Changing workloads and corporate priorities have caused data center builders and operators to take a closer look at UPS runtimes. With the explosive market growth in AI comes different and often reduced backup power requirements. Meanwhile, larger facilities mean the difference between 3 minutes and 7 minutes of battery runtime likely doesn’t impact the end result if a generator fails to start. Focusing specifically on achieving enough runtime to ride through generator starting and load transfers allows our data center clients to reduce their capex, ease their supply chain burden, and reduce equipment footprint, allowing them to maximize the sellable space in their facilities.”

Gary Russinko, PE & Managing Principal,
kW Mission Critical Engineering

Right-Sizing Battery Banks for Faster UPS Runtimes

Right sizing your battery bank begins with selecting a battery that can be optimized for shorter battery runtimes. Data center designers and operators are reducing their runtime requirement to less than 5 minutes, with some targeting 3 minutes at 10 years as an optimized solution.

Let’s look at a common industry example of a 1MW UPS design for a side-by-side comparison. In this example, the data center can switch over to generator power in 1 minute or less, and because of this, it is determined that they only require a 3-minute (at 10 years) battery runtime design.

Figure 2 shows how this setup might work using different types of batteries.

Most data centers today use either traditional lead acid or lithium-ion batteries as their backup power source but these batteries may not be the best option for UPS systems with faster automatic failover. Due to limits in their battery chemistries, lead acid and lithium-ion batteries may not be optimized for UPS runtimes under 5 minutes whereas nickel-zinc solutions can.

Lead acid batteries have a high discharge rate but a low energy density. This means you need a larger number of batteries to handle your data center load.

Lithium-ion battery cabinets, on the other hand, have a higher energy density but have a lower discharge rate. This is due to limitations in maximum current and the safety risk an overcurrent scenario poses to thermal runaway risks. If the discharge current goes above this limit, the BMS that controls the lithium-ion batteries will shut the cabinet down. These overcurrent protection limits increase the risk of cutting off the system in the middle of battery runtime, preventing the system from providing the high-power discharge required to support the critical load.

In either case, with lead acid or lithium-ion batteries, you have to oversize your battery bank to reach your target load of 1 MW and provide a sustained, short-duration uninterrupted, high-power discharge. This means the data center must buy a larger number of battery cabinets to serve a UPS system with a runtime of less than 3 minutes.

On the other hand, nickel-zinc batteries offer three times the power density of lead acid, two times the current carrying capability of industry leading lithium-ion batteries, and a BMS that is passive during a discharge and will not interrupt the operation as there is no safety risk to do so.

Using nickel-zinc batteries, data centers can safely optimize their battery load for runtimes of 5 minutes or less. In the example above they would only need 3 cabinets of nickel-zinc batteries to reach 1 MW target load, and provide a 3-minute short-term, high-power discharge.

Many data centers are selecting runtimes less than 5 minutes to optimize cost and minimize footprint. These customers have realized long ago that 5 minutes of UPS battery runtime does not provide added protection. If a generator fails to start, the problem cannot be corrected in 5 minutes or even 15 minutes.”

Harry Handlin, U.S. Data Center Segment
Leader at ABB

Achieving Total Cost of Ownership Savings with Shorter Runtimes

Today’s UPS systems can provide TCO savings for data center owners who take advantage of shorter battery runtimes. If one knows how to optimize a battery bank to provide a sustained discharge for less than 5 minutes, one can reduce the quantity of batteries required to meet the facility’s load.

Since NiZn batteries can be optimally sized for runtimes of 5 minutes or less, one only needs to buy the minimum number of battery cabinets to support the load and runtime requirements.

This reduction in batteries that nickel-zinc battery technology can offer for shorter runtimes has large implications to the overall total cost of ownership decrease. The largest contributing factors are the footprint reduction, installation costs and lower overall maintenance costs over the life of the system. These can be compounded – and the savings even larger – if the UPS and battery system are containerized in a modular power system where footprint reduction can translate to larger savings.

Conclusion

Shorter runtimes may not suit every data center’s needs. Many facilities maintain longer battery runtimes during automatic failover for various reasons. For such applications, nickel-zinc batteries continue to offer a robust and dependable backup power source.

Nickel-zinc batteries are a smart choice for the future. As UPS systems and generators improve on transfer time, this failover will become even faster and more reliable as data centers are able to switch over to generator power (or from one utility feed to another) more seamlessly than ever before. With these improvements, there will be a high demand for a battery that can be right-sized and optimized for shorter runtimes, while still delivering a high-power discharge.

The nickel-zinc battery offers the power density, reliability, safety, and reduced runtime capabilities that the data centers require.

Introduction

This white paper provides an informational guide to the United States Codes and Standards regarding Energy Storage Systems (ESS), including battery storage systems for uninterruptible power supplies and other battery backup systems. There are several ESS technologies in use today, and several that are still in various stages of development.

While various technologies, such as flywheels, fuel cells, compressed gas, and others, are either in use or development, the primary focus of most of the jurisdictional Authority Having Jurisdiction (AHJ) is currently being placed on electrochemical storage systems.

These same Codes – and many of the Standards cited – cover all of the currently available ESS technologies, and in some cases, there are additional Codes and Standards cited to cover those specific technologies. For the sake of brevity, electrochemical technologies will be the primary focus of this paper due to being the ESS choice of the vast majority of users. Technologies such as pumped hydro, compressed gas, fuel cells, and various methods of gravity storage are not covered in this paper.

Why do we have Codes and Standards?

A code repository is necessary to increase awareness and improve safety in the energy storage industry. Electrochemical energy storage has a reputation for concerns regarding the ventilation of hazardous gases, poor reliability, short product life, substantial cooling requirements, and high levels of periodic maintenance.

Like the newer lithium battery technologies, the traditional lead-acid technology has developed a stigma. While generally a safe product, damage presents the risk of fire. When damaged, lead-acid, lithium, and some other battery technologies may also leach corrosive chemicals such as mercury, cadmium, and lead into landfills and may contaminate water supplies and ecosystems. These chemicals are dangerous to human health and are expensive to clean up. Users desire a safe, cost-effective alternative that requires less maintenance and is environmentally sustainable.

Definition of Codes and Standards

What differentiates Codes from Standards is the usage. Codes are an overarching statement of best (and safest) practices for an entire industry or technology.

As seen in the provided definitions, a Code is a set (or collection) of mandatory regulations that are established and enforced by a governmental authority. This is generally a collection of safety practices and other associated Standards enforced for the purposes of safety and reliability.

Standards, on the other hand, are technology or product specific, and provide a method to verify that the technology or product meets or exceeds the minimum acceptable level of safety.

What we refer to as “Codes” are actually model Codes, published by one of several nationally recognized organizations, with the purpose of presenting the “best practice” for subjects covered by the intended area of application. Each model code presents the latest consensus information on its related subject. These model Codes are then reviewed and adopted by the various jurisdictions, and when accepted become the legal Code for that jurisdiction. There are several separate model Codes, covering a variety of applications. For the purposes of this paper, only the applicable fire Codes and their related Standards will be considered.

In some cases, these Codes are adopted on a statewide basis. In others, they may be adopted on a county-by-county, or city-by-city basis. In county and local cases, the jurisdiction is allowed to amend the Code to make it more stringent, but in no case may the jurisdiction reduce the Code if it is adopted in full by a higher authority.

There are two primary organizations in the United States that publish these model Codes:

Every jurisdiction in the US has adopted either the NFPA Codes or International Code Council’s I-Codes. Currently (2023), there are eight states that adopt the NFPA 1 Fire Code, and forty-two that adopt the International Fire Code. Interestingly, although there are much more advanced Codes available, there are still several jurisdictions that are using the 2015 revisions of the Codes, and at least one that is still using the 2009 revision.

As one gains understanding of the increasing number of new battery chemistries, and the associated risk factors, it is hard to justify maintaining an outdated Code base unless that Code is regularly amended to maintain the intended safety aspects of the more recent Codes.

Storage Technologies and Electrochemistries

There are several basic groups of ESS technologies in use today. They are usually broken into categories such as Mechanical, Gravity, Electrical and Electrochemical. While some of these may somewhat bridge gaps between technology groups, these are generally recognized as the major categories for Code purposes. It should be noted that technologies falling under the “Gravity” section have very little mention in the Fire Codes.

How Codes Are Applied

Starting with the NFPA 1-2018 and IFC 2018, the Codes introduced some measure of regulation to ensure the safety of installed Energy Storage Systems (ESS). Although there was limited regulation regarding batteries and their installation, maintenance, and explosion/fire control in earlier revisions of the Codes, it was relatively rudimentary compared to later Codes.

As the adoption of newer battery technologies increased, both NFPA and ICC recognized that these technologies required more stringent regulation. The AHJs were unfamiliar with lithium, sodium, and other new technologies and had limited or no understanding of their potential problems. NFPA had several conflicting Codes and Standards cited in their fire code and recognized that creating a unified Energy Storage System (ESS) Standard was the best way to deal with that issue. This led to NFPA 855, the single ESS Standard NFPA now recognizes. The IFC 2021 revision deals with ESS slightly differently but has a chapter (Chapter 12) dedicated to Energy Systems and a section (Section 1207) dedicated to Electrical Energy Storage Systems. The IFC 2021 essentially copies the Maximum Allowable Quantities table (1206.9) from the IFC 2018, adds Nickel-Metal Hydride and Flow batteries to the table, and reprints almost the same data. There are additional stipulations in the requirements of the IFC 2021, but ICC did not go to the same lengths as NFPA in their 2021 edition. The 2024 edition of the IFC is anticipated to be more closely aligned with NFPA 855-2023.

Both the NFPA 1 National Fire Code and the ICC International Fire Code have adopted significant enhancements to the earlier code language in their 2021 revisions. Many jurisdictions still utilize the 2018 and earlier revisions of the Fire Codes. Still, many jurisdictions with high data center density are adopting the 2021 revisions and/or implementing NFPA 855-2020 as their ESS standard. While the 2021 revisions of the International Fire Code is the most recent available today (Q3, 2023), the new NFPA 1-2024 Fire Code has just been released in September of 2023, and NFPA 855 has already been updated with the NFPA 855-2023 Standard.

In the case of a jurisdiction adopting either of these Fire Codes and their related Standards, there are significant restrictions on some Energy Storage technologies. Any technology not explicitly listed in the relevant tables (Table 9.4.1 in NFPA 855-2023, and Table 1207.5 in IFC 2021), and even some of those listed but not specified as having an unlimited allowable quantity in a given fire area, will have to be approved on a case-by-case basis. Some jurisdictions will not allow a lithium battery to be installed under any circumstances. In contrast, some will allow them with severe restrictions on where they are installed, how much capacity is allowed, and what type of lithium battery is being proposed.

As mentioned earlier, the cited Standards significantly influence the Codes and their implementation. It is necessary to highlight some of these Standards (other than NFPA 855) due to their importance to both Codes. Three of particular importance are UL 1973, UL 9540, and UL 9540A.

List of Major U.S Codes and Standards

This section provides a comprehensive list of major U.S. Codes and Standards. See Appendix A for a concise cheat sheet for Codes and Appendix B for Standards.

---

---

Looking Ahead

New Codes and Upcoming Code Updates

Many Codes are due for updated releases next year, which are very important in the ESS world. Among these are the NFPA 1 National Fire Code and the ICC International Fire Code, International Building Code, International Mechanical Code, and several more from ICC, all expected to be released late in 2023. In addition, the NFPA 101 Life Safety Code will be released in late 2023.

Many of the NFPA and UL Standards are either due for a revision release or are in the development cycle for release soon.

As a reminder, checking for the latest amendments or additions to the existing Codes and Standards is extremely important. For example, NFPA will issue Technical Interim Amendments (TIA), which supersede the language in the published document. UL issues Certification Requirement Decisions (CRD) to update the language in their Standards and add or remove requirements for compliance with the Standard in question. In both cases, these changes will be resolved by the Technical Committee responsible for the specific document and incorporated in the document’s next release.

Conclusion

In Summary

It is difficult to overemphasize the importance of maintaining an awareness and understanding of the relevant Codes and Standards affecting modern ESS systems’ selection, use, installation, and maintenance. There are many choices from a broad spectrum of solutions available. Some of these may present challenges in being acceptable to the local AHJ having jurisdiction over Fire Code enforcement, or potentially, even with the AHJ having jurisdiction over the Building Code that applies to their jurisdiction.

Knowing and understanding these Codes and Standards’ impact on implementing a given ESS could have significant financial and scheduling issues for the end user. There is a value in maintaining a knowledge of the most recent revisions and additions to these documents, as this is a rapidly changing landscape, and what is considered acceptable in the Codes and Standards today may change to an unacceptable risk in a very short time.

Works Cited and Useful Links

International Code Council (ICC), https://www.iccsafe.org/
National Fire Protection Association (NFPA), https://www.nfpa.org/
American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE), https://www.ashrae.org/
Underwriters Laboratories (U.L.), https://www.ul.com/-Underwriter
Battery University, https://batteryuniversity.com/
Energy Systems and Energy Storage, http://www.eseslab.com/index
DOE/EPRI 2013 Electricity Storage Handbook in Collaboration with NRECA, https://www.sandia.gov/app/uploads/sites/163/2021/09/SAND2013-5131.pdf
Environmental and Energy Study Institute, https://www.eesi.org/papers/view/issue-brief-energy-storage

---

Appendix A: Codes

Org.	Document Number	Title	Publication Date	Next Revision Due
NFPA	1	National Fire Code	2021	2024
NFPA	30	Flammable and Combustible Liquids Code	2021	2024
NFPA	70	National Electrical Code	2023	2026
NFPA	72	National Fire Alarm and Signaling Code	2022	2025
NFPA	101	Life Safety Code	2021	2024
NFPA	400	Hazardous Materials Code	2022	2025

Org.	Document Number	Title	Publication Date	Next Revision Due
ICC	IBC 21	International Building Code	2021	2024
ICC	IEBC 21	International Existing Building Code	2021	2024
ICC	IFC 21	International Fire Code	2021	2024
ICC	IFGC 21	International Fuel Gas Code	2021	2024
ICC	IMC 21	International Mechanical Code	2021	2024
ICC	IPC 21	International Plumbing Code	2021	2024
ICC	IPMC 21	International Property Maintenance Code	2021	2024
ICC	IRC 21	International Residential Code	2021	2024
ICC	IWUIC 21	International Wildland-Urban Interface Code	2021	2024

Org.	Document Number	Title	Publication Date	Next Revision Due
IEEE	C2	National Electrical Safety Code	2023	2028

Appendix B: Standards

Org.	Document Number	Title	Publication Date	Next Revision Due
NFPA	12	Standard for Carbon Dioxide Extinguishing Systems	2022	2025
NFPA	12A	Standard for Halon 1301 Fire Extinguishing Systems	2022	2025
NFPA	13	Standard for Installation of Sprinkler Systems	2022	2025
NFPA	15	Standard for Water Spray Fixed Systems for Fire Protection	2022	2025
NFPA	17	Standard for Dry Chemical Extinguishing Systems	2024	2027
NFPA	68	Standard on Explosion Protection by Deflagration Venting	2023	2026
NFPA	69	Standard on Explosion Prevention Systems	2019	2024
NFPA	76	Standard for Fire Protection of Telecommunications Facilities	2020	2024
NFPA	110	Standard for Emergency and Standby Power Systems	2022	2025
NFPA	111	Standard on Stored Electrical Energy Emergency and Standby Power Systems	2022	2025
NFPA	750	Standard on Water Mist Fire Protection Systems	2023	2026
NFPA	770	Standard on Hybrid (Water and Inert Gas) Fire Extinguishing Systems	2021	2024
NFPA	855	Standard for the Installation of Stationary Energy Storage Systems	2023	2026

Org.	Document Number	Title	Publication Date	Next Revision Due
UL	263	Fire Test of Building Construction and Materials	2021	UL does not publish new editions on a schedule.
UL	268	Smoke Detectors for Fire Alarm Systems	2016
UL	1741	Inverters, Converters, Controllers and Interconnection Systems Equipment for Use with Distributed Energy Resources	2021
UL	1778	Uninterruptible Power Systems	2023
UL	1973	Batteries for Use in Stationary, Vehicle Auxiliary Power and Light Electrical Rai (LER) Applications	2022
UL	1974	Evaluation for Repurposing Batteries	2018
UL	9540	Energy Storage Systems and Equipment	2023
UL	9540A	Test Method for Evaluating Thermal Runaway Fire Propagation in Battery Energy Storage Systems	2019

Org.	Document Number	Title	Publication Date	Next Revision Due
IEEE	1491	IEEE Guide for Selection and Use of Battery Monitoring Equipment in Stationary Applications	2012	IEEE documents are updated on a ten-year cycle unless there is an amendment or corrigenda published.
IEEE	1635	IEEE/ASHRAE Guide for the Ventilation and Thermal Management of Batteries for Stationary Applications	2012
IEEE	1657	Recommended Practice for Personnel Qualifications for Installation and Maintenance of Stationary Batteries	2018
IEEE	1679	IEEE Recommended Practice for the Characterization and Evaluation of Energy Storage Technologies in Stationary Applications	2020