Battery Passport and Online Diagnostics for Lithium-Ion Batteries: A Technical Review of Materials–Diagnostics Interactions and Online EIS

Abstract

Digital battery passports are being adopted to provide traceable records of lithium-ion batteries across their lifecycle, credible performance, and durability. However, it requires continuous diagnostics rather than lab-based tests and conditions. This review establishes a materials-informed system that links (i) battery-passport frameworks, (ii) cell-level design, and (iii) online electrochemical impedance spectroscopy (EIS) observables. Therefore, a chemistry-aware indicator set is proposed for passport reporting that relies on capacity and impedance indices, each accompanied by explicit tests. A review of the common and commercial LIBs (LCO, NCA, NMC, LMO, LFP) explains differences and characteristics. In addition, online EIS is reviewed, and different techniques for battery online diagnostics and state estimation are described, with details on how this online analysis is incorporated into the battery passport framework. This review covers the battery passport framework, the materials used in commercial batteries that must be documented and traced, and how these materials evolve throughout the degradation process. It concludes with the state of the art in online battery cell inspection, which enables comparable health reporting, conformity assessment, and second-life grading. Finally, it outlines key implementation priorities related to the reliability and accuracy of battery passport deployment and online battery diagnostics.

1. Introduction

Over the past few years, battery adoption has significantly accelerated, proven by the rapid growth in electric vehicles (EVs) relative to gasoline and diesel cars and by increasing deployment of utility-scale battery energy storage systems (BESSs). Previously, BESS lagged the expansion of large renewable plants because of high battery costs; as this bottleneck eases, storage is scaling with the broader energy transition [1], as shown in Figure 1.

The price reduction in the lithium-ion batteries (LIBs) is supporting the mission to achieve electrification across transport and stationary systems [2,3,4,5], yet the scale and diversity of deployments introduce persistent challenges in safety assurance, durability verification, and life-cycle sustainability [6,7]. Meeting these challenges requires decision-quality information that is traceable across the full battery life history, manufacturing data, first use, repurposing, and end-of-life, rather than relying solely on sporadic laboratory characterization or static nameplate data [8].

In addition, the global battery management system (BMS) market is projected to grow from USD 10.2 billion in 2025 to USD 23.3 billion by 2035 with a compound annual growth rate (CAGR) of 8.6%. The BMSs for LIBs represent 44% of market value in 2025, driven by EV, BESS, and consumer electronics adoption, while the automotive segment is expected to expand at a CAGR of 7.4%, as shown in Figure 2 [9].

On the other hand, the battery passport is a requirement for ensuring transparency, traceability, and sustainability across the entire battery value chain from raw material sourcing to end-of-life management. Within the EU battery passport policy context, a battery passport is a digital, machine-readable record that travels with each battery across its life cycle, standardizing identity, provenance, and verified condition to satisfy regulatory, sustainability, and market-transparency objectives [10]. It acts as an information layer that supports informed decisions and links the broader Digital Product Passport (DPP) value ecosystem [11], while translating stakeholder and regulatory requirements into actionable, interoperable data structures and governance choices (e.g., Distributed Ledger Technology (DLT) design) [12]. From a circularity perspective, passport data enables credible second-life allocation and end-of-life recycling by providing reliable health and history at transfer points [13,14].

The passport’s value depends on what it records and how entries are produced. Materials choices at the cathode, anode, electrolyte, and separator determine performance, safety, and degradation pathways as well as environmental burdens [15,16]. Layered oxide cathodes (e.g., Co- and Ni-rich) offer high energy density yet face interfacial reactivity, oxygen loss, and transition-metal dissolution, manifesting as capacity fade and increases in ohmic/charge-transfer impedance components [15,16]. Spinel lithium manganese oxide (LMO) trades energy for power and exhibits thermal-exposure sensitivity [17], whereas olivine lithium iron phosphate (LFP) provides thermal robustness and supply-chain advantages but typically requires transport enhancements via carbon coating, doping, and nano-structuring [18]. On the anode side, solid electrolyte interphase (SEI) formation, transport limitations, and volume-change mechanics in alloying/conversion systems govern efficiency and impedance evolution [19,20]. Electrolyte and separator choices further set transport properties, stability windows, interphase chemistry (SEI/ cathode electrolyte interphase (CEI)), and abuse tolerance [21,22]. Because these materials process degradation associates drive environmental impacts and service decisions, the passport should connect materials tracing (what the cell is and how it is made) with diagnostic observables (how it ages and performs) [23].

Electrochemical impedance spectroscopy (EIS) links the measured impedance to the cell’s material properties and internal processes. High-frequency reflects collector/electrolyte paths; mid-frequency charge-transfer/interphase; low-frequency diffusion and porous-electrode effects. EIS can run online, using small planned excitations or natural current/voltage variations, to produce condition snapshots. These snapshots can be reduced to a small set of chemistry-aware indicators and equivalent-circuit model parameters (e.g., R_Ω, R_ct, and diffusion time constants) with defined test windows and temperature normalization. These indicators complement capacity metrics and yield passport entries comparable across chemistries and operating conditions. Therefore, there is a critical need to evaluate LIB materials and integrate their performance data with the battery passport using online EIS measurements. This paper reviews these methodologies and establishes a baseline framework to guide future research, standardization, and regulatory development.

This review establishes an integrated framework that links the battery passport concept discussed in Section 2, the materials and cell design considerations presented in Section 3, the lithium plating and interfacial layer formation mechanisms detailed in Section 4, and the online diagnostic approaches (with a focus on EIS) outlined in Section 5. Specifically, Section 3 correlates the dominant different commercial lithium-ion chemistries with their associated degradation mechanisms and identifies the capacity and impedance features that most reliably represent these effects. Section 4 explores the critical phenomena of lithium plating and interfacial layer evolution, which strongly influence battery performance, safety, and lifetime. Section 5 reviews developments in online EIS and related in-use diagnostic techniques for real-time battery health assessment. Lastly, Section 6 concludes the review by synthesizing these insights into a materials-informed, diagnostics-driven framework that supports credible health reporting, environmentally responsible decision-making, and scalable circularity for LIBs.

2. Battery Passport

2.1. Battery Passport Purpose and Regulatory Framework

The battery passport is a digital, machine-readable record that accompanies each battery and standardizes identity, provenance, and verified condition information. Under Regulation (EU) 2023/1542, passports become mandatory from 18 February 2027 for EV, industrial, and light means of transport (LMT) batteries (>2 kWh), linking compliance (e.g., carbon footprint, recycled content, model number) with market transparency [24,25]. Figure 3 represents the timeline of the EU plan for the battery passport policy. Beyond compliance, the passport acts as an information provider for professionals and consumers [3] and provides the stakeholder operational data needs within the wider digital-product-passport ecosystem [11,26]. It also enables credible second-life allocation and recycling by making health and history data available at transfer points [13,27].

2.2. Data Model and Materials Traceability

The battery passport follows a data structure that distinguishes between static and dynamic condition information, as shown in Figure 4 [28,29] (identifiers, form factor, composition, and manufacturing data) from dynamic condition records (state of health (SoH) and capacity fade, impedance indices, cycle count, etc.) [30,31,32].

An Asset–Administration–Shell (AAS)-based approach [33], together with standardized templates, keeps the battery cell’s properties traceable and ready for Life Cycle Assessment (LCA) work [34] and the need to keep the data format unified through the different platforms to keep the same meaning [6,35]. By viewing the battery cell’s passport profile, important information can be observed, such as expected lifetime and the CO₂ footprint [3].

Traceability in the battery passport means to implement a distributed shared record replicated across participating organizations and validated by consensus (ledger), where new entries are added rather than overwritten, and any attempt to change past entries is detectable [24]. This ledger captures multi-party provenance events [12], while physical–digital identifiers (e.g., RFID or QR) bind each cell to its ledger entries, maintaining a persistent, unique identifier and end-of-life identification [36,37]. At the system level, passports enable circular business models by providing verifiable, portable data that support responsible sourcing and higher-yield recycling [25,27,38,39].

2.3. Condition Reporting, Chemistry-Aware Indicators, and Lifecycle Updates

Because the condition evolves during the battery lifetime, the battery passport system should be fed by sensors existing in the BMS to feed the battery indicators (e.g., cumulative energy throughput, counters, fault events) [40]. Chemistry-neutral proxies such as round-trip energy efficiency support periodic logging [41], while generalized SoH models provide universal health parameters across chemistries [42]. Method-neutral data definitions allow indicators from laboratory tests or online methods (e.g., EIS during idle windows) to be reported comparably [6]. Lifecycle policy should prefer event-based and periodic updates (service, ownership transfer, repurposing intake) with clear responsibilities and authenticated data paths [2,35]. Collaboration studies show that fair value sharing and higher data standardization accelerate stable multi-actor deployment [43]. For second-life viability, viability is sensitive to new battery price thresholds, operating limits, and process automation. These motivate passport fields such as SoH, lifetime energy throughput (LET), and duty-cycle descriptors for standardized intake and grading [13].

2.4. BMS Interoperability Challenges

The interoperability of BMS from different manufacturers also presents a considerable obstacle in the creation of standardized health reports within a digital battery passport. Variations exist in data acquisition rates, sensor configurations, and state of charge (SoC)/SoH estimation algorithms, including differences between impedance-based, coulomb-counting, and model-based methods. These differences can lead to health report entries that are non-comparable across systems. Therefore, a manufacturer’s agnostic data ingestion layer should be introduced. It should follow the Battery Pass data model (DIN DKE SPEC 99100:2025-02) [44] and contain metadata such as the BMS manufacturer, firmware version, sensor setup, and test conditions. In this way, health reports should be recorded in the passport that remains transparent, auditable, and interoperable across the entire life cycle.

2.5. Economic Feasibility

Transitioning from conventional capacity-based monitoring methods (e.g., coulomb counting, open circuit voltage (OCV) correction, and periodic capacity checks) to online EIS diagnostics offers significant diagnostic improvements but also introduces economic challenges. Online EIS requires additional excitation and sensing circuitry, higher sampling resolution, and increased computational resources within the BMS [45,46,47]. These hardware and integration demands raise implementation costs compared with capacity-based approaches, which rely on existing voltage and current sensors and are well-established for low-cost systems.

While EIS-based diagnostics enable deeper electrochemical insight, early fault detection, and improved SoH accuracy, their financial viability depends on application scale and criticality. For instance, in premium EVs or large stationary storage systems, where downtime, degradation, or safety incidents incur high costs, and the enhanced predictive capability of online EIS can offset the higher initial investment through extended battery lifetime and reduced maintenance frequency. On the other hand, for budget-constrained applications such as two-wheelers, light EVs, or small residential energy storage systems, conventional capacity-based methods remain more economical. Hybrid approaches, combining low-cost health estimation with periodic or offline EIS testing, may therefore offer a balanced compromise between accuracy and cost [48].

The current literature provides a limited quantitative assessment of these tradeoffs. Most studies focus on technical feasibility, accuracy improvement, or algorithmic enhancement rather than comprehensive cost–benefit analyses. Future work should therefore incorporate techno-economic modeling to evaluate the lifecycle cost implications of online diagnostics relative to battery degradation savings and operational risk mitigation.

2.6. Comparison of EU, USA, and Chinese Frameworks for Battery Passports

The EU Battery Regulation (EU 2023/1542) provides the first mandatory and unified battery passport to be implemented from 18 February 2027, defining standardized fields for identity, origin, sustainability, and verified health indicators. In contrast, the United States and China do not yet mandate a formal battery passport, but several national programs influence data reporting and traceability practices.

In the U.S., federal policy is distributed across different programs such as the DOE Battery Recycling R&D Center, the DOE National Blueprint for Lithium Batteries 2021–2030, the Federal Consortium for Advanced Batteries (FCAB), and the Inflation Reduction Act (IRA) reporting schemes for critical minerals and component traceability [49,50]. These frameworks prioritize supply-chain documentation and eligibility verification (e.g., origin of critical minerals) but do not prescribe a mandatory digital battery passport nor standardized online health-indicator terminology. Industrial consortia, including UL, SAE, and NAATBatt, are still developing voluntary guidelines for data schemas and diagnostic indicators.

China, through the Ministry of Industry and Information Technology (MIIT), the China Automotive Technology and Research Center (CATARC), and the Traceability Management Platform for New Energy Vehicle (NEV) Power Batteries, requires detailed tracking of battery manufacturing, safe operation, repurposing, and recycling [51]. In addition, China has deployed a national NEV battery ID and lifecycle-tracking platform [10]. However, the system focuses on traceability and safety reporting rather than harmonized electrochemical health-indicator definitions. According to the author’s knowledge, existing regulations do not specify any standardized impedance-derived or chemistry-aware battery health indicators.

Based on the above-mentioned discussion, the EU battery passport represents the most comprehensive and prescriptive framework currently available. The differing levels of regulatory maturity across the U.S. and China lead to variability in minimum required data fields, online state of X (SoX) indicators, and diagnostic granularity. This lack of global alignment reinforces the need for international standardization of health-indicator nomenclature, data formats, and minimum diagnostic fidelity to ensure interoperability and comparability of battery health reporting across markets.

Following the battery passport framework and regulations, Section 3 maps the dominant commercial lithium-ion chemistries to their corresponding degradation mechanisms and material-level impacts, while Section 4 focuses on lithium plating and interfacial layer formation, elucidating their influence on cell performance, safety, and lifetime. Section 5 then examines online diagnostic techniques for tracking capacity and impedance evolution, providing measurement-level guidance for implementing dynamic and data-driven entries in the battery passport.

3. Fundamentals of Li-Ion Battery and Its Categories

A battery is a complex electrochemical device consisting of multiple intricately coupled parts for energy storage. Its main components are the cathode, anode, electrolyte, separator, current collectors, battery case, terminals, and pressure vent, as shown in Figure 5 for a lithium cobalt oxide (LCO) battery [52]. In this electrochemical system, an electrolyte separates the anode and cathode. These two electrodes store electrical energy and transform it into chemical energy. Electronic and ionic components are involved in the subsequent chemical reaction. As the electronic component traverses an external circuit, the electrolyte facilitates a smoother flow for the ionic component within the cell. Notably, the chemical process is reversible in rechargeable batteries, enabling the periodic transfer of chemical energy into electrical energy.

There are many types of LIBs based on the materials used in the cathode, anode, and electrolyte. However, the cathode and anode are considered crucial elements in LIBs and play a decisive role in their commercialization as they directly influence the electrochemical performance [18]. LIBs are mainly based on different cathode materials, with each combination having distinct advantages and disadvantages in terms of power/energy density, safety, cost, performance, and stability. The characteristics of the cathode materials are illustrated in Figure 6 [53,54]. Lithium cobalt oxide (LCO) is the predominant battery technology for consumer electronics. However, LCO technology is unstable due to the over-lithiation problem; therefore, it is unsuitable for EV applications. Other cathode materials, such as lithium nickel cobalt aluminum oxide (NCA), lithium nickel manganese cobalt oxide (NMC), spinal lithium manganese oxide (LMO), and lithium iron phosphate (LFP), are dominant for EV applications compared to LCO because of their abundant resources, stable technology, and low price [55]. The chemical reaction during charging (cathode) and discharging (anode) of the LIBs is listed in

Table 1. An overview of the cathode and anode electrochemical reactions during charging and discharging for different LIB chemistries.

Symbol	Cathode	Anode	Chemical Reaction at Anode (Discharging)	Chemical Reaction at Cathode (Charging)
LCO	LiCoO2	Graphite	$Li{C}_6\stackrel{Li-ionextraction}{\to }{C}_6+{Li}^{+}+e^{-}$	${LiCoO}_2\stackrel{Li-ioninsertion}{\to }{Li}_{a}Co{O}_2+{aLi}^{+}+{ae}^{-}$
NCA	LiNi1−x−yCoxAlyO2	Graphite	$Li{C}_6\stackrel{Li-ionextraction}{\to }{C}_6+{Li}^{+}+e^{-}$	$Li{Ni}_{1-x-y}{Co}_x{Al}_y{O}_2\stackrel{Li-ioninsertion}{\to }{{Li}_{a}Ni}_{1-x-y}{Co}_x{Al}_y{O}_2+{aLi}^{+}+{ae}^{-}$
NMC	LiNi1−x−yMnxCoyO2	Graphite	$Li{C}_6\stackrel{Li-ionextraction}{\to }{C}_6+{Li}^{+}+e^{-}$	$Li{Ni}_{1-x-y}{Mn}_x{Co}_y{O}_2\stackrel{Li-ioninsertion}{\to }{{Li}_{a}Ni}_{1-x-y}{Mn}_x{Co}_y{O}_2+{aLi}^{+}+{ae}^{-}$
LMO	LiMn2O4	Graphite	$Li{C}_6\stackrel{Li-ionextraction}{\to }{C}_6+{Li}^{+}+e^{-}$	${LiMnO}_2\stackrel{Li-ioninsertion}{\to }{Li}_{a}Mn{O}_2+{aLi}^{+}+{ae}^{-}$
LFP	LiFePO4	Graphite	$Li{C}_6\stackrel{Li-ionextraction}{\to }{C}_6+{Li}^{+}+e^{-}$	${LiFePO}_4\stackrel{Li-ioninsertion}{\to }{Li}_{a}FeP{O}_4+{aLi}^{+}+{ae}^{-}$

[56,57].

These LIB chemistries were compared in terms of five aspects: power, energy, safety, cost, and stability. Nowadays, automakers such as BMW, Nissan, Tesla, Volkswagen, Chevrolet, BYD, and Mercedes-Benz use NMC, NCA, LMO, and LFP technology to reduce CO₂ emissions and oil dependence, as tabulated in

Table 2. LIB chemistries (NMC, NCA, LMO, and LFP) adopted by major automobile manufacturers.

Cell Manufacturer	Cell Type	Cell Chemistry		Cell Voltage (V)	Cell Capacity (Ah)	Energy Density (WhKg−1)	Installed in
		Cathode	Anode				Company	EV Model
A123	Pouch	LFP	C	3.3	20	131	Chevy	Spark
BYD	Prismatic	LFP	C	3.2	216	166–140	BYD	Tang electric
AESC	Pouch	NMC532	C	3.65	56.3	130	Nissan	Leaf
AESC	Pouch	LMO-NCA	C	3.75	33	155	Nissan	Leaf
LG Chem	Pouch	NMC721	C	2.08	130	160	Renault	Zoe
LG Chem	Pouch	NMC721	C	1.85	145	164	Volkswagen	ID.3
LG Chem	Pouch	LMO-NMC	C	3.70	16	-	Ford	Focus
LG Chem	Pouch	NMC721	C	3.65	64.6	263	Audi	e-Tran GT
Samsung SDI	Prismatic	NMC111	C	3.7	37	185	Volkswagen	e-Golf
Samsung SDI	Prismatic	NMC622	C	3.68	94	148	BMW	i3
Panasonic	Cylindrical	NCA	C	3.6	3.4	236	Tesla	Model S
Panasonic	Cylindrical	NCA	C or SiOC	3.6	3.4	236	Tesla	Model X
Panasonic	Cylindrical	NCA	C or SiOC	3.6	4.75	260	Tesla	Model 3
Li-Energy Japan	Prismatic	LMO-NMC	C	3.7	50	109	Mitsubishi	i-MiEV
SK Innovation	Pouch	NMC811	C	3.56	180	250	Kia	Niro
SK Innovation	Pouch	NMC811	C	3.56	180	250	Kia	Soul

. In 2022, NMC remained the dominant battery technology with a market share of 60%, followed by LFP with a share of 30%, and NCA with a share of 8% [58]. The mining of the raw materials of these cathode materials is going to increase from 180 Ktons in 2016 to 400 Ktons in 2025.

3.1. LCO Chemistry

The LCO is one of the oldest cathode materials commercialized by Sony Corporation to power the first portable phone in 1991 [52]. The layered structure LCO is a cathode material with a theoretical capacity of 274 mAhg⁻¹. LCO batteries are primarily used in consumer electronics, specifically smartphones, IT devices, and power tools [15].

A hexagonal symmetry characterizes LCO due to the alternating placement of lithium (Li) and cobalt (Co) within octahedral sites spanning multiple layers. This structural arrangement emphasizes the ordered and hexagonally symmetric aspect of LCO and is consistent with the R-3m space group [55]. Therefore, LCO has relatively high theoretical specific and volumetric capacities, around 274 mAhg⁻¹ and 1300 mAhcm⁻³, respectively [59]. Indeed, LCO batteries undergo a series of phase transitions during the lithium insertion (charging) and desorption (discharging) processes. Initially, an insulator-to-metal phase transition occurs during charging at low voltage. Then, when charged to 4.2 V, a reversible hexagonal to monoclinic to O3 phase change occurs. Beyond 4.5 V, the shift from O3 to H1-3 or O6 occurs, culminating in an O1 phase. While reversible, these changes lower Li⁺ diffusivity and produce mechanical stress, resulting in strains and microcracks [60]. These elements all contribute to rapid capacity loss during Li_xCoO₂ deep charging. To overcome these challenges, early LiCoO₂-based batteries limited charging voltage to 4.2 V. Various strategies, like surface coating [61,62] and element doping [63,64], are proposed to solve these challenges and cater to the long-cycling stability of LCO at high voltages above 4.2 V.

3.2. NCA Chemistry

Lithium nickel cobalt aluminum oxide (LiNi_1−x−yCo_xAl_yO₂, NCA) chemistry has existed since 1999 [57]. Panasonic played a pivotal role in launching the commercial use of NCA by introducing the NCR18650 series of LIBs, specifically for Tesla’s EVs [65,66]. NCA is obtained by doping cobalt (Co) and aluminum (Al) elements, endowing it with exceptional electrochemical properties [67]. This composition contributes to NCA’s ability to provide high energy and power density, making it a valuable choice for EV applications. Moreover, NCA exhibits high stability, which is crucial for the long-life and safety of batteries.

Despite these advantages, NCA chemistry still has room for further development. While it achieves high energy density in cycles, reports suggest significant capacity loss during long-term cycling [68]. The layered structure of NCA can go through phase transformation because of repetitive lithium movements or exposure to high temperatures. This transformation can lead to the crystalline phases inside the material from α-NaFeO₂ layered structure with a space group of R3m to different phases, including spinel with a space group of Fd-3m and rocksalt with a space group of Fm-3m [69]. Phase changes indicate a change in the atoms’ configuration within the crystal lattice, which affects the NCA electrode’s electrochemical efficiency. These structural changes may impact battery parameters like capacity, voltage, life, and general stability during charge and discharge cycles [70]. One of the reasons for this degradation is the release of Ni²⁺ ions from the cathode and Ni-metal formation on the anode surface [71]. Therefore, to develop long-life and high-performance battery systems based on NCA cathodes, coating of the nanoparticles is required to protect from side reactions with the electrolyte, avoid dissolution of transition metal in the electrolyte, prevent the loss of oxygen, and stabilize the layer surface [72,73].

LiNi_0.8Co_0.15Al_0.05O₂ is the most prominent composition of NCA chemistry. The current focus is to increase the ratio of the Ni content in its design to reduce the Co content utilization, which helps improve the energy density and lower the battery cost. However, it can deteriorate the thermal stability and cycling performance. The NCA chemistry is mainly employed by Tesla (Model-S2020, Model3-2017, Model X-2015), Toyota RAV4 hybrid, Mercedes-Benz S400, and VW E-golf [53].

3.3. NMC Chemistry

Commercialization of the lithium nickel manganese cobalt oxide (NMC) cathode began in late 2004 [74]. This chemistry, which combines metals nickel, cobalt, and manganese, shows great potential in terms of better specific energy density, power density, stability, and low internal resistance. The success of LiNi_1−x−yMn_xCo_yO₂ is the strategic combination of nickel, cobalt, and manganese: nickel contributes high specific energy but has poor stability, manganese improves structural and thermal stability while providing lower specific capacity due to electrochemical inactivity, and cobalt improves electric conductivity, allowing for high-rate capability. However, a considerable trade-off occurs, as a significant increase in cobalt concentration dramatically raises material costs. The stoichiometry ratio of metal components allows for the creation of several types of NMC. To address cost concerns, several formulations with low-cobalt NMC cathodes and higher nickel content have been developed; for example, LiNi_1/3Mn_1/3Co_1/3O₂ (NMC-111, 160 mAhg⁻¹), LiNi_0.5Mn_0.3Co_0.2O₂ (NMC-532, 170 mAhg⁻¹), LiNi_0.6Mn_0.2Co_0.2O₂ (NMC-622, 180 mAhg⁻¹), and LiNi_0.8Mn_0.1Co_0.1O₂ (NMC-811, 200 mAhg⁻¹) at 4.3V [53,75,76]. Moreover, NMC-811 has higher ionic diffusivity (~10⁻⁸–10⁻⁹ Scm⁻¹) and electronic conductivity (~10⁻⁵ Scm⁻¹) compared to NMC-111 (ionic diffusivity: (~10⁻¹¹–10⁻¹² Scm⁻¹) and electronic conductivity: (~10⁻⁸ Scm⁻¹)) [77]. Therefore, NMC-811 has enhanced battery performance, including faster charging and discharging rates. However, the commercialization of NMC-811 and NMC-622 will be implemented in the near future due to their low cost and high performance for EVs [78]. A drawback of NMC-811 is its surface reactivity, primarily attributed to the instability of nickel ions in liquid organic electrolytes in a dehydrated state [16]. Despite this, NMC-811 poses thermal safety concerns owing to its exothermic self-heating rate and low onset temperature for exothermic reactions. To mitigate this, supplementary cathode components are required to protect the surface integrity and prevent inadvertent parasitic reactions that may occur between the electrode and the electrolyte. The utilization of electrode coatings is a prevalent approach; nevertheless, these post-treatments frequently give rise to issues concerning coverage and interfaces because of foreign material incompatibility [59]. Consequently, these complications can cause segregation to occur during procedures like chemical synthesis and electrochemical cycling. Core–shell structures have gained recognition as a more efficient approach to uniformly encapsulating Ni-rich NMC materials, in contrast to the challenges posed by conventional surface coatings. These structures are characterized by Ni-rich regions that predominate within the particle interior, guaranteeing favorable electrochemical characteristics, while Mn-rich regions are present on the surface (shell) to regulate stability [79]. Indeed, significant improvements in safety and stability are essential before feasibly employing enhanced Ni-rich cathodes for widespread commercial applications.

NMC-cathodes with 60% Ni content have succeeded in full commercialization. Notable commercial applications include Toshiba’s NMC/LTO prismatic cell for the Honda Fit EV (2013), Panasonic/Sanyo and LG’s NMC/C prismatic cells for the VW e-Golf (2015), and Chevrolet Bolt (2016)/Renault Zoe (2017), respectively [78]. LG Chem’s investment in an NMC cell production facility in Michigan, United States, specifically for the Bolt and Chevy Volt, underscores the industry’s commitment [55]. Various companies, such as General Motors, Gotion, SK Innovation, BASF, REPT, Li Energy Japan, AESC, TODA, CATL, Li-Tec, and SVOLT, actively manufacture Ni-rich NMC and NCA cathode technologies to meet the growing demands of the EV market. SK Innovation and BASF have forged a collaboration agreement for the production of cathode-active materials. This collaborative effort extends beyond materials production to encompass the entire value chain, including recycling, in the North American and Asia-Pacific markets.

3.4. LMO Chemistry

The lithium manganese oxide (LMO) electrode was introduced in the early 1980s and marked a pivotal moment in power battery technology, although its commercial viability took around 15 years to be realized [80]. The spinel structure LMO is a cathode material with a theoretical capacity of 148 mAhg⁻¹, a working potential of around 4.1 V vs. Li/Li⁺, and higher ionic (~10⁻⁶ Scm⁻¹) and electronic conductivity (~10⁻⁴ Scm⁻¹). The unique three-dimensional architecture of LMO significantly improves ion flow across the electrodes, leading to a substantial reduction in internal resistance [81]. This characteristic contributes to enhanced battery performance and efficiency. Research has highlighted the design flexibility of LMO, allowing engineers to develop battery banks with extended service lifespans and high specific power [57]. This adaptability makes LMO versatile for various applications, ranging from portable electronics to EVs. A significant focus in the realm of manganese and lithium oxide electrode research involves the development of composite electrodes. These electrodes incorporate layers of Li₂MnO₃, contributing to improved structural stability during electrochemical cycling [81]. The synergy of these materials allows for greater capacity and higher discharge rates, addressing key challenges in power battery technology [17].

On the other hand, LMO batteries experience significant capacity degradation due to manganese surface dissolution in the electrolyte at elevated temperatures (above 60 °C) [82]. Indeed, LMO exhibits similarities with LFP in terms of being cost-effective compared to other cathode materials and having a lower energy density. However, LMO distinguishes itself in its ability to deliver high power, a critical attribute for applications in EVs. The high-power capability of LMO makes it well-suited for the rapid energy demands typical in EVs and power grid applications. Despite these advantages, LMO does face limitations. Notably, it has a low capacity, with a theoretical capacity of 148 mAh/g. Additionally, its relatively short lifetime is attributed to manganese dissolution, causing structural instability during cycling in the electrolyte. This limitation underscores the challenges in achieving long-term durability and reliability, especially in applications where the battery undergoes frequent charge and discharge cycles.

To address these challenges, a strategic method entails combining 70% NMC with cathodes of the LMO type. This integration capitalizes on the advantages of both systems, prolonging the overall lifespan and improving the NMC component’s capacity while using the cost-effectiveness and high-rate capability of LMO-type cathodes. The LMO/NMC composite has been effectively utilized in multiple EVs, including prominent models like the Nissan Leaf, Chevy Volt, and BMW i3 [23]. The cohesive incorporation of various cathode materials enhances efficiency and fulfills the distinct criteria of EVs, hence promoting the extensive implementation of this composite in the automotive sector.

3.5. LFP Chemistry

Lithium iron phosphate (LFP), a cathode with an olivine structure, exhibits enhanced safety, excellent cycle life, and elevated thermal and electrochemical stability owing to the strong bond energy of its PO₄ tetrahedral units [18,83]. Moreover, it exhibits notable attributes such as moderate operating voltage (3.4–3.5 V vs. Li/Li⁺), a flat voltage plateau, moderate theoretical capacity (170 mAhg⁻¹), low material cost, abundant material supply, reversibility, environmental friendliness, and high thermal stability, making LFP a favorable cathode material for commercial LIBs [84,85,86]. Despite these strengths, LFP does face challenges associated with poor electronic conductivity (~10⁻¹⁰ Scm⁻¹), poor low-temperature performance, relatively slow lithium-ion diffusion (~10⁻¹¹–10⁻¹⁰ Scm⁻¹), and low energy density [87]. Nevertheless, these limitations are mitigated by applying the two strategies that were employed to enhance the conductivity of LFP electrodes. The first involved coating the electrodes with a conducting agent, typically carbon, to improve surface and structural conductivity [88,89]. This coating reduces electrical resistance, facilitating more efficient electron transport and enhancing battery performance. The second strategy included metal doping, introducing metal ions into the LFP electrode. This process created defects in the crystal lattice, promoting electron mobility and improving charge transfer kinetics within the electrode [90]. Moreover, using nanoscale materials further improved electronic conductivity by reducing the lithium-ion diffusion pathway, enabling faster charging and discharging rates [91]. These combined strategies aim to optimize LFP electrode performance in batteries.

In EV applications, safety, performance, and high lifetime are the essential factors [92]. Therefore, the LPF battery reached its highest share in the past decade. This trend is mainly driven by the preferences of Chinese original equipment manufacturers (OEMs). Approximately 95% of the EV production in China has used LFP batteries. Significantly, BYD emerges as a prominent contender, accounting for 50% of the total demand for LFP batteries. Tesla, which holds a 15% market share, has observed a significant surge in the proportion of LFP batteries employed, which escalated from 20% in 2021 to 30% in 2022 [58]. This observation underscores a noticeable tendency towards implementing LFP technology, specifically within the swiftly expanding EV industry.

3.6. Anode Materials for Battery Technology

Initially, lithium metal was utilized as the anode material due to its status as the lightest element in the periodic table. However, serious dendrite formation during charging and discharging cycles raised safety concerns and led to substantial lithium consumption. In the past two decades, graphite has become a widely adopted anode material due to its safety and environmental friendliness, although with a lower specific capacity (372 mAhg⁻¹) that falls without meeting the demand for high-performance LIBs [93]. To address this, significant research efforts have been directed towards finding alternatives that can deliver enhanced performance. Maintaining the thermodynamic stability of LIBs requires ensuring that the lowest unoccupied molecular orbit (LUMO)-highest occupied molecular orbit (HOMO) energy gap of the electrolyte is greater than the potential energy difference between the cathode and the anode [94]. Simply, LIB imposes specific requirements on the potential of their electrode materials. The potential energy of the anode must be lower than the LUMO of the electrolyte, while the potential energy of the cathode must be higher than the HOMO of the electrolyte. These requirements protect the electrolyte, contributing to an extended service life for the LIBs.

Lithium-alloy anode materials have attracted much attention due to their low operating voltage and high theoretical capacity. The common alloying materials are silicon and Germanium. The specific capacity (3580 mAhg⁻¹) of silicon-based materials is more than ten times that of commercial graphite anodes (372 mAhg⁻¹) [95]. However, silicon anodes result in irregular and unstable electrical contact (delamination), particle fractures, dynamic solid electrolyte interphase (SEI) layer formation, and low coulombic efficiency [96]. In practical applications, silicon anodes exhibit significantly lower capacity than their theoretical values, and their cyclability is inadequate for practical use. Addressing the challenges posed by silicon anodes requires the adoption of expensive nanostructures, the creation of silicon-carbon composites, or the use of electrolyte additives to stabilize SEI formation. While these approaches can enhance energy density, the associated procedures are cost-intensive. Consequently, fully replacing graphite with silicon is deemed unrealistic. However, Panasonic emerged as one of the pioneering cell manufacturers to implement silicon-carbon composite anodes in Tesla Models (S and 3). Germanium exhibits superior electronic conductivity (10⁴ times higher) and lithium-ion diffusion rates (400 times faster) compared to silicon [97]. Moreover, the surface oxide layer on germanium is thinner, and the Coulomb efficiency of germanium-based negative electrode materials is often higher than silicon-based anodes [98]. Like other alloying anodes utilized in LIBs, Germanium experiences a substantial volume expansion induced by lithium. When fully lithiated to the Li₂₂Ge₅ state, crystalline Ge expands in volume by 370% [97]. This significant volume expansion causes deep crack formation, resulting in the generation of unstable SEI layers and the pulverization of the electrode. Several strategies are proposed to address these challenges, such as low-dimensional nanostructures [99], introduction to porous structures [100], and generation of a composite [101]. However, there is still potential for further exploration, especially in the realm of germanium materials. Future studies are expected to emphasize the dynamic behavior of composite anode materials, considering both economic and safety considerations.

Conversion-type transition-metal compounds (CTAMs) have gained importance as highly promising anode materials for next-generation LIBs due to their appealing compositions and high theoretical specific capacity [102,103,104,105]. Examples of CTAMs include transition-metal oxides, sulfides, phosphides, nitrides, fluorides, and selenides. The natural presence of CTAMs provides advantages over alloy anode materials, potentially leading to lower production costs. Pyrite (FeS₂), magnetite (Fe₃O₄), and pyrolusite (MnO₂) are noteworthy examples. Moreover, compared to graphite anodes, CTAMs exhibit a reduced potential for lithium intercalation, mitigating the formation of lithium dendrites and enhancing the safety of LIB operation [106]. Despite these advantages, CTAMs face challenges such as low ionic and electronic conductivity, continuous electrolyte degradation, and relatively high-volume expansion (<200%) [20]. However, ongoing technological advancements, particularly those involving nano-engineering methods, have demonstrated improvements in increasing the capacity of CTAMs for lithium storage over time.

3.7. Electrolyte Materials for Battery Technology

The electrolyte is one of the key components that serves a critical role in facilitating the Li-ion transportation between the battery’s electrodes during charging and discharging operations. The cell capacity, cycle-life, safety, and temperature range of LIBs are significantly influenced by electrolytes. According to the physical state, electrolytes are divided into two categories: Solid electrolytes (SEs) and liquid electrolytes (LEs). Most of the LEs are composed of lithium hexafluorophosphate (LiPF₆) salt dissolved in one or more organic solvents such as ethylene carbonate (ECs), dimethyl carbonates (DMCs), diethyl carbonates (DECs), propylene carbonates (PCs), ethyl methyl carbonates (EMCs) [21,107]. The LEs’ composition has a significant impact on the batteries’ electrochemical performance. Because of the safety risks, such as decomposition, leakage, and even explosion, associated with typical flammable organic-based electrolytes, the utilization of non-flammable LEs is of particular interest. The identification and testing of non-flammable LEs have been the subject of numerous investigations [108,109]. It is challenging to compare the performance of these electrolytes because they have been investigated under different conditions. Phosphorus-free fluorinated solvents outperform phosphate and phosphonate-based solvents, resulting in greater capacity retention after cycling [21]. Ionic liquids (ILs) have also been explored as alternative electrolytes for LIBs due to their wide electrochemical stability window, high ionic conductivity, and non-flammability [110]. Hybrid solid–liquid electrolytes (HEs) have also been researched as alternate electrolyte systems to address safety concerns with liquid electrolytes [111]. The overall goal of research in this domain is to increase the safety and electrochemical performance of LEs for next-generation energy storage applications.

The SEs have great electrochemical characteristics and are very safe; therefore, research is underway to substitute LEs in LIBs for EVs. Different types of solid electrolytes have been studied, including Li_1+xAl_xTi_2−x(PO₄)₃ (LATP), Li_1+xAl_xGe_2−x(PO₄)₃ (LAGP), and Li_7−3xAl_xLa₃Zr₂O₁₂ (LLZ) [112]. These solid electrolytes have demonstrated good lithium-ion conductivity at ambient temperature, making them potential materials for all-solid-state LIBs for high-energy density applications like EVs [113]. Specifically, Li-stuffed garnet electrolytes (LLZO: Li₇La₃Zr₂O₁₂) have attracted the interest of researchers due to their high total Li-ion conductivity, chemical stability, safety, and electrochemical stability window [114]. The use of solid-state electrolytes can also enable the operation of lithium-ion cells at higher voltages, increasing their specific energy [115]. Furthermore, composite solid electrolytes have been investigated to increase the overall performance of solid electrolytes by combining the advantages of various components [116]. Thus. SEs have the potential to improve the safety and performance of LIBs used in EVs.

3.8. Separator Materials for Battery Technology

One of the most important LIB components for maintaining LIB safety is the separator, a thin, porous membrane that physically separates the anode and cathode. The main function of the separator is to facilitate ion transportation within the LIB’s cell while preventing direct physical contact between the anode and cathode [117]. In addition, the separator’s physical structure and chemical characteristics directly affect the internal resistance and cycle performance [118]. Nevertheless, thermal runaway and explosion incidents of LIBs are brought on by separator damage from localized overheating, external impact punctures, and internal and external short circuits [119]. Therefore, the requirements of high-performance separators usually include thickness, porosity (ratio of pore volume to the total volume of separator), pore size, wettability, thermal stability, mechanical strength, chemical/electrochemical stability, and ionic conductivity rate [120]. Usually, the thickness, porosity, and pore size of the separator used in EVs are around 25–40 µm, 40–60% and <1 µm, respectively [121,122,123,124].

Polymeric materials, specifically polypropylene (PP) and polyethylene (PE), serve as separators in batteries for commercial applications, including Tesla’s EVs and Asus’s electronic devices [125]. These separators are single-layer or multilayer polymer sheets constructed of polyolefins [22,126]. However, polymer-based separators encounter practical challenges related to properties such as limited mechanical strength, low melting points, and reduced chemical resistance. These factors can impede overall efficiency. Although polyethylene terephthalate (PET) and polyvinylidene fluoride (PVdF) have been employed in commercial separators, they are significantly less prevalent than polyolefin films. These separators are highly porous, with typically >40% porosity, are around 25 µm thick, have low ionic resistivity (1.5–2.5 Ωcm⁻²), and bulk puncture strengths of more than 300 g/mil [127]. From a performance standpoint, separators should be much thinner than 25 µm, and there are examples of separators as thin as 12–20 µm, but the very thin membranes lose a lot of mechanical strength.

Commonly used nanomaterials for enhancing separators include carbon nanotubes (CNTs) [128], graphene, graphene oxide [129], nanofibers, silicon oxide [130], lithium titanate [131], metal–organic [132], and polymer nanocomposites [133].

Table 3. An overview of different nanoparticles, highlighting their main characteristics and the manufacturers producing them.

Nanomaterial	Material Production Companies	Material Properties
Carbon Nanotubes (CNTs)	BYK Additives, Cabot Corporation, Arkema, Nanocyl, OCSiAl, LG Chem	Used to enhance the mechanical strength and electrical conductivity of separators.
Graphene and Graphene Oxide	Graphenea, NanoXplore/XG Sciences	Incorporated to improve mechanical properties, thermal stability, and ion conductivity.
Nanofibers (Polymer, Ceramic, or Carbon)	Asahi Kasei, Hollingsworth and Vose	Electrospun nanofibers are used to create separator mats with increased porosity, improving electrolyte penetration.
silicon Dioxide (SiO2)	Cabot Corporation	Used to enhance the thermal stability of separators and improve their mechanical properties.
Lithium Titanate (Li4Ti5O12)	NEI Corporation, Umicore	Used to improve ion conductivity and thermal stability.
Metal–Organic Frameworks (MOFs)	BASF	Used for their porosity to enhance electrolyte penetration and ion transport.
Polymer Nanocomposites	LG Chem, Arkema	Nanoscale additives are incorporated into polymer separators to improve their overall mechanical properties.

lists several nanoparticles, along with their characteristics and the companies that produce them. These materials can either function as coatings for polymer separators or be added as components in polymer-based composites.

4. Lithium Plating and Interfacial Layer Formation

4.1. Lithium Plating

Lithium plating is one of the major degradation factors of LIBs during fast charging. During the charging process, lithium ions are typically intercalated into the anode (usually made of graphite). However, under certain circumstances, lithium ions can deposit on the surface of the anode as metallic lithium rather than being intercalated into the graphite due to a Faradaic side reaction. The formation of these lithium metals on the anode surface is called lithium plating. The lithium plating is more likely to occur at low temperatures (reduced Li-ion movements), higher C-rate charging (insufficient time to intercalate), and anode overcharged (an unusual case). In addition, when lithium metal is deposited on the anode, it can form needle-like structures known as dendrites. However, these dendrites pose a risk of penetrating the separator between anode and cathode, causing potential short circuits, thermal runaway, and safety hazards within the LIBs, as shown in Figure 7 [134]. Moreover, lithium plating has the potential to impact performance, accelerate aging, and compromise operational safety through various mechanisms. Firstly, capacity fading may ensue because of the depletion of active lithium inventory caused by the deactivation of plated lithium. Secondly, internal resistance could rise due to pore blockage caused by plated lithium, impeding ion transport within the porous electrode. Thirdly, there is a heightened risk of short circuits attributed to the formation of Li-metal dendrites [135].

4.2. Solid Electrolyte Interphase (SEI)

The solid electrolyte interphase (SEI) is a crucial interface on anode electrode surfaces in LIBs, produced during the first charging/discharging process by the decomposition of electrolytes. The SEI layer (usually has 3–10 nm thickness) is critical in promoting Li-ion movement while restricting electron flow, avoiding additional electrolyte degradation, and ensuring long-term electrochemical reactions [136,137]. Its function is critical to the functionality of LIBs in oxidation/reduction reactions. Initially, a dense and intact SEI limits electron movements, protecting the battery’s chemical and electrochemical stability [19]. However, SEI growth depletes active lithium and electrolyte materials over time, resulting in capacity fading, increased battery resistance, and decreased power density. This process begins when the redox potential of the electrodes falls outside the electrochemical window of the electrolyte. Several factors influence the creation of the SEI layer, including the active material used, binder material, electrolyte composition, salt, solvents, and other chemical features inherent in the battery’s design. Addressing SEI-related issues is critical for improving LIB performance and lifespan.

4.3. Cathode Electrolyte Interphase (CEI)

Cathode electrolyte interphase (CEI) layer appears at the interface between the cathode and electrolyte due to electrolyte decomposition, as shown in Figure 7. The CEI layer is an extremely thin layer (0.5–1 nm), and the elements that make up the CEI layer include ROCO₂Li, LiF, Li₂CO₃, and Li₂O [137,138]. Cathode deintercalation causes phase changes during the charging process, which causes disordering in the crystal structure. Additionally, the development of the CEI layer accelerates gas evolution and electrolyte decomposition, which raises battery impedance. It contributes to the depletion of lithium inventory by acting as a passivation layer [139]. It is possible to establish a stable CEI layer and improve overall cycling stability by cycling below a critical voltage (4.05 V vs. Li in the case of LCO). On the other hand, the CEI layer becomes dynamically unstable, and the interfacial reactions and degradation start continuously if the critical potential is in line with the cycling voltage range [140]. This deterioration ultimately leads to an increased battery impedance by accelerating the evolution of gases and the decomposition of electrolytes. As a result, this increase in impedance causes power fading. To address challenges and improve the overall performance, lifetime, and efficiency of LIB systems, several approaches in material design, electrolyte composition, and structural considerations must be adopted.

5. Online EIS for Lithium-Ion Battery Diagnostics

5.1. Degradation Indicators from EIS

The lifetime and degradation behavior of LIB cells and hybrid lithium-ion capacitors (LICs) can be modeled and estimated through the monitoring of current and temperature profiles, as reported in [141,142]. However, such empirical or semi-empirical degradation models are inherently limited, as they are unable to capture the underlying physicochemical transformations occurring within the cell materials. The EIS is one of the non-destructive and accurate methods to probe the battery’s internal processes across frequency sweeping, measuring Ohmic conduction, interfacial charge-transfer, double-layer behavior, and charge transfer effects in order to be translated into a meaningful state estimation [143,144]. The Nyquist plot provides a comprehensive visualization of the impedance behavior of a LIB cell, correlating each frequency domain with specific electrochemical processes. At high frequencies, the intercept on the real axis corresponds to the ohmic resistance (R_Ω), and inductance L representing the electrolyte, current collectors, and contact resistances. The small arc in the high-frequency region reflects the solid electrolyte interphase (SEI) resistance (R_SEI) and its associated capacitance (C_SEI), as shown in Figure 8. R_SEI and C_SEI create a better fit in the equivalent circuit model (ECM). The wider semicircle at medium frequencies represents the charge-transfer resistance (R_ct) and the double-layer capacitance (C_dl), which dominate the electrode kinetics. At low frequencies, the impedance increases linearly with a slope of approximately 45°, known as the Warburg impedance (W), corresponding to lithium-ion diffusion within the electrode material, as shown in Figure 8a. As degradation progresses, both R_SEI and R_ct tend to increase, leading to a rightward shift and enlargement of the Nyquist semicircle, as shown in Figure 8b. This behavior is commonly used as a diagnostic indicator of aging and loss of electrochemical activity in LIBs, and the SEI/CEI growth, loss of active material, electrolyte decomposition, or lithium plating alter these parameters in characteristic (e.g., increased R_Ω and R_ct) [145,146], shifts in characteristic time constants, and increase in the low frequency component representing the diffusion response, providing a degradation mark and an indicator to capacity fade [147]. Empirical studies on LIBs are motivating the use of impedance-derived indicators in operational settings.

5.2. EIS Indicators for Comparison of LIB Chemistries

Early lithium plating can be observed in the higher-frequency region in the EIS. Fitting the EIS to its ECM parameters and recording the values of the RSEI and CSEI, as well as the different parameters, facilitates battery cell development monitoring. For passport integration, it is proposed to have a record of these EIS-derived ECM parameters together with temperature metadata, including a temperature normalization for comparability over time and across operating conditions. Additionally, these records can include some parameters, such as R_Ω, by taking the voltage response to the current step, then using this parameter in addition to the low-frequency extracted parameters to create the full range EIS analytically [148].

The Nyquist plots comparing five different LIB chemistries at the beginning-of-life (BOL) and after 100 cycles are illustrated in Figure 9 [149] under fast charging conditions. From Figure 9a, it can be seen that LFP shows the smallest post-aging growth of the mid-frequency semicircle (R_ct) and the low-frequency diffusion tail. On the other hand, LMO remains comparatively stable, as shown in Figure 9a,c, but exhibits a more vertically low-frequency signature than LFP after cycling, while layered oxides (LCO, NMC811, NCA) degrade more in terms of interfacial impedance, shown by large arc growth in the Nyquist plots, as illustrated in Figure 9b–d. Furthermore, LFP (olivine) and LMO (spinel) exhibit a smaller increase in their impedance arcs, meaning they degrade less (in terms of impedance) under the same fast-charging stress. Therefore, these trends are essential for the battery passport record for tracking the LIB health over time. For thermally sensitive chemistries such as LMO, where Mn-related interfacial changes accelerate at higher temperature, while LFP shows a low impedance change and is thermally stable, LMO shows greater sensitivity in the low-frequency diffusion region and mid-frequency charge-transfer arc, and this sensitivity is amplified at elevated temperature due to Mn-related lattice effects and dissolution, so impedance-based indicators should be normalized to a reference temperature and reported with measurement-temperature metadata. Practically, a shorter update cadence is recommended for LMO-rich packs during hot operation, while LFP’s higher thermal stability allows longer intervals [150,151].

Comparison of NCA and NMC: EIS Indicators

Ni-rich layered oxides such as NCA and NMC-811 are among the most promising cathode materials for high-energy LIBs, appreciated for their ability to deliver higher specific energy densities. The EIS plot of NMC811 and NCA is illustrated in Figure 9d. Nevertheless, despite their compositional resemblances, these materials exhibit notably different impedance responses when subjected to thermal stress and high C-rate cycling, which can influence both their long-term stability and performance under real-world conditions.

According to [152], a study of five nickel-containing positive electrodes found that at 40 °C, the area-specific impedance (ASI) increased by 0.5% per cycle for NCA and 1.5% per cycle for NMC811 under minor aging (100 cycles). This indicates that both chemistries are more sensitive to higher temperatures than other nickel-based materials. In [153], the authors measured the interfacial impedance of NMC111 and NMC811 at high SoC (84%) with LP57 electrolyte and found the impedance value for NMC811 to be 4.5 times that of NMC111 at the same Li content. While this comparison does not include NCA, it demonstrates the significantly higher impedance of Ni-rich cathodes under high SOC and emphasizes the need for frequent impedance monitoring.

Based on insights and empirical observations, NCA and NMC-811 cathodes exhibit distinct impedance evolution pathways when exposed to thermal and high-rate cycling stress. For NCA, rapid increases in impedance with temperature can be attributed to accelerated nickel dissolution and oxygen evolution at the cathode interface, which promotes the formation of interfacial and bulk defects; these processes drive marked growth in both R_ct and interfacial impedance components under higher thermal conditions. On the other hand, NMC-811 demonstrates greater sensitivity to high-rate cycling, where higher C-rates intensify lithium-ion transport limitations, induce micro-cracking, and contribute to the progressive thickening of the CEI. Therefore, these effects are most evident in the increased mid- and low-frequency impedance components, such as those associated with the Warburg element and diffusive transport.

For enhanced battery passport diagnostics tailored to Ni-rich cathode degradation, it is recommended to systematically log high-frequency R_Ω and R_ct at regular intervals, like every 40–50 full cycles or every 100 h of operation, particularly when cells operate at temperatures of 40 °C or above, or under high C-rates (≥2C). In addition, capturing low-frequency diffusion impedance (W) parameters is essential for cells undergoing fast-charging or sustained high-rate cycling (>2C), as these elements provide early diagnostic warning of emerging transport limitations and associated degradation mechanisms. Routine documentation of temperature, C-rate history, and SoC window, mapped against interval impedance data, is also crucial for supporting predictive models of cathode health and for lifetime assessment in mission-profile applications.

This enhanced data-collection strategy enables early detection of interface and structural degradation in Ni-rich chemistries, thus improving SoH estimation, facilitating meaningful battery passports, and ultimately supporting second-life/recycling decision frameworks.

5.3. Standardization Gaps in Converting Raw EIS to Chemistry-Aware Indicators

EIS is a powerful and non-destructive technique for probing the internal electrochemical processes of LIBs and LICs. Despite its diagnostic potential, the research community lacks a harmonized framework that can reliably translate raw impedance spectra into standardized, chemistry-aware indicators across various cell formats, electrode chemistry, and OEM implementations. Current performance and safety standards, such as the IEC 62660 series for automotive cells, define procedures for evaluating capacity, life, and safety but do not establish a standardized mapping from EIS spectra to chemical indicators suitable for real-time diagnostics, digital twins, or battery passports [154,155,156]. This lack of standardization prevents interoperability and traceability of impedance-based health analysis, in due course delaying the integration of EIS into data-driven platforms.

5.3.1. Causes of Non-Harmonization

The lack of harmonized EIS interpretation arises from a combination of measurement, modeling, and data-management challenges. First, measurement and fixturing variability remain a major contributor. The impedance spectrum of a LIB cell is highly sensitive to the measurement setup, including fixture impedance, cable routing, and contact resistances. Experimental studies have demonstrated that these effects can introduce substantial systematic errors, particularly in the low-to-mid frequency range [157]. In the absence of a unified standard experimental setup, calibration methods, and metadata reporting requirements, the extracted parameters vary significantly between laboratories and cannot be reliably compared across datasets or OEMs.

Second, state dependence and missing metadata further complicate comparability. The spectral response of a LIB cell shifts noticeably with different states of SoCs and temperature, while relaxation history and current load also affect the observed impedance. Unfortunately, many published datasets omit crucial metadata such as exact SoC, temperature, prior rest time, and cycling profile. Without this information, it becomes impossible to reproduce or chemically interpret impedance features in a consistent way [157].

A third source of irregularity lies in model and parameter extraction diversity. Researchers employ a wide range of modeling approaches, from ECM with diverse topologies to DRT reversal and Warburg-based diffusion models. However, inconsistencies in definitions, for example, using the constant-phase element (CPE-Q) instead of true capacitance, or employing different Warburg formulations, lead to parameters that are not directly comparable. Agreement on canonical definitions, units, and fitting frequency ranges is essential to ensure reproducibility.

Beyond modeling, chemistry and geometry dependencies limit the generalization of any single mapping. The same impedance signatures may reflect distinct inherent mechanisms depending on cathode/anode chemistry, electrolyte, electrode porosity, mission profile loading, and aging mechanisms. Therefore, mapping EIS parameters to chemical meaning without explicit chemistry-specific and geometry-specific metadata risks misdiagnosis.

Finally, progress is slowed by a lack of interpreted datasets and standardized toolchains. Although research community-driven infrastructures such as the Department of Energy (DOE)/National Renewable Energy Laboratory (NREL) battery_data_tools and the battery-data-toolkit are emerging, comprehensive repositories linking EIS spectra to independent chemical or morphological characterization remain limited [158,159]. The absence of such benchmark datasets impedes data-driven modeling and delays convergence toward universally accepted diagnostic indicators.

5.3.2. Metadata Needs and Parameters

To demonstrate these above challenges,

Table 4. Common EIS parameters, their physical significance, and standardization requirements.

Parameter	Diagnostic Use	Minimum Metadata Required	Standardization Gaps
RΩ	Ohmic resistance of electrolyte and contacts	Fixture description, cable correction, geometry, T, SoC	Fixture-dependent corrections, non-uniform area normalization
Rct	Interfacial kinetics; sensitive to SEI/CEI, metal dissolution	ECM topology, frequency range, SoC, T, rest time	Variable ECM topologies and fitting ranges, SoC/T normalization lacking
CPE	Non-ideal double-layer capacitance in porous electrodes	Electrode area, porosity, conversion model	Non-uniform capacitance conversion
W	Solid-state or electrolyte diffusion limitation/linked to mass transport	Model type, fit range, electrode thickness	Finite vs. semi-infinite definitions and mixed units

provides key EIS parameters, their diagnostic meaning, major standardization gaps, and the minimum metadata needed for reproducibility. For example, the R_Ω, obtained from the high-frequency capture, is often reported without reference to cell area or fixture corrections, making cross-study comparison unreliable. The R_ct, an indicator of interfacial reaction kinetics, varies strongly with SoC and temperature, yet publications rarely normalize or report these dependencies. Likewise, parameters derived from CPEs and W differ among ECMs. Therefore, synchronized metadata and reporting formats are needed for reliable interlaboratory comparison between different LIBs chemistries, manufacturer, and their internal structure.

5.3.3. Existing Standards and Initiatives

Several international and research community-driven initiatives are trying to address these above-mentioned challenges, though none yet provide a complete solution. The IEC 62660 series remains the most widely used reference for performance, life testing, abuse testing procedure, and safety testing of automotive LIBs [154,155,156]. However, it does not specify any formal procedure or metadata schema for EIS-based diagnostics. Likewise, open-source toolkits such as DOE/NREL’s battery_data_tools and the battery-data-toolkit project are valuable for data handling and pre-processing but have yet to define standardized chemistry-aware EIS indicators that could be adopted by formal standards bodies [158,159]. At the policy level, European Commission (JRC), ISO, SAE, and EU Battery Passport initiatives are increasingly emphasizing traceability, data interoperability, and second-life evaluation. However, most of these efforts still focus on policy and supply-chain data exchange rather than harmonizing the underlying EIS measurement methodologies [10,160].

5.3.4. Practical Recommendations for Future Standardization

Achieving harmonization of EIS-based diagnostics requires coordinated action between research institutions, industry stakeholders, and standardization bodies. As a first step, a minimum EIS metadata schema should be defined for inclusion in battery passports, mandating information on cell geometry, electrode area, temperature, SoC, rest and current history, fixture description, and measurement frequency points. Equally important is the standardization of parameter definitions and units, ensuring consistent interpretation of R_Ω, R_ct, CPE-to-capacitance mappings, and W across laboratories and applied applications. The research community should also adopt shared tool frameworks and curated datasets, such as those provided by DOE/NREL, to establish benchmark references and enable machine-learning-based cross-chemistry validation. Lastly, coordination with international standards organizations, including IEC, ISO, SAE, and EU regulatory frameworks, will be essential to integrate these harmonized definitions into the global standards landscape, enabling cross-OEM comparability and trustworthy EIS-based battery passports.

5.4. Real-Time Implementation on Operational Systems

Classical EIS requires dedicated AC perturbations and full frequency sweeps, which are hard to deploy on vehicles due to hardware overhead, measurement time, and noise sensitivity [143,145,146]. Recent work demonstrates feasible online implementations: On-board sensing modules inject small excitations and fit parametric models in situ, enabling periodic scans during operation [161].

In standard EIS testing in the lab, a small-amplitude sinusoidal signal is applied to the cell at sweeping frequencies in a range from high to low (typically from ~10 kHz down to ~10 mHz) [162]. This method is called AC-EIS and is usually used for standard laboratory testing, not for online in situ techniques. AC-EIS is presented in Figure 10a. Another method is the Square-Potential EIS (SP-EIS), where instead of a sine wave, a small square-wave perturbation is used at a fixed frequency (e.g., 40 Hz and 10 mV p–p) as in Figure 10b. A 50%-duty square wave at f0 produces odd harmonics at (f0, 3f0, 5f0, …). By sampling v(t) and i(t), performing a fast Fourier transform (FFT), and computing Z(f) = V(f)/I(f) at those harmonic frequencies, the impedance is measured at multiple odd-harmonic frequency points. AC-EIS delivers a high-resolution spectrum but needs a frequency-response analyzer and longer measurement time; SP-EIS yields one or a handful of accurate impedance points quickly with simple hardware. Both methods require small perturbations (linearity) and control of temperature and SoC [163]. Impedance is estimated in multicell stacks using only cell voltages and known circuit parameters, removing dedicated current transducers and simplifying integration [164,165]. The direct-synthesis ternary (DST) sequence is a broadband excitation signal that alternates between three discrete levels (e.g., 0, 1, 2) to generate a wide range of frequency components with uniform spectral energy. This makes it ideal for fast, accurate impedance measurements with minimal distortion. DST is resilient to quantization noise and transients, but its accuracy depends on stable, well-calibrated current measurements. However, the sensor bias, offset, or gain drift can corrupt the sequence and bias extracted impedance parameters [166,167,168]. On the other hand, sensorless and observer-based current estimation (e.g., EKF, UKF, nonlinear adaptive observers) eliminates the physical sensor but is more sensitive to model mismatch, inverter ripple, and sampling jitter. However, robust observer designs mitigate these effects but increase computational load and require careful tuning [169,170,171]. In general, DST with high-quality sensing yields more reliable passport-grade parameters, while sensorless approaches demand advanced observers to achieve comparable robustness.

In a multi-cell stack system, each cell is connected to a simple switch–resistor circuit that can inject small excitation signals into the cell. The excitation is a pseudo-random or DST sequence, which alternates between three discrete voltage or current levels to produce a broadband frequency response as shown in Figure 11a. A DST signal, denoted as $u_{DST\left(t\right)}$, contains evenly distributed spectral energy across many frequencies while suppressing harmonic multiples of two and three, making it highly suitable for impedance measurements of nonlinear systems such as LIBs, as shown in Figure 11b [164].

The estimated current and measured voltage are transformed into the frequency domain using FFT, and the impedance spectrum is obtained as Z(f) = V(f)/I(f). This current-sensorless approach simplifies on-board implementation, reduces hardware cost, and has been experimentally validated to produce accurate results over 1 Hz–2.5 kHz. It allows practical on-board EIS in multi-cell battery packs for real-time SoH and temperature estimation. On the other hand, model-based observer approaches, such as Kalman filter designs, are employed to estimate impedance-related states; however, they rely on highly accurate system dynamics and impose significant computational demands on embedded platforms [172].

Another technique is the distribution of relaxation times (DRTs), which enables finer separation of electrochemical processes; however, it involves solving ill-posed inverse problems and requires substantial data processing and regularization, which pose challenges for real-time implementation [173,174]. The DRT requires explicit regularization techniques such as Tikhonov or Bayesian. Tikhonov regularization provides a smoothing approach that fits the DRT while adding a penalty to discourage sharp and asymmetric peaks. In addition, the smoothing level is selected using a rule such as the L-curve or cross-validation in Tikhonov regularization [175,176,177,178]. On the other hand, Bayesian regularization provides a framework where prior probability distributions (priors) are placed over the model’s parameters to express assumptions about their likely values. This prior Gaussian process is specified to enforce smoothness, and the method returns both the DRT and uncertainty bands [179,180]. While DRT is sensitive to noise measurement and short records, ECM-based analysis is less sensitive to noise, as it embeds model bias [181,182]. Therefore, in EV application and onboard EIS, a combined solution of the ECM-DRT-based method is satisfactory to trend SEI-related parameters and achieve battery state estimation in the onboard conditions [183].

5.5. Online EIS: Architecture and Workflow

Implementing in situ EIS testing in existing battery systems is challenging, as retrofitting additional hardware is often infeasible, and updating the embedded firmware to enable switching-based perturbation for impedance measurement and state estimation is typically complex. Therefore, a cloud-based computational solution for EIS analysis utilizing available low-frequency measurements and trained models for state estimation is needed.

Furthermore, the recent developments indicate that several major automakers have started integrating online battery diagnostic systems to improve real-time state estimation and predictive maintenance scheduling. According to the global market insights report, companies such as Tesla, Ford, Rivian, BYD, and General Motors have implemented onboard diagnostic solutions that combine impedance-based sensing and machine learning based voltage/current analytics for continuous monitoring of battery health throughout the EV lifetime [48]. Although OEMs do not disclose explicit SoH/SoC accuracy metrics, published studies suggest that EIS-based diagnostic systems typically achieve SoH estimation errors below 3%, while SoC estimation errors are around 2% under controlled conditions [181,184,185,186]. These advancements illustrate the ongoing transition from laboratory diagnostics to online diagnostic frameworks, connecting the gap between passive monitoring and active system intelligence for battery lifecycle management.

5.5.1. Passive EIS Testing

The main challenge of passive EIS is that the mission profile (voltage–current logs) does not excite the full range of frequencies required for EIS testing. With a limited sample rate and a finite window length, only a subset of the spectrum is observable. Therefore, high-frequency semicircles cannot be measured directly, and estimation must be noise-robust and restricted to the frequencies that are actually present.

A practical solution is to compute impedance as the ratio of averaged voltage and current spectra formed from many short, overlapped, windowed segments (Welch-style averaging of periodograms), then validate against laboratory EIS at matched SoC and temperature to ensure comparability during on-board operation [187].

Passive EIS, when equipped with rich current harmonics and high-frequency logging, effectively extends the accessible measurement bandwidth to about 0.2 Hz–3 kHz, allowing the partial reconstruction of the impedance spectrum without the requirement for active perturbation [188,189,190]. This makes passive EIS specifically valuable for real-time monitoring in embedded or automotive applications, where continuous impedance tracking is valuable and operational invasiveness must be minimized. Coupling this approach with DRT stabilization and observer-based state estimation (e.g., Kalman-filter–based designs) further enhances robustness and interpretability, supporting the extraction of physiochemical signatures linked to battery SoH and state-of-temperature (SoT) [191].

However, the reliability of passive EIS for precise SoH and SoT estimation is intrinsically system-dependent, with its fidelity relative to active EIS influenced by operating parameters such as C-rate, temperature, and drive-cycle profile. Whereas active EIS permits controlled excitation and high signal-to-noise performance across a broader frequency range, including low-frequency processes critical for degradation monitoring, passive EIS may demonstrate reduced sensitivity under certain dynamic conditions, and its accuracy may be limited by the operational profile or noise present during typical practice. Accordingly, comparative validation studies are essential to understand the regimes in which passive EIS provides reliable diagnostics and where supplementary active EIS or hybrid methods are demanded. Such studies, targeting representative driving and fast-charging scenarios, are needed to establish best practices for integrating EIS-based features into digital battery passports and predictive battery management.

Furthermore, signal-processing techniques can also be employed to recover impedance from time-domain voltage and current records using FFT-based analysis. Leakage and aliasing effects can be mitigated through established correction methods that are now standard in passive measurement workflows [192].

5.5.2. Cloud-Based EIS Battery Diagnostic

A cloud-based EIS battery diagnostic system for predictive cell-state estimation and smart energy management integration is presented in [193]. In this approach, BMS voltage and current sensors sample and stream operational signals to a per-cell digital twin, where cloud-hosted digital signal processing applies FFT to extract low-frequency impedance features and identify ECM parameters.

The extracted parameters are subsequently utilized within a machine learning framework, where trained models like linear regression (LR), support vector regression (SVR), random forests (RF), gaussian process regression (GPR), recursive least squares (RLS), three-point extraction (TPE), chaotic particle swarm algorithm (CPSA), genetic algorithm (GA), long short-term memory (LSTM), and segment averaging scheme algorithms (SASA), and other compact neural networks to process either the raw parameters or their low-dimensional embeddings. Through this data-driven approach, the models predict key SoX indicators, including SoH, SoC, and SoT, enabling adaptive and predictive battery management. Both the parameters and the state estimates are stored in a battery-passport database for analytics and lifecycle traceability [194]. The database maintains versions, continuously updated training data, and passport records, and supports periodic model retraining to keep the estimator current, as shown in Figure 12. The cloud-based battery monitoring system can be a reliable component in the Holistic, data-driven energy management [195], where the battery is an asset that needs monitoring and smart management in addition to other variables such as electricity and fuel prices [196]. In this way, the cloud-based EIS diagnostic framework provides a scalable solution that supports battery manufacturers and related industries by enabling real-time impedance analysis, data-driven health diagnostics, and predictive maintenance across large fleets of cells and systems.

5.6. Adaptation of Machine Learning Models for Real-World Battery Diagnostics

Machine learning models trained on controlled laboratory datasets commonly face challenges maintaining predictive accuracy once deployed under variable real-world conditions due to inconsistencies in temperature dynamics, stochastic load profiles, and sensor noise. In addition, laboratory data, in general, exhibit idealized charge/discharge cycles and limited aging modes, whereas field data comprise non-stationary behaviors such as partial cycling, highly variable current patterns, and calendar aging. These differences induce distributional shifts that damage model generalization and reliability in practice.

Recent developments focus on frameworks like transfer learning, domain adaptation, and incremental learning that allow models to adjust their parameters using newly acquired field data without requiring full retraining [197,198,199]. Hybrid approaches combining electrochemical model constraints with data-driven updates have enhanced robustness in predicting SoH and SoC across temperature ranges and dynamic operating profiles [200,201,202,203]. Integration of online adaptive learning and active data selection within embedded BMS assists continuous recalibration, which is essential for real-time monitoring and lifetime prediction.

In parallel, the diversity of LIB chemistries and their distinct degradation mechanisms prevents a single universal characterization of EIS-based SoH estimation performance. As summarized in

Table 5. Comparison of different impedance measurements used to estimate LIB SoH. TC = time constants.

Ref.	Battery Chemistry	Part of Impedance	Frequency Range	Used Model for SOH Estimation	SoH Estimation Error
[204]	NMC	RΩ	-	RLS	4%
[205]	LFP	Rct	-	TPE	6.1%
[206]	NMC	RΩ, Rct and RSEI	Determine via TC	DRT	<10%
[184]	NCR and LFP	RΩ, Rct and RSEI	20 Hz–1 kHz	DRT	<3%
[207]	LFP	Rct	0.01 Hz–1 kHz	CPSA/GA	<15%
[208]	LFP, LTO	RΩ, Rct and CPE, W	0.01–100 kHz	LS/PF/LR	7%
[209]	LCO	RΩ, Rct RSEI, CSEI CPE, W	0.02 Hz–20 kHz	DRT/LSTM	2.68%
[210]	LFP	Zreal	0.1 Hz–10 kHz	RPR	4.46%
[181]	LCO	RΩ, Rct RSEI, CSEI CPE, W	0.02 Hz–20 KHz	GPR	2.95%
[211]	NMC	RΩ, Rct, CPEs	0.01 Hz–10 kHz	Empirical	5%
[187]	NMC	Zreal	25 kHz	SASA	2.3%
[212]	NMC	RΩ, Rct RSEI	180 mHz–2.7 kHz	Empirical	-

, different chemistries have a unique impedance signature based on the employed frequencies and SoH estimation methods [146]. Therefore, existing EIS-based SoH estimation methods, including RLS, TPE, CPSA/GA, GPR, LSTM, and SASA, exhibit a broad yet expected accuracy range of approximately 2–15%. This variation does not indicate methodological inconsistency; rather, it reflects that each technique is tailored to specific LIB chemistries, frequency ranges, modeling assumptions, and application requirements. Consequently, the literature consistently shows that no single EIS-based model performs best across all LIB chemistries or use cases. Furthermore, reported accuracies are highly dependent on the chemistry-specific impedance characteristics and the chosen SoH estimation method. Therefore, model performance must always be taken within this contextual framework.

5.7. Cybersecurity Considerations for Battery Passports

The integration of impedance-based health indicators within battery passports and cloud-connected diagnostic platforms certainly raises critical cybersecurity challenges that must be effectively addressed to protect data integrity, confidentiality, and interoperability across OEMs and energy management systems. The bidirectional data exchange among the BMS, vehicle gateway, and cloud infrastructure exposes vulnerabilities to cyber threats such as data tampering, spoofing, and unauthorized access to sensitive electrochemical parameters. Ensuring secure data handling and transmission is crucial for the large-scale deployment of these technologies.

Emerging automotive cybersecurity standards, including ISO/SAE 21434 for road vehicle cybersecurity and IEC 62443 for industrial network security, offer foundational guidance for implementing security by design (SbD) principles in connected vehicle systems [213,214]. These standards underline securing communication protocols, enforcing strict access controls, and employing cryptographic authentication for software and firmware updates. Recent developments encourage blockchain-based traceability, edge computing, and digital twin encryption techniques to enhance diagnostic data protection and prevent malicious interference within online monitoring environments [215,216,217,218,219,220].

Furthermore, robust cybersecurity strategies must be included within BMS architectures to mitigate risks such as denial-of-service attacks (DoS), man-in-the-middle (MITM) attacks, fault injection, fuzzing, unauthorized remote access (URA), and firmware tampering. For these attacks, several critical measures can be implemented across different layers of the BMS to improve cybersecurity and ensure data integrity. One key part involves securing communication and data transfer processes. End-to-end encryption techniques such as AES-256 and TLS 1.3 should be applied to protect communication between the BMS and cloud systems, while cryptographic authentication can safeguard firmware updates from malicious modifications [221,222]. Additionally, the use of secure communication protocols, including MQTT and HTTPS, can further strengthen remote monitoring and control functions [213,223]. Another essential measure is strengthening access control and authentication mechanisms. Multi-Factor Authentication (MFA) should be adopted for all remote access points. And Role-Based Access Control (RBAC) can be implemented to limit data access according to predefined user roles. Furthermore, digital signatures can serve as an added layer of protection for verifying firmware authenticity before installation [224].

Moreover, intrusion detection and anomaly monitoring also play an important role in maintaining system security. Artificial intelligence-based anomaly detection can identify unusual battery behavior patterns that may signal cyberattacks, while a real-time Intrusion Detection System (IDS) can continuously monitor network traffic for potential threats [225]. Another aspect is to secure network interfaces and connectivity. It is important to isolate BMS networks from public networks using firewalls and VPNs. A secure API gateway should also be deployed to validate and monitor third-party interactions, and regular penetration testing can help identify and mitigate emerging vulnerabilities.

Finally, firmware and hardware security enhancements further reinforce system resilience. Implementing secure boot mechanisms ensures that only authorized firmware is executed on the BMS, while integrating a Hardware Security Module (HSM) provides a secure environment for storing cryptographic keys, protecting the system from unauthorized access and tampering [226].

Regulatory frameworks such as the United Nations Economic Commission for Europe’s (UNECE) regulation on cybersecurity and software updates, together with the European Union’s General Safety Regulation, mandate the implementation of robust cybersecurity measures for connected and automated vehicles from 2024 onward [227]. These frameworks emphasize an industry-wide commitment to safeguarding vehicle communication networks and protecting battery-related data flows against cyber threats. Future research should focus on designing lightweight cryptographic protocols and privacy-preserving architectures tailored to the high-frequency impedance data streams common in battery health monitoring, thereby ensuring trustworthy and resilient battery passports.

6. Conclusions

The battery passport is necessary to ensure transparency, sustainability, and reliability across the entire battery value chain from raw material extraction to recycling. In addition, a battery passport is a digital record that attaches to each battery throughout its life and keeps its identity, origin, and verified condition authenticated. The battery passport integrity depends on two important components: (i) materials and manufacturing documentation, which provide information about how the battery is built and what impact it leaves on the environment; (ii) online inspection during system operation, which establishes how the cell behaves under real duty cycles and the battery state. These components are essentially connected, where materials and design topologies set electrochemical pathways and degradation modes, while online EIS translates operating signals into chemistry-aware and material change indicators that keep the passport’s dynamics updated. When combined with interoperable data structures and governance, these materials–diagnostics–passport integration enables reliable lifecycle health reporting and supports safety, durability, and circularity decisions across the value chain.

On the materials side, commercial chemistries (e.g., LCO, NCA/NMC, LMO, LFP) and anode/electrolyte/separator choices shape impedance signatures through ohmic, charge-transfer, double-layer, and diffusion processes. Aging modes, including SEI/CEI growth, active-material loss, oxygen release, transition-metal dissolution, and lithium plating, map to stable impedance features (for example, growth in R_Ω and R_ct, shifts in time constants, and change in low-frequency diffusion response).

On diagnostics, a clear progression has been established from laboratory AC-EIS toward in situ/on-board methods operating under mission constraints. In addition to classical sweeps and square-wave EIS, sensorless current measurement and switching-based approaches provide broadband excitation with minimal hardware. Passive, FFT-based EIS information extraction exploits the native voltage–current fluctuations to recover low-frequency complex impedance without signal injection or perturbation and relying only on BMS sensors.

The cloud-based EIS diagnostic and battery-passport framework enables condition-aware battery management, impedance-informed decision, predictive maintenance, and standardized second-life grading. Extracted parameters are processed by machine learning models directly to predict key SoX indicators. Both parameters and state estimates are stored in a battery-passport database, supporting benchmarking, lifecycle traceability, and periodic model retraining. This integrated system provides scalable, real-time diagnostics and predictive management, linking materials, sensing, analytics, and lifecycle decisions for optimized battery and energy management.