Battery Second Life: A Review of Experimental Studies

arXiv:2610.11632 · eess.SY, cs.SY · Submitted 2026-10-08 · Read on arXiv

Listen

Transcript

Introduction to the show: ident: Robotics Radio. Generated commentary on the latest robotics and control papers.

Rosa: Today's paper: "Battery Second Life: A Review of Experimental Studies".

Dev: As a meticulous AI researcher, I have thoroughly analyzed both provided texts (A and B) concerning the paper "Battery Second Life:

Rosa: First, who's behind it and why it matters.

Title and authors: Rosa: So we're looking at this paper called "Battery Second Life: A Review of Experimental Studies." It’s by Marwan Hassini, Eduardo Redondo-Iglesias, and Pascal Venet. The title itself tells you that this isn't just a casual look at battery reuse; it’s a deep dive into the actual experimental work done on retired batteries.

Dev: Yeah, and what this review is really setting out to do is fill a gap in the literature. It points out that most existing reviews talk about the possibilities and barriers for reusing batteries, but they don't actually synthesize the experimental results from studies on retired cells.

Rosa: Exactly. The authors are saying that this synthesis of more than one hundred fifty articles presenting experimental data is really important because it moves us away from just talking about potential applications to looking at what's actually been measured in the labs <ref:2610.11632#pg2,of more than 150 articles>.

Dev: And the main finding they highlight right up front is their proposal of a new vocabulary for batteries. They want to create a way to properly distinguish between different states of battery health and usage, which is something that's currently quite confusing out there.

Rosa: That makes sense because when you look at what these one hundred fifty plus studies show, the performance can vary wildly, from zero percent capacity loss all the way up to over one hundred percent capacity retention <ref:2610.11632#pg2>.

Dev: It highlights that we can't just treat every retired battery as if it’s in the same category; there is a significant spread in performance levels that needs to be addressed.

The paper's summary: Rosa: So, let's talk about what this review actually summarizes. The authors are summarizing the experimental results from these many studies to show where we stand right now regarding retired batteries and their second life potential.

Dev: They’re saying that while some of these retired batteries can still perform quite well, there’s a real spread in how good they are, and they need specific management strategies because of that heterogeneity.

Rosa: They go into the details about degradation too. They note that aging can be slow and safe if you manage the experimental settings correctly, but they push to define what "deeply degraded" batteries actually mean, suggesting a threshold beyond just eighty percent State of Health.

Dev: That’s interesting because it challenges the idea that we only need to look at capacity loss as the main indicator of whether a battery is ready for reuse. They feel that measuring capacity alone isn't enough to make decisions about reusing batteries.

Rosa: And they strongly argue against just using State of Health as the sole metric, proposing instead something much more detailed, like a battery passport that includes degradation mechanisms and specific defects.

Dev: That means we need a way to look beyond simple health metrics and actually understand what's happening chemically inside the cell to predict its long-term behavior in a second life application.

The paper's improvements: Rosa: The paper also suggests some big improvements for how we approach this whole topic. They are suggesting we stop relying on just one way of looking at battery health and start using a more comprehensive framework.

Dev: They advocate for adopting a systematic approach to literature review, basically following PRISMA principles, so that future research can be more reliable and comprehensive when they are gathering data on these batteries.

Rosa: And they urge the community to adopt a cumulative research approach, meaning future studies should look at previous results and use consistent health indicators to set benchmarks so we don't see such variability in the literature.

Dev: It’s about reducing that noise in the field by making sure everyone is using the same language and metrics when they are testing these batteries.

Rosa: They also point out that current characterization techniques often focus too much on electrical measurements, but they identify a lot of promising rapid methods, like incremental capacity analysis or electrochemical spectroscopy.

Dev: The authors conclude that the key to long-term, linear aging during a second life is really about controlling those experimental conditions and having a solid battery management strategy in place.

Conclusion: Rosa: So wrapping up what we've heard, this paper, "Battery Second Life: A Review of Experimental Studies," really pushes the field toward a more unified understanding of how to assess retired batteries. It’s about getting past the individual studies and seeing the bigger picture.

Dev: They emphasize that the main implication is moving from just observing performance to building a proper decision-making tool, like that battery passport they suggest, which aggregates all those different metrics.

Rosa: And they lay out clear paths for future research, specifically calling for a registry of experimental studies for each battery model so we can standardize those health indicators and aging conditions across the board.

Dev: They want systematic procedures for determining if a retired battery is actually suitable for reuse, which helps bridge the gap between lab findings and real-world application.

Taro: From an autonomy researcher point of view, what I find interesting is how this review forces us to think about what happens when the world misbehaves during a second life; they are pushing for predictions about degradation under new operational constraints.

Rosa: That's a crucial point, Taro. We need to make sure that when we reuse batteries, we aren't introducing unexpected failure modes because the aging dynamics change based on how we use them.

Dev: And I agree with that, Rosa. If the degradation is nonlinear, they tend to be affected by their new operating conditions, so usage change should preferably happen during a linear degradation phase of the battery.

Marwan Hassini, Eduardo Redondo-Iglesias, Pascal Venet

Université Gustave Eiffel · ENTPE · EMob-Lab

eess.SY, cs.SY

Submitted: 2026-10-08

Updated: 2026-10-08

The gist: As a meticulous AI researcher, I have thoroughly analyzed both provided texts (A and B) concerning the paper "Battery Second Life: A Review of Experimental Studies." My objective is to synthesize

Key concepts

Performance Dispersion: Retired batteries show highly varied performance levels when tested. Instead of one consistent result, the data ranges from 0% to over 110%, with an average around 84.5%. This wide spread means management strategies must account for these inconsistencies.
Battery Passport: This proposed framework is a comprehensive record for a battery that goes beyond just its current State of Health (SOH). It includes detailed information about the battery's history, degradation mechanisms, and usage changes. This holistic view is necessary to accurately assess if a battery is safe for reuse.
Second Life Battery: A second-life battery is one that has been used in its first application (like an electric vehicle) and then repurposed. This repurposing can involve physical changes, such as altering the module count or configuration, or changing the management system to suit a new use case like stationary storage.

Terminology

State of Health — 67× in this paper
Accuracy — 12× in this paper · explained in 1 episode
Clustering — 11× in this paper · explained in 2 episodes
Benchmark — 6× in this paper · explained in 2 episodes
IEEE Vehicle Power and Propulsion Conference — 5× in this paper
Consistency — 3× in this paper · explained in 3 episodes
Convolutional Neural Network — 2× in this paper
Feature Extraction — 2× in this paper · explained in 2 episodes
Feature Selection — 2× in this paper · explained in 4 episodes
Long Short-Term Memory — 2× in this paper
Aggregation — 1× in this paper · explained in 1 episode
Attention Mechanism — 1× in this paper · explained in 3 episodes
Calibration — 1× in this paper · explained in 7 episodes
Conv Block Attention Module — 1× in this paper
Core — 1× in this paper · explained in 2 episodes
Energy Efficiency — 1× in this paper · explained in 3 episodes
Failure Modes — 1× in this paper · explained in 3 episodes
Homogeneity — 1× in this paper · explained in 2 episodes
IEEE Energy Conversion Congress and Exposition — 1× in this paper
Lithium Cobalt Oxide — 1× in this paper
Lithium Fer Phosphate — 1× in this paper
Lithium Manganese Oxyde — 1× in this paper
Predictive Maintenance — 1× in this paper · explained in 2 episodes
Research Review — 1× in this paper · explained in 6 episodes
Transformer — 1× in this paper · explained in 2 episodes

Summary

As a meticulous AI researcher, I have thoroughly analyzed both provided texts (A and B) concerning the paper Battery Second Life: A Review of Experimental Studies. My objective is to synthesize these summaries into a single, comprehensive, and highly detailed description of the review article, ensuring accuracy and capturing all critical nuances.

Here is the detailed synthesis:

Comprehensive Summary of Battery Second Life: A Review of Experimental Studies

This review article provides a rigorous and critical synthesis of experimental studies concerning the performance, aging mechanisms, and safety considerations for retired and reused batteries, focusing specifically on their potential in second-life applications. The authors aim to clarify the current state-of-knowledge by addressing significant gaps in the existing literature.

Core Contributions and Scope:

Conceptual Framework & Terminology: A primary contribution is the proposal of a new, clarified vocabulary for battery end-of-life management. This involves distinguishing between terms such as 'fresh,' 'aged,' 'retired,' and 'highly degraded' batteries, establishing consensus based on both life cycle and performance levels.
Performance Characterization: The review synthesizes experimental data from over 104 articles presenting characterization tests on more than 10,000 retired cells. The findings indicate that while the performance of these retired batteries can remain high, there is significant dispersion in performance levels—ranging from 0% to over 110%, with a median of 84.5% and a mean of 79.9%. This wide dispersion necessitates specific management strategies due to heterogeneity between cell capacities or the presence of outlier cells.
Degradation and Aging: The review evaluates aging studies, noting that degradation can be slow and safe under controlled conditions, provided experimental parameters are managed appropriately. A key area of focus is defining deeply degraded batteries, suggesting a threshold beyond the conventional 80% State of Health (SOH) for further research into how deep degradation affects the speed and dispersion of aging.
State-of-Health (SOH) Evaluation: The authors strongly argue against the current, narrow use of SOH—which is often used merely to track performance or stop experiments—as the sole metric for reuse suitability. Instead, they propose a more comprehensive evaluation framework that systematically incorporates a battery passport and an in-depth analysis of degradation mechanisms and defects.
Second Life Definition: The review proposes a clear definition for second-life batteries: those that have undergone either usage changes (e.g., first life in a vehicle vs. reuse in stationary storage) or structural changes (e.g., module count modification, series-parallel configuration alteration, or BMS/thermal management system updates).
Methodological Synthesis: The review adopts the PRISMA principles for systematic literature review to ensure reliability and comprehensiveness, investigating a total of 157 articles [20–177]. It emphasizes the need for a cumulative research approach, urging future studies to consider previous results and use consistent metrics (health indicators) to set benchmarks, thereby reducing variability in feature names and experimental conditions.
Characterization Techniques: While many current efforts focus on electrical measurements, the review identifies a diversity of promising rapid characterization techniques, including incremental capacity analysis, electrochemical spectroscopy, charge-discharge cycles, and current/voltage curve analysis. It concludes that controlling experimental conditions and implementing an adequate battery management strategy are essential for achieving long-term, linear aging during second life.
Future Directions: The paper outlines crucial future research needs:

The establishment of a registry of experimental studies for each battery model to standardize health indicators and aging conditions.
Systematic evaluation procedures (as illustrated in Figure 5) for determining the suitability of a retired battery for reuse.

Supporting Data and Appendices:

The review is supported by extensive datasets shared under an open license, including:

Appendix A: The primary dataset used for the review.
Appendix B: Experimental conditions from aging campaigns leading to short-lifespan studies.
Appendix C: Experimental conditions for long-lifespan studies.
Figure 5 & Figure 11: Syntheses of ideal procedures and the evolution of performance dispersion across four aging studies, respectively.
Tables (4, 7, 8, 9): Present specific data distributions (e.g., performance distribution in Table 4), health indicators derived from different testing methods (Table 7 for charge-discharge cycles; Table 8 for processed curves), and pioneer study results using non-destructive techniques (Table 9).
Figure 12 & Figure 13: Illustrate degradation curves down to 35% SOH and the effect of various usage changes across multiple studies.

Contextual Relevance:

The review highlights the timeliness of this work, noting that while experimental studies on retired batteries existed previously, significant interest has only recently gained traction (since 2021), with 102 articles published in the last five years (2021-2026). The literature reviewed spans various battery chemistries and applications, including Li-ion cells for electric vehicles, grid storage systems, and lead-acid batteries.

**In essence, this paper serves as a foundational synthesis that moves the field from disparate experimental findings to a unified understanding of how to reliably assess the performance and predict the safe reuse potential of retired batteries.

Improvements for AI systems

A new vocabulary for battery classification will be implemented to clarify ambiguous terms like second life battery by using definitions based on position on life cycle and performance level, specifically distinguishing between Remanufactured, Repurposed, and Deeply degraded batteries.
The AI system can perform a more realistic evaluation of end-of-life thresholds by moving away from the common assumption that battery end of first life is reached after a capacity loss of 20-30% [191, 192] toward using performance fade related to usage, as suggested in Section 5.1: Favoring such definition would lead to the fact that measuring capacity is also an insufficient method for assessing performance in make decision related to batteries end-of-life and reuse.
The system can rapidly assess battery health using a more robust health indicator approach by prioritizing looking for health indicators highly correlated with state of health is the most promising approach rather than relying on correlations between capacity and resistance, as noted in Section 5.2: Instead, looking for health indicators highly correlated with state of health is the most promising approach for selecting batteries for reuse.
The AI can predict degradation dynamics under specific operational constraints by analyzing experimental conditions, learning that if the degradation is nonlinear they tend to be affected according by their new operating conditions, and suggesting that usage change should be preferably done during a linear degradation of the battery (Section 4.6).
The system can identify potential safety risks in reused batteries by linking them directly to aging mechanisms, specifically by ensuring abusive use and aging conditions that encourage lithium plating should be avoided, as implied by the findings in Section 4.7: Safety risks can be minimized by avoiding conditions that accelerate aging.
The AI system can develop a decision-making tool, such as a battery passport, which aggregates diverse metrics to inform reuse decisions, addressing the review's call for a more complete document that could help in the decision-making (Section 5.1).