Second-Life EV Batteries in Stationary Storage: Techno-Economic and Environmental Benchmarking vs. Pb-Acid and H2
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThis study conducts a comprehensive techno-economic and environmental assessment of the reuse of lithium-ion batteries, lead-acid batteries, and hydrogen energy storage systems for electric vehicles, employing an optimised scheduling framework and a unified functional unit. The advantage of the proposed method lies in its transparent integration of circular economy pathways and life cycle assessments, which clearly define the value of reuse. However, the manuscript's shortcomings include insufficient quantification of uncertainty in the health status of secondary batteries, potential limitations in conclusions drawn from specific regions, and inadequate analysis of long-term ageing effects.
The content of the manuscript is within the scope of the journal and can be of broad interest to readers. However, in terms of specific content, there is still room for improvement. Therefore, I decided to make the decision of major revision. It is recommended that the author properly absorb the reviewers' comments and make corresponding improvements and enhancements.
1. For the keywords, 'hydrogen storage', 'Levelized Cost of Storage (LCOS)', 'circular economy pathway', and 'Optimal dispatch' should be added to attract a broader readership.
2. Page 3, 'The technical studies highlight that the State of Charge (SoC), cell heterogeneity, degradation history and safety constraints strongly influence the achievable capacity, power limitations and lifetime during reuse. Practical implementation of second-life requires screening/diagnostics, module reconfiguration, adaptation of the Battery Management System (BMS) and compliance with stationary standards; these steps introduce additional costs and environmental bur-dens that need to be consistently considered in the techno-economic frameworks and LCA.'
I consider the author overlooked a crucial point regarding the utilisation of secondary lithium batteries: the need for fast identification of micro-health parameters. For example, a second-use application is the optimal solution for retired EV batteries to effectively avoid energy waste and use the remaining value of retired batteries. However, the long-time performance tests increase the second-use application cost, and the regular classification basis (battery capacity or internal resistance) cannot guarantee the consistency of regrouped retired batteries, which will accelerate the battery performance degradation. The micro-health parameters stand for the performance of active material and electrolyte inside the battery, and the changes in the micro-health parameters can present the battery's internal health state. A rapid method for identifying micro-health parameters can improve testing efficiency and provide a basis for classifying retired batteries (10.3390/batteries9010064). This issue should be further clarified and introduced.
3. When comparing the three different technological approaches, the system boundary definitions for each system need to be more precise and unified. In particular, the multiple energy conversions (electricity-hydrogen-electricity) and auxiliary energy consumption involved in hydrogen energy storage systems should be clearly explained in detail to maintain comparability with other technologies within the LCA framework, avoiding issues related to functional unit equivalence.
4. Cycle life predictions for reusable lithium-ion batteries lack a systematic discussion of the impact of historical usage conditions, differences in battery chemistry (e.g., NMC vs LFP), and screening/recombination processes. It is recommended to introduce more refined degradation models and quantify the impact of these uncertainties on LCOS and LCA results through sensitivity analysis.
5. Page 4, 'HES-based storage pathways are typically proposed for long-term and seasonal applications due to scalable energy capacity, via storage tanks/caves, and the potential for cross-sector connectivity.'
I think this section on HES challenges is too brief and does not seem to cover any specific technical challenges. I believe that at the very least, this system should involve an electrolysis hydrogen production system and a fuel cell power generation system, and the durability of these systems is the key issue that needs to be addressed, as it is also the biggest challenge.
For example, fuel cells will experience a gradual decline in performance during long-term operation. This performance degradation is caused by a variety of complex factors, including the degradation of electrode materials, a loss of catalysts, mechanical damage to the membrane electrode assembly, and fluctuations in operating conditions. Performance degradation not only affects the efficiency and output power of fuel cells but also shortens their service life and increases maintenance and replacement costs. The authors may refer to 'degradation prediction of proton exchange membrane fuel cell performance based on a transformer model'. This challenge should be introduced briefly, and a similar challenge also happens in the electrolyser section.
6. The study focused too much on the GWP indicator and failed to comprehensively reflect the environmental trade-offs of different technologies. It is recommended to expand the scope of LCA assessments to include dimensions such as resource scarcity (e.g., critical minerals like lithium and cobalt), human toxicity, and ecosystem quality, and to discuss the potential risks of burden shifting.
7. The economic model only considers simplified price signals and feed-in tariff subsidies, neglecting the diverse value streams in which energy storage can participate in modern electricity markets (such as frequency regulation, capacity markets, and congestion management). It is recommended to construct a more comprehensive market participation strategy model and analyse the impact of different market designs on the ranking of technological and economic competitiveness.
8. Section 3.3.2 of the manuscript mentions the complexity of the hydrogen energy model (electrolyser-storage tank-fuel cell), but does not discuss in depth the impact of fuel cell parameter estimation on system efficiency. It is recommended that the authors refer to [Energies 2024, 17(12), 2917], which proposes a high-precision parameter estimation method for the proton exchange membrane fuel cell (PEMFC) voltage model, which can directly improve the accuracy of fuel cell performance simulation in its hydrogen energy storage model.
It is suggested that the original text's 3.3.2 Hydrogen Model section be supplemented with the following explanation: "Fuel cell performance has a significant impact on the system's round-trip efficiency, and the uncertainty in its parameter estimation may lead to deviations from the optimised results in actual scheduling. It is recommended to adopt advanced parameter estimation methods to improve the reliability of model predictions."
9. The unique safety risks (such as inconsistency and thermal runaway probability) and reliability issues of reusable battery systems have not received sufficient attention. It is recommended to supplement this with a quantitative or semi-quantitative safety risk assessment, discuss the potential impact of failure modes on total lifecycle costs and environmental impact, and conduct a system comparison with new batteries and lead-acid batteries.
10. The manuscript's conclusions lack specific policy recommendations. It is recommended that, based on the research findings, specific and feasible policy interventions be proposed, such as a certification standard system for reusable batteries, a recycling responsibility allocation mechanism, and the design of circular economy incentive policies, to enhance the research's practical value for policymakers.
Author Response
We sincerely thank Reviewer 1 for the careful reading of our manuscript and for the highly constructive, technically insightful, and practically relevant comments. The reviewer’s suggestions helped us improve the manuscript substantially in terms of literature coverage, boundary definition, uncertainty treatment, environmental interpretation, and policy relevance. In the revised version, we have strengthened the discussion of second-life battery screening and uncertainty, clarified the harmonized system boundaries across all storage pathways, expanded the treatment of hydrogen subsystem challenges, improved the discussion of burden shifting beyond GWP, and added more specific policy implications. We believe that these revisions have significantly enhanced the rigor, clarity, and practical value of the manuscript.
Comment 1: For the keywords, “hydrogen storage”, “Levelized Cost of Storage (LCOS)”, “circular economy pathway”, and “Optimal dispatch” should be added to attract a broader readership.
Response:
We thank the reviewer for this useful suggestion. We agree that the discoverability and thematic positioning of the manuscript can be improved by refining the keywords. In the revised manuscript, we have added the suggested keywords, including hydrogen storage, Levelized Cost of Storage (LCOS), circular economy pathway, and optimal dispatch. This modification broadens the visibility of the study to readers working in energy systems, storage economics, circularity assessment, and optimization-based operational analysis.
Comment 2: The manuscript overlooks the need for fast identification of micro-health parameters in second-life lithium batteries...
Response:
We thank the reviewer for this important observation. We fully agree that rapid identification of micro-health parameters is highly relevant for the effective repurposing of retired EV batteries. In practical second-life applications, screening based only on conventional state-of-health indicators may be insufficient to ensure consistency among regrouped cells and modules, which can in turn accelerate imbalance, non-uniform aging, and performance degradation during stationary operation.
In response, we have expanded the literature review and discussion to explicitly address micro-health-aware screening and classification as an important aspect of second-life battery deployment. We now clarify that fast identification of internal condition indicators can improve sorting efficiency, reduce regrouping uncertainty, and enhance the reliability of repurposed systems. We have also incorporated the reviewer-suggested reference and connected this issue more directly to repurposing cost, system consistency, and uncertainty reduction.
At the same time, we clarify that the present study is intentionally framed as a system-level techno-economic and environmental benchmarking framework, rather than a cell-level electrochemical diagnostics model. Detailed micro-health identification is therefore not modeled explicitly, but is now clearly acknowledged as an important limitation and a valuable direction for future research.
Comment 3: System boundary definitions should be more precise and unified, particularly for hydrogen systems with multiple energy conversions and auxiliary energy consumption.
Response:
We thank the reviewer for this important and constructive comment. We fully agree that boundary harmonization is essential for a fair comparison among fundamentally different storage technologies.
In the revised manuscript, we have strengthened the methodological description to ensure a fully harmonized cradle-to-grave system boundary across all assessed technologies. We now state more explicitly that all systems are evaluated under the same functional unit, namely 1 MWh of electricity delivered to the load, and that all energy and material flows are normalized accordingly.
For the hydrogen pathway in particular, we have expanded the description of the full conversion chain to explicitly include:
(i) electricity-to-hydrogen conversion via electrolysis,
(ii) hydrogen storage,
(iii) hydrogen-to-electricity conversion via fuel cell, and
(iv) auxiliary energy demands associated with compression, balance-of-plant, cooling, and standby loads.
These additions improve comparability and remove ambiguity regarding functional equivalence across the three storage pathways.
Comment 4: Cycle life predictions for reusable lithium-ion batteries lack a systematic discussion of historical usage, chemistry differences, and screening/regrouping processes. Please quantify the effect of these uncertainties on LCOS and LCA.
Response:
We thank the reviewer for this insightful and highly relevant comment. We fully agree that the performance and remaining lifetime of second-life Li-ion batteries depend not only on residual capacity, but also on prior first-life usage history, chemistry type, and the quality of screening and regrouping prior to stationary deployment.
In the revised manuscript, we have therefore strengthened both the qualitative discussion and the quantitative uncertainty treatment of second-life batteries. More specifically, we now discuss the role of historical degradation, chemistry-related variation, and regrouping quality in the literature and methodology sections, and we substantially extend the uncertainty analysis to capture their system-level effects through scenario-based parameter variation.
The revised analysis now varies residual state-of-health at second-life entry, second-life duration, round-trip efficiency, cycle life, and utilization factor. These uncertainties are propagated through both the techno-economic and environmental assessment. The results indicate that second-life uncertainty can produce variations of approximately 10–25% in LCOS and 5–15% in GWP, while the overall comparative ranking remains stable in the majority of tested scenarios. We also explicitly clarify that the current study uses aggregated system-level parameters rather than chemistry-specific electrochemical degradation sub-models, and we identify detailed chemistry-resolved aging analysis as an important direction for future work.
Comment 5: The discussion of HES challenges is too brief. Durability of electrolyzers and fuel cells should be addressed more explicitly.
Response:
We thank the reviewer for this valuable comment. We agree that the original manuscript did not sufficiently elaborate the technical and durability challenges associated with hydrogen energy storage subsystems.
In the revised manuscript, we have expanded the discussion to explicitly address electrolyzer and fuel cell degradation, including catalyst loss, membrane aging, electrode deterioration, performance decline under variable loading, and the resulting consequences for efficiency, replacement frequency, and system cost. We now clarify that these degradation mechanisms directly affect both LCOS and environmental impacts, since they may increase energy demand and material replacement over the system lifetime.
We also note more clearly that the present framework represents these effects through aggregated lifetime and efficiency parameters rather than detailed degradation-aware dynamic sub-models. This limitation is now explicitly acknowledged in the manuscript.
Comment 6: The study focuses too much on GWP and should better reflect broader environmental trade-offs and burden shifting.
Response:
We thank the reviewer for this important comment. We fully agree that GWP alone is not sufficient to capture the full environmental complexity of energy storage technologies.
In the revised manuscript, we clarify more explicitly that the environmental analysis includes not only GWP, but also CED, ARD, AP, and EP, thereby broadening the environmental perspective. We have further strengthened the discussion of resource criticality, toxicity-related concerns, and burden-shifting effects. In particular, we now explain that Li-ion systems may appear favorable in GWP terms while relying on critical materials such as lithium, cobalt, and nickel; hydrogen systems may shift burdens toward energy demand and infrastructure requirements; and Pb-acid systems benefit from mature recycling chains but remain associated with lead-related toxicity concerns.
We also explicitly acknowledge that a full multi-impact LCA with richer toxicity and ecosystem indicators would be a valuable future extension.
Comment 7: The economic model neglects additional market value streams such as ancillary services, capacity markets, and congestion management.
Response:
We thank the reviewer for this important observation. We agree that stationary storage systems can derive value from multiple market services beyond pure energy arbitrage.
In the revised manuscript, we clarify that the current model intentionally focuses on energy-based market participation using time-varying electricity prices and export remuneration as a transparent and comparable baseline across technologies. At the same time, we now explicitly discuss the relevance of ancillary services, capacity remuneration, congestion management, and other stacked revenue streams. We note that battery systems, especially Li-ion technologies, are often well suited for fast-response ancillary services, whereas hydrogen systems may be more aligned with long-duration and capacity-oriented value propositions.
We also state more clearly that incorporating stacked revenues and multi-market participation would require an expanded optimization framework and is therefore left for future work.
Comment 8: Fuel cell parameter estimation uncertainty should be discussed more explicitly in Section 3.3.2.
Response:
We thank the reviewer for this valuable suggestion. We agree that fuel cell parameter uncertainty affects hydrogen-system efficiency estimates and, consequently, optimization outcomes.
In the revised manuscript, we have extended the hydrogen model discussion to explicitly acknowledge that fuel cell performance parameterization is an important source of uncertainty. We further note that improved parameter estimation methods can enhance the fidelity of hydrogen system simulations and reduce the gap between idealized model behavior and actual operation. We have also included the reviewer-suggested reference in the revised discussion.
At the same time, we clarify that, for tractability at the system level, the present framework represents fuel cell behavior using aggregated efficiency assumptions rather than detailed electrochemical parameter-estimation sub-models.
Comment 9: Safety and reliability issues of reusable battery systems should be addressed more explicitly, ideally with a quantitative or semi-quantitative treatment.
Response:
We thank the reviewer for this important comment. We fully agree that safety and reliability are especially relevant for second-life battery systems, where prior degradation history, cell mismatch, and regrouping quality may introduce additional risk compared with new batteries.
In the revised manuscript, we have expanded the discussion to address safety-related issues such as heterogeneity, imbalance, thermal stress, and increased uncertainty in second-life operation. To provide a semi-quantitative treatment consistent with the scope of the study, we now represent safety- and reliability-related uncertainty through scenario-based variation in lifetime, efficiency, and replacement frequency, which act as system-level proxies for derating, mismatch, and reliability loss. We further clarify that these effects can influence both economic and environmental outcomes through increased OPEX, more frequent replacement, and additional embodied burdens.
We also explicitly state that detailed probabilistic safety modeling lies beyond the scope of the present paper, but constitutes an important avenue for future work.
Comment 10: The conclusions lack specific policy recommendations.
Response:
We thank the reviewer for this important suggestion. We agree that the practical contribution of the manuscript is strengthened by more explicit policy implications.
In the revised manuscript, we have expanded the concluding section to include specific policy recommendations, including:
(i) the need for certification and qualification frameworks for second-life batteries,
(ii) clearer responsibility structures for reuse, recycling, and end-of-life handling, and
(iii) targeted policy incentives that support circular-economy pathways and technology-appropriate deployment.
We also emphasize that policy support should be technology-differentiated rather than technology-neutral, since second-life Li-ion, Pb-acid, and hydrogen systems serve distinct operational niches and generate different combinations of economic, environmental, and circularity benefits.
Once again, we sincerely thank Reviewer 1 for the detailed, technically rich, and highly constructive comments. The reviewer’s observations helped us strengthen the manuscript substantially in terms of methodological consistency, treatment of uncertainty, environmental interpretation, and policy relevance. We believe that the revised version addresses all concerns raised and is significantly clearer, more rigorous, and more impactful as a result of these improvements.
Reviewer 2 Report
Comments and Suggestions for AuthorsThis paper proposes a comprehensive techno-economic and environmental assessment framework to compare the performance of second-life Li-ion batteries, Pb-acid batteries, and hydrogen systems in PV-assisted applications. The structure is clear and the modeling is rigorous, particularly with the inclusion of explicit recycling and second-life credits in the LCA. However, there is room for improvement in the description of certain modeling details and the standardization of figures and tables.
1：The manuscript mentions in Section 5.6 that simplified conversion factors were used to calculate CED, AP, and EP, and notes that full integration of ecoinvent or ELCD databases is pending. It is recommended to add a table in the "Materials and Methods" section explicitly listing the specific impact factor values and their original references used in the current calculations to enhance reproducibility.
2：The paper uses a discount factor to reflect the reduced CAPEX of repurposed batteries. However, the specific value or range for this coefficient is not explicitly stated. Given that is a key variable determining the economic advantage of second-life batteries, the author should provide the rationale for this parameter and its specific value in the baseline scenario.
3： The results indicate that the GWP of the hydrogen system is significantly affected by auxiliary power consumption, such as compression and thermal management. Please clarify whether these auxiliary loads are calculated as a constant fraction or if non-linear characteristics relative to the electrolyzer's load rate are considered. Further elaboration of the mathematical expression in Section 3.3.2 is suggested.
4：Several figures (e.g., Figure 4) show negative LCOS values, explained as export revenues exceeding total costs. While explained in the discussion, LCOS is conventionally a cost metric. It is suggested to explicitly state that "negative LCOS" in this context is equivalent to "net revenue per delivered MWh," or use more precise terminology like "Levelized Profit of Storage" to avoid reader misunderstanding.
5：Section 3.5 utilizes a system-expansion approach for second-life credits, reflecting the displacement of new stationary battery production. However, the determination of the substitution factor (s) and utilization factor (u) in Equation (6) appears somewhat subjective. It is recommended that the authors provide the specific value ranges for these key parameters in Section 3.5 or an appendix.
Author Response
We sincerely thank Reviewer 2 for the careful and methodologically focused evaluation of our manuscript. The reviewer’s comments were especially helpful in improving the transparency, reproducibility, and technical precision of the study. In response, we have clarified the simplified environmental impact factors, explicitly stated key economic and LCA parameters related to second-life batteries, improved the mathematical description of hydrogen auxiliary loads, and strengthened the interpretation of negative LCOS results. These revisions have significantly improved the methodological clarity of the manuscript.
Comment 1: Please add a table listing the specific impact factor values and references used for CED, AP, and EP.
Response:
We thank the reviewer for this constructive suggestion. We agree that explicitly reporting the simplified impact factors improves transparency and reproducibility.
In the revised manuscript, we have added a dedicated table in the Materials and Methods section listing the conversion factors used for CED, AP, and EP, together with the corresponding references. We also clarify that these factors are simplified proxies intended to support comparative system-level analysis, while full integration with detailed inventory databases remains a possible future enhancement.
Comment 2: The discount factor for reduced second-life battery CAPEX is not explicitly stated.
Response:
We thank the reviewer for this important comment. We agree that the second-life CAPEX reduction factor is a key economic assumption and should be stated explicitly.
In the revised manuscript, we now define the second-life CAPEX reduction factor αSL, provide its baseline value, and report the sensitivity range considered in the robustness analysis. We also clarify that this factor reflects the reduced energy-related CAPEX of repurposed batteries relative to new stationary Li-ion systems, while still accounting for screening, testing, repackaging, and integration requirements. This addition improves the transparency of the economic assumptions and allows the reader to better interpret the cost sensitivity of second-life deployment.
Comment 3: Please clarify whether hydrogen auxiliary loads are constant or load-dependent, and elaborate the mathematical expression in Section 3.3.2.
Response:
We thank the reviewer for this important technical comment. We agree that the treatment of hydrogen auxiliary loads must be made fully explicit.
In the revised manuscript, we now clarify that auxiliary loads are modeled using a combination of load-dependent and fixed components. Specifically, compression electricity is represented as proportional to hydrogen production, while auxiliary electrical consumption is modeled as fractions of electrolyzer and fuel-cell power, together with a constant standby component. We have also expanded the mathematical description in Section 3.3.2 to make this formulation explicit and easier to follow.
In addition, we acknowledge that real hydrogen systems may exhibit more complex non-linear part-load and thermal-management behavior, but note that the adopted formulation provides a tractable approximation suitable for system-level optimization.
Comment 4: Negative LCOS values may be misleading and should be clarified more explicitly.
Response:
We thank the reviewer for this important clarification. We agree that negative LCOS values may be misinterpreted if they are presented without sufficient explanation.
In the revised manuscript, we now explicitly state that negative LCOS values correspond to net economic benefit per MWh delivered, i.e., situations in which discounted export revenues exceed discounted system costs under the assumed tariff structure. We retain the LCOS formulation for comparability with the literature, but clarify in the text and figure discussion that, in such cases, the metric effectively behaves as a levelized net benefit / net revenue indicator rather than a pure cost measure. This clarification improves interpretability and prevents confusion.
Comment 5: The substitution factor (s) and utilization factor (u) in the second-life credit formulation appear somewhat subjective. Please provide the value ranges explicitly.
Response:
We thank the reviewer for this important and highly relevant comment. We fully agree that the second-life credit formulation must be parameterized transparently in order to avoid the impression of arbitrary allocation.
In the revised manuscript, we have therefore clarified the parameterization of the second-life credit in the LCA section and in the associated sensitivity analysis. In particular, the utilization factor u is now explicitly varied within the range 0.70–0.95, while the second-life duration is varied between 5 and 10 years. In addition, we now state explicitly the baseline assumption and tested range for the substitution factor s, together with alternative methodological treatments including cut-off and split-allocation formulations. This makes clear that the second-life credit is not treated as a single fixed assumption, but as a transparent modeling choice whose effect is tested under multiple plausible parameterizations.
These additions substantially improve the transparency, robustness, and reproducibility of the LCA treatment for second-life batteries.
We sincerely thank Reviewer 2 for the precise and highly constructive methodological comments. These suggestions led to important improvements in the clarity of assumptions, parameter reporting, and model interpretation. We believe that the revised manuscript is now substantially more transparent and reproducible, and that it addresses all of the reviewer’s concerns in a clear and rigorous manner.
Reviewer 3 Report
Comments and Suggestions for AuthorsThis paper evaluates the role of second-life lithium-ion batteries repurposed from electric vehicles for stationary applications, compared to lead-acid batteries and power-to-hydrogen-to-power systems. Please further revise the paper based on the following comments:
1. The research problem remains insufficiently focused. The current manuscript attempts to address several objectives simultaneously, which weakens the depth of the analysis. A clearer definition of a single core research question would significantly improve the overall coherence of the study.
2. The logical structure of the theoretical framework appears underdeveloped. Although multiple relationships between variables are mentioned, the underlying mechanisms and causal pathways are not clearly articulated. Providing a conceptual model or structural illustration would strengthen the theoretical foundation.
3. Several key variables lack precise conceptual clarification. In particular, the distinction between independent and dependent variables is not always clearly presented, and the operationalization of certain constructs requires further specification.
4. The justification for the selection of control variables is not sufficiently convincing. It remains unclear whether the current model adequately accounts for potential confounding factors, and additional explanation regarding the rationale for these variables would be beneficial.
5. The manuscript does not adequately address issues related to sample representativeness. A more detailed description of the sampling procedure, together with a discussion of potential selection bias or non-response bias, would help readers better evaluate the reliability of the empirical results.
6. The authors correctly identify that the performance of hydrogen storage systems is highly dependent on component sizing, utilization rates, and operational strategies. However, the current literature review in this section seems to focus primarily on stationary applications. I suggest broadening the scope to include insights from the automotive sector, specifically Fuel Cell Electric Vehicles (FCEVs). The challenges mentioned here—optimizing component size and managing efficiency under variable loads—have been extensively studied in the context of FCEV Energy Management Strategies and powertrain sizing (e.g.,doi.org/10.1016/j.etran.2025.100537). Citing these works would strengthen the argument by demonstrating that these optimization challenges are universal across hydrogen applications, rather than unique to stationary storage.
7. The robustness of the empirical findings remains somewhat limited due to the reliance on a single model specification. Additional analyses based on alternative variables, subsample tests, or different model settings would considerably strengthen the credibility of the results.
8. The interpretation of the results occasionally suggests causal implications that are not fully supported by the analytical design. Greater caution is required when distinguishing between statistical association and causal inference.
9. The connection between empirical findings and the proposed theoretical framework is not sufficiently elaborated. A more thorough discussion linking the results back to the theoretical assumptions would enhance the academic contribution of the study.
10. The conclusion section currently focuses mainly on summarizing the findings and lacks deeper reflection. Expanding this section to discuss broader implications, limitations of the study, and potential directions for future research would improve the overall impact of the manuscript.
Author Response
We sincerely thank Reviewer 3 for the thoughtful, conceptually oriented, and academically valuable comments. The reviewer’s observations helped us sharpen the focus of the study, improve the articulation of the theoretical and analytical framework, clarify the role of variables and controls, strengthen the discussion of representativeness and robustness, and moderate the interpretation of model-based findings. As a result, the revised manuscript is now more focused, conceptually coherent, and theoretically grounded.
Comment 1: The research problem remains insufficiently focused.
Response:
We thank the reviewer for this important observation. We agree that the original version addressed multiple objectives too broadly.
In the revised manuscript, we now formulate a single primary research question and position all methodological elements—optimization-based dispatch, techno-economic analysis, and life-cycle assessment—as integrated components of one unified evaluation framework. The comparative assessment of Li-ion, Pb-acid, and hydrogen storage systems is now presented more clearly as an application of this central analytical framework rather than as a set of loosely connected objectives. This revision improves the coherence and focus of the study.
Comment 2: The theoretical framework is underdeveloped and would benefit from a conceptual model or structural illustration.
Response:
We thank the reviewer for this valuable comment. We agree that the analytical logic of the study needed to be stated more explicitly.
In the revised manuscript, we have strengthened the conceptual framing by clarifying the structural link between exogenous inputs, optimization-driven operational decisions, technology-specific conversion pathways, and the resulting techno-economic and environmental outcomes. This revised framing makes the analytical structure of the manuscript more explicit and better aligned with the comparative objective of the study. Where relevant, this conceptual logic is also described in a more structured and transparent manner in the revised text.
Comment 3: Several key variables lack conceptual clarification; the distinction between independent and dependent variables and the operationalization of constructs should be improved.
Response:
We thank the reviewer for this very helpful observation. We agree that the manuscript benefits from a clearer distinction between the different classes of variables used in the framework.
In the revised manuscript, we now explicitly distinguish between:
(i) exogenous input variables, such as PV generation, load demand, electricity prices, and grid carbon intensity;
(ii) decision variables, determined by the optimization model, such as storage sizing, charge/discharge schedules, hydrogen production/use, and import/export flows;
(iii) control variables, used in scenario analysis, such as discount rate, feed-in factor, efficiency assumptions, and energy-to-power ratio; and
(iv) dependent outcome variables, including LCOS, GWP, autonomy, self-consumption, and related indicators.
We also clarify how these constructs are operationalized across the optimization, economic, and environmental modules. This revision improves both conceptual precision and methodological transparency.
Comment 4: The justification for the selection of control variables is not sufficiently convincing.
Response:
We thank the reviewer for this important methodological comment. We agree that the rationale for the selected control variables should be made more explicit.
In the revised manuscript, we now explain more clearly that the selected controls—such as electricity price profile, feed-in remuneration, discount rate, efficiency assumptions, and energy-to-power ratio—were chosen because they are among the dominant drivers of economic and environmental performance in stationary storage systems. We further clarify that additional sources of uncertainty, such as degradation variation, second-life heterogeneity, and hydrogen auxiliary demand, are not ignored but are instead addressed through dedicated sensitivity and robustness analysis.
This revision strengthens the logic of the analytical design and makes the selection of control variables more convincing.
Comment 5: The manuscript does not adequately address issues related to sample representativeness.
Response:
We thank the reviewer for this important comment. We would like to clarify that the present study is not based on a statistical sample or survey, but rather on a simulation-based case study using real measured time-series data from a single site.
However, we fully agree that the limits of generalizability should be stated more clearly. In the revised manuscript, we now explicitly clarify that the study is representative of a specific class of applications, namely SME-type PV-coupled stationary systems, rather than of a statistically representative population. We also acknowledge the potential influence of site-specific bias and explain that scenario and sensitivity analysis were used to partly mitigate this limitation. This clarification improves the transparency of the study’s scope and interpretive boundaries.
Comment 6: Please broaden the hydrogen literature review by including insights from FCEV applications.
Response:
We thank the reviewer for this valuable suggestion. We agree that optimization and degradation issues relevant to hydrogen systems are not unique to stationary applications.
In the revised manuscript, we have broadened the hydrogen literature discussion by incorporating insights from fuel cell electric vehicle (FCEV) studies, particularly regarding component sizing, utilization patterns, energy management, efficiency under variable load, and degradation-aware system operation. This broader perspective supports the view that many of the identified hydrogen-system challenges reflect cross-domain characteristics rather than purely stationary-system-specific issues.
Comment 7: The robustness of the findings is limited by reliance on a single model specification.
Response:
We thank the reviewer for this important comment. We agree that robustness must be demonstrated more explicitly.
In the revised manuscript, we have strengthened the robustness analysis by incorporating multiple forms of sensitivity testing, including parameter variation, alternative second-life credit formulations, adverse-end scenario testing, and expanded uncertainty analysis for second-life batteries. We now state more clearly that the analysis examines result stability across multiple plausible parameter configurations rather than relying on a single deterministic model specification.
This revision strengthens confidence in the comparative conclusions of the study.
Comment 8: The interpretation of the results occasionally suggests causal implications that are not fully supported.
Response:
We thank the reviewer for this important observation. We fully agree that causal language must be used carefully in a model-based comparative study.
In the revised manuscript, we have revised the discussion to avoid overstatement and to distinguish more clearly between scenario-dependent model-based relationships and universally generalizable empirical causal claims. We now clarify that the framework captures structured causal pathways within the analytical model itself, linking exogenous conditions, operational decisions, conversion efficiency, and resulting performance indicators. However, these relationships are now explicitly presented as model-based and scenario-dependent, rather than as universally validated empirical causal laws.
This adjustment improves interpretive precision and academic rigor.
Comment 9: The connection between empirical findings and the proposed theoretical framework is not sufficiently elaborated.
Response:
We thank the reviewer for this important comment. We agree that the contribution of the study is stronger when the reported findings are interpreted explicitly through the lens of the proposed analytical framework.
In the revised manuscript, we therefore expand the discussion to link the main results back to the conceptual structure of the study. We clarify that the observed differences in LCOS and environmental performance do not arise as isolated descriptive findings, but rather from the interaction between exogenous inputs, optimization-based dispatch decisions, technology-specific conversion characteristics, and circularity-related allocation assumptions. In this way, the empirical findings are now more clearly positioned as evidence illustrating the usefulness of the integrated framework.
Comment 10: The conclusion should discuss broader implications, limitations, and future research directions in greater depth.
Response:
We thank the reviewer for this helpful suggestion. We agree that the conclusion should go beyond summary and articulate the broader significance of the study.
In the revised manuscript, we have expanded the concluding discussion to include broader implications for technology selection, circular-economy strategy, environmental trade-offs, and policy design. We also strengthen the discussion of study limitations, including case-specific generalizability, simplified degradation representation, and the absence of richer multi-impact toxicity/ecosystem indicators. Finally, we outline several future research directions, including chemistry-specific degradation modeling, broader multi-site validation, richer market participation models, and more detailed second-life allocation and safety treatments.
These additions improve both the academic depth and the practical relevance of the manuscript.
We sincerely thank Reviewer 3 for the thoughtful and conceptually rich comments. The reviewer’s suggestions helped us sharpen the focus of the paper, improve the theoretical and analytical framing, clarify the interpretation of variables and findings, and strengthen the discussion of robustness and generalizability. We believe that the revised manuscript is now considerably more coherent, theoretically grounded, and persuasive as a result of these revisions.
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThe authors have fully responded to the initial review comments, showing significant improvements in methodological transparency (e.g., optimising framework details), lifecycle boundary definition (recycling and secondary lifespan allocation mechanisms), and the depth of sensitivity analysis (electricity price fluctuation scenarios). This review identified several remaining issues requiring refinement; minor revisions are recommended before acceptance.
1. The empirical basis for cycle life modelling is insufficient. The manuscript emphasises that the performance of retired batteries depends on "historical degradation paths and recombination heterogeneity" (p.4), but the optimised model does not reflect this dynamic degradation mechanism. Please supplement: Is an electrochemical-mechanical coupled degradation model used? How to quantify the impact of the capacity/internal resistance dispersion between modules after recombination on the system lifetime?
2. Pages 3 and 5 of the manuscript emphasise that PEMFC degradation (such as catalyst degradation, membrane thinning, and thermal sensitivity) reduces system efficiency, shortens lifetime, and increases LCOS. Song et al. (2025) [Journal of Power Sources 648 (2025): 237227] systematically reviewed the inhibitory effects of liquid-cooled thermal management techniques on PEMFC degradation, demonstrating that optimised thermal management can significantly improve efficiency and extend lifetime (especially under dynamic loads). Adding relevant content to your degradation discussion would provide empirical support, strengthen the feasibility analysis of the PtHâ‚‚P system, and demonstrate practical mitigation strategies.
It is suggested that the following be added to Section 2.4 when discussing degradation mechanisms: "For example, Song et al. (2025) demonstrated that liquid-cooled thermal management can effectively mitigate the degradation of PEMFCs, improve system efficiency and lifetime, which is of significant reference value for optimizing the LCOS of PtHâ‚‚P [Journal of Power Sources 648 (2025): 237227]."
3. The source of the recovery rate parameter is missing. The environmental assessment relies on the assumption of "high recovery rate" (lead acid >95%, Li-ion unclear), but: the source of the lead acid recovery rate data is not explained; the differences in recovery efficiency between Li-ion cathode materials (LFP/NMC) are not distinguished.
4. Sensitivity analysis did not cover key policy variables. The manuscript analysis of the sensitivity of "electricity price and discount rate" (p. 1) neglected: the impact of carbon price fluctuations on the competitiveness of the PtHâ‚‚P system; and the impact of different regional "mandatory blending ratios of recycled materials" on LCOS.
5. Page 5 of the manuscript points out that degradation modelling is crucial for LCOS and LCA assessments, but the current framework uses "aggregated efficiency and lifetime parameters," and suggests that future work should "incorporate detailed degradation models" (e.g., "incorporating detailed degradation models remains an important direction for future work"). Meng et al. (10.1016/j.jpowsour.2024.235634) proposed an innovative lifetime prediction model that considers the recovery of reversible voltage loss, which can more accurately predict fuel cell lifetime (especially under cyclic load conditions). Related content can fill the modelling gaps in the manuscript, improve the accuracy of techno-economic assessments, and provide specific directions for the "future work" recommendations.
It is suggested that the following be added to the "Conclusion" or "Future Work" section when summarizing degradation challenges or proposing improvements: "Future research can integrate more advanced degradation models, such as the lifetime prediction framework proposed by Meng et al., which considers the recovery of voltage loss and can improve the robustness of the assessment" (10.1016/j.jpowsour.2024.235634).
6. Safety costs are not included in the economic model. The manuscript points out the "thermal runaway risk" of retired batteries (p.4) and that hydrogen systems need to address "proton exchange membrane degradation" (p.5), but the LCOS calculation does not include: the additional costs of battery safety monitoring; and the regular maintenance costs of fuel cells. A complete demonstration of the operation and maintenance cost model is needed.
7. The dynamic grid carbon factor is missing. The environmental assessment assumes a constant grid emission factor, but actual photovoltaic output is negatively correlated with grid carbon intensity in time and space, and the impact of dispatch strategies on implicit carbon emissions is ignored. It is recommended to couple real-time carbon intensity data to optimise charge and discharge strategies.
8. The technical comparison benchmarks are inconsistent. Hydrogen energy storage is benchmarked against "long-cycle applications" (p.5), but lead-acid/Li-ion is not designed for equivalent energy storage duration: should the same energy capacity (kWh) be compared instead of the same delivered power (MWh)? In addition, it is not stated whether the PtHâ‚‚P system includes compression/liquefaction energy consumption, and the comparison premise needs to be unified, and the details of the hydrogen chain need to be supplemented.
Author Response
We would like to sincerely thank the Reviewer for the careful reading of our manuscript and for the constructive and technically insightful comments. We greatly appreciate the reviewer’s recognition of the manuscript’s potential and the specific recommendations provided to improve its methodological transparency, consistency of comparison, and practical relevance. We have carefully revised the manuscript in response to all comments. The revised version includes expanded methodological explanations, additional sensitivity analyses, improved clarification of second-life battery assumptions, more explicit treatment of hydrogen-system auxiliaries and benchmarks, and an enriched future-work discussion. We believe these changes have substantially strengthened the manuscript, and we are grateful for the opportunity to improve it further.
Comment 1
The empirical basis for cycle life modeling remains insufficiently transparent. The paper mentions using a “cycle life correction factor based on state-of-health diagnostics”, but it is unclear how degradation pathways after module regrouping were modeled. Is an electrochemical-mechanical coupled degradation model used? How is the impact of capacity/internal resistance dispersion between recombined modules quantified?
Response 1
We thank the Reviewer for this important comment. We agree that the treatment of second-life battery degradation and heterogeneity must be clarified explicitly. In the revised manuscript, we now state clearly that the present framework does not employ an explicit electrochemical-mechanical coupled degradation model at the cell or module level. Instead, second-life battery degradation is represented through aggregate system-level parameters, including round-trip efficiency, effective lifetime, replacement frequency, and scenario-based uncertainty ranges.
To address the issue of post-regrouping heterogeneity, we expanded both the methodological explanation and the sensitivity analysis. The revised text now explains that dispersion in residual capacity and internal resistance after recombination is not modeled directly at cell level, but is represented through conservative derating assumptions. In particular, Section 3.7.7 now introduces an additional robustness treatment in which effective lifetime is reduced by up to 30%, round-trip efficiency is perturbed by ±5%, and replacement frequency is increased relative to the baseline. This provides a tractable system-level proxy for mismatch-driven aging, imbalance effects, and reliability-related derating in repurposed battery systems. These clarifications have been added in the second-life modeling discussion and in the dedicated uncertainty subsection.
Comment 2
The author is encouraged to cite Song et al. (2025), which shows that liquid-cooled thermal management can effectively mitigate PEMFC degradation, improve system efficiency, and extend lifetime. This is directly relevant to the LCOS discussion.
Response 2
We thank the Reviewer for this valuable recommendation. In the revised manuscript, we expanded the PEMFC degradation discussion to explicitly include the role of advanced thermal management, particularly liquid-cooled configurations, in mitigating catalyst degradation, membrane thinning, and thermal stress. We have incorporated the suggested reference and linked it directly to the techno-economic discussion by noting that improved thermal management can enhance PEMFC efficiency and extend lifetime, which is directly relevant for reducing replacement needs and improving LCOS in PtHâ‚‚P systems under dynamic operating conditions.
Comment 3
The source of the recovery rate parameter in Tables 3 and 4 is still not specified. In particular, the basis for assuming a >95% recovery rate for lead-acid batteries remains unclear. In addition, the paper does not distinguish the differences in recovery rates for lithium-ion chemistries (e.g., LFP vs. NMC).
Response 3
We thank the Reviewer for highlighting this point. In the revised manuscript, we have clarified that the recovery-rate assumptions are based on published literature and current industrial recycling practice. We now explicitly state that lead-acid batteries are modeled using a high-recovery baseline above 95%, reflecting their mature closed-loop recycling infrastructure.
We also revised the discussion of lithium-ion recycling to make the chemistry dependence explicit. The manuscript now distinguishes between NMC- and LFP-based systems, noting that NMC batteries enable higher recovery of valuable metals such as nickel and cobalt, whereas LFP systems generally have lower economic recovery potential because they do not contain these high-value metals. In addition, we clarify that lithium-ion recovery is treated as chemistry-dependent and is linked to the broader sensitivity and policy analysis rather than being represented by a single universal recovery value. We believe these revisions substantially improve the transparency and realism of the recycling assumptions.
Comment 4
Although the study includes local sensitivity analysis for electricity price and discount rate, it does not address policy-related parameters such as carbon price fluctuations and mandatory recycled content ratios, which are increasingly relevant under EU battery regulations.
Response 4
We thank the Reviewer for this important suggestion. We agree that policy-related variables can materially influence both economic and environmental performance and therefore deserve explicit attention. In response, we added a new subsection, Section 3.7.8 (“Policy sensitivity: carbon price and recycled-content constraints”), which introduces an additional sensitivity analysis for two policy variables: carbon price and recycled-content requirements.
In this new subsection, carbon price is varied over the range 0–150 EUR/tCOâ‚‚e and represented as an additional cost associated with grid electricity consumption, which particularly affects electricity-intensive pathways such as PtHâ‚‚P. In parallel, recycled-content requirements are represented through a technology-specific compliance factor affecting material-related cost and recycling-credit assumptions. The revised discussion explains how these policy scenarios modify absolute LCOS and GWP values while leaving the main comparative conclusions unchanged in the majority of tested cases. These additions were made to improve the practical and regulatory relevance of the manuscript under emerging circular-economy and decarbonization requirements.
Comment 5
The future work section could benefit from citing Meng et al. (2024), which proposes a fuel-cell lifetime prediction model under cyclic conditions accounting for the recovery of reversible voltage losses.
Response 5
We thank the Reviewer for this excellent recommendation. In the revised manuscript, we expanded the future-work discussion to explicitly acknowledge the need for more advanced degradation modeling for hydrogen systems. The new text now states that future research should integrate degradation-aware lifetime prediction models for fuel cells, including approaches that account for reversible voltage-loss recovery under cyclic operation. This addition directly reflects the Reviewer’s suggestion and strengthens the discussion of how future system-level optimization can be improved for PtHâ‚‚P technologies. The relevant reference has been incorporated into the manuscript.
Comment 6
The discussion of safety costs remains mostly qualitative. Since second-life batteries require additional fire suppression, diagnostics, and repackaging control, these safety-related expenditures should be included in the economic model. A complete demonstration of the O&M cost model is also needed.
Response 6
We thank the Reviewer for this important comment. In the revised manuscript, we clarified the treatment of O&M and safety-related costs within the LCOS framework. We now state explicitly that annual O&M costs are represented in aggregated form and include routine servicing, monitoring, safety-related supervision, and technology-specific auxiliary maintenance. For second-life battery systems, this includes battery management, thermal supervision, inspection, and safety-related monitoring. For hydrogen systems, the O&M representation includes scheduled maintenance of fuel-cell stacks, compressors, cooling equipment, and other balance-of-plant components.
To improve transparency further, we added an explicit aggregated representation of annual O&M cost in the economic assessment section. This was done to show more clearly how these recurring expenditures are treated in the model while preserving comparability and tractability across all technologies. We believe this revision addresses the Reviewer’s concern and makes the cost model substantially clearer.
Comment 7
The environmental assessment still assumes a constant grid carbon factor, but in reality, the marginal carbon intensity of charging electricity depends strongly on time of day and season. It is recommended to couple real-time carbon intensity data with the optimization model.
Response 7
We thank the Reviewer for this valuable comment. We agree that coupling storage dispatch with time-resolved grid carbon intensity would improve the realism of the environmental assessment. In the revised manuscript, we clarified this point explicitly. We now state that, although hourly operational dispatch is optimized using time-resolved demand, PV generation, and electricity-price signals, the environmental accounting in the present study is based on an average grid emission factor for tractability and comparability across technologies.
We also make clear that this simplification does not capture the interaction between dispatch timing and time-varying grid carbon intensity, and we identify a full coupling of dispatch optimization with time-resolved carbon-intensity data as an important direction for future work. We believe that this clarification resolves the ambiguity in the original text and transparently communicates the present modeling scope and its limitations.
Comment 8
The technical comparison benchmarks remain inconsistent. Hydrogen storage is compared as a long-cycle application, but lead-acid/lithium-ion batteries are not designed for equivalent storage duration. Should the same energy capacity be compared instead of the same delivered power? In addition, it is not specified whether the PtHâ‚‚P system includes compression/liquefaction energy use. The comparison premise should be unified and the hydrogen-chain details expanded.
Response 8
We thank the Reviewer for this important clarification request. In the revised manuscript, we further unified the comparison basis and expanded the hydrogen-system description. We now state explicitly that the benchmark is based on equivalent delivered service rather than equal nominal storage duration or equal installed energy capacity. This choice was made because the analyzed technologies are inherently suited to different duration regimes: Li-ion and Pb-acid systems are primarily short- to medium-duration storage technologies, whereas PtHâ‚‚P systems are intended for long-duration applications. Using a harmonized delivered-energy basis therefore provides a more methodologically consistent and unbiased comparison.
We also expanded the hydrogen model description to specify that the reference PtHâ‚‚P configuration is based on compressed gaseous hydrogen storage. The revised text clarifies that compression energy and auxiliary loads are included explicitly in the model, whereas liquefaction is not considered because it falls outside the scope of the analyzed stationary configuration. In addition, the manuscript now states more clearly that the PtHâ‚‚P pathway includes electrolysis, hydrogen storage, fuel-cell reconversion, cooling, standby demand, and other balance-of-plant consumption. We believe these revisions directly address the Reviewer’s concern and significantly improve the consistency and transparency of the comparison framework.
Comments on the Quality of English Language
English quality is good throughout. It can still be further improved for clarity and readability.
Response
We thank the Reviewer for this positive assessment. The manuscript has been carefully edited again to improve clarity, consistency of terminology, and overall readability.
Once again, we would like to thank Reviewer 1 for the highly constructive and technically valuable comments. The suggestions helped us improve the manuscript substantially in terms of methodological clarity, transparency of assumptions, treatment of uncertainty, and consistency of comparison across technologies. We hope that the revised manuscript satisfactorily addresses all concerns and that the changes made have strengthened both the scientific quality and the practical relevance of the study.
Reviewer 2 Report
Comments and Suggestions for AuthorsAll questions have been addressed, no other suggestions
Author Response
We would like to express our sincere appreciation to the Reviewer for the careful evaluation of our manuscript and for the positive final assessment. We are grateful for the time and attention devoted to the review process, as well as for the constructive comments provided during revision. We especially appreciate the Reviewer’s confirmation that all previously raised questions have been satisfactorily addressed and that no further suggestions remain. This favorable evaluation is highly valued and confirms that the revisions have improved the clarity, consistency, and overall quality of the manuscript.
Reviewer 3 Report
Comments and Suggestions for AuthorsThe authors have made all the necessary revisions.
Author Response
We would like to express our sincere appreciation to the Reviewer for the careful evaluation of our manuscript and for the positive final assessment. We are grateful for the time and attention devoted to the review process and especially appreciate the Reviewer’s confirmation that all necessary revisions have been made. This encouraging assessment is highly valued and indicates that the revised manuscript has been substantially improved in response to the review comments.
