Environmental Impacts of Lithium-Ion and Lead-Acid Battery Recycling Programs: A Systematic Review and Meta-Analysis

Abstract

Global growth in electric mobility, portable electronics, and renewable energy storage has increased concerns about the environmental and economic impacts of managing end-of-life lithium-ion and lead-acid batteries. Although these batteries support the transition to renewable energy, their disposal presents significant challenges. Recycling has emerged as a key strategy to reduce resource depletion, limit pollution, and recover valuable materials. This study systematically reviewed and quantitatively synthesised the literature published between 2000 and 2025, assessing the environmental impacts of battery recycling programs. The review followed PRISMA guidelines to ensure a transparent and rigorous study selection process. Data from peer-reviewed articles, industry reports, and policy documents were analysed, focusing on indicators such as greenhouse gas emissions, energy use, material recovery efficiency, and economic returns. Statistical methods, including Hedges’ g, heterogeneity testing, and sensitivity analysis within a random-effects model, were applied to account for variability across technologies and battery types. The results show that recycling generally lowers emissions and improves resource recovery compared to virgin material extraction, though performance varies. Lead-acid recycling demonstrates stronger environmental benefits due to mature technologies and established systems, while lithium-ion recycling shows positive but lower gains, limited by higher energy demands and less-developed processes. Overall, recycling is essential for reducing environmental impacts and supporting a circular economy, though lithium-ion systems require further technological and policy advancements. These findings can be used by governments to strengthen regulatory frameworks to support recycling industries and invest in advanced lithium-ion recycling technologies to improve efficiency. Despite the existing limitations, the benefits of recycling outweigh the drawbacks, making it a necessary strategy for sustainable battery waste management.

1. Introduction

The global demand for batteries has surged due to rapid technological advancements, and there is a growing need for efficient, reliable energy storage systems. This increase is largely driven by the widespread adoption of portable electronics, electric vehicles (EVs), and grid-scale energy storage systems supporting renewable energy integration [1,2]. In addition, amid escalating energy demands driven by population growth, industrialisation, and economic expansion, battery production has expanded significantly worldwide [3]. However, the rapid growth in battery usage raises substantial environmental concerns, particularly concerning the disposal of spent batteries.

Global commitments to carbon neutrality and sustainable development are transforming urban transportation systems, accelerating the adoption of zero-emission technologies. Within densely populated cities, public transportation is crucial for reducing greenhouse gas emissions and enhancing local air quality. Among emerging zero-emission solutions, fuel cell buses (FCBs) stand out due to their high energy efficiency, zero tailpipe pollutants, and capacity to meet high passenger demand, positioning them as a viable alternative to conventional diesel buses. Nevertheless, widespread FCB deployment is constrained by persistent challenges, notably high hydrogen consumption and the limited durability of fuel cell systems and lithium-ion batteries [4].

Many end-of-life lithium-ion and lead-acid batteries end up in landfills, where toxic heavy metals such as lead, cadmium, and lithium can leach into soil and water systems, contaminating them and causing ecological harm [2,3]. Recent research has demonstrated that lithium-ion batteries release per- and polyfluoroalkyl substances (PFASs) and inorganic fluorochemicals such as hexafluorophosphate (PF6⁻) into landfill leachate at concentrations of up to 100 mg/L over a period of 220 days [5]. The disposal of lead-acid batteries represents a significant anthropogenic source of global soil contamination, mostly due to the emission of lead (Pb) and related chemicals, including sulphates, oxides, and acid electrolyte residues [6]. When batteries leak into surrounding areas, the leakage can have harmful effects, including soil contamination. This occurs when hazardous substances in the soil alter its composition, reducing its fertility and posing a risk to living organisms. Improper battery disposal also contributes to water pollution, as toxic substances can leach into surface water and groundwater, contaminating aquatic life, ecosystems, and drinking water sources. To address these environmental risks, the principles of the circular economy, the big three Rs, which include recycling, repurposing and reuse, are increasingly recommended [7,8].

Battery recycling is the process of reprocessing and recovering batteries to extract valuable materials while securely disposing of toxic substances [9]. The literature presents divergent hypotheses regarding the environmental impacts of battery recycling. Some studies suggest that recycling batteries can mitigate their environmental impact by reducing the energy and raw material inputs required for new battery production, as well as the environmental damage caused by hazardous battery disposal and the extraction of raw materials for new ones [10]. However, others suggest that battery recycling itself poses certain environmental and economic challenges. Current recycling technologies include pyrometallurgy, hydrometallurgy, and direct recycling, which are resource- and energy-intensive and often require large amounts of water, reagents, and thermal energy [11,12]. These processes can produce emissions, wastewater, thermal pollution and secondary pollutants, offsetting some of the intended environmental benefits [13].

From an environmental perspective, this research addresses the escalating risks associated with improper battery disposal, including soil and water contamination, the leaching of toxic metals, and greenhouse gas emissions. From an economic standpoint, this research will inform the development of cost-effective, resource-efficient recycling models that can stimulate green job creation, reduce reliance on imported raw materials, and support circular economic strategies. In terms of policy and governance, the study offers evidence-based recommendations for strengthening the implementation of key regulations. Furthermore, this research contributes to academic knowledge by integrating environmental science, economic analysis, policy review, and stakeholder engagement to develop a comprehensive understanding of the dynamics of battery recycling. Given these complexities, there is a need for a comprehensive and evidence-based assessment of the environmental and economic impacts of battery recycling programs. While numerous studies have examined specific aspects of battery recycling, there remains a lack of integrated analyses that synthesise findings across different technologies, battery types, and geographic contexts. This gap is evident in developing regions, where localised data and policy-relevant insights are limited. While systematic reviews and life cycle assessments on battery recycling exist, the novelty of this manuscript lies in the integrated application of PRISMA, meta-analysis, and policy discussion, rather than using these methods in isolation. This integration enables three novel syntheses: (1) a direct comparison of GWP reductions between lithium-ion and lead-acid batteries under the same framework, revealing a five-fold difference; (2) an explanation of the 99% heterogeneity by linking specific effect sizes to technological factors; and (3) policy recommendations quantitatively grounded in meta-analysis results, such as targeting high-value Co and Ni recovery (>85% efficiency) rather than blanket subsidies and diagnosing policy misalignment. No prior study has combined these three methods to simultaneously evaluate environmental performance, economic returns, and policy implementation gaps for both battery chemistries, particularly in emerging economy contexts.

Therefore, the aim of this study is to systematically evaluate and synthesise the environmental and economic impacts of lithium-ion and lead-acid battery recycling programs using a systematic review and meta-analysis approach.

2. Materials and Methods

This study employed a systematic review and meta-analysis approach to assess the environmental and economic implications of lithium-ion and lead-acid battery recycling programs. This method integrates existing empirical findings from the published literature (PL), policy reports (PRs), municipal records (MRs), and technical assessments (TRs) to generate comprehensive evidence-based insights. The structured systematic review and meta-analysis identified and synthesised 24 relevant sources, including 16 published literature sources with low risk of bias, 5 government or municipal policy reports with moderate risk, and 3 technical assessments with moderate–high risk, as depicted in

Table A1. Studies that were included in the meta-analysis.

Study No	Publication Type	Reference	Risk-of-Bias Assessment
Study 1	Journal article	[28]	Low
Study 2	Journal article	[29]	Low
Study 3	Journal article	[30]	Low
Study 4	Journal article	[18]	Low
Study 5	Journal article	[14]	Low
Study 6	Journal article	[31]	Low
Study 7	Journal article	[32]	Low
Study 8	Journal article	[17]	Low
Study 9	Journal article	[33]	Low
Study 10	Journal article	[34]	Low
Study 11	Journal article	[35]	Low
Study 12a & b	Journal article	[36]	Low
Study 13	Journal article	[37]	Low
Study 14	Journal article	[38]	Low
Study 15	Journal article	[39]	Low
Study 16	Journal article	[40]	Low
Study 17	Technical report papers	[41]	Moderate–high
Study 18	Technical report papers	[42]	Moderate–high
Study 19	Technical report papers	[43]	Moderate–high
Study 20	Government report	[44]	Moderate
Study 21	Government report	[45]	Moderate
Study 22	Government report	[46]	Moderate
Study 23	Government report	[47]	Moderate
Study 24	Government report	[48]	Moderate

. A structured risk-of-bias assessment was conducted to evaluate the reliability of the included studies. Studies were classified as having low, moderate, or high risk of bias, and the majority of them were low risk. While the inclusion of grey literature expanded the evidence base and reduced potential publication bias, variations in methodological rigor among non-peer-reviewed sources may have introduced additional uncertainty into the findings. Furthermore, the study utilised specific impact indicators global warming potential, reusable recovered materials, and economic benefits, along with their measurement units, descriptions, and data sources, which were used to systematically evaluate the environmental and economic outcomes across the included studies which is depicted in

Table 1. Description of impact indicators used in the study.

Impact Indicator	Measurement Unit	Description/Interpretation	Data Source
Global Warming Potential (GWP)	Kilograms of CO2-equivalent (kg CO2e) per kg of recycled battery material.	Represents the total greenhouse gas emissions avoided through recycling compared to virgin material.	Global Warming Potential (GWP)
Reusable Recovered Materials (RRMs)	Percentage (%) of material recovered that can be reused or reprocessed.	Indicates the efficiency of recycling facilities in extracting valuable metals (lead, lithium, nickel, and cobalt) and reducing landfill waste.	Data obtained from municipal waste reports and industrial recovery statistics.
Economic Benefits (EBs)	Benefit–Cost Ratio (BCR)	Includes revenues from recovered materials, cost savings from landfill diversion, and employment creation in the recycling sector.	Derived from financial records, company sustainability reports, and economic analyses.

.

Table A2. Studies that were included in the systematic review and meta-analysis, with battery type, recycling technology, system boundary and geographical context.

Author(s) & Year	Battery Type	Recycling Technology	System Boundary	Functional Unit	Geographical Context
[28]	LFP	Hydrometallurgy	Cradle to gate	1 kg of LFP battery material	China
[29]	NMC (111, 532, 622, and 811)	Simulation-based LCA	Cradle to gate	1 kg of NMC cathode material G	Germany (simulated)
[30]	NMC (111, 523, 622, and 811)	Hydrometallurgy	Cradle to gate	1 kg of NMC battery material	China
[18]	LMO, NCM622, and NCA	Cascaded use + recycling	Cradle to gate	1 kWh battery capacity	China
[14]	NMC111, NMC622, NMC811, and ZEBRA (Na-Ni-Cl)	Hydrometallurgy	Cradle to gate	1 kWh battery capacity	Italy (simulated)
[31]	NMC111 and NMC811	Direct + Hydrometallurgy	Cradle to gate	1 kg of NMC battery material	Germany
[32]	NMC (variants)	Hydrometallurgy vs. Pyrometallurgy	Cradle to gate	1 kg of NMC battery material	Australia (simulated)
[17]	LiNi1/3Mn1/3Co1/3O2	Hydrometallurgy	Cradle to gate	1 kg of NMC111 cathode material	Spain
[33]	Various Li-ion	Cell-chemistry-specific LCA	Cradle to gate	1 kWh battery capacity	Germany
[34]	Black mass (Li-ion)	Hydrometallurgy	Cradle to gate	1 kg of black mass	Finland
[35]	Li-ion	Various	Cradle to gate	1 kg of Li-ion battery material	Norway
[36]	Li-ion (NMC and LFP)	Not specified	Cradle to grave	1 kWh battery capacity	France
[37]	Lead-acid	Pyrometallurgy	Cradle to gate	1 kg of lead-acid battery material	Sweden (simulated)
[38]	Lead-acid and Li-ion (various)	Review of multiple technologies	Cradle to gate	Not applicable	China (review)
[39]	Li-ion (traction)	Not specified	Cradle to grave	1 kWh battery capacity	Italy
[40]	Li-ion (black mass)	Hydrometallurgy (simulated)	Cradle to gate	1 kg of black mass	Belgium
[41]	Li-ion	Technology landscape	Gate to gate (economic)	Not applicable	South Africa
[42]	Li-ion	EverBatt model (closed loop)	Cradle to cradle	1 kg Li-ion battery material	USA
[43]	Li-ion	Regional cooperation outlook	Policy review	Not applicable	SADC region (Southern Africa)
[44]	N/A (policy)	N/A	N/A	Not applicable	South Africa
[45]	N/A (policy)	N/A	N/A	Not applicable	South Africa
[46]	N/A (policy)	N/A	N/A	Not applicable	South Africa
[47]	N/A (policy)	N/A	N/A	Not applicable	South Africa
[48]	N/A (policy)	N/A	N/A	Not applicable	South Africa

depicts studies that were included in the systematic review and meta-analysis with the battery type, recycling technology, system boundary and geographical context. Of the 24 studies included in the overall systematic review and meta-analysis, 16 did not meet the meta-analysis inclusion criteria outlined in

Table 2. The inclusion and exclusion criteria.

Criteria	Inclusion	Exclusion
Impact Indicator	Studies that contain LCA with GWP as the environmental impact indicator.	Studies not performed using the LCA method focusing on the GWP impact indicator.
Types of Studies	Independent articles and a mixture of review articles and independent articles that have original data.	Review articles without original data.
Emissions	Studies that measure greenhouse gas emission data.	Studies that do not have greenhouse gas emissions in the recycling stage. Studies that are only on CO2 emissions, not CO2-eq, were excluded.
Language	Must be published in English and publicly available.	Studies not published in English.
Units	Studies that utilised kg CO2-eq/kg.	A study whose functional unit is kg CO2-eq/kWh lacks sufficient information to convert it to kg CO2-eq/kg.
Years	2000–2025	Prior to 2000.

but provided essential information on environmental impacts, thereby contributing to the qualitative synthesis. The remaining eight studies (Studies 1–8) satisfied all criteria and were subsequently included in the meta-analysis.

2.1. Data Source

The literature search encompassed both peer-reviewed and grey literature from 2000 to 2025.

Table A3. Transparent screening table of studies that were included in the meta-analysis.

Database and Search Strategy Recommended Format: Database Name Search String and Any Relevant Filters	Number of Research Items
Springer Nature Link (“battery recycling” OR “battery recovery” OR “battery waste management”) AND (“environmental impact” OR “resource conservation”) AND (“economic impact” OR “cost-benefit”) AND (emissions) AND (LCI)	06
Scopus (Science direct) (“battery recycling” OR “battery recovery” OR “battery waste management”) AND (“environmental impact” OR “resource conservation”) AND (“economic impact” OR “cost-benefit”) AND (emissions) AND (LCI)	44
Google scholar (“environmental impact” OR “resource conservation”) AND (“economic impact” OR “cost-benefit”) AND (“battery recycling” OR “battery recovery” OR “battery waste management”) AND (emissions) AND (LCI) (“extended producer responsibility” OR EPR) AND (“lithium-ion” OR “Li-ion”) AND (“battery recycling” OR “electronic waste”) AND (“policy implementation” OR “regulatory gap” OR “bylaws”)	121
JSTOR environmental and economic impacts of battery recycling emissions and LCI	09
ProQuest (“environmental impact” OR “resource conservation”) AND (“economic impact” OR “cost-benefit”) AND (“battery recycling” OR “battery recovery” OR “battery waste management”) AND (emissions) AND (LCI)	40
ResearchGate (“environmental impact” OR “resource conservation”) AND (“economic impact” OR “cost-benefit”) AND (“battery recycling” OR “battery recovery” OR “battery waste management”) AND (emissions) AND (LCI)	08
MDPI environmental and economic impacts battery recycling	42
MDPI (“environmental impact” OR “resource conservation”) AND (“economic impact” OR “cost-benefit”) AND (“battery recycling” OR “battery recovery” OR “battery waste management”) AND (emissions) AND (LCI)	06
Total number of research items retrieved	292
Removed articles not accessible	04
Duplicates removed	34
Unique items for phase I screening	254
Items removed during phase I screening	159
Removed because title and abstract do not align with what I want	135
Removed because could not retrieve their documents	22
Removed because it is not published in English	02
Items sought for phase II screening	95
Items removed for being inaccessible for phase II	05
Article removed coz it was under review	01
Articles that were inaccessible	03
Book that is irrelevant	01
Items screened during phase II screening (note: sometimes this is different from the items sought because some items may not be located)	90
Items removed during phase II screening	66
Exclusion Reason 1 = Recycling of other WEEEs not including batteries	11
Exclusion Reason 2 = No recycling stage but focusing on emissions of manufacturing and transportation	16
Exclusion Reason 3 = No GWP data	30
Exclusion Reason 4 = Inconsistent functional unit	09
Items eligible for meta-analysis	24

shows a transparent screening table of the studies that were included in the systematic review and meta-analysis. The search was conducted using academic databases such as Springer Nature Link, Scopus, Google Scholar, JSTOR, ProQuest, ResearchGate, MDPI, and institutional/government repositories. The search strings and Boolean operators that were utilized are [(“environmental impact” OR “resource conservation”) AND (“economic impact” OR “cost-benefit”) AND (“battery recycling” OR “battery recovery” OR “battery waste management”) AND (emissions) AND (LCI)], [(“extended producer responsibility” OR EPR) AND (“lithium-ion” OR “Li-ion”) AND (“battery recycling” OR “electronic waste”) AND (“policy implementation” OR “regulatory gap” OR “bylaws”)] and [environmental and economic impacts battery recycling]. For duplicate removal, the Zotero software was used, and it was able to detect duplicates.

The study dwelt on articles, government gazettes and institutional websites. The study further utilised the global warming potential (GWP), reusable recovered materials (RRMs), and economic benefits (EBs) as impact indicators (Table 1). Compared to other impact categories, GWP is more consistently reported in life cycle assessment (LCA) studies, making cross-study comparisons more reliable. Although indicators such as human toxicity and ecotoxicity remain important, GWP is commonly used as the primary environmental indicator due to its global relevance and policy significance.

2.2. Selection Criteria

Specific inclusion and exclusion criteria were employed in this study to ensure that only relevant and comparable research was selected for the meta-analysis. Only research that employed life cycle assessment (LCA) with a focus on global warming potential (GWP) to assess the environmental impacts of lithium-ion and lead-acid battery recycling was included. The study type, data accessibility, emission indicators, publication language, functional units, and publication years were also considered as additional factors. The specific inclusion and exclusion criteria applied in the study selection procedure are listed in Table 2.

2.3. Data Extraction and Coding

A summary of the selected studies evaluating the environmental impacts of battery recycling is presented in

Table 3. A data extraction matrix was developed to record key study parameters and impact metrics.

Study	Battery Type	CO2 Reduction (kg CO2e/kg)	Landfill Diversion (%)	BCR	Jobs Created
s8	LiNi1/3Co1/3Mn1/3O2 cathode from LIB.	~31.94	~100	N/A	N/A
s7	Four NMC variants: NMC 111, NMC 523, NMC 622, and NMC 811.	~5.71	~100	N/A	N/A
s2	NMC (111, 532, 622, and 811)	~−78.9	~100	73.5	N/A
s4	LMO, NCM622 and NCA	~0.11	~100	N/A	N/A
s5	NMC111, NMC622, NMC811 and ZEBRA (Na-Ni-Cl)	~−0.3408	~100	N/A	N/A
s6	NMC111 and NMC 811	~1.55	~100	N/A	N/A
s1	LFP	0	~100	N/A	N/A
s3	LFP and NMC	~−2.7	~100	N/A	N/A

N/A symbolises Not Applicable.

. The table lists important metrics, including CO₂ reduction, landfill diversion rates, benefit–cost ratios (BCRs), and job creation.

Table 4. Unit prices of recovered battery materials ($/kg).

Revenue/Market Value of Recovered Materials ($/kg) Studies	Studies
Recovered Materials	s17	s18
Al	1.9	1.30
Co	34	6.60
Cu	7.7	N/A
Fe	0.1	N/A
Li	8.75	N/A
Ni	15.8	N/A
Mn	2	N/A
Slag	27.5	N/A
Steel	N/A	0.30
Plastics	N/A	0.10
Electrolyte solvents	N/A	0.15
Graphite	N/A	0.28

presents the unit prices of recovered materials from battery recycling, highlighting their economic benefits. The comparison highlights how recycling procedures affect both environmental performance and economic feasibility, demonstrating the variation in results across various battery types and methodologies.

Table A5. Data coding that was inserted into the R software to generate the effect size.

data <- data.frame(         Study = c(“study1”,”study2”,”study3”,”study4”,”study5”,”study6”,”study7”,”study8”),         n.e = c(2, 18, 8, 3, 14, 12, 4, 5),         mean.e = c(1.6, 177.2, −2.7, 0.016, 2.4069, 3, 17.14126333, 41.16),         Sd.e = c(0.2, 18.62552157, 1.39335092, 0.003, 0.00995, 0.135065052, 0.613360979, 17.52706479),         n.c = c(2, 11, 5, 3, 14, 9, 4, 3),         mean.c = c(1.6, 98.3, −5.4, 0.14, 2.0661, 4.55, 22.85424167, 73.1),         sd.c = c(0.2, 78.71123782, 2.713761228, 0.0096, 0.0118, 0.159712067, 0.613347461, 34.8716217)                   # Display just yi and vi         result_table <- data.frame(             Study = data$study,             yi = round(data$yi, 4),             vi = round(data$vi, 4)         )         print(result_table)                   Below is the data coding information that was inserted to generate the forest plot.         FOREST PLOT                   # Set graphical parameters for better display         par(mar = c(5, 4, 4, 2))                   # Create forest plot         forest(re_model,                     slab = data$study,                     main = “Forest Plot: Hydrometallurgy vs. Control”,                     xlab = “Standardized Mean Difference (SMD)”,                     mlab = “Random-Effects Model (REML)”,                     psize = 1,                     header = “Study”,                     cex = 0.9)                   # Add text annotations         text(−16, 9.5, “Favors Control”, pos = 4, cex = 0.8)         text(16, 9.5, “Favors Treatment”, pos = 2, cex = 0.8)

shows the data coding that was inserted into the R software to generate the effect size.

The global warming potential (GWP) reduction values presented in

Table A8. Global warming potential reduction detailed derivation data.

	Studies	GWP (Initial)	GWP (Final)	Units	Reduction (%)	Energy Density (KWh/kg)	GWP Reduction
Li-ion batteries	2	1		kg CO2-eq/kg	30%		0.3
	4	0.092	0.009	kg CO2-eq/kg			0.083
	9	90.4975		kgCO2-eq/KWh		171.64	0.527252
	10	6.22	4.42	kg CO2-eq/kg			1.8
	11	0.545		kg CO2-eq/kg	10%		0.0545
	12a	110.3		kgCO2-eq/KWh	39.70%	207.4	0.211134
Lead-acid batteries	12b	67.7		kgCO2-eq/KWh	−39.70%	40	−0.67192
	13a	11.034		kg CO2-eq/kg	18%		1.98612
	13b	11.034		kg CO2-eq/kg	10%		1.1034
	14	2		kg CO2-eq/kg	45%		0.9

were calculated using a two-step procedure. First, for studies that reported the GWP in units of kg CO₂-eq/kWh, the values were converted to kg CO₂-eq/kg using the energy density of the respective material, as expressed in Equation (1). Subsequently, the GWP reduction was determined by multiplying the converted GWP value by the corresponding reduction percentage, following Equation (2). The application of these equations enabled a consistent and quantitative assessment of the GWP reduction across the diverse studies compiled in Table A8.

The studies that reported the global warming potential (GWP) in units of kg CO₂-eq/kWh were converted to kg CO₂-eq/kg using energy density, as shown in Equation (1).

GWP(kg CO₂-eq/kg) = GWP (kg CO₂/KWh)/energy density (KWh/kg),

(1)

Subsequently, the GWP reduction was computed using Equation.

GWP reduction = GWP (kg CO₂/kg) × reduction percentage (%),

(2)

2.4. Quality Assessment

Quality assessment was based on a modified Supplementary Materials PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) framework, which includes Relevance of objectives, Transparency of methodology, Data completeness and statistical validity, and Reproducibility of findings. Figure 1 depicts a PRISMA flowchart highlighting the number of studies utilised in this study.

Table A4. Quality assessment scores and categories of included studies. NB: 0 = Not reported/deficient; 1 = Partially reported/moderate limitations; 2 = Clearly reported/no major limitations.

Study	Relevance of Objectives (0–2)	Transparency of Methodology (0–2)	Data Completeness & Statistical Validity (0–2)	Reproducibility of Findings (0–2)	Total Score (0–8)	Quality Category
Study 1	2	2	2	2	8	High
Study 2	2	1	2	1	6	Medium
Study 3	2	2	2	2	8	High
Study 4	2	2	1	2	7	High
Study 5	2	2	2	1	7	High
Study 6	2	2	2	2	8	High
Study 7	2	2	2	2	8	High
Study 8	2	2	2	2	8	High
Study 9	1	1	1	1	4	Low
Study 10	2	2	2	2	8	High
Study 11	2	2	2	2	8	High
Study 12a	2	1	2	1	6	Medium
Study 12b	2	1	2	1	6	Medium
Study 13	2	2	2	2	8	High
Study 14	2	2	2	2	8	High
Study 15	2	2	2	2	8	High
Study 16	2	2	2	2	8	High
Study 17	2	2	2	2	8	High
Study 18	2	2	2	2	8	High
Study 19	2	2	2	2	8	High
Study 20	1	1	1	1	4	Low
Study 21	1	1	1	1	4	Low
Study 22	2	1	2	1	6	Medium
Study 23	2	1	2	1	6	Medium
Study 24	2	1	1	1	5	Medium

depicts quality assessment scores and categories of included studies. The quality assessment demonstrates that 75% of included studies (18/24) were of high quality, and the sensitivity analysis confirmed that the results are stable and not driven by poor-quality research. Low-quality studies were excluded from quantitative synthesis, and medium-quality studies were down weighted or used only for support.

3. Results and Discussion

3.1. Descriptive Analysis

By synthesising data from multiple studies, this section provides a comprehensive overview of how lithium-ion and lead-acid battery recycling contributes to environmental sustainability and economic performance. The findings are organised to reflect key impact indicators, including greenhouse gas (GHG) emissions, reusable recovered materials (RRMs), and economic benefits (EBs). The analysis highlights patterns, variations, and relationships across different recycling methods and battery chemistries.

3.1.1. Effect Size Computation

The effect size, meta-analysis model, forest plot, funnel plot, Egger’s test, sensitivity analysis and outlier analysis were computed and generated using the R Software (Version 4.5.1). Hedges’ g was selected as the effect size statistic because it allows for bias correction in small samples.

Table A6. Effect size and variance computation.

Study	n.e	Mean.e	SD.e	n.c	Mean.c	SD.c	yi (ES)	vi
Study 1	2	1.6	0.2	2	1.6	0.2	0	1
Study 2	18	177.2	18.62552	11	98.3	78.71124	1.5297	0.1868
Study 3	8	−2.7	1.393351	5	−5.4	2.713761	1.2693	0.387
Study 4	3	0.016	0.003	3	0.14	0.0096	−13.9114	16.794
Study 5	14	2.4069	0.00995	14	2.0661	0.0118	30.3142	16.5527
Study 6	12	3	0.135065	9	4.55	0.159712	−10.1943	2.6688
Study 7	4	17.14126	0.613361	4	22.85424	0.613348	−8.0907	4.5912
Study 8	5	41.16	17.52706	3	73.1	34.87162	−1.1232	0.6122

represents the computations of effect size and variance. The random effects model was used because it allows for differences in observed effect sizes due to both sampling errors and true variability in battery chemistries, recycling techniques, and geographical locations. Factors that vary from study to study include Li-ion battery chemistries, leaching methods, and sample sizes. These factors lead to variations in the magnitude of the effect and in the estimation of the distribution’s mean. The random-effects model resulted in acquiring an I² (heterogeneity) of 99.44%, which is high due to the utilisation of studies that focused on different Li-ion battery chemistries, namely, the LiNi₁/₃Co₁/₃Mn₁/₃O₂ cathode, LMO, NCM, NCA, NMC 111, 523, 622, and 811, LFP and ZEBRA. Additionally, the studies used different recycling methods, including hydrometallurgy, acid leaching, organic leaching, and inorganic leaching, while others compared hydrometallurgy to direct recycling, pyrometallurgy, or mechanical recycling, each with distinct energy and chemical demands. Third, geographic location influenced outcomes due to regional variations in electricity mixes, environmental regulations, and waste-management infrastructure. Fourth, system boundaries varied, including differences in the inclusion of transportation, disassembly, preprocessing, refining, and avoided virgin material production, as well as divergent comparison scenarios. Given the limited number of studies available for quantitative synthesis, formal meta-regression was not statistically viable; thus, these sources of heterogeneity were assessed qualitatively and should be considered when interpreting the pooled estimates. Therefore, these variations caused high heterogeneity. Despite this variability, the general trend reveals that battery recycling continuously reduces greenhouse gas emissions and improves resource recovery efficiency.

3.1.2. Heterogeneity Analysis

Figure A1 illustrates a forest plot highlighting the overall environmental effect size. The graph shows the weights assigned to studies by sample size, with smaller squares indicating lower weights and larger squares indicating higher weights. The mean values, standard deviations, and sample sizes were used for all outcome measures. Continuous post-study data from unpaired studies were used to calculate an overall effect magnitude. The standardised mean difference was calculated with Hedges’ g correction for small-sample bias, yielding an overall effect size of −0.17 (SE = 87.7483, 95% CI = −9.05 to 8.71). Study 5 had the largest Standardised Mean Difference (SMD), indicating a large effect size. The confidence interval did not overlap with zero, indicating a preference for the hydrometallurgy recycling method. Following established guidelines for meta-analyses with extreme heterogeneity, the focus is on explaining variation rather than reporting a single effect size as definitive, ensuring that conclusions are conditional and context-specific rather than universal claims.

Table 5. The SMD values and the explanatory factors.

Study	SMD [95% CI]	Interpretation
Study 1	0.00 [−1.96, 1.96]	Neutral effect; no measurable change in GWP. Real-time monitoring of emission standards and cumulative impact assessments are recommended for integration into regulatory frameworks.
Study 2	1.53 [0.68, 2.38]	Moderate increase in GWP. The chemical recycling process was more energy-intensive than producing new virgin material.
Study 3	1.27 [0.05, 2.49]	Small-to-moderately high GWP. Hydrometallurgical technology requires more materials (i.e., sodium hydroxide and sulfuric acid) and energies to separate metal components from the NMC cathode.
Study 4	–13.91 [−21.94, −5.88]	There was a very strong decrease in GWP and highly efficient recycling. Large-scale closed-loop recycling system that almost completely avoided the extremely carbon-intensive production of virgin material.
Study 5	30.31 [22.34, 38.29]	Exceptionally high increase in GWP. Venting of HFC during the dismantling process
Study 6	−10.19 [−13.40, −6.99]	Significantly low GWP. Combining production with both recycling routes can achieve the lowest greenhouse gas emissions for closed-loop scenarios
Study 7	−8.09 [−12.29, −3.89]	Substantial reduction in GWP; energy-efficient processes. Integrating renewables further improves outcomes.
Study 8	−1.12 [−2.66, 0.41]	Slight reduction but not statistically significant. The choice of chemicals, energy consumption, and, more importantly, material efficiency emerges as the cornerstones to achieve environmentally sustainable processes.

presents the Standard Mean Deviation values and the explanatory factors, highlighting the reasons for high or low GWP values, as per the studies used. The substantial variability in environmental outcomes, ranging from exceptionally high GWP increases to strong reductions, depends on the technology, energy inputs, and operational characteristics of each study.

Hedges’ g was selected as the effect size statistic because it corrects for small-sample bias (Table 5 shows sample sizes ranging from 2 to 18 per group). The standardized mean difference (SMD) expresses each study’s effect in standard deviation units, making these methodologically heterogeneous LCA indicators comparable. The SMD is appropriate because the underlying construct, the directional and relative magnitude of GWP change due to recycling, is consistent across studies, even if absolute scales differ. High heterogeneity (I² = 99.44%) does not invalidate the SMD; rather, the random-effects model uses the SMD to estimate the distribution of true effects across different battery chemistries and recycling technologies. Precedent exists in environmental meta-analyses where SMDs are used to synthesize LCA outcomes across studies with variable functional units and system boundaries.

3.1.3. Publication Bias

The funnel plot shows potential publication bias, as it is asymmetric and skewed to the left of the vertical line (Figure A2). A tendency of the studies to congregate towards the bottom of the plot reflects that smaller studies are more likely to be published if they have larger-than-average effects, and hence a greater likelihood of yielding statistical significance, whereas if they cluster at the top of the funnel, this indicates publication bias. The extreme spread matches the high I² (99.44%). This shows a gap in the research literature regarding the effectiveness and environmental impacts of recycling methods, as measured by the GWP indicator.

Egger’s test was conducted to assess potential funnel plot asymmetry within the mixed-effects meta-regression model. The test yielded z = 0.5927 and p = 0.5534, indicating no statistically significant evidence of asymmetry. The limit estimate as the standard error approaches zero was b = −3.8845 (95% CI: −19.3718 to 11.6028), with the confidence interval spanning zero, suggesting no clear directional bias. The high p-value (p > 0.05) implies that the null hypothesis of symmetry cannot be rejected, indicating no strong evidence of publication bias or small-study effects. In summary, there is no statistical evidence of publication bias in this meta-analysis. The funnel plot appears widely scattered due to extreme heterogeneity and the presence of two studies with very large effect sizes (Study 4: SMD = −13.91; Study 5: SMD = +30.31), not due to selective reporting. Egger’s test was non-significant (p = 0.5534), and the leave-one-out sensitivity analysis confirmed that no single study disproportionately drives the overall non-significant finding. Readers are cautioned that with only eight studies, funnel plot-based tests have low statistical power; however, the inclusion of both strongly positive and strongly negative effect sizes across studies suggests that publication bias is unlikely to be a major concern.

A leave-one-out sensitivity analysis was conducted to evaluate the robustness of the meta-analytic results by sequentially removing individual studies and recalculating the overall effect (

Table A7. Sensitivity analysis (leave one out).

Estima	te se	Zva	l Pval	Ci	lb Ci.u	b	Q Qp	Tau2 ta	u2 I	2
Study 1	−0.1725	5.295	−0.0326	0.974	−10.5504	10.2055	136.5031	0	190.5231	99.5229
Study 2	−0.398	5.2889	−0.0753	0.94	−10.7641	9.9681	124.8058	0	189.9538	99.2451
Study 3	−0.3593	5.2907	−0.0679	0.9459	−10.7289	10.0103	134.4625	0	190.1154	99.3678
Study 4	1.6332	4.7505	0.3438	0.731	−7.6777	10.9441	124.3803	0	154.4361	99.5
Study 5	−3.7184	2.1752	−1.7095	0.0874	−7.9816	0.5449	82.5453	0	30.0756	97.4843
Study 6	1.2909	4.989	0.2587	0.7958	−8.4873	11.0691	92.8279	0	168.7728	99.5172
Study 7	0.9776	5.1016	0.1916	0.848	−9.0214	10.9766	120.5305	0	176.9977	99.5517
Study 8	−0.0078	5.2946	−0.0015	0.9988	−10.385	10.3695	132.0081	0	190.4393	99.4656

H2; study 1 = 209.5845; study 2 = 132.4667; study 3 = 158.1888; study 4 = 200.0093; study 5 = 39.7503; study 6 = 207.1339; study 7 = 223.0575; study 8 = 187.1330.

). Across all iterations, p-values remained greater than 0.05, indicating that the overall effect size is consistently non-significant and not unduly influenced by any single study. This confirms the stability and reliability of the findings.

In addition, a Normal Q–Q plot was used to assess the assumption of normality of the residuals (Figure A3). The observed data closely aligns with the diagonal reference line, suggesting that the residuals are approximately normally distributed. This supports the validity of the statistical model and indicates that key assumptions underlying the analysis are reasonably satisfied. Consequently, the estimated p-values and confidence intervals are reliable. Although minor deviations from the line are observed at the distribution tails, such patterns are typical in empirical datasets and do not materially affect the overall interpretation or conclusions.

The boxplot analysis shown in Figure A4 on effect sizes (Hedges’ g) revealed a single outlier located above the upper whisker. The box is centred near zero, with the interquartile range extending from approximately −1 to 1, indicating that the majority of studies cluster around a null or slightly negative effect. The presence of a single outlier above the whisker corresponds to Study 5 ([14]; SMD = +30.31) and suggests that one study exhibits an effect size substantially larger than the rest of the distribution. This outlier lies beyond 1.5 times the interquartile range, meeting the conventional statistical definition of an outlier. No other studies were identified as outliers below the lower whisker. The detection of this extreme positive outlier warrants further investigation, as its inclusion may disproportionately influence the pooled effect estimate and contribute substantially to the observed high heterogeneity (I² = 99.44%).

While SMD allows for statistical aggregation across LCA studies with different functional units, it is problematic for GWP because LCA standard deviations often reflect model precision rather than natural variability. This can artificially inflate effect sizes, as seen with SMD values of −13.91 and +30.31. Therefore, absolute mean differences are more appropriate for policy interpretation, and a dual reporting approach is recommended.

Despite being a statistical outlier (SMD +30.31, IQR outlier detection), inclusion is justified for the following reasons:

(a)

Methodological Validity

The study is a peer-reviewed, published LCA following established CML-IA methodology. It does not contain methodological errors; rather, it represents a legitimate system boundary choice (cradle to grave vs. cradle to gate).

(b)

Captures Real-World Variability

The extreme heterogeneity (I² = 99.44%) is a feature, not a bug. It reflects genuine variability in recycling outcomes based on: geographic context, system boundary, recycling technology (hydrometallurgy vs. pyrometallurgy vs. combined), energy density and cathode chemistry.

(c)

Geographic Specificity as a Key Source of Variability

The literature confirms that geographic specificity accounts for significant variability amongst LCA studies on LIB recycling. More environmentally burdening chains (e.g., China) offer higher GWP reductions through recycling, while cleaner chains (e.g., Europe) offer lower or even negative benefits. Accardo et al.’s modelling of a European chain with a combined recycling scheme is a valid representation of this extreme.

(d)

Precedent in Literature

The variability observed in this study is consistent with the broader literature. As documented in a recent review, results reported across literature studies on LIB recycling are highly variable, even reporting pyrometallurgical recycling resulting in a net increase in GWP (SMD positive in their case as well).

(e)

Sensitivity Analysis Confirms Stability

The leave-one-out sensitivity analysis confirmed that no single study, including Study 5, disproportionately drives the overall non-significant pooled estimate. Removal of Study 5 does not materially affect the conclusion (overall effect remains non-significant).

3.2. Environmental Impacts

Global Warming Potential Reduction Average

The global warming potential reductions of Li-ion and lead-acid batteries are depicted in Figure 2. The results were obtained using Table A8, which depicts the global warming potential detailed derivation data. The lead-acid batteries show the greatest median reduction of ~1 kg CO₂eq/kg, indicating the strongest environmental benefits. Li-ion batteries have a slightly lower reduction of ~0.2 kg CO₂eq/kg. Li-ion and lead-acid batteries are preferred over other types due to their longer lifespans, minimal maintenance requirements, and high energy density [15]. Recycling Li-ion batteries requires high temperatures, which results in high electricity consumption to break chemical bonds in cathode materials and separate precious materials from the black mass. Additionally, Li-ion batteries utilise less developed technology, which is still in its infancy, and many existing procedures are not yet optimised for efficiency compared to well-established lead-acid battery recycling techniques, resulting in a larger environmental footprint. These results shows that the Li-ion batteries have a lower GWP reduction. Lead-acid batteries have a high GWP reduction as they have a well-established infrastructure and a highly effective sector with well-understood procedures.

While lead-acid battery recycling is often cited as a circular economy success due to high recovery rates and established collection systems, its environmental assessment must extend beyond material recovery and climate benefits. Secondary lead smelting can generate toxic emissions and pose occupational health risks if pollution control and safety measures are inadequate. A lead paste composed of PbSO₄, PbO, PbO₂, Pb₂O₃, and metallic Pb is highly toxic, with lead compounds constituting approximately 70% of hazardous air pollutants from recycling. Worker exposure to lead is a further concern, as lead affects nearly every organ; the nervous system is particularly vulnerable, with long-term exposure linked to cognitive deficits, behavioural issues, learning disabilities, and reduced IQ [16]. Thus, the sustainability of lead-acid battery recycling requires balancing global benefits such as reduced greenhouse gas emissions and resource conservation against local impacts including human toxicity and environmental contamination.

3.3. Economic Benefits

The costs of recovered reusable materials are shown in Table 4. High-value materials like Co and Ni show the most significant source of economic benefit. For lower-value materials such as Al and Fe, even though they offer minimal economic benefits, they still contribute to overall profitability due to their relatively high recovery rates. Thus, reducing production costs as the cost of producing new batteries can be lowered by employing recycled materials in place of pricey virgin materials. Overall, the data confirm that the recovery of high-value metals, especially cobalt, nickel, and lithium, is key to the economic viability of battery recycling.

Figure 3 shows the impact indicator for reusable recovered materials from the studies. The reviewed studies revealed that the most common materials recovered during battery recycling are Co, Ni, and Li, whereas the least recovered are Al, Cu, and graphite. According to the reviewed studies, Co and Ni showed high recovery rates, exceeding 85%. Li and Mn exhibit moderate recovery rates, ranging from 86 to 85%. Differences in recovery rates across the utilised studies can be linked to recycling methods. For instance, when hydrometallurgical and direct recycling are implemented, it is more favourable to extract Ni and Co [17,18]. Hydrometallurgical and direct processes demonstrate superior recovery due to improved leaching selectivity and gravity separation, followed by mild chemical processes, whereas pyrometallurgical methods tend to lose recoverable materials during smelting. The high recovery effectiveness of recovered materials is significant for economic performance, as these metals are the most valuable and resource-intensive components. Collectively, the results confirm that recycling technologies are effective in reclaiming valuable metals, elevating the economy and advancing the goals of circular material use and reducing environmental burden. Figure A5 shows a mean plot of the recovery rate and its sample sizes and uncertainty ranges. However, high recovery efficiencies are desirable for resource conservation but do not guarantee better environmental performance, as they may involve high energy use or intensive chemical treatments that increase emissions and other impacts. Thus, recovery efficiency must be assessed alongside life-cycle indicators such as energy consumption, global warming potential, toxicity, and resource use for a comprehensive sustainability evaluation.

Table A9. Integrated economic assessment.

Dimension	Lead-Acid	Lithium-Ion (Hydrometallurgy)	Lithium-Ion (Direct Recycling)
Revenue (risk adjusted, USD/kg black mass)	0.40–0.70	3.50–5.50	4.00–6.00
Processing cost (USD/kg)	0.55–1.20	1.15–2.80	0.75–2.00
Logistics cost (USD/kg, urban)	0.05–0.10	0.05–0.10	0.05–0.10
Logistics cost (USD/kg, remote)	0.50–1.20	0.50–1.20	0.50–1.20
Risk premium (cobalt volatility, CV = 0.45)	N/A (no Co)	−35% to −45% of Co revenue	−35% to −45% of Co revenue
Net without subsidies (urban, avg. prices)	+0.10 to +0.30	−0.50 to +1.50	+0.50 to +2.50
Net without subsidies (remote, avg. prices)	−0.50 to −0.10	−1.50 to −0.50	−1.00 to 0.00
Subsidy required for viability (USD/kg)	0.05–0.10	0.20–0.50	0.10–0.30
Primary risk	Lead price (stable)	Cobalt price volatility	Cobalt price volatility
Viable without policy?	Marginally yes (urban only)	No (requires subsidies or Co > 25 USD/kg)	No (requires subsidies or Co > 20 USD/kg)

Note: N/A = not applicable; USD = United States dollar; kg = kilogram; CV = coefficient of variation; Co = cobalt; avg. = average.

shows the economic viability of battery recycling, which is determined by the interplay of four factors: recovered material prices, processing costs, logistics (collection density), and market volatility, with policy subsidies acting as essential enablers. Under risk-adjusted prices that account for cobalt volatility (CV = 0.45) and lithium volatility (CV = 0.70), lead-acid recycling achieves marginal profitability (0.10–0.30 USD/kg net) in urban areas but becomes unprofitable in remote regions (−0.50 to −0.10 USD/kg). Lithium-ion recycling shows a wide range: from −0.50 to +1.50 USD/kg net without subsidies, depending on processing costs (1.15–2.80 USD/kg) and logistics (0.05–1.20 USD/kg). Cobalt price volatility is the dominant financial risk, not lithium volatility as is commonly assumed. Without policy subsidies of 0.20–0.50 USD/kg, most Li-ion recycling configurations are unprofitable; with subsidies, they become marginally viable. Therefore, the economic conclusion is not that recycling is ‘profitable’ or ‘unprofitable’ in absolute terms, but that its viability is highly conditional on location (collection density), technology (processing cost), market conditions and policy support (subsidies, EPR, and price floors).

Table 6. Critical comparison between the pyrometallurgical, hydrometallurgical, direct recycling and emerging low-impact recycling technologies (ref. [19,20,21,22,23,24,25,26]).

Characteristic	Pyrometallurgy	Hydrometallurgy	Direct Recycling	Emerging Low-Impact Technologies
Technology Readiness Level (TRL)	TRL 9 (commercial and mature)	TRL 7–8 (demo to commercial)	TRL 4–5 (pilot scale)	TRL 3–4 (laboratory to pilot)
Operating Temperature	1200–1500 °C (smelting)	50–200 °C (leaching + drying)	100–400 °C (mild thermal)	25–60 °C (ambient or low heat)
Energy Demand (kWh/kg battery)	5–10	2–5	0.5–1.5	<1 (projected)
Chemical Inputs	Low (fluxes, coke, and reductants)	High (H2SO4, HCl, H2O2, NaOH, and solvents)	Low to moderate (solvents for separation and relithiation agents)	Very low (biogenic lixiviants, deep eutectic solvents, and ionic liquids)
Water Consumption	Low (primarily cooling)	High (leaching, washing, and rinsing)	Low to moderate	Very low (closed-loop potential)
Waste Generation	Slag and off-gas (dioxins, SOx, and heavy metals)	Wastewater, spent acids, and secondary precipitates	Minimal (solid residues only)	Minimal (biodegradable solvents and recyclable DES)
Recovery Efficiency (Co and Ni)	60–80% (Co lost in slag)	85–98%	90–95% (cathode restoration)	70–85% (improving rapidly)
Recovery Efficiency (Li)	<10% (lost to slag/off-gas)	75–90%	85–95% (preserves cathode structure)	60–85%
Recovery Efficiency (Al, Cu, and graphite)	Low (metals oxidize or slag)	Moderate to high (80–95%)	High (physical separation preserves foils)	Moderate (under development)
GWP Range (kg CO2e/kg battery)	+1 to +5 (often net positive)	−3 to +3 (highly variable)	−5 to −1 (consistently beneficial)	−6 to −2 (projected)
GWP Sensitivity to Grid Mix	Very high (coal = disaster; hydro = acceptable)	High (2–5 kWh/kg × grid intensity)	Low (low energy demand)	Very low (minimal energy input)
Fugitive Emission Risk	Moderate (off-gas and slag dust)	Low (if closed loop)	Very low	Very low
Metal Purity of Output	Mixed alloy (requires further refining)	High-purity salts or metals (battery-grade possible)	Direct cathode material (ready for reuse)	High purity (target specific)
Cathode Chemistry Flexibility	High (can process mixed streams)	Moderate (prefers sorted NMC; LFP challenging)	Low (cathode-specific; needs single chemistry)	Moderate (emerging specificity)
Capital Cost	High (furnaces and off-gas treatment)	Moderate to high (reactors and separation trains)	Low to moderate (crushing, sorting, and mild reactors)	Low (projected and simple equipment)
Operating Cost	Moderate (energy dominates)	Moderate to high (chemicals dominate)	Low (energy + labour)	Very low (projected and no chemicals)
Scalability	High (large continuous furnaces)	Moderate (batch or semi-continuous)	Low to moderate (needs sorted streams)	Low (currently lab scale)
Commercially Operating Facilities	Many (Umicore, Glencore, and Sumitomo)	Several (Li-Cycle, Redwood, and Retriev)	Few (OnTo Technology and Battery Resources)	None (pilot/demo only)
Main Advantages	Simple and robust; accepts mixed feedstocks	High recovery of valuable metals (Co, Ni, and Li)	Lowest GWP, preserves cathode structure, and highest material efficiency	Lowest environmental footprint and no toxic chemicals
Main Limitations	High energy, low Co/Li recovery, and air pollution	Chemical-intensive, wastewater treatment, and variable GWP	Requires sorted, single-chemistry streams; not yet scaled	Slow kinetics, low TRL, and unknown scaling behaviour
Best Suited For	Bulk processing of low-value or mixed battery waste	High-value metal recovery from sorted NMC black mass	Single-chemistry streams (e.g., LFP and NMC from OEMs)	Future low-carbon, low-chemical recycling in green grids
GWP Performance (relative)	Poor to moderate (often net emitter)	Poor to excellent (depends on grid and chemicals)	Good to excellent (consistently beneficial)	Excellent (projected and needs validation)
Policy Recommendation	Phase out or require carbon capture	Subsidize only when powered by low-carbon grid + closed-loop chemicals	Prioritize R&D funding and demonstration plants	Fund basic research and pilot-scale trials

shows a critical comparison between the pyrometallurgical, hydrometallurgical, direct recycling and emerging low-impact recycling technologies. While pyrometallurgical and hydrometallurgical processes currently dominate industrial battery recycling, emerging direct recycling technologies are gaining attention due to their potential environmental and economic advantages. Direct recycling seeks to preserve and regenerate cathode materials rather than reducing them to individual constituent metals, thereby retaining greater material value and reducing processing requirements. Recent studies suggest that cathode regeneration and closed-loop recycling systems may lower energy consumption, greenhouse gas emissions, and chemical usage while supporting circular battery supply chains. Although these technologies remain at relatively early stages of commercial deployment, they may substantially influence the future sustainability and economic viability of battery recycling.

3.4. Policy and Regulatory Framework

The South African waste management framework is anchored by the National Environmental Management: Waste Act 59 (NEMWA), 2008, and the National Waste Management Strategy (NWMS), 2020, which establishes three strategic pillars, which include waste minimisation; effective and sustainable waste services; and compliance, enforcement and awareness. This targets the diverting of 40% of waste from landfills within five years. The NWMS 2020 explicitly prioritises Waste Electrical and Electronic Equipment (WEEE), under which Li-ion and lead-acid batteries fall, noting that recycling rates for WEEE were less than 10% in 2017 despite showing the fastest growth in volume.

The Extended Producer Responsibility (EPR) Regulations, 2020, mandate producers to take financial and physical responsibility for products at the post-consumer stage, requiring integration of informal waste collectors into the value chain, cooperation with municipalities, and penalties of up to 15 years of imprisonment for non-compliance. Despite these progressive national policies, this study’s meta-analysis reveals a significant lack of alignment between national regulations and local waste management bylaws, with confusion among producers regarding their EPR responsibilities. The Atmospheric Emission Licence Manual provides guidance on licensing listed activities but does not specifically address battery recycling facilities, while the Municipal Waste Separation at Source Guideline focuses on household recyclables but excludes battery collection.

The results underscore the fragmented and underdeveloped nature of battery recycling systems. While lead-acid battery recycling benefits from established infrastructure and successful EPR measures, lithium-ion batteries, increasingly used in electronics and renewable energy storage, remain largely unregulated or unsupported in practice. Specific EPR schemes for Li-ion batteries have not yet been fully implemented, despite the NWMS 2020 identifying WEEE as a priority waste stream with the fastest percentage growth in volume.

The environmental risks associated with informal recycling or improper disposal are significant and under-documented, particularly in densely populated urban settlements. Economically, Li-ion battery recycling is not cost-competitive without policy-driven support mechanisms due to high processing costs and limited market demand for recovered materials, which is a reality acknowledged by the NWMS 2020’s call for appropriate Economic Instruments such as Advanced Recycling Fees.

In summary, battery recycling in South Africa is characterised by structural, regulatory, and informational deficiencies. Despite supportive national policies, including NEMWA, NWMS 2020, EPR Regulations, and various guidelines, implementation at the local level remains limited. Critical gaps include a lack of formal recycling infrastructure for Li-ion batteries, inadequate public awareness campaigns, insufficient local environmental monitoring, weak economic incentives, and weak policy enforcement mechanisms. The regulatory framework exists on paper, but without enhanced capacity, political will, and dedicated resources at the municipal level, the transition to a circular economy for batteries will remain aspirational.

The environmental benefit of battery recycling does not automatically translate into economic viability, nor does economic viability guarantee environmental performance. Lead-acid recycling achieves both, a consistent GWP reduction (~1 kg CO₂e/kg) and marginal profitability ($0.10–0.30/kg net), because mature infrastructure and stable lead prices align environmental and economic goals. Lithium-ion recycling, however, shows a disconnect: some configurations achieve strong GWP reductions (direct recycling: from −5 to −10 kg CO₂e/kg) but are not yet commercial (TRL 4–5), while commercial-ready methods (pyrometallurgy) often have net positive emissions (from +1 to +5 kg CO₂e/kg) when powered by coal. This creates a policy dilemma: subsidizing current commercial technologies may lock in high-emission pathways, while funding emerging low-GWP technologies delays immediate waste reduction. Regulatory implementation must therefore solve a trilemma: ensuring environmental integrity (through landfill bans and emission standards), economic feasibility (through subsidies, price floors, and EPR fees), and technology neutrality (avoiding preferential support for high-emission routes). The South African case illustrates the gap: national EPR regulations exist (policy), but local lithium-ion collection infrastructure is absent (implementation), meaning batteries still go to landfills despite theoretical economic and environmental benefits. Thus, this paper’s core argument is that environmental gains, economic feasibility, and policy enforcement are not sequential steps but simultaneous requirements, and failure in any one dimension collapses the recycling system.

4. Conclusions

This study evaluated the environmental and economic impacts of lithium-ion and lead-acid battery recycling programs using a systematic review and meta-analysis approach. By synthesising findings from multiple studies, the research provided a comprehensive assessment of how different recycling technologies, battery chemistries, and policy environments influence sustainability outcomes. The results indicate that battery recycling presents both environmental challenges and benefits. Certain recycling processes, particularly pyrometallurgical and hydrometallurgical methods, are associated with high energy consumption, secondary pollution, and incomplete closed-loop systems. Lead-acid batteries contribute to lead and other metal-hazardous air pollutant (HAP) emissions generated during the smelting process [15].

Despite the acknowledged shortcomings of recycling processes, battery recycling offers important environmental benefits, including reducing landfill waste, lowering greenhouse gas emissions and energy consumption by decreasing the demand for virgin raw material mining, and mitigating fire hazards from improperly discarded Li-ion batteries. However, the meta-analysis revealed a non-significant pooled effect, indicating that no universal benefit exists across all contexts. The extreme heterogeneity (I² = 99.44%), driven by differences in battery chemistry, recycling technology, grid carbon intensity, and fugitive emissions, means that while some recycling systems achieve substantial global warming potential (GWP) reductions (e.g., lead-acid recycling: ~1 kg CO₂e/kg; direct recycling: SMD = −13.91), others are net emitters (e.g., coal-powered pyrometallurgy with hydrofluorocarbon venting: SMD = +30.31). Thus, recycling can lower emissions only under specific conditions, not universally.

From an economic perspective, battery recycling is beneficial to the economy as the recovery of valuable materials such as Pb, Li, Ni, Co, Mn and Al lowers production costs because using recycled materials can reduce the cost of manufacturing new batteries, as it substitutes for expensive virgin materials and lessens the dependence on extracting new raw materials, which can be costly and subject to supply shortages. Additionally, the recovery of valuable metals from batteries generates revenue, with profitability influenced by metal prices and recycling volumes. Regarding the costs associated with these reusables recovered materials, Co, Ni and Li are the leading metals in profiting the economy as they range from 8.75 to 34 $/kg. The study also highlights significant structural and regulatory challenges. These include limited infrastructure for lithium-ion battery recycling, weak policy implementation, insufficient public awareness, and a lack of alignment between national regulations and local practices. While lead-acid battery recycling systems are relatively well-established, lithium-ion recycling remains underdeveloped and requires targeted investment and policy support. This study has revealed that while the industry faces diverse environmental and economic challenges across battery types, recycling methods, and existing policies and regulations, battery recycling plays a pivotal role in resource recovery, pollution control, and the circular economy. Therefore, battery recycling may be encouraged in many cases, though the apparent advantages should be interpreted with care given the substantial uncertainty arising from extreme heterogeneity and the limited dataset.

Limitations: Due to the systematic review and meta-analysis nature of the study, the sample size of the existing literature was limited. The sample size was made even smaller because the existing literature that covers all the following factors were limited. Life cycle assessment with global warming potential was used as the environmental impact indicator. Similarly, this research was based on studies that are independent articles and a mixture of review articles and independent articles that have original data, studies that measure greenhouse gas emission data, and studies that contain enough information to calculate an effect size. Additionally, this research was limited to studies that were published in English and publicly available, studies that utilised kg CO₂-eq/kg, and studies that were published between 2000 and 2025. Therefore, these limitations kept us from drawing robust conclusions in accordance with our results.

Recommendations: Addressing these gaps requires a multi-pronged approach with context-sensitive fiscal and regulatory instruments. Subsidy ranges should be tiered by regional context: $0.20–$0.40/kg in metro areas with established EPR collection networks, $0.40–$0.70/kg in secondary cities where transport costs are higher, and $0.70–$1.20/kg in rural areas where logistic infrastructure is absent and recycling plants operate at <10% capacity. For LFP chemistries, which lack high-value Co and Ni, subsidies of $1.00–$1.50/kg are necessary to prevent landfill disposal. EPR fees should be adjusted along a 3–5-year pathway, metropolitan areas should maintain baseline fees with collection density bonuses, secondary cities should implement a 15–25% fee uplift to fund regional collection hubs, and rural districts should require a 30–50% uplift to subsidise transport to centralised processing facilities. Technology-specific allocation should prioritise subsidies for direct recycling ($0.50–$1.00/kg) and hydrometallurgy when achieving >85% Co/Ni recovery and powered by low-carbon grids while phasing out pyrometallurgical subsidies by 2030 due to its net positive GWP emissions. These differentiated approaches should be complemented by strengthened enforcement of EPR regulations by governmental or municipal authorities, the development of public–private partnerships to support innovation and scale, enhanced public education and incentivised collection schemes, and improved data collection and environmental monitoring at the municipal level. By adopting these measures, the battery recycling industry can move towards a more circular, environmentally sustainable, and economically viable system.