The Battery’s Environmental Footprint: How Clean Is the Technology?
The premise of this book is that increased access to battery storage technologies will catalyze the transition to a low-carbon economy. But in fully evaluating this question, it is imperative to also understand what the real environmental impact is of widespread battery deployment. Determining the answer to this question requires a closer look at the technology in question. Scientists and engineers have developed an accounting methodology that assesses the total environmental impact of a specified technology—across a variety of categories—from “cradle to grave.” This accounting protocol tracks a device through its manufacture, use, and disposal to determine its effects on the environment across its entire life cycle. The protocol is, fittingly, called a life-cycle assessment (“LCA”), and applying this methodology to the battery will help inform its possibilities for success. It is important to more fully understand the life-cycle footprint to know what unintended consequences may ensue due to widespread deployment of battery technologies, and to mitigate these consequences to the extent possible.
By employing the life-cycle assessment prior to widespread adoption of various technologies, relevant industries and government agencies gain a more detailed understanding of the potential effects and unintended consequences associated with the technology. This step is a particularly important for technologies aimed at reducing carbon emissions and improving environmental quality. In some cases, technologies that purport to be better for the environment during use actually cause greater harm to the environment during their manufacture or disposal stages. It is imperative for policy makers and consumers to be aware of these types of solutions that merely shift environmental burdens temporally or geographically rather than mitigate such burdens.
Understanding the Life-Cycle Environmental Impact
Life-Cycle Assessment Standards
There are currently no laws or regulations that create mandatory standards for life-cycle assessments in the United States. Nevertheless, there are well-established international standards that are widely accepted and used. In particular, the International Organization for Standardization (ISO) developed LCA standards in response to “heightened awareness of the importance of environmental protection, and the possible impacts associated with products manufactured and consumed.”1 ISO created these standards to provide organizations with “guidelines on the principles and conduct of LCA studies [which ultimately inform ways] to reduce the overall environmental impact of its products and services.”2 Corporations and other organizations frequently use the ISO standard as a foundation for building their LCAs, sometimes modifying or adding procedures to better assess specific products. Despite any tweaks that individual organizations may implement, “ISO standards represent an international consensus on the state of the art in the technology or good practice concerned.”3
ISO develops LCA standards within its 14000 Environmental Management Series (specifically in ISO 14040–14043). These standards define four general phases for life-cycle analyses: goal and scope definition, life-cycle inventory analysis, life-cycle impact assessment, and interpretation.4 LCA studies are most useful when the impact of one product can be compared to the impact of another. Yet, “there is no scientific basis for reducing LCA results to a single overall score or number.”5 In fact, the detailed methodologies for carrying out a LCA may vary greatly across industries and applications. This makes product and system comparisons difficult unless the assumptions and context of each study are the same. The requirement for such levels of comparability necessitates complete transparency of the scope, assumptions, descriptions of data quality, methodologies, and outputs for each LCA.6 Therefore, maintaining transparency throughout all four phases of a LCA study is likely the most important objective for researchers.
In the first phase of the LCA, researchers clearly define the goal and scope of study and ensure that these are consistent with its intended application. The “goal of an LCA study shall unambiguously state the intended application, the reasons for carrying out the study and the intended audience.”7 Additionally, the “scope should be sufficiently well defined to ensure that the breadth, the depth and the detail of the study are compatible and sufficient to address the stated goal.”8 In addressing the study’s scope, researchers must focus on defining the functional unit (a specific unit of output that can be compared across products and processes), the product or system boundaries, the data requirements and data quality, and any assumptions or limitations of the study.9 Again, these components should be well defined to increase the relevance and usefulness of the study, as well as to provide transparency such that researchers can compare various studies and improve upon them in the future.
The second phase of the LCA—the life-cycle inventory analysis—“involves data collection and calculation procedures to quantify relevant inputs and outputs of a product system.”10 System inputs include raw materials and resources that are used to manufacture a product. Typically, data are available to define the embodied energy, toxicity, and other properties for each of the materials used to create the product being studied. System outputs typically include emissions to the natural environment as a result of product use, as well as the waste stream that results from disposal of the product being studied. By defining the system boundaries in the first phase, researchers identify precisely how much information they will include in the system input/output calculation (see Figure 3.1).
The third phase of the LCA—the life-cycle impact assessment—“is aimed at evaluating the significance of potential environmental impacts using the results of the lifecycle inventory analysis.”11 This phase attempts to put the results of the inventory analysis in a broader context, allowing researchers to assess and understand the actual impact of the product. For example, a product may require a raw material with a particularly high embodied energy or it may result in a radioactive waste stream. But if the product only requires a small amount of the energy-intensive material or if the radioactivity is lower than the baseline levels of radiation occurring naturally in the environment, then the overall impact of the product may still be minimal.
Finally, the fourth phase of the LCA—the life-cycle interpretation—requires a more thorough evaluation of the inventory analysis and the impact assessment to allow researchers to form conclusions that may inform decision makers.12 This phase of the process is often iterative and may require researchers to reexamine different phases of the assessment based on incorporation of different data such that the results are consistent with the defined goal and scope of the study. Often, this phase of the study includes a comparison between products to determine whether a new product has a lower environmental impact than its incumbent competitors. Similarly, this phase might reveal particularly impactful stages of a product’s life cycle and inform companies of specific areas in which they should make improvements.
Figure 3.1 A sample process flow diagram used by the EPA for conducting a LCA comparing electric vehicles with plug-in hybrid vehicles using LIBs with different battery chemistries. (Reprinted from U.S. EPA.)
To help explain each of these phases in more detail, the rest of this chapter will refer generally to the life cycle of a lithium-ion battery designed for use in vehicles. Although much of this book focuses on myriad different battery chemistries, configurations, and uses, LIB electric vehicles are an emerging technology that will likely be with us for the foreseeable future. Lithium-ion batteries are also sufficiently established that data exist to enable a detailed life-cycle assessment. Nevertheless, it is important to note that the more detailed a study, the more it will vary based on the particular variables and data used. The LCA of LIBs described in this chapter is not necessarily applicable to every LIB in every electric vehicle, as the results will vary depending on the particular manufacturer, geographic location of use, and other factors.
Goal and Scope: Establishing Boundary Conditions
One of the most important components of the initial phase of a LCA is establishing boundary conditions. This step requires researchers to define which aspects of the studied device’s life cycle they will include in the analysis. Most LCAs strive to analyze the total impact of the device from “cradle to grave”—that is, from manufacture and production, through use, and ultimately disposal. Yet, this approach still leaves many variables. For example, should the analysis begin with the raw material inputs and sum the impact from the point of extraction for every individual widget that comprises the whole device? Many academics attempt to do just that, as this approach provides the most complete assessment. Unfortunately, it is difficult and time consuming to track down the data to use in such a detailed assessment. Therefore, what happens more frequently is that researchers will conduct this type of detailed analysis for only a handful of the most resource-intensive (or otherwise impactful) components of a device.
Other questions must also be answered about a LCA’s boundary conditions. For instance, in today’s globalized economy many devices are the culmination of manufacturing and production processes all over the world. A complete LCA would include all impacts, regardless of where the impact is felt. Yet, some corporate studies—particularly studies of devices that spend most of their life cycle in one location—will confine their assessment to the impact felt in one dominant location. For example, the study of a device that is manufactured and used in the United States but is recycled by a company in India might omit the impacts of the recycling process. The American company may do this either because it is less concerned about the impacts felt in another country or because it does not have the resources to determine the impacts. Similarly, a company conducting an assessment of a device that is predominantly American-made, with the exception of a small component (the production of which the company subcontracts to a manufacturing plant in China), may omit the impact of that particular component.
Additionally, the globalized effort necessary to create a single device also requires researchers to determine whether to include the impact of transportation in their assessment. Again, a thorough LCA should include all impacts. Yet, some studies omit transportation, either to reduce the burden of identifying accurate data or to constrain the assessment to energy and other impacts that are directly related to the device itself. The inclusion or exclusion of transportation offers a good example of why it is important to understand a study’s boundary conditions. Anyone seeking to compare two devices must ensure, to the extent possible, that the assessments prepared for the devices employ the same boundary conditions. Food production offers a simple case study: A consumer in New York is looking to purchase an apple that is produced with the least impact to the environment. She has a choice between an apple grown with conventional methods and an apple grown organically. Most studies show that organic apples have a lower environmental impact because the production process omits harmful pesticides, fertilizers, and other chemicals. The organic apple, however, was grown in Chile and the conventional apple was grown in New York. Considering the environmental consequences of transportation changes the equation and may make her decision more difficult.13
Finally, the first phase of a LCA also requires researchers to identify the precise output they are studying. A “functional unit” is the specific unit of output the LCA analyzes. Defining a functional unit provides a means of normalizing data such that the results can be compared across products and systems.14 For example, if researchers are trying to identify the total impact of a lithium-ion battery for vehicles, they will not sum the impact of Tesla’s entire inventory. Instead, they will define the functional unit of one kilometer driven by a Tesla Model S. Therefore, the impact of the battery may be accurately compared to one kilometer driven by a conventional vehicle of a similar size and weight under similar conditions. This is one example of a functional unit; other studies may employ other methodologies. The functional unit assessed may differ across studies as long as it offers a reasonable comparison to other products.
With prescribed boundary conditions and a well-defined functional unit, researchers may compile data about a device and its components such that they can determine its environmental impact per functional unit.
Determining Life-Cycle Inventory Analysis
The second phase of a LCA requires researchers to compile and analyze data related to the studied device’s material and energy inputs and outputs.15 Depending on the assessment’s boundary conditions, this may include the following: primary materials; primary energy; recycled materials; transportation; outputs associated with use, maintenance or replacement; and disposal. Thus, the data required to conduct a thorough and useful LCA are substantial.
Ideally, researchers obtain data related to each phase of a device’s life cycle directly from the company responsible for that phase. While it is best to rely on these types of highly specific data—particularly from production, manufacturing, and recycling companies—these data may be protected by patent laws or other confidentiality arrangements or may simply not exist. More often, researchers rely on generalized data from private companies that may be publicly available. Where data gaps persist in the information provided by private companies, the public sector may provide some answers. Government agencies and other research institutions often make available generalized data on popular manufactured products and processes. Alternately, both private and public entities have developed life-cycle assessment databases, which allow researchers to generate models to simulate the production, manufacturing, and recycling phases of particular materials and devices. A thorough LCA will likely rely on data from all of these sources and will inevitably require modeling software to compile all of the acquired data.
The life-cycle inventory analysis of our sample lithium-ion battery would likely focus on the primary materials used to create the battery cell. This analysis may include the energy and other impacts associated with material extraction, particularly for the cathode material in the prescribed battery chemistry. Other material extraction data may also be included—most notably data related to aluminum (for housing the battery cell), copper (for the electrical circuits), and materials required for the prescribed electrolyte.16 Although the material extraction for battery components takes place in various countries all over the world, these processes are uniform and data are typically available to model the energy and other associated impacts with reasonable accuracy.
The assessment would also include primary energy and other impacts associated with material processing and component manufacturing.17 This refers to the part of the process in which raw materials are combined with energy and other products to create the individual components of a LIB. The dominant input in these phases is energy, since processes such as material refining, electroplating, and die casting often require immense amounts of energy. Yet, this phase may be particularly difficult to assess. Many LIB components are manufactured in Asia and processing and manufacturing protocols may differ substantially between companies. It is frequently difficult to obtain data from private international companies. Thus, this phase of the LCA would most likely suffer from data gaps. Nevertheless, if consumers or consumer-facing battery companies begin to demand more thorough LCAs to better inform purchasing decisions, these data may become more accessible.
Researchers would also consider the impacts of product manufacture.18 This part of the process is predominantly comprised of assembling the manufactured components into the finished product. Product assembly tends to require fewer inputs than other elements of the production process. Therefore, this phase of the analysis tends to be straightforward. This is also the phase that consumer-facing battery companies typically control, so access to data tends to be greater. Nevertheless, this phase may also include quality control, validation, and other testing, which can require significant energy inputs, so it should not be discounted.19
Next, a battery’s use phase will inevitably rely on modeled data, since each battery will result in slightly different inputs and outputs, depending on a wide array of variables. The most important consideration in a battery’s use phase is the composition of the grid electricity in the region in which the battery operates. Electricity is a necessary input for any battery technology, so the battery’s impact is highly dependent upon the impact of electricity generation.20 The U.S. Environmental Protection Agency and the Energy Information Administration offer detailed data describing the environmental impact of the electric grid in different states and across different electricity generation methods.21 The data required for this phase of the inventory analysis are typically available through these types of government agencies, though it may be generalized.
Finally, most LCAs also consider the battery’s end of life, or disposal. This phase depends heavily on the company responsible for the battery and the decisions made by the consumer. Battery disposal may involve complete reuse or recycling, material recovery, landfilling, or incineration. Each of these options will result in substantially different environmental impacts.
Not only is it difficult to predict the life-cycle assessment of batteries because of the vast array of variables, it is also difficult because the technologies are constantly changing. Innovations are occurring rapidly not just in battery chemistries—potentially resulting in reduced impacts from material extraction and processing—but also in manufacturing processes and disposal techniques. Each of these innovations significantly affects the end product’s ultimate impact. In order to improve LCAs, battery companies should meticulously gather data on each phase of their product’s life cycle. This information will allow companies to identify areas of high environmental impact and work toward reducing those impacts.
Evaluating Total Environmental Impact
Lifecycle assessments may be capable of evaluating many different components of a device’s environmental impact. Researchers may decide to look solely at one aspect of environmental impact, depending on the objectives of the study and the boundary conditions they impose. For instance, researchers from a Riverkeeper organization may be most interested in the impacts that a particular device has on a local watershed. Alternately, consumer protection groups might care more about exposure to toxics or radioactivity throughout the device’s life cycle. Finally, researchers interested in evaluating a device’s impact on the climate may focus on the device’s global warming potential (GWP). Notwithstanding a particular research bias, a thorough LCA will assess all of a device’s various environmental impacts. It is then up to the company, government agency, or other research organization to determine the relative importance of the various impacts and to intervene accordingly to reduce the most egregious impacts.
LCA modeling software may sort environmental impacts into various categories based on the medium or community affected. Typically, these categories include air pollution, terrestrial and marine pollution, and carcinogens. The granularity with which each of these categories is evaluated depends on the amount and quality of the available data. If lots of highly reliable data are available, then a wide variety of environmental impacts may be quantified.
Air pollution is one of the most common aspects of a device’s environmental impact that LCA researchers seek to understand. Therefore, this category is often further divided based on more specific impacts. For example, a LCA may report a device’s impact on particulate matter formation,22 photochemical oxidation potential (or the potential for smog formation),23 or ozone depletion potential,24 in addition to its global warming potential.25 Each of these analyses is based on a complex model of molecular outputs and their subsequent reactions with the environment. Overwhelmingly, a LIB’s use phase has the greatest impact on global warming and smog formation, whereas its material extraction and product manufacturing phases have the greatest impact on ozone depletion potential.26 This result reinforces the concept that the way a battery is used, including the electric grid fuel sources and efficiency in which it operates, may have the greatest effect on the technology’s ability to combat climate change.
Terrestrial and marine pollution are also relevant in most LCAs. This broad category may be further divided into ecotoxicity,27 acidification potential,28 or eutrophication potential,29 among others. The results of these models indicate serious implications, such as diminishing water quality, reducing soil productivity, propagating heavy metals or other pollutants through ecosystems, and generally reducing environmental health for the local flora and fauna. Once again, LIBs predominantly impact toxicity, acidification, and eutrophication through their use phase.30 These effects result from the fuel combustion processes related to electricity generation. Additionally, the material extraction phase in the production of LIBs also contributes substantially to ecotoxicity, driven by the extraction of metals.31 This information indicates that battery chemistries and the grid composition of the areas in which batteries are used also play a major role in ecosystem health.
Finally, many researchers and consumers are also interested in the carcinogenicity of the studied device. This metric may be reported in terms of the carcinogen hazard,32 or it may be integrated into the broader metric of human toxicity.33 Predictably, the use phase and materials extraction phase of a LIB life cycle contribute most to the release of carcinogens and other toxins. This is predominantly due to the extraction of the metals required to build LIBs. One study actually reports that the human toxicity potential may be more than 200 percent greater for electric vehicles than for traditional internal combustion vehicles.34 Although not all of the results are positive, LCAs for lithium-ion batteries help consumers, legislators, and innovators understand where the environmental impacts are worst and may help them develop means of mitigating those impacts.
Conducting Sensitivity Analyses
One additional component of every thorough LCA is a sensitivity analysis. A sensitivity analysis is a mathematical tool that helps researchers determine how sensitive the results of the study are to various factors. For instance, the result of our sample study of lithium-ion batteries for electric vehicles is very sensitive to the particular battery chemistry employed.35 The type and magnitude of environmental impacts vary dramatically even among different lithium-ion chemistries. Additionally, use phase parameters, such as the lifetime of the battery and the composition of the electricity supplying the energy, significantly impact the results.36 Sensitivity analyses might also assess the impact of transportation on the product’s environmental impact. For instance, in a LCA comparing batteries manufactured in the United States with those that utilize components from all over the world, transportation impacts may play a major role.
Sensitivity analyses also allow researchers to identify notable areas of uncertainty. If total environmental impacts vary dramatically depending on certain inputs, it is important for researchers to be able to accurately define those inputs. If they cannot, then the results of the study may be subject to considerable uncertainty. While it is impossible to reduce uncertainty to zero, the most compelling studies—those that will inspire change—will avoid uncertainties to the extent possible.
Life-Cycle Environmental Impact of Lithium-Ion Batteries
Because there are myriad variables input into life-cycle assessments, different researchers may reach somewhat different conclusions about a given subject based on their individual methodology, boundary conditions, and data sources. Many researchers have studied lithium-ion batteries, particularly in electric vehicle applications. While these studies employ slightly different assumptions and draw a variety of different comparisons, there are some consistent findings over the past several years.37
The recent studies used in this chapter were all conducted using similar boundary conditions. All of the studies included battery production, use, and disposal. Several of the studies expanded their boundary conditions to include resource extraction and transportation. The studies tended to use the full life of a battery deployed in an electric vehicle as the functional unit, typically measured in total distance driven. Therefore, the results were most often presented in units of environmental impact per kilometer (or per mile).
Most of the studies also conducted sensitivity analyses related to end of life disposal techniques, useful life or cycle number of the battery, and composition of the grid electricity during use. Assumptions about these three factors seemed to consistently lead to the greatest variation in total life-cycle impact of the lithium-ion battery. Additionally, several of the studies included comparisons of various lithium-ion battery chemistries. Lithium-based chemistries tended to perform similarly, with greater disparity in comparisons between various other battery chemistries and between LIB-based electric vehicles (EVs), partial hybrid electric vehicles (PHEVs), and internal combustion engine vehicles (ICEVs).
Total Life-Cycle Environmental Impacts
Each study offered both good news and bad news for battery storage advocates. The bad news is that lithium-ion batteries used in electric vehicles have a greater environmental footprint than hybrid electric vehicles or traditional internal combustion engines when it comes to human toxicity, ecotoxicity, and some types of air pollution. The good news is that, under most conditions, they have a substantially lower global warming potential than other vehicle types.
Predominantly the supply chain and production phase add significantly to the environmental impacts of LIBs in electric vehicles. In fact, one study found that production phase of EVs is “more environmentally intensive than that of ICEVs for all impact categories with the exception of terrestrial acidification potential.”38 The most significant impact related to LIB electric vehicles is increased sulfur oxide (SOx) emissions, particularly for chemistries that require nickel, copper, or cobalt.39 Several studies found that SOz emissions were up to four times greater for EVs than for ICEVs over their lifetimes.40 Increased SOx emissions may lead to acid rain, respiratory problems, and decreased visibility, harming humans, ecosystems, and physical infrastructure.41 In addition, some models predict that various EV chemistries have from 180 to 290 percent greater human toxicity impacts compared to ICEV alternatives.42 Nonetheless, a shift away from nickel- and cobalt-based battery chemistries, responsible raw material sourcing, and recycling can all help to reduce the environmental impacts of increased SOx emissions.43 Particularly, recovering materials at the end of a battery’s life cycle “significantly reduces overall life-cycle impacts, as the extraction and processing of virgin materials is a key contributor to impacts across battery chemistries.”44
Similarly, EV contributions to global warming potential also predominantly occur in the production phase. “In contrast with ICEVs, almost half of an EV’s life-cycle GWP is associated with its production,” with one study finding that EVs create roughly two times more GWP during production than ICEVs.45 This is due in large part to the fact that the fuels used to create the primary energy used in EV production are hard coal, crude oil, natural gas, and lignite.46 The primary energy generated by these fuels is used to extract and process the metals used in LIBs, specifically aluminum and the particular metal used in the cathode.47
Importantly, “for some environmental impact categories, lower emissions during the use phase compensate for the additional burden caused during the production phase of EVs.”48 The use phase results in the majority of the GWP reductions for EVs over a battery’s lifetime, but emissions will vary substantially based on the composition of the electric grid.49 In most circumstances, LIB EVs create substantially fewer greenhouse gas emissions and result in a dramatically reduced global warming potential during the use phase. Yet, the reductions are most pronounced when the electric grid is comprised of natural gas and renewables rather than coal.50 A recent study by the Union of Concerned Scientists (UCS) found that “modeling of the two most popular BEVs available today and the regions where they are currently being sold, excess manufacturing emissions are offset within 6 to 16 months of average driving.”51
Another study found that when powered by average European electricity, EVs are found to reduce GWP by 20 to 24 percent compared to gasoline ICEVs and by 10 to 14 percent relative to diesel ICEVs under the base case assumption of a 150,000-km vehicle lifetime.52
For a United States comparison, one study found that EV greenhouse gas emissions are approximately 25 percent lower when using the California electric grid compared to the average grid composition across the whole country.53 That means that an average EV in the United States will result in approximately 33 percent less GWP compared to an ICEV, whereas an average EV in California will result in approximately 50 percent less GWP than an ICEV.54 According to the UCS study, “driving the average electric vehicle in any region of the country produces lower global warming emissions than the average new gasoline car achieving 29 MPG” and “about 66 percent of Americans—up from 45 percent just three years ago—live in regions where powering an EV on the regional electricity grid produces lower global warming emissions than a 50 mpg gasoline car.”55 And in looking to the future, the UCS study found that in “a grid composed of 80 percent renewable electricity, manufacturing a BEV will result in an over 25 percent reduction in emissions from manufacturing and an 84 percent reduction in emissions from driving—for an overall reduction of more than 60 percent (compared with a BEV manufactured and driven today).”56
Based on the current state of research, it is difficult to make blanket statements regarding the total environmental impacts of battery storage technologies. Yet, as the focus of this book is combating climate change, increased use of battery technologies seems like an important step in the right direction.
Life-cycle assessments are extremely important in understanding the total impact of batteries and other devices. But even these studies cannot account for everything. This is especially true in the case of innovations that are meant to catalyze a paradigm shift in their respective fields. In these instances, there are many spillover effects in addition to the direct impacts of the device itself. As discussed in other chapters of this book, the integration of battery technologies into different facets of society is poised to have a variety of spillover effects, both positive and negative.
To begin with, broader integration of battery storage into grid operations will support the expedited development of grid-scale renewable energy technologies and electric vehicles. This is one of the main objectives of grid-scale battery storage: broader use of renewable energy technologies to offset existing fossil-fueled resources, which are more carbon-intensive. Therefore, even if aspects of a LIB’s life cycle such as production and transportation generate additional global warming potential, those effects will likely be offset by reduced reliance on traditional energy resources.
Complicating this analysis is the current research and development underway exploring whether end of life batteries can effectively be deployed as grid storage (discussed in Chapter 6), thereby lengthening the useful life of transportation LIBs by another decade or so with electric grid storage use. This dual use reduces both the production and disposal LCA impacts through a more comprehensive utilization of the battery. Furthermore, increasing efforts to recycle LIBs will also improve the analysis in favor of LIBs. Lithium can be effectively recycled, but currently it is not cost effective for manufacturers to do so given the market price for lithium.
These types of spillover effects—that improve the efficiency and reduce the carbon intensity of the grid—tend to improve the total life-cycle impacts of battery technologies. Yet, there may also be spillover effects that worsen the total impacts of battery technologies.
In addition to failing to address regional effects on the grid, life-cycle assessments typically “[do] not address the economic or social aspects of a product.”57 These impacts may be significant depending on the nature of the product and may include things like material scarcity, social impacts of minimally regulated labor forces, and other socioeconomic issues. Unfortunately, these effects tend to worsen the total impacts of battery storage technologies.
To begin with, many of the materials necessary for increasing the efficiency and energy density of battery storage technologies are scarce and more costly.58 In particular, some studies estimate that stores of nickel and cobalt may be strained, with the demand from lithium-ion batteries representing between 4 percent and 9 percent of the world reserve base.59 Another study estimates that the demand for cobalt, in particular, is growing at 5 percent per year.60 Unfortunately, “more than half of the world’s supply of cobalt, used in LIBs, comes from the [Democratic Republic of the Congo].”61
This geographic distribution leads to another issue that is not captured by life-cycle analyses. Particularly in instances where components of a particular device are made internationally, there may be labor issues, human exposures, or other circumstances that U.S. consumers would consider violations. For instance, the cobalt mines of the Democratic Republic of the Congo (DRC) are replete with known and suspected human rights violations. At least 80 artisanal miners died underground in southern DRC between September 2014 and December 2015 alone.62 Mines in the DRC are also known to employ child laborers. In 2014, approximately 40,000 children worked in mines across southern DRC, many of them mining cobalt.63
Fortunately, other processes are in place to begin tracking some of these impacts. Companies are now required to disclose sourcing of “conflict minerals” from the DRC to the Securities and Exchange Commission (SEC).64 These regulations cover things like gold and the derivatives of tin and tungsten, but they do not cover cobalt and other materials used in the production of batteries.65 Yet, even for materials that must be reported to the SEC, the regulations have few embedded consequences for companies that are unable to fully identify countries of origin for their necessary minerals and materials.66 For example, many battery companies cannot trace the origin of the materials used in their products.67 Therefore, social impacts, like those that result from cobalt mining in the DRC, often go unaccounted for by any policy or procedure.
These social spillover effects represent negative impacts. Yet, there may be many other social impacts, depending on the battery chemistry, manufacturing processes, and geographic locations involved in the supply chain. These will not be included in most LCAs. Manufacturers, legislators, and other clean technology leaders, however, should do their best to identify these issues and mitigate the negative impacts to the extent possible.
Conclusion: Is the Electric Battery a Clean Technology?
In short, the electric battery, like other technologies, has both positive and negative environmental impacts. Studies consistently indicate, however, that the electric battery offers a lower carbon solution to transportation and they can help reduce the carbon intensity of the electric grid. There are also many ways in which we can amplify the positive impact of battery technologies and reduce their negative impacts. The most important thing that manufacturers and researchers can do to reduce the impact of battery technologies is to improve data collection for battery production processes. The greater the magnitude and reliability of battery data, the more accurate LCAs become. This accuracy allows researchers to more precisely identify areas of the battery life cycle that can most benefit from improvement. In many cases, increased transparency will be required relating to the sourcing and supply chain of particular battery components, namely, the cathode materials.68 Importantly, data collection will improve more quickly with pressure from consumers. While we build better data collection practices and accumulate more accurate data, there are some general improvements that the battery industry might strive for, such as improving the efficiency of batteries’ use phase as well as improved efforts at recycling. Innovations such as new uses for end of life EV batteries as well as exploring alternatives for increasing the recycling of LIBs will begin to address some of the environmental impacts and negative spillover effects of battery production.