Life-cycle environmental impacts of typical plasticizers for plastics and their sustainable transformation potential

Yijun Li† ^a, Kai Zhao†^a, Chi Wang^a, Hang Fu^b and Shanying Hu*^a
aDepartment of Chemical Engineering, Tsinghua University, Beijing 100084, P. R. China. E-mail: hxr-dce@tsinghua.edu.cn
^bSchool of Resources and Environment, Nanchang University, Nanchang, 330031, P. R. China

First published on 24th July 2025

---

---

Green foundation 1. This study systematically evaluates the environmental impacts of 12 typical plasticizers during the production stage by constructing a unified “equivalent performance” life cycle functional unit. It advances the standardized application of green chemistry evaluation methods in additives and provides scientific data support for the sustainable transformation of industry. 2. Future scenario modeling based on learning curves shows that new plasticizers have a reduction potential of up to 12% towards key indicators such as carbon footprint and toxicity. A multidimensional comprehensive assessment reveals that di-2-ethylhexyl terephthalate performs well across environmental, health, and economic aspects, demonstrating outstanding potential for sustainable substitution. 3. This analytical framework and methodology will be extended to the study of other types of additives in the future to promote broader green transformation assessments.

1. Introduction

Plastics are composed of primary polymers and various additives, making them one of the most important materials in modern industries.^1–3 With the continuous expansion of plastic production and consumption,^4,5 numerous regulations and policy tools have been introduced internationally to regulate their use.^6–8 Plasticizers are the most widely produced additives in the plastics industry, primarily used in thermoplastic polymers such as polyvinyl chloride. By lowering the glass transition temperature (T_g), they substantially improve material flexibility and processability, serving critical functions in the construction, automotive, medical, consumer goods, and food packaging sectors.^9–12 For example, unplasticized PVC typically remains rigid with a T_g of around 80 °C, limiting its applicability. The addition of 30 parts of dibutyl phthalate can reduce the T_g to below 0 °C, making the material soft and bendable for complex product requirements.¹³ Due to this key role in tuning material properties, plasticizers are indispensable for the development of functional plastic products. In 2020, the global market for plasticizers reached USD 19745 million, accounting for 33% of the total additive market.¹⁴

However, existing studies using life-cycle assessment (LCA)^15–17 have quantitatively identified that the environmental impacts of plasticizers during the production stage are significant and cannot be ignored. (For instance, the contribution of diethylhexyl phthalate to carbon emissions in PVC films reached 7.14%.)¹⁸ Therefore, reducing the life cycle environmental burden of plasticizers is a key factor in the sustainable transformation of the plastics industry.

Currently, the plasticizer market is primarily dominated by traditional plasticizers such as diethylhexyl phthalate and dibutyl phthalate (accounting for 45% of the total plasticizer production in China),¹⁹ which benefit from mature processing technologies, abundant raw materials, and relatively low prices. However, the emergence of new additives (such as acetyl tributyl citrate and epoxidized soya bean oil) is breaking the traditional pattern and providing new options for plasticizer use. Notably, Fig. 1(b) illustrates that among the 25 LCA studies focused on the production stage of plasticizers, 80% focus on individual plasticizers in bio-based plastics (e.g., glycerol and polyethylene glycol).^20–39 Only five studies focus on the unitary production processes (per ton) of typical plasticizers (mainly diethylhexyl phthalate and phthalic anhydride) used in petroleum-based general plastics,^18,40–43 highlighting the lack of comprehensive LCA analysis for new plasticizers. Moreover, each of these five studies investigates no more than five plasticizers (Fig. 1(b)). These findings highlight a critical research gap. Despite the large production volume, wide application, and substantial environmental impacts of plasticizers, most existing LCA studies provide limited coverage of petroleum-based plasticizers and insufficient assessment of newer alternatives. Additionally, the strategy of substituting plastic additives under equivalent functional requirements has been proven to be the most effective mitigation approach for reducing the environmental impact of plastic additives during the production stage.¹⁸ However, the replacement of different types of plasticizers is not functionally equivalent. Fig. 1(b) shows that none of the existing studies has considered the differences in the amounts of additives required for achieving equivalent material performance (e.g., the amount of long-chain chlorinated paraffins required to achieve the same plasticizing effect is 1.6 times that of dibutyl phthalate).¹⁰ Furthermore, to explore the substitution potential of new plasticizers, it is necessary to consider the evolution of production processes, optimization of raw material paths, and progress in technological maturity, evaluating the potential of both traditional and new plasticizers in future green transformations. However, existing research remains at the static assessment stage, with insufficient forecasting of future emission reduction potential. Therefore, it is necessary to adopt prospective analysis to promote the updating of relevant scientific information, facilitate pollution control and green transformation, and provide support for policy formulation and resource allocation.

This study selects 12 widely used plasticizers based on their prevalence and typicality (covering multiple categories, accounting for over 95% of total production).¹⁹ It systematically collects key environmental emission data for these plasticizers during the production stage. Based on this, an LCA model is developed, and functional units are defined according to equivalent performance to quantify the environmental impacts of different plasticizers, with the results being analyzed from multiple perspectives. Additionally, future scenario modeling based on learning curves is conducted to identify the potential for resource and energy efficiency improvements and environmental emission reductions as these plasticizers evolve. The aim of this study is to reveal the environmental performance differences of various types of plasticizers and their key influencing factors, providing scientific information and data support for the green design of plastic-related chemicals and plasticizer substitution strategies, contributing to plastic pollution control and sustainable management.

2. Methodology

Fig. 1 illustrates the technical approach of this study, providing a detailed description of the equivalent performance LCA framework for plasticizers and the specific analytical methods.

2.1 Scope of research subjects

This study selected 12 typical plasticizers as the main research focus, which account for over 95% of the total plasticizer production in China, providing representative samples. Their detailed information (including their name, molecular formulas and Chemical Abstracts Service registry numbers) is provided in Table 1.

---

2.2 Description of the LCA method

2.3 Typology-based analysis

To further explore the impact of plasticizer properties on environmental performance differences, this study performs binary grouping of the samples from four dimensions to identify system correlations between categories. Specifically, these dimensions include (i) chemical structure (whether it contains an aromatic ring), (ii) raw material source (petroleum-based vs. bio-based), (iii) market introduction order (traditional vs. new), and (iv) functional attributes (whether it is a general-purpose plasticizer). Meanwhile, to enable horizontal comparison between different environmental indicators, the average values of all environmental impact results are standardized to a relative scale ranging from 1 to 10. For specific classification methods, please refer to Table S29 of the SI.

2.4 Prospective analysis

This study introduces a learning curve-based prospective analysis method to evaluate the potential for gradual reduction of environmental impacts throughout the life cycle of typical plasticizers, driven by technological advancements and improvements in resource efficiency. The learning curve initially aimed to describe the pattern where unit costs decrease progressively with cumulative production output in industrial manufacturing.^52–55 As the method developed, its application expanded from early manufacturing management to environmental system modeling. In recent years, researchers such as Bergesen and Suh have introduced the learning curve method into the life-cycle assessment framework to describe the dynamic trends of unit raw material and energy consumption as production scales expand in future scenarios.⁵⁶ The basic assumption of the study is that, with the continuous increase in cumulative production of a particular plasticizer, technological maturity, process optimization, and system integration will lead to a gradual decrease in the unit input of raw materials or energy.

The basic formula for modeling the input of the LCA process using the power law form of the experience curve is as follows:

where n_i,j,t1 and n_i,j,t2 represent the unit input amount of raw materials or energy i required for the production of 1 ton of plasticizer j at time t₁ and t₂, respectively. X_j,t1 and X_j,t2 represent the cumulative production of plasticizer j from the reference time to t₁ and t₂ (In this study, the reference time is set to 2015, with t₁ as 2020 and t₂ as 2030, and specific production data can be found in SI Table S30.) b_i,j represents the learning index, which is related to the learning rate (LR) through the following conversion:

b = −log2(1 − LR).

To prevent the input from decreasing infinitely due to the learning curve, which would violate mass conservation or physical limits, this study introduces a minimum input threshold, , for constraint and adjustment:

Due to the difficulty in obtaining detailed historical data, this study follows the “learn-by-component” hypothesis proposed by Thomassen, combining the empirical learning rate ranges for raw materials (4%–7%) summarized by them.⁵⁷ A classification-based setting strategy is adopted to reflect the differences in technical maturity and learning potential across different input types. Considering that new plasticizer products (such as ATBC) are still in the optimization and scaling-up phase within their industrial chains, and their upstream raw materials have not yet reached the resource utilization efficiency typical of mature products, a relatively higher learning rate is set for these inputs. In contrast, traditional plasticizers (such as DOP and DBP) have long-established production paths and stable experience with raw material usage, and thus a lower learning rate is set to reflect their limited improvement potential. Additionally, if the input value calculated by the learning curve falls below the theoretical limit, the learning rate is appropriately adjusted downward to ensure the physical plausibility of the simulation results.

For background energy inputs such as electricity and steam, this study follows the approach of Bergesen and Suh in prospective life cycle modeling,⁵⁶ using a uniform learning index (b) of 0.15, corresponding to a learning rate of approximately 9.875%. This reflects the potential resource efficiency improvements achievable by emerging technologies such as combined heat and power (CHP) steam, wind power, and others in the future (specific input learning rates can be found in SI Table S31). By introducing such parameter settings, the prospective life-cycle assessment can reflect the evolution of input structures driven by technological learning. This ensures that the environmental impact results are time-dynamic and sensitive to future development trends.

To interpret the extent of environmental performance improvement under future scenarios, we define the concept of Environmental Impact Mitigation Potential as the relative reduction in each environmental impact indicator compared to the baseline scenario. For each plasticizer, the mitigation value of each indicator is calculated as

where I^baseline_i and I^scenario_i denote the values of the i-th environmental indicator under the baseline and future scenarios, respectively. This approach enables consistent comparison of environmental improvement potential across alternative plasticizers, and the corresponding results are discussed in section 3.3.

3. Results and analysis

3.1 Environmental impact and contribution analysis of plasticizers

Among the 12 plasticizers studied, over 90% of the environmental impact (excluding toxicity indicators) stems from the contribution of raw material inputs (SI Tables S34 to S45). Four phthalate-based plasticizers (DOP, DBP, DPHP, and DINP) show relatively similar performance across six environmental indicators. While they perform well on most indicators, their toxicity is generally high (exceeding 400 CTU in all cases, Fig. 2f). This is primarily due to the significant contribution of the phthalic anhydride process (Fig. 3a—d) and electricity inputs to three toxicity-related midpoint indicators. DOA has lower impacts on resource consumption (Fig. 2c) and toxicity (Fig. 2f) compared to the sample average; however, it still has significant environmental impacts in other areas, with adipic acid and octanol being its core contributors (Fig. 3i). Similarly, bio-based DOS, a fatty acid ester, shows adverse impacts across all environmental indicators except for GWP, particularly in toxicity (869 CTU) and freshwater ecosystem impact (2.4 × 10⁻⁶ species year), which are much higher than other categories. This is mainly due to its high dependence on castor oil and other agricultural-based raw materials, which introduce higher environmental loads from fertilizer inputs and extraction processes (Fig. 3h). Bio-based ATBC performs in stark contrast to DOS, having the highest carbon emissions (6840 kg CO₂, Fig. 2a); however, its overall environmental impact on other indicators is relatively mild, indicating that its citric acid pathway offers certain environmental advantages in non-carbon metrics (Fig. 3j). Another bio-based plasticizer, ESO, exhibits polarized environmental characteristics. It performs well in GWP (1702 kg CO₂-eq), resource consumption (35692 MJ), and toxicity (39 CTU), but shows poor performance in the other three indicators, which is closely related to the heavy use of organic fertilizers in soybean oil production (Fig. 3k). The environmental performance of diphenyl (DOTP) and terephthalate (TOTM) plasticizers is consistent. Although they perform well across all six indicators, the upstream use of octanol and terephthalic acid (terephthalic anhydride) remains key factors contributing to their environmental impact (Fig. 3e and f). Additionally, DOTP shows slightly lower environmental impacts than TOTM across most indicators, suggesting that DOTP has greater potential as a green substitute. ASE exhibits the highest resource consumption (255149 MJ), with its paraffin oil process being the key area requiring optimization (Fig. 3g). LCCPs show the lowest impact levels across all six environmental indicators, which may be attributed to their relatively low reliance on organic raw materials during synthesis, with liquid chlorine production being the main source of environmental load.

3.2 Environmental impact trends of different plasticizer groups

In the structural classification dimension (Fig. 4a), plasticizers containing aromatic rings exhibit higher median toxicity values than non-aromatic plasticizers. This is due to the introduction of more highly toxic intermediates or by-products (such as phthalic anhydride, sulfonates, etc.) during the production stage. In contrast, non-aromatic plasticizers exhibit significantly higher environmental loads in terms of freshwater ecosystem impacts, owing to their reliance on agricultural raw materials such as fatty acids, castor oil, and phosphorus-nitrogen fertilizers during synthesis. These raw materials often come with higher risks of water nutrient load and biological disturbance in upstream production processes.

In the raw material source dimension (Fig. 4b), both the median and mean values indicate that bio-based plasticizers exhibit higher environmental loads on most indicators, except for resource consumption, with a particularly significant difference in the FE indicator. This result suggests that, although bio-based raw materials have advantages in terms of renewability, their raw material utilization rate is lower, and the energy inputs and indirect emissions associated with the agricultural supply chain are more complex (e.g., land-use change, pesticide application, etc.). Additionally, although DOS and ATBC are categorized as bio-based plasticizers, their upstream raw material structures still contain a certain proportion of petroleum-based components, and the environmental loads associated with this portion contribute significantly to the overall life cycle impact.

In the market entry time dimension (Fig. 4c), traditional plasticizers (such as phthalates) show small environmental performance differences across all indicators, demonstrating strong consistency. In contrast, the environmental impact values of new plasticizers exhibit greater variability, reflecting their diverse production pathways and varying levels of process maturity. Furthermore, the toxicity indicators for traditional plasticizers are generally higher than those for new types, as their core structures still rely on aromatic intermediates.

In the functional characteristics dimension (Fig. 4d), general-purpose plasticizers show comparable or even superior environmental performance across all six environmental indicators compared to non-general plasticizers. This indicates that general plasticizers, represented by DOP, DINP, and DOTP, not only have advantages in terms of production costs and process maturity but also that the robustness of their environmental impacts may be a key factor supporting their long-term dominance in the market.

3.3 Environmental impact mitigation potential of plasticizers driven by technological evolution

This study employs a learning curve approach to simulate the dynamic evolution of raw material input structures and predict the changes in the life cycle environmental impacts of various plasticizers by 2030 (Fig. 5). The results indicate that, under the assumed learning rates and key raw material improvement pathways, the reduction in environmental impacts (i.e., the mitigation potential) varies across plasticizers, ranging from 1.9% to 12.3%.

Traditional plasticizers (including phthalate-based and LCCPs) exhibit limited mitigation potential overall, with average reductions across six endpoint indicators remaining below 6%. This is primarily due to the high technological maturity of their production processes and the limited room for energy efficiency improvements in key upstream inputs (such as phthalic anhydride and liquid chlorine), resulting in relatively low learning rates.

In contrast, new plasticizers such as ATBC, ESO, and ASE demonstrate greater mitigation potential, with average reductions typically between 7% and 10% and some indicators exceeding 12%. Among them, ATBC shows the highest average mitigation rate across all impact categories. ASE, however, displays significant variability in its mitigation potential across different indicators (e.g., a 4.8% reduction in land ecosystem impacts versus a 12.3% reduction in toxicity), which can be attributed to the varying contributions of its raw materials to different environmental dimensions.

Additionally, certain plasticizers (such as TOTM, DOS, and DOA) are already approaching the defined material efficiency limits under current assumptions, meaning that their environmental impact levels are nearing the optimal boundary achievable through existing pathways. Further reductions for these plasticizers would require structural innovation in upstream production routes, such as raw material substitution, synthesis route redesign, or fundamental adjustments to process energy systems.

4. Discussion

The substitution of plastic additives is considered a key approach to reducing plastic pollution and promoting sustainability. However, how to scientifically select appropriate alternatives, especially balancing performance requirements with reduced environmental impacts and achieving economic benefits, remains a critical issue to be addressed. This study builds an LCA framework around the production stage life cycle of 12 typical plasticizers, filling key gaps in current life cycle research, including the insufficient environmental data for multiple types of plasticizers and the limited assessment of future emission reduction potential. Unlike existing studies that focus on individual plasticizer types and use product-based functional units, this study adopts a unified functional unit based on achieving the same performance. This approach enhances the comparability of environmental impact results between different plasticizers and identifies their key load sources and structural differences.

On this basis, further grouping analysis and prospective LCA modeling reveal the driving mechanisms of plasticizer environmental performance and their future evolutionary pathways, ultimately providing scientific data support for promoting plasticizer substitution and the sustainable transformation of the industry. Additionally, the methodology used in this study is highly versatile and can be extended to the environmental assessment of other types of plastic resins and products, as well as to the evaluation of other additives such as thermal stabilizers and flame retardants. This provides a methodological reference and data support for life cycle management of plastic-related chemicals and the formulation of environmental policies for green chemical products (Fig. 6).

Given the central role of functional equivalence in this study, it is important to evaluate the applicability and limitations of the substitution factor approach. Previous studies and industrial practices have demonstrated that various plasticizers can be functionally interchangeable in specific polymer systems. For example, in flexible PVC, many plasticizers exhibit adequate compatibility in terms of molecular polarity and interactions with polymer chains and can effectively improve flexibility and processability.¹¹ Additionally, the Handbook of Plasticizers by Wypych provides extensive guidance on the recommended use levels of different plasticizers in various polymer systems, further supporting the feasibility of functional substitution.⁵⁸

Nevertheless, it should be noted that functional equivalence is typically established based on a specific performance criterion, such as achieving a Shore A hardness of 80 at room temperature, as used in this study. The SF approach inherently assumes that plasticizers with different chemical structures can be compared based on equal functional output. However, significant differences may exist in migration behavior, volatility, thermal stability, and compatibility with polymer matrices. Moreover, SF values are derived under standardized conditions, which may not fully reflect performance in actual processing or application environments. Despite these limitations, the substitution factor method provides a practical normalization strategy for life cycle assessment, enabling fair comparisons across plasticizers based on equivalent functionality. This is consistent with the core LCA principle of functional-unit-based comparison and strengthens the robustness and applicability of the results in this study.

In the typology-based analysis section, the sample distribution is relatively balanced across the grouping dimensions of structure, market entry time, and functional attributes. However, an imbalance exists in the raw material source dimension, with a 3:9 ratio between bio-based and fossil-based plasticizers, which may affect the representativeness of inter-group comparisons for certain environmental indicators. To mitigate the influence of sample size differences, boxplots are used in Fig. 4 to visualize the environmental impact characteristics of each group. Compared to approaches that only present means or error bars, boxplots simultaneously display medians, quartiles, and outliers, providing a more comprehensive view of data distribution. This is particularly beneficial under unbalanced sample conditions, as the median serves as a robust statistic that reveals typical trends and minimizes the influence of outliers.

We also recognize that such bias cannot be entirely eliminated, as it stems from the limited availability of life-cycle data for bio-based plasticizers. This limitation reflects the fact that these products are still in the early stages of industrial scaling, with relatively low levels of data transparency and accessibility. Future research could address this issue by collaborating with industry stakeholders to supplement high-quality life-cycle data for a broader range of bio-based plasticizers, thereby improving the representativeness and generalizability of grouping analyses.

This study conducts a qualitative uncertainty analysis of data sources based on the criteria specified in Table S28 in the SI. The uncertainty assessment results for each process in the inventory are presented in Tables S5–S25 in the SI. All plasticizers performed well in terms of sample completeness, temporal representativeness, and geographical representativeness, which is mainly due to the relatively easy availability and strong consistency of the relevant data. In terms of source reliability, most data are sourced from irregularly updated literature or specialized books (such as environmental impact assessment reports). Some key processes, such as carbon dioxide emissions, rely on periodically updated data from authoritative sources, which are of higher quality. However, some data, such as the emissions from the three wastes of DOP and DBP, are based on estimates or expert judgments. In terms of technical representativeness, most data reflect the current level of mainstream companies or equivalent processes, with only a few cases where comparable data were used as substitutes due to missing information. Overall, some phthalate-based plasticizers (DOP, DBP, and DINP) still have room for improvement in certain direct environmental emission data, while the inventory data for other categories already have a solid quality foundation (Fig. 7).

To further assess the impact of database selection on LCA results, this study compares the carbon emission results for plasticizer production under the same model input structure using both localized Chinese background data and the Ecoinvent generic database. The results show that, for most plasticizers, the carbon emissions calculated using localized data are lower than those using the Ecoinvent version, with reductions ranging from 1% to 58%. This may be due to the lack of specific process data for some plasticizer raw materials (e.g., octanol used for DOTP) in the Ecoinvent database. The use of a generic production scenario, such as “chemical production, organic”, which represents the average production of many organic chemicals, has likely led to overestimated carbon emissions. A notable exception is DOS, where carbon emissions are higher (+13%) when using localized data. This is due to more detailed modeling of upstream agricultural inputs (such as fertilization and irrigation) related to castor oil in the Chinese context, indicating that the carbon emissions of bio-based pathways are more sensitive to agricultural inputs and regional modeling differences.

Overall, localized data better reflect the raw material structure and energy system characteristics within China, improving the representativeness of the assessment results for actual production scenarios. However, the relatively limited openness of the CLCD database presents certain limitations in applications requiring high academic transparency, reproducibility, and methodological consistency.⁵⁹ In contrast, although the Ecoinvent database has some issues with process substitution, its complete data system and transparent methodology make it widely applicable in international life cycle research and comparative analysis (Fig. 8).

Sections 3.1–3.3 focus primarily on the environmental impacts of plasticizers during the production phase, but this analysis is not exhaustive. To evaluate the development potential and mutual substitutability of different types of plasticizers from a more systematic and multidimensional perspective, this study selects three key dimensions for quantitative analysis: (i) environmental load during the production phase (represented by GWP), (ii) potential health risks during the use phase (i.e., the possible toxicological effects on human health due to exposure during the use phase, represented by chemical toxicity factor), and (iii) economic viability (represented by market prices). The calculation method for the toxicity factor and the selection criteria for market prices are detailed in the SI.

Phthalate-based plasticizers, as a class of products with mature processes and widespread applications, have relatively low carbon emissions during the production phase and cost advantages in terms of market prices. However, they not only cause residues in typical environmental media but also pose significant ecological and health risks in certain usage scenarios,^60,61 thus necessitating the identification of feasible substitutes.

ATBC, as an emerging bio-based plasticizer, performs well in terms of use safety but has a relatively high carbon footprint due to its high production energy consumption. In addition, due to the incomplete scaling-up of its preparation route, its price remains high, limiting its large-scale market application (other new plasticizers such as TOTM, ASE, DOS, and DOA also face similar price issues). The bio-based plasticizer ESO (epoxidized soybean oil), with a similar positioning, performs well across all three dimensions and shows strong substitution potential. However, as noted earlier, ESO exhibits excessively high values in other environmental impact indicators, so its overall applicability still requires further evaluation in specific application scenarios.

DOTP, in terms of carbon emissions and market prices, is comparable to traditional phthalate-based plasticizers, but its chemical toxicity factor (CTF) is significantly lower. As indicated in section 3.1, its production process also has relatively low environmental impacts, making DOTP a promising alternative that balances environmental, health, and economic considerations. Chlorinated paraffins show low carbon emissions, low toxicity factors, and price advantages in this study, but it should be noted that short-chain chlorinated paraffins (SCCPs) have been included in the Stockholm Convention as persistent organic pollutants (POPs) that are banned from use.⁶² This ban focuses on the high persistence (P), bioaccumulation (B), and toxicity (T) exhibited by these chemicals in the environment, as well as their potential for long-range environmental transport (LRTP). The chemical toxicity factor used in this study is primarily based on publicly available toxicological test data, covering only some of the PBT characteristics but not systematically incorporating environmental behavior parameters (e.g., environmental half-life, bioaccumulation factor, and mobility),⁶³ which may underestimate the high persistence risk of long-chain chlorinated paraffins after use. Given that LCCPs and SCCPs share certain structural and environmental behavior similarities, it is necessary to incorporate more external toxicological and environmental behavior research results for a comprehensive assessment.

5. Conclusion and outlook

In summary, this study systematically evaluates the environmental impacts of typical plasticizers during the production stage, explores their sustainability performance from multiple dimensions, and proposes forward-looking trends and development priorities for their transformation. It ultimately defines a technical strategy for the reasonable substitution of plastic additives based on equivalent functional requirements. However, the study still has certain limitations, such as the reliance on publicly estimated data and empirical rules for learning rates and future production data, introducing some uncertainty. Additionally, the current model only applies dynamic learning curve evolution to major raw material inputs and does not comprehensively cover all input pathways, which may underestimate the mitigation potential of certain environmental indicators, especially for plasticizers that depend on complex additives or multi-stage synthesis routes.

Despite these limitations, valuable insights can still be drawn from the results. Based on the comprehensive results from the three-dimensional analysis and the forward-looking analysis in section 3.3, it can be concluded that although traditional phthalate-based plasticizers currently dominate the market due to their mature industrial systems and cost advantages, their potential for further improvement is limited, and their development is also constrained by potential health risks and regulatory trends. At the same time, emerging categories like DOTP, which have demonstrated outstanding performance, exhibit more balanced advantages in terms of environmental impact, health, and economics, providing a solid foundation to become the next generation of mainstream alternatives. With the optimization of future technological pathways and the expansion of market scale, more new plasticizers are expected to stand out in multi-dimensional sustainability evaluations, forming a more diverse and robust green substitution system.

The future sustainable development of plasticizers can be further advanced in several key areas. First, more precise environmental impact assessments should be conducted during the production stage, with continuous improvement of regional life cycle data and enhanced transparency and representativeness of background databases, enabling accurate identification of high-impact stages and the optimization of cleaner processes. Scenario analysis indicates that new plasticizers still have significant potential for transformative development. By optimizing raw material pathways and energy systems, promoting technological maturity, and scaling up production, both environmental benefits and unit costs can be improved, enhancing market feasibility. Moreover, considering the trade-offs among environmental friendliness, performance, and economics, future efforts should focus on building a sustainable design framework based on a full life cycle perspective and multi-criteria synergy, promoting the green substitution of traditional high-risk products with green plasticizers from the source. Regulatory bodies should also play a guiding role in accelerating the industry's green transformation, while companies should incorporate environmental performance into core decision-making alongside cost-effectiveness. Policy measures should include fiscal support and technological incentives to reduce barriers to the initial promotion of emerging green additives and enhance their accessibility and application in practical plastic systems. Ultimately, through the collaboration of industry, research, and regulation, the green transformation and high-quality development of the plasticizer system can be accelerated.

Author contributions

Conceptualization, Y. L., K. Z.; methodology, Y. L., K. Z.; validation, Y. L., K. Z.; formal analysis, Y. L., K. Z.; investigation, Y. L., K. Z.; resources, S. H.; writing – original draft, Y. L., K. Z., F. H.; writing – review & editing, Y. L., K. Z., C. W., F. H.; writing – revision & organization, Y. L., K. Z., C. W., F. H. Visualization, Y. L., K. Z.; supervision, S. H.

Conflicts of interest

The authors declare no competing financial interest.

Data availability

The data presented in this article (including life cycle inventory data, environmental impact analysis results, contribution analysis data, etc.) are included as part of the SI. The datasets and related analysis results can be accessed through this SI.

The Supplementary Information includes detailed life cycle inventory datasets, environmental impact assessment results, contribution analysis data, and comparative analysis with existing plasticizer studies. The SI also contains additional figures, tables, and methodological descriptions supporting the main text. See DOI: https://doi.org/10.1039/d5gc01960e.

Acknowledgements

We would like to express our sincere gratitude to Xin Wang, a PhD candidate in the Department of Chemical Engineering at Tsinghua University, for her valuable assistance in data collection and processing for this manuscript.

References

1. R. Geyer, J. R. Jambeck and K. L. Law, Sci. Adv., 2017, 3, e1700782 CrossRef PubMed .
2. M. Schiller, in PVC Additives, ed. M. Schiller, Hanser, 2015, pp. I–XVI,  DOI:10.3139/9781569905449.fm .
3. H. Wiesinger, Z. Wang and S. Hellweg, Environ. Sci. Technol., 2021, 55, 9339–9351 CrossRef CAS .
4. A. S. Pottinger, R. Geyer, N. Biyani, C. C. Martinez, N. Nathan, M. R. Morse, C. Liu, S. Hu, M. de Bruyn, C. Boettiger, E. Baker and D. J. McCauley, Science, 2024, 386, 1168–1173 CrossRef CAS .
5. P. Stegmann, V. Daioglou, M. Londo, D. P. van Vuuren and M. Junginger, Nature, 2022, 612, 272–276 CrossRef CAS .
6. European Commission, A European Strategy for Plastics in a Circular Economy, Brussels, 2018 Search PubMed .
7. UNEP, Single-Use Plastics: A Roadmap for Sustainability, United Nations Environment Programme, Nairobi, 2018 Search PubMed .
8. UNEP, Intergovernmental Negotiating Committee to Develop an International Legally Binding Instrument on Plastic Pollution, Including in the Marine Environment, UNEP/PP/INC.1/7, United Nations Environment Programme, Geneva, 2022 Search PubMed .
9. D. F. Cadogan and C. J. Howick, in Kirk–Othmer Encyclopedia of Chemical Technology, 2000,  DOI:10.1002/0471238961.1612011903010415.a01 .
10. E. A. Coleman, in Applied Plastics Engineering Handbook, ed. M. Kutz, William Andrew Publishing, 2nd edn, 2017, pp. 489–500,  DOI:10.1016/B978-0-323-39040-8.00021-3 .
11. M. Rahman and C. S. Brazel, Prog. Polym. Sci., 2004, 29, 1223–1248 CrossRef CAS .
12. K. Zhao, Z. Zhang, Y. Li, H. Fu, X. Peng, R. Tao, Z. Li, W. Wen, X. Xie and S. Hu, Resour., Conserv. Recycl., 2025, 217, 108213 CrossRef CAS .
13. C. E. Wilkes, J. W. Summers and C. A. Daniels, PVC Handbook, Hanser Gardner Publications, Munich, 2005 Search PubMed .
14. S. Rajaram, Global Markets for Plastics Additives, BCC Research, Wellesley, MA, 2020 Search PubMed .
15. G. Finnveden, M. Z. Hauschild, T. Ekvall, J. Guinée, R. Heijungs, S. Hellweg, A. Koehler, D. Pennington and S. Suh, J. Environ. Manage., 2009, 91, 1–21 CrossRef .
16. J. B. Guinée, R. Heijungs, G. Huppes, A. Zamagni, P. Masoni, R. Buonamici, T. Ekvall and T. Rydberg, Environ. Sci. Technol., 2011, 45, 90–96 CrossRef .
17. S. Hellweg and L. Milà i Canals, Science, 2014, 344, 1109–1113 CrossRef CAS .
18. K. Zhao, X. Wang, Z. Zhang, Y. Li, S. Li, E. Tian, X. Xie, H. Fu and S. Hu, Environ. Sci. Technol., 2024, 58, 16386–16398 CrossRef CAS .
19. Plastic Additives, in China Plastic Industry Yearbook, ed. Z. Ma, China Light Industry Press, Beijing, 2021, pp. 256–258. ISBN: 978-7-5184-3764-1 Search PubMed .
20. M. Ayala, Ø. Arlov, K. Nøkling-Eide, M. Sœther, C. Dore, J. Vidal, Q. Zhou, S. Wang, L. Michalak, A. Kyvik, B. Jolain, L. Aubel, S. Strand Jacobsen and M. Pizzol, J. Cleaner Prod., 2024, 444, 141248 CrossRef CAS .
21. M. Ayala, M. Thomsen and M. Pizzol, Algal Res., 2023, 71, 103036 CrossRef .
22. H. Baniasadi, L. Äkräs, Z. Madani, F. Silvenius, M. Fazeli, S. Lipponen, J. Vapaavuori and J. Seppälä, Int. J. Biol. Macromol., 2024, 279, 135173 CrossRef CAS .
23. B. Chalermthai, K. Nootong, B. D. Olsen, S. Assabumrungrat and P. Charoensuppanimit, Environ. Res., 2024, 251, 118622 CrossRef CAS .
24. W. Chen, T. L. Oldfield, P. Cinelli, M. C. Righetti and N. M. Holden, J. Cleaner Prod., 2020, 269, 122366 CrossRef CAS .
25. Y. Deng, W. M. J. Achten, K. Van Acker and J. R. Duflou, Biofuels, Bioprod. Biorefin., 2013, 7, 429–458 CrossRef CAS .
26. A. Etxabide, I. Leceta, S. Cabezudo, P. Guerrero and K. de la Caba, ACS Sustainable Chem. Eng., 2016, 4, 4626–4634 CrossRef CAS .
27. G. Gadaleta, C. Ferrara, S. De Gisi, M. Notarnicola and G. De Feo, Sci. Total Environ., 2023, 871, 161958 CrossRef PubMed .
28. C. Ingrao, C. Tricase, A. Cholewa-Wójcik, A. Kawecka, R. Rana and V. Siracusa, Sci. Total Environ., 2015, 537, 385–398 CrossRef .
29. I. Leceta, A. Etxabide, S. Cabezudo, K. de la Caba and P. Guerrero, J. Cleaner Prod., 2014, 64, 218–227 CrossRef .
30. N. Méité, E. G. Zoro, B. I. H. G. Doubi, A. Sanou, L. P. M.-S. Kouakou, N. Fenzl, L. O. do Canto Lopes and L. K. Konan, Results Chem., 2024, 11, 101853 CrossRef .
31. A. Mohammed, K. Ward, K.-Y. Lee and V. Dupont, Green Chem., 2023, 25, 5501–5516 RSC .
32. J. Müller-Carneiro, M. C. B. D. Figueirêdo, C. Rodrigues, H. M. C. D. Azeredo and F. Freire, Resour., Conserv. Recycl., 2023, 188, 106637 CrossRef .
33. J. Müller-Carneiro, C. Rodrigues, L. C. Dias, C. Henggeler Antunes, A. L. A. Mattos and F. Freire, J. Cleaner Prod., 2023, 427, 139268 CrossRef .
34. A. Riofrio, M. Cornejo and H. Baykara, Int. J. Life Cycle Assess., 2022, 27, 834–848 CrossRef .
35. P. M. Rojas-Bringas, G. E. De-la-Torre and F. G. Torres, Sci. Total Environ., 2021, 769, 144869 CrossRef PubMed .
36. C. Ruini, P. Neri, G. Cavalaglio, V. Coccia, F. Cotana, A. M. Raspolli Galletti, D. Morselli, P. Fabbri, A. M. Ferrari and R. Rosa, ACS Sustainable Chem. Eng., 2023, 11, 12014–12026 CrossRef .
37. K. Sirivechphongkul, N. Chiarasumran, M. Saisriyoot, A. Thanapimmetha, P. Srinophakun, K. Iamsaard and Y.-T. Lin, Catalysts, 2022, 12, 1150 CrossRef .
38. A. Surendren, Y. Hasan, A. K. Mohanty, B. Abbassi and M. Misra, Sustainability, 2024, 16, 7406 CrossRef .
39. C. J. R. Verbeek, D. W. Van der Merwe and J. M. Bier, Sustainability, 2023, 15, 3455 CrossRef CAS .
40. E. Shittu, V. Stojceska, P. Gratton and M. Kolokotroni, Energy Build., 2020, 217, 110007 CrossRef .
41. H. Stripple, R. Westman and D. Holm, J. Cleaner Prod., 2008, 16, 1764–1776 CrossRef CAS .
42. C. Webb, L. Anguilano, G. Troisi and X. Schmidt Rivera, J. Cleaner Prod., 2024, 467, 142903 CrossRef CAS .
43. G. Wernet, S. Hellweg and K. Hungerbühler, Int. J. Life Cycle Assess., 2012, 17, 720–728 CrossRef CAS .
44. C. M. Vanek and M. T. Berard, in Handbook of Vinyl Formulating, 2008, pp. 315–326,  DOI:10.1002/9780470253595.ch11 .
45. ISO, Environmental management – Life cycle assessment – Principles and framework (ISO 14040:2006), International Organization for Standardization, Geneva, Switzerland, 2006 Search PubMed .
46. X. Liu, H. Wang and J. Chen, Chinese Life Cycle Assessment Database (CLCD) (2024), IKE Environmental Technology Co., Ltd, the PRC, 2024 Search PubMed .
47. X. Liu, H. Wang, J. Chen, Q. He, H. Zhang, R. Jiang, X. Chen and P. Hou, J. Environ. Sci., 2010, 30, 2136–2144 CAS .
48. M. A. J. Huijbregts, Z. J. N. Steinmann, P. M. F. Elshout, G. Stam, F. Verones, M. Vieira, M. Zijp, A. Hollander and R. van Zelm, Int. J. Life Cycle Assess., 2017, 22, 138–147 CrossRef .
49. IPCC, Climate Change 2021: The Physical Science Basis, Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 2021 Search PubMed .
50. O. Jolliet, M. Margni, R. Charles, S. Humbert, J. Payet, G. Rebitzer and R. Rosenbaum, Int. J. Life Cycle Assess., 2003, 8, 324–330 CrossRef .
51. R. K. Rosenbaum, T. M. Bachmann, L. S. Gold, M. A. J. Huijbregts, O. Jolliet, R. Juraske, A. Koehler, H. F. Larsen, M. MacLeod, M. Margni, T. E. McKone, J. Payet, M. Schuhmacher, D. van de Meent and M. Z. Hauschild, Int. J. Life Cycle Assess., 2008, 13, 532–546 CrossRef CAS .
52. A. C. Köberle, D. E. H. J. Gernaat and D. P. van Vuuren, Energy, 2015, 89, 739–756 CrossRef .
53. G. Lavidas, Renewable Energy, 2019, 132, 1290–1300 CrossRef .
54. C. Sharma, A. K. Sharma, S. C. Mullick and T. C. Kandpal, Energy Sustainable Dev., 2018, 42, 121–128 CrossRef .
55. L. Stamford and A. Azapagic, Energy Technol., 2018, 6, 1148–1160 CrossRef .
56. J. D. Bergesen and S. Suh, Appl. Energy, 2016, 169, 721–728 CrossRef CAS .
57. G. Thomassen, S. Van Passel and J. Dewulf, Renewable Sustainable Energy Rev., 2020, 130, 109937 CrossRef .
58. V. Y. Senichev and V. V. Tereshatov, in Handbook of Plasticizers, ed. and G. Wypych, ChemTec Publishing, 3rd edn, 2017, pp. 135–164,  DOI:10.1016/B978-1-895198-97-3.50008-1 .
59. J. Guo, R. Li, R. Zhang, J. Qi, N. Li, C. Xu, A. S. F. Chiu, Y. Wang, H. Tanikawa and M. Xu, J. Ind. Ecol., 2025, 29, 766–776 CrossRef .
60. M. Bi, W. Liu, X. Luan, M. Li, M. Liu, W. Liu and Z. Cui, Environ. Sci. Technol., 2021, 55, 13980–13989 CrossRef CAS PubMed .
61. Y. Cui, J. Chen, Z. Wang, J. Wang and D. T. Allen, Environ. Sci. Technol., 2022, 56, 11006–11016 CrossRef CAS .
62. Secretariat of the Stockholm Convention, Stockholm Convention on Persistent Organic Pollutants: Text and Annexes, Revised in 2023, United Nations Environment Programme (UNEP), Geneva, 2023 Search PubMed .
63. H. Wang, Z. Wang, J. Chen and W. Liu, Environ. Sci. Technol., 2022, 56, 6774–6785 CrossRef CAS PubMed .

---

Footnote

† These authors contributed equally to this paper.

This journal is © The Royal Society of Chemistry 2025