The worldwide demand for battery cells is increasing exponentially, driven particularly by the growing success of electric vehicles (EVs). Current studies predict a demand of approximately 2600 GWh/a for 2030 [1]. Others even predict demands of up to 10 000 GWh/a, but without referencing to an exact year [2]. As a result, battery cell production capacity is being rapidly expanded worldwide; e.g., by the end of 2020, 800 GWh of battery cell production capacity was announced or planned in Europe alone [3]. A change in this trend is not expected at present. Such a scenario has led to a strong ongoing price competition among battery manufacturers. The literature reports on different effects that contribute to the cost reduction of battery technology. For example, it has been reported that economies of scale—cost reduction through learning rate after increasing production quantity (in this case doubling)—can reduce the price as much as 6%–9% per battery pack [4]. The development of optimized materials for use in lithium-ion cells, such as active materials for cathodes, anodes, and electrolytes, as well as corresponding inactive materials, can help achieve higher energy densities and lifetimes, thereby indirectly reducing costs at the cell level since less material is required per cell [5]. In addition, price advantages in terms of more efficient methods of material synthesis can be passed on to cell-producing customers. Another crucial cost-reducing factor is the optimization of the battery cell production process. Studies have reported high cost sensitivity of electrode production, advanced formation and aging processes, continuous process control, and process substitution as targets for innovative process technologies [6]. The integration of new production technologies is required to reduce prices even further [2]. However, for many production researchers and manufacturing engineers, it is rather difficult to identify the part of the process chain for which new production technologies are needed the most and the economic impact these new approaches might bring. One major reason is that the battery cell cost and, in particular, the exact cost structure is somewhat complex and involves variables other than those from usual cell and process design activities. The general obstacles that inhibit production of reliable battery cells that are cost efficient are as follows.

1. Costs usually depend on industry activity and are, therefore, strongly confidential.

2. The battery market is not fully developed yet. Battery cells, especially for EVs, are being produced in significant amounts only for the last couple of years. Thus, the current knowledge and experience base for EV applications is still inadequate.

3. As the market is relatively young and growing strongly, production know-how is being learned continuously, leading to a decrease in production costs.

4. There is no stable ratio between supply and demand. Both are changing rapidly and have a substantial impact on costs and prices. Other influencing factors include market structures such as possible oligopolies and lack of information availability, the discovery of new raw material sources and development of efficient refining methods, and the use of new cell chemistries (which changes the demand for existing materials and thus costs). Note that the terms “price” and “cost” are often used simultaneously in existing studies.

5. Battery cell costs depend strongly on material prices, and which in turn depend significantly on the quantity of purchase, which varies as well.

As every business-oriented organization aims to maximize profit while reducing costs under the given conditions, the same is true for battery cell production. A literature analysis by Duffner et al. [7] from 2020 shows an average cost decrease of approximately 50% for lithium-ion battery cells between 2015 and 2020. More importantly, even today, the cost assumptions range from 84 [8] to 138 €/kWh [3] for state-of-the-art lithium-ion NMC cell chemistry. Thus, there is a large variation in reported costs, in general and over time. In addition, many costs for battery cells are not split in terms of material and production costs. There are studies that try to analyze the cost structure, but strong differences can be observed here as well. The German Engineering Association (VDMA) reports, for example, material cost proportions of 60% [3], while BMW has reported material cost proportions of 80% [9]. Therefore, there are large differences in quotable battery cell production costs, and it is difficult to elaborate on their validity and how they are calculated. Similarly, there are papers and publications that explain the breakup of production costs through different methodological approaches (e.g.,[3], [7], [10], [11], [12], [13]). Most of these studies gathered information from various other sources (secondary data). However, the combination of these data from different sources for a cost model is not perfect because of the following.

1. The primary source and its credibility are mostly unknown.

2. The locations of the analyzed production sites differ, which has a significant effect on production costs.

3. The exact time of data collection is unknown. Owing to the rapid developments in battery cell production, the life of available information diminishes fast.

4. The exact cell chemistry is unknown or not named, even though it has a significant effect on cost per kWh.

5. The cell format is unknown, which also has a significant effect on the cost per kWh.

6. The cell design is unknown. If a cell design has a high-power layout or a high-energy layout, it also has a significant effect on the cost per kWh.

Despite the abovementioned shortcomings, a cost model involving real production-related large-scale data on state-of-the-art battery production technology, allowing for full transparency and access to primary data, is missing. In addition, the effects of new battery cell manufacturing technologies on production economics are largely unknown.

The present study aims to contribute toward answering the following question by considering battery cell production as a specific case: How can the manufacturing costs be improved by new process technologies or general improvement measures?

This study is settled in the field of technology assessment and offers implications for literature on production economics and technology planning. Our work contributes in particular to reducing complexity of technology assessment problems in the light of integrating new and emerging technologies. This is achieved by methodically disclosing the cost structure of battery cell production (as exemplary processing industry) in a generalizable way and elaborating on the impact, which results from the integration of new and emerging technologies. It furthermore adds to the robustness of bottom-up cost modeling by providing validation for the case of real battery cell fabrication. Implications for technology managers, researchers, and policy responses are discussed as well.

The rest of this article is organized as follows. Section II introduces the research work of the present study by first referring to the literature on production economics. Subsequently, existing preliminary work on cost modeling of battery cell production is recapitulated, and special features of the field of battery research are highlighted. This emphasizes the relevance of the topic from the perspective of the academic discipline of innovation and technology management. Section III outlines the case study of a lithium-ion battery cell manufacturing facility analyzed in this study by first discussing its specific background (funding and task profile). Subsequently, actual planning data on the facility layout, the production throughput, cost-relevant equipment specifications, and the production balancing are provided. Section IV describes the chosen methodological approach for cost modeling and discusses it with alternative approaches. Section V provides an overview of production costs and discusses the influence of production-related optimization factors. Subsequently, the influence of technological innovations on production is described, and an outlook on recommendations for action derived from these is given. Finally, Section VI concludes the article with the contributions and limitations as well as recommendations for future research.

SECTION II.

Literature Review

Meeting cost targets is, besides achieving quality and throughput targets, arguably an essential component of decisions on the use of technologies in the production process. Work in the scientific field of engineering and technology management, which is closely related to the subject of this study, focuses on the development and testing of methodology for problems of technology assessment [14] and decision-making [15], [16]. Although the studies discussed in the following originate from multidisciplinary scientific streams, they provide a solid overview of cost structures in battery cell production as an exemplary manufacturing sector, which is a basis for understanding possible cost-reducing measures in the context of technology assessment.

Even though recent studies have reported annual cost reductions of approximately 8% (between 2007 and 2014) [4] and promising cost levels of 70–90/kWhforcellsand150/kWh per pack [17], [18], there are a variety of variables that can lead to the realization of these cost reduction potentials. Different strategies exist to achieve these goals, which have been discussed and investigated in various publications. Thus, in terms of the cost structures of lithium-ion battery cells, it is well known that a large part is due to the sourcing and further processing of battery materials (especially cathode, anode, and electrolyte). Certain studies even see material price as the central limiting factor for the price reduction potential of lithium-ion batteries [12]. Purchasing large quantities to reduce costs per unit quantity are suggested as strategies to achieve cost targets. In addition, in-house production of materials or precursors is under discussion for backward integration along the value chain. This helps extend value creation processing within the company [19], which also involves the establishment of strategic partnerships [6]. Certain researchers have addressed their cost forecasts on battery cells and packs accordingly, examining the impact of battery materials utilized [5], [20], [21]

In addition to the strategic approaches to input factors aimed at increasing cost efficiency, there is a wide scope for actions required to realize cost degression effects. These concern the actual production process, for example, by balancing the production size. Studies that address the size of production scale define a corridor from 200 to 300 MWh/year to 2 GWh/year as the most efficient so far [10], [22]. With the prospect of positive effects of balancing (in this case, the capacitive load of electrode production), larger factories can also be operated in a cost-optimized manner in the future. This effect was quantified as a cost advantage of >$5 kWh-1 [10]. Studies that address the impact of compound effects assume savings of between 9% and 21% per battery pack based on bottom-up calculations [11]. In addition to the design of the size scaling of production processes, the production location, and associated costs due to employee wages, energy prices, building prices, and tax levies are other key issues that influence production costs. Studies attribute a cost difference of $6.4/kWh to the location decision and address factors indirectly affecting production, such as GHG emissions through energy mix, knowledge structure, labor market, and industrial development [23], [24], [25]. Closely related to the approach of using business strategies to reduce material costs is the approach of fostering technological efficiency to optimize resource allocation. For example, studies address critical resources and their alternatives [26] on the one hand, and material recycling opportunities or reusability concepts [27], [28] on the other. Optimization work with the aim of reducing energy consumption [29], as well as elaboration on the digitalization of the production environment and quality assurance have so far been primarily simulation based [30], [31], [32], [33], [34]. The establishment of continuous process control enables a more uniform design of manufacturing stages, potentially leading to lower buffer inventories, and thus, enabling a lean, cost-optimized manufacturing environment [7], [35].

Another way that is widely discussed in academic literature is the integration of technological innovations to substitute or streamline the required inputs. For example, it has been shown that investments in plant equipment show potential savings by eliminating solvent recovery equipment (when processing with NMP). This is enabled by alternative processing routes using aqueous solvents [36]. The use of dry coating processes that eliminate solvents can lead to substantial savings in terms of equipment investment and ongoing costs due to energy consumption [37].

The cell finishing, the process section at the end of the manufacturing chain for lithium-ion battery cells, comprises the steps of formation and cell aging. The formation step refers to the formation of the solid electrolyte interface (SEI), an interface layer at the electrodes of the battery cell that allows physical and chemical thermodynamic processes to take place by applying voltage and running a time- and phase-dependent program [38], [39]. The formation step is particularly quality determining and has an enormous impact on cell performance [6], [40]. However, formation is considered one of the most cost-intensive processes in battery cell manufacturing due to its enormous energy consumption. Therefore, cost reduction strategies often deal with reducing the time of the forming cycles [41]. This holds implications for the energy costs required and involves forming strategies to accelerate the formation of the necessary boundary layer between the electrode and electrolyte [42], [43]. The following examples, among others, are known in the literature as strategies to save time.

1. Narrowing the voltage window of the formation cycle can effectively shorten the formation time [40].

2. The application of pulsed current charging enables a higher charging rate in the formation process, and thus, shortens the formation time [44].

3. The application of simultaneous elevated temperature and mechanical loading resulted in time reductions in the formation process [41].

4. The approach of applying an artificial SEI layer by thin film techniques such as atomic layer deposition, rather than its formation out of the components of the cell (primarily the electrolyte) by formation programs, has so far only been demonstrated on a laboratory scale [45], [46], [47], but may provide a crucial time and energy advantage in the future.

However, these strategies are all still subject to debate in the scientific literature and definitely require further intensive, in-depth understanding of the formulation-dependent physicochemical processes within the cell before they can be applied on a scaled-up, industrial scale [48]. In this regard, the uncertain formation mechanism and composition of the SEI are major obstacles in understanding formation and aging. Studies illustrate that formation programs aiming at time optimization are in tension between minimizing impedance rise, improving capacity maintenance, and avoiding lithium plating [49]. Improper forming processes that are not matched to the conditions inside the cell can lead to uneven formation of SEI or undesirable lithium plating, which result in negative effects on lifetime due to premature capacity degradation and safety risks due to dendrite formation and short-circuit hazards [49], [50], [51], [52]. The development of advanced characterization techniques can contribute to further understanding of the formation, property formation, and aging behavior of SEI, and thus, represents the starting point of time-shortening or energy-saving measures.

18650 21700 26650 32650 સિલિન્ડ્રિકલ સેલ પ્રોડક્શન લાઇન

As part of the manufacturing process, cell aging is an end-of-line test that has the task of detecting irregular capacity degradation of individual cells over time, and thus, identifying defective cells [42]. Cost-reducing innovations are primarily concerned here with measurement technology and data-based modeling, which are able to detect defective cells more quickly, and thus, contribute to streamlining the overall process

In addition to the possibility of innovating the existing process through technological innovations, a final way to be mentioned is to increase the performance of the cells through product innovations (in this case, optimization of the cell chemistry). This reduces the number of cells required, which indirectly contributes to an increase in efficiency. Solid-state batteries, although highly uncertain due to the lack of real production data, have already been addressed by bottom-up cost calculations and predict, competitive costs in the long term, at least for their subcategory of solid-state sulfide cells [56]. For oxide batteries, the energy-intensive processing step of sintering is considered an obstacle. The aerosol deposition method is shown to be promising and could bring cell technologies down to a cost level of up to $150/kWh [57]. Implications for their processability and possible optimization potentials by increasing the throughput or reducing the residence time are also discussed [58]. Depending on the intended application, the cell designs can be varied. For example, particularly thick electrode layers, especially for high-energy applications in all-electric automobiles, could lead to a reduced BEV pack cost by an additional 8% [22]. In addition to the variation in the electrode properties, there is further potential for innovation in the design of current collectors. Recent examples consider scientific studies as well as announcements by automotive companies regarding multitab or “tabless” design” [59], [60].

The findings of investigations into the various subaspects of battery cell production are aggregated in studies on the modeling of production costs. For lithium-ion battery cell manufacturing, however, this is not a completely unaddressed topic; thus, earlier works can be referenced. One key dataset used by a variety of studies, particularly for building bottom-up models to replicate battery cell manufacturing costs, is the BatPaC database from Argonne National Lab. BatPaC is a freely available LIB design and cost model that allows the evaluation of different LIB designs and chemistries to be predicted using simulation. This is done based on laboratory data and pack-level metrics. By evaluating the cost of battery packs at specified production levels, it can be used to predict material and energy requirements and to identify opportunities for cost reduction [61], [62]. Duffner et al. [53] provided an excellent overview of the previous research landscape on battery cell manufacturing cost modeling, highlighting the BatPaC model as being the most influential database in this scientific field to date.

Previous studies have highlighted a variety of individual content facets on the part of the impact of cost sensitivity of the selected product or process innovations. However, we observed here that the sources and references cited in many studies contain outdated data or data that have not yet progressed beyond a bottom-up approach [21], [63]. Datasets based on real battery cell production planning data are urgently needed to validate these planning-conceptual approaches. In this study, we follow up on the recommendations for future research by Duffner et al. [64]. Thus, the use of quantitative models for cost calculation, the provision of model architecture and input data, as well as the thematization of novel technologies and calculation of a standard format (21700) are carried out. In addition, our work responds to the call for additional integration of process technology innovations and validation through concrete planning/operational data of a cell manufacturing research factory [7].

Furthermore, since the field of battery cell manufacturing has high innovation dynamics, studies based on current datasets are urgently needed to advance the field and conduct research toward successful technology transfer. Studies indicate that it is more difficult for battery research, exemplified as a subdiscipline of energy technologies encompassing diverse knowledge sectors [65], [66], to integrate innovative technologies into production processes than in comparable manufacturing sectors. This is due to the additional coordination effort of knowledge and its transfer to other use environments (use environments, [67], [68]) (e.g., from research to industrial application, or even use in different end products) [69]. This underpins the relevance of the topic of battery cell manufacturing to the knowledge field of technology and innovation management. Furthermore, it provides a case [70] that is suitable for deriving generalizable conclusions for similar industries involving multisectoral knowledge fields [71]. Discussions on the manifestations of technological change in industry, such as battery cell manufacturing as a multisectoral industrial field have scarcely been discussed so far. The analysis of concrete case studies can offer an added value to built-up knowledge, such as the identification of correlations between innovations of different origins and their influence on production efficiency at the system level. Such insights are helpful in taking a holistic view without disregarding individual technological innovations and their specific contributions. For example, different inputs can often be substituted for each other. Technological innovations that influence production efficiency should be discussed here, as well as novel application conditions that influence the appearance of the end-product, i.e., the battery cell.

In Germany, the federal government is completely funding the construction of a large-scale factory for lithium-ion battery cell production with more than 700 million €, solely for research reasons. Fig. 1 shows a rendered picture of the factory during the earlier planning phase. In this factory, namely the “Research Fab Battery Cells FFB,” battery cells shall be produced on an industrial scale, production problems will be identified, and new production technologies will be tested [72]. By setting up this unique technology accelerator, Germany and Europe aim to catch up with their Asian counterparts, which dominate the global battery cell market today. The factory, which is run by the “Fraunhofer-Gesellschaft,” will be technically capable of producing up to 7.0 GWh/a of electrodes. This is close to industrial scale (e.g., SK Innovation, Georgia-US, 11.7 GWh/a [73]; Northvolt Zwei, Salzgitter-Germany, 16 GWh/a [74]; and CATL, Erfurt-Germany, 14 GWh/a [75]). Therefore, the “Research Fab Battery Cells FFB” will probably be the largest demonstration factory of its kind worldwide. The factory, which is the basis for this study, is going to be located in Münster, in the west of Germany. Two large state-of-the-art coating lines will be located at the elongated side of the canal. Right next to it, the assembly and formation as well as aging and testing for different cell formats (cylindrical, pouch, and prismatic) will be located. As the mission of this research fab is to gain knowledge about the processing of battery cells, good comparability through the use of a standardized format (such as the 21700 cell) is beneficial toward achieving this goal and the same is also recommended by corresponding literature for the development of further insights into cost modeling and the identification of complex cause–effect relationships [6], [7]. The output of the assembly and forming lines, however, are much smaller than that of the electrode lines, capable of producing 30 cylindrical cells per minute.