A systematic review of environmentally persistent free radical (EPFR) formation, characteristics, and health effects: are there sufficient data for risk assessment?

Abstract

Environmentally persistent free radicals (EPFRs) are stable free radicals formed on particulate matter (PM) through processes such as combustion and pyrolysis. These free radicals are generated on transition metal oxide surfaces in the presence of aromatic precursors. Exposure to EPFRs occurs primarily via inhalation of PM deriving from combustion, traffic, industrial activities, and both indoor and outdoor burning. Other environmental factors that might generate EPFRs are radon, electronic and tobacco cigarettes. EPFRs exhibit unexpectedly long half-lives, ranging from several weeks to, in some cases, several years. EPFRs may be carbon-centered, oxygen-centered or mixed, identified by g-values exhibited in electron paramagnetic resonance analysis. The radicals undergo redox cycling within aqueous solutions and in biological tissues/fluids triggering production of reactive oxygen species (ROS), comprised primarily of hydroxyl, superoxide, and peroxyl radicals. The stability of EPFRs, their association with PM_2.5, and their ability to generate ROS may pose significant concerns for human health. To determine whether there are sufficient data for risk assessment, recent advances were examined in the following important aspects of EPFR research: (1) atmospheric chemistry, (2) human exposures, (3) animal toxicity, and (4) epidemiology. Our review found insufficient epidemiological and exposure studies; however, toxicological data in animals suggested that EPFR inhalation contributes to cardiovascular, respiratory, and metabolic diseases. Although EPFRs are not currently surveyed by a regulatory monitoring system, data indicate their widespread presence in the environment and their potential to initiate/exacerbate diseases.

Graphical Abstract

Introduction

According to the World Health Organization (WHO), ambient air pollution is responsible for approximately 7 million premature deaths worldwide each year (Lelieveld et al. 2015). A key contributor to the adverse health effects associated with exposure to air pollution is combustion-derived particulate matter (PM) (Chauhan et al. 2025; Gatto et al. 2014). Particulate matter comprises a heterogeneous mixture of inhalable, thoracic, and respirable solid particles and liquid aerosols suspended in the atmosphere (Sidwell, Smith, and Roper 2022). Both natural and anthropogenic activities might produce PM, and the most important sources of PM are wildfire, industrial combustion processes, vehicle exhaust, household fuel combustion, cigarette (and e-cigarette) smoking, and the thermal treatment of organic wastes (Barros, Oliveira, and Morais 2023; Cormier et al. 2006; Lewtas 2007; Valavanidis, Fiotakis, and Vlachogianni 2008).

Exposure to PM has been linked to the development of non-communicable diseases such as chronic obstructive pulmonary disease (COPD) (Tsai et al. 2014), asthma (Cheng, Chen, and Chiu 2014), cardiovascular diseases (Hamanaka and Mutlu 2018), vascular disorders (Shkirkova et al. 2020) diabetes (He et al. 2017), neurological diseases (Chauhan et al. 2025; J. Lee et al. 2023), and lung cancer (Hamra et al. 2014). Notably, PM exposure aggravates respiratory diseases such as lower respiratory tract infections and asthma (Horne et al. 2018; Krall et al. 2017; Tsai et al. 2014). Substantial research and evaluation by national and international agencies, including the US EPA, ATSDR, CDC, and WHO, have significantly advanced our understanding of PM-associated adverse health risks, including formation, environmental fate, and health impacts (ATSDR 2024; EPA 2019; Orellano et al. 2024). These assessments provide a strong foundation for ongoing investigations into PM-related toxicity. The precise etiology underlying PM-induced adverse health effects remains unclear but is believed to involve interrelated mechanisms, including oxidative stress, autonomic nervous system imbalance, endothelial dysfunction, and alterations in immune and metabolic pathways (Gangwar et al. 2020; Glencross et al. 2020; Hamanaka and Mutlu 2018; Rajagopalan et al. 2024; Ying et al. 2014).

Combustion processes generate PM that often contains environmentally persistent free radicals (EPFRs). EPFRs are surface-stabilized organic radicals formed during combustion processes when organic precursors, such as substituted benzenes or polycyclic aromatic hydrocarbons (PAHs), chemisorb onto metal oxides or transition metal–containing particulate surfaces (Khachatryan et al. 2014; Vejerano, Lomnicki, and Dellinger 2011). EPFRs were detected in emissions from various sources, including thermal treatment of hazardous waste, open burning of biomass, vehicle exhaust, and cigarette smoke (Khachatryan et al. 2011; Z. Zhao et al. 2024). EPFR concentrations and types differ by particle size. Chen et al. (2020a) reported in Linfen, China, that fine (< 2.1 μm) and coarse particles (2.1–10 μm) contain distinct types of EPFRs, with seasonal differences in concentration patterns where higher levels in coarse PM were found during summer but higher amounts in fine PM during winter. Combustion, especially coal burning, was the major source of EPFRs in winter, while other fuels dominated in summer (Chen et al. 2020a). Measurement and characterization of EPFRs are frequently performed using electron paramagnetic resonance (EPR) spectroscopy, which provides information regarding radical type, concentration, and stability, complemented by Fourier-transform infrared (FTIR) spectroscopy and mass spectrometry (MS) to elucidate surface chemistry and metal-organic interactions (Patterson et al. 2017; M. Xu et al. 2019; X. Zhang et al. 2025).

These long-lived radicals are also found in wildfire soot, another important global source of EPFRs (Sigmund et al. 2021). Worldwide landscape fires and wildfires produce approximately 256 teragrams of charcoal each year, contributing to their redistribution and potentially producing harmful effects on both the natural environment and human health. The lifetime of EPFRs ranges from days to months or even years (Sigmund et al. 2021), and in biological systems, EPFRs become reactive, producing hydroxyl radicals (OH) in a catalytic cycle (Khachatryan et al. 2014), thus promoting oxidative stress. Oxidative stress plays a major role in the pathogenesis of a variety of human diseases including diabetes, atherosclerosis, hypertension, liver diseases, and cancer (Cichoż-Lach 2014; Pitocco et al. 2013; Reddy 2023; Reuter et al. 2010; Shkirkova et al. 2020; Vogiatzi, Tousoulis, and Stefanadis 2009). Further, several investigators demonstrated that oxidative stress plays a key role in promoting adverse health effects associated with PM exposure (Gangwar et al. 2020; Miller 2020; Natarajan et al. 2019).

Research on EPFRs is still in its early stages, and while their associations with adverse health outcomes were observed; however, the underlying cellular, biochemical, and organ-system level mechanisms are not yet well understood. To the best of our knowledge, no systematic review has yet been conducted to comprehensively summarize the formation, characteristics, and health impacts associated with exposure to PM containing EPFRs. It is crucial to determine the health implications of EPFR exposures, because EPFRs are not specifically monitored in the environment, and a reference dose for EPFR exposure has yet to be established. To fill this knowledge gap, advances were comprehensively examined in the following important aspects of EPFR science: (1) atmospheric chemistry, (2) human exposures, (3) animal toxicity, and (4) epidemiology. With this information, this review aimed to test whether sufficient risk assessment data were available to suggest the need for regulatory strategies.

Methodology

This systematic review was conducted by an inter-disciplinary research team as part of a graduate course designed to educate students on professional skills relevant to environmental science and toxicology. In the first phase of the study, 4 sub-themes – atmospheric chemistry, animal toxicology, epidemiology, and exposure science – were identified. Three to four subject-matter experts per subtheme conducted a literature search and screening process following the guidelines set forth by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2021 (Page et al. 2021). The selection of a database by each group was based upon the research discipline or specialty area. More than one database was examined by each team to minimize the chances of missing a publication which might be indexed in one database but absent in another. The process consisted of 4 steps – identification, screening, eligibility, and inclusion – that were applied to all publications that met the search criteria. Each group then determined the principal keywords, along with the screening, eligibility, inclusion, and exclusion criteria, to select relevant journal articles (Figure 1), as detailed in the following sections.

Figure 1.

Stepwise diagram illustrating the records search methods used in this systematic review.

Figure 2 presents a generally accepted mechanism for the formation of environmentally persistent free radicals (EPFRs) from simple precursors like hydro-xyphenol, chlorophenol, and dichlorobenzene. First, the precursor molecule undergoes chemisorption onto the metal oxide (CuO) surface, followed by an electron transfer from the precursor to the metal. The resulting radical is then stabilized by monomerization. The stabilized radical (III) can undergo deprotonation to produce a radical anion (V), which is reduced in the presence of oxygen to peroxides. This is followed by the Fenton reaction, in which the peroxide formed reacts with iron, generating highly reactive hydroxyl radicals, which are at the origin of oxidative stress. Species (VII) is then catalytically reduced by NADPH back to species (II) to complete the EPFR life cycle

Figure 2.

EPFR formation pathways and redox cycling mechanisms.

In addition to the specific inclusion and exclusion criteria used by each group (Figure 1), all groups excluded review articles from their findings and limited the multi-step screening process to journal articles in English. For quality assurance, screening of each record was conducted by 3–4 members of the relevant group using the established criteria.

Atmospheric chemistry

Atmospheric chemistry studies related to the ubiquity of EPFRs (source and concentration), their lifetime, detection, and links to reactive oxygen species (ROS) formation were included for review. Studies lacking robust statistical analysis or sufficient experimental data to corroborate their findings were excluded. The selected studies were grouped based upon EPFR source and concentration, lifetime, and association with ROS generation. The databases used for sourcing included Web of Science^™ as a primary reference, with Google Scholar and PubMed^® as secondary references, employing these keywords: “EPFRs,” “Environmentally Persistent Free Radicals,” “Environmentally Persistent Free Radicals lifetime,” “Environmentally Persistent Free Radicals Particulate Matter,” and “Environmentally Persistent Free Radicals Reactive Oxygen Species.” All available years in each database were included in the search. A study was determined to meet the inclusion criteria if it featured experimental research addressing EPFR formation, lifetime, source analysis and quantification, and links to ROS generation. Inclusion also required sufficient data and comprehensive statistical analysis. Three reviewers initially worked independently to retrieve and screen records and finally compiled all records, collecting both qualitative and quantitative data from the reports.

Outcomes sought by this search included understanding the lifetimes of EPFRs in environmental conditions. The effects measured for each outcome were lifetime data to confirm EPFR persistence, statistical differences in EPFR concentrations in PM collected from differing sampling sites, and experimental data confirming the ability of EPFRs to redox cycle to produce ROS, independent of their interaction with biological matrices. Any discernable reporting bias was thoroughly discussed to evaluate the source’s reliability, and methodological certainty was assessed based upon the article’s error reporting and repeatability of experiments. In this group a total of 156 publications were consulted, 126 were removed and 30 were used in the analysis.

Exposure

Studies sought for this theme focused on the presence of EPFRs in areas of human activity with a potential for inhalation. The exposure group conducted their search using the terms “EPFRs and exposure” OR “Environmentally Persistent Free Radicals and exposure” OR “EPFRs, exposure and PM2.5” across four databases: PubMed^®, Google Scholar, SciFinder, and Web of Science^™. All available years in each database were included in the search. The initial search results were then refined by incorporating the term “particulate matter” and removing duplicate and unrelated references using Web of Science^™. The retained articles were then screened using the following exclusion criteria: non-peer reviewed records and irrelevant topics, such as cell or animal toxicology, laboratory experiments, epidemiology, and other topics not related to exposure. The exposure group identified 317 potential articles related to their search terms, excluded 270 using the exclusion criteria, and retained 47 that were used in the study.

Animal toxicology

The animal toxicology group applied inclusion criteria that focused on full-text research articles investigating the toxicological effects/health risks resulting from EPFR exposures in animals. Search strings TS= (“environmentally persistent free radical*” OR ‘EPFR*’) AND TS= (‘animals’ OR ‘toxicity’ OR ‘health effect*’)” were used to identify exposure studies across all available years using three databases: PubMed^®, Google Scholar, and Web of Science^™. A few review articles suggesting their potential risks were included at this selection stage solely to identify any possible missed original research articles. Investigations focusing on formation or physicochemical properties of EPFRs without direct relevance to human exposures were excluded. Selected articles underwent a stringent secondary screening to ensure they were well-designed, with a clear presentation of material and methods, data, and results. In addition, the study groups (e.g., control vs exposed) used in the research needed to be clearly defined, with the well-characterized pollutants, dose, regimens, and routes of exposure relevant to ambient conditions. The literature search yielded 9527 articles from which 3444 duplicates were excluded. Of the remaining 6083 publications, 939, or 15.4%, met the rigorous criteria required for inclusion. The most significant findings from 47 selected publications are discussed in this article.

Epidemiology

The epidemiology group conducted two rounds of searches. In the first round, the topic string “TS= (‘environmentally persistent free radical*’ OR ‘EPFR*’) AND TS= (‘human health’ OR ‘health outcome’ OR ‘health effect*’)” was searched for all available years across three databases: Google Scholar, PubMed^® and Web of Science^™ (Figure 1). Over 2000 publications were identified through these broad searches: 64 from Web of Science^™, 678 from PubMed^®, and over 1700 from Google Scholar (screening was limited to the first 20 pages).

First, articles that did not mention EPFRs and some aspect of human health in the title or abstract were excluded. After merging articles from the three databases and removing duplicates, the remaining articles were further analyzed, focusing on those where EPFRs were the main stressors and contained certain types of human health-related studies (excluding cell studies or animal models). Publications that met all criteria outlined in the above processes were included. However, almost no article could be identified using these stringent search terms. To address this limitation, the search was broadened to include studies that measured some component of PM related to EPFRs, including the ability of the PM to produce ROS, and/or combustion-related PM using the three search engines. The broader search yielded over 10,000 results, including 212 from PubMed^®, 39 from Web of Science^™, and 200 records from the first 20 pages of Google Scholar. All papers meeting these eligibility criteria were then included. In total, 9 articles were selected by the group for analysis.

Results

Atmospheric chemistry and EPFRs

Principal findings related to the atmospheric chemistry of EPFRs are presented in Table 1. Generally, about 80% of the articles describing the atmospheric chemistry of EPFRs were based upon lab studies involving model pollutants and/or metal oxide catalysts. EPFRs were found in ambient air and soils (Chen et al. 2020b; Jia et al. 2023) and correlated with submicron particles such as redox-active transition metal oxide (TMO) particles and PM (Chen et al. 2019, 2020b). Studies indicate that EPFRs are formed through several common processes including combustion (in the post-flame and cool zone) and pyrolysis, which might occur from sources such as car engines, wildfires, and waste burning (Hwang et al. 2021). Various organic precursors, including 2,4-dichloro-1-naphthol, phenol, mono-chlorobenzene, 2-mono-chlorophenol, 1,2-dichlorobenzene, hydroquinone and catechol, were linked to formation of EPFRs on different transition metal oxides, such as CuO, Al₂O₃, ZnO, NiO, TiO₂, and Fe₂O₃ (Arangio et al. 2016; Hwang et al. 2021; Li et al. 2023; Lomnicki et al. 2009; Y. Wang et al. 2019; J. Zhao et al. 2022).

Table 1.

Principal characteristics of EPFR generated in laboratory settings.

Pollutants used for EPFR generation	Transition metal oxide	[EPFR] (spins/g)	g-value	Principal findings	Reference(s)
Hydroquinone, catechol, phenol, 2-chlorophenol, mono-chlorobenzene, and1,2-dichlorobenzene	Copper oxide nanoparticle	N/A	2.0011–2.0070	g-Value depends on temperature and precursor molecule.	Lomnicki et al. (2009).
Mono-chlorophenol, dichlorobenzene	Copper oxide nanoparticle	N/A	N/A	EPFRs in aqueous suspension redox cycle to form hydroxyl radicals.	Kelley et al. (2013).
2,4-Dichloro-1-naphthol	CuO, Al2O3, ZnO, NiO	1020 – 1022 spins/g	2.0029–2.0063	Metal oxide nanoparticles increased EPFR concentrations more than micrometer-sized particles. Radicals on metal surfaces exhibited half-lives from 68–108 d.	Yang et al. (2017).
Phenol	ZnO, CuO, Fe2O3, and TiO2 nanoparticles	1015 – 1018 spins/g	2.00315–2.00447	Oxidation potential of the metal cation influences its ability to catalyze EPFR formation. EPFR concentration and g value depend on type of metal oxide.	Sakr et al. (2021).
Catechol, hydroquinone, phenol, mono-chlorophenol, mono-chlorobenzene, dichlorobenzene	Transition metal oxides	1018 spins/g	2.003–2.004	EPFR half-life (up to months) depends on the adsorbate and the transition metal used in its generation.	Vejerano, Lomnicki, and Dellinger (2012)
Phenol	Iron (III)-exchanged calcium montmorillonite clay	7.5×1017 spins/g	~2.0033	Reduction of the metal center in a clay system was observed via x-ray analysis, confirming the role the metal center plays in EPFR formation.
benzo[a]anthracene, pyrene, and benzo[a]pyrene)	Fe, Cu, Ni, Co, Zn oxides loaded into montmorillonite clays	1016spins/g	2.0028–2.0040	The yields and lifetimes of EPFRs depend on parent PAH and the metal ion.	Jia et al. (2018).
Anthracene pyrene, benzo[a] pyrene, benzo[ghi] perylene, naphthoquinone, anthraquinone	No transition metal used. Particles generated by reaction with ozone	1018 radicals/g	2.001–2.0056	EPR spectra exhibit three radical types, two assigned as semiquinone species and one a PAH-derived, carbon-centered radical.	Borrowman et al. (2016).
Phenol	ZnO	N/A	2.0044	Only minor differences detected when nano ZnO particles are exposed to phenol at room temperature compared to 250°C	Patterson et al. (2017).
Hexane	hexane-generated soot	1.04–2.37×1019		There was no difference in degradation rates between wetted and non-wetted soot.	Runberg et al. (2020).
Soil system of phenol	Iron (III)-exchanged calcium montmorillionite	1×1017	~2.0034	Reduction of the metal center in a clay system confirmed pivotal role of metal center in EPFR formation.	Nwosu et al. (2016)
Mono-chlorophenol	Copper oxide	~1017	2.0036	Through its metal oxide, EPFRs exhibit a surface-catalyzed redox cycle that generates surface-bound, rather than free, hydroxyl radicals.	Khachatryan et al. (2011)., Khachatryan et al. (2011)

EPFRs were also detected in soil and air samples containing a larger variety of metals, including Mn, Pb, and V (Chen et al. 2018; Jia et al. 2023; J. Zhao et al. 2021). Depending upon the aromatic precursor and transition metal oxide, the concentration of radicals in both lab and environmental samples ranged from 10¹⁵ to 10²² spins/g (Arangio et al. 2016; Chen et al. 2018; Jia et al. 2023; Li et al. 2022, 2023; Morrison et al. 2022; Qian et al. 2020; P. Wang et al. 2018; Xiao et al. 2021; Y. Xu et al. 2021; Yang et al. 2017; J. Zhao et al. 2021, 2022). The oxidation potential of the metal cation and size of the metal oxide significantly influenced the generation of radicals on the particle surface (J. Zhao et al. 2022). However, transition metal oxides without an aromatic precursor do not produce EPFRs (Arangio et al. 2016). EPFRs exhibit significant longevity compared to typical radicals, with lifetimes ranging from days to months and years (Hwang et al. 2021; Li et al. 2022, 2023; Sly et al. 2019; J. Zhao et al. 2022) and positively correlate with standard reduction potential of associated metal. Similar long lifetimes of EPFRs and their dependence upon transition metal concentrations were reported for radicals from field studies, especially on road dust and wildfire charcoal samples (Li et al. 2023; Sigmund et al. 2021). Typical radicals such as the hydroxyl (.OH), peroxyl, and oxygen radicals exhibit extremely short lifetimes, in the order of milliseconds, and, therefore, are far less stable compared to EPFRs (Pryor 1986). Among the transition metal oxides studied, EPFRs formed on ZnO exhibited the longest half-lives (Li et al. 2023). UV radiation degrades adsorbed organics on the surface, transforming them into other organic compounds which may lead to additional EPFR formation (Feng et al. 2022). This was exemplified in the study of (Chen et al. 2018), where UV exposure produced secondary radicals and/or new EPFR species, with yields ranging from 15% to 60% of the original EPFR content, despite their relatively short lifetimes (30 min–1 hr).

EPFR formation is likely to increase with the rising frequency of catastrophic wildfire events. Elevated levels of PM were associated with increased total ROS levels and a significant rise in harmful hydroxyl radicals in air. For example, one study reported an approximate 50-fold elevation in hydroxyl radicals between “clean” and “megacity” samples (Chen et al. 2018). Xiao et al. (2021) found that EPFRs from biomass combustion, coal combustion, and airborne PM exhibited significant differences in particle size distributions, with the highest concentrations of EPFRs found in PM < 1.1 μm. EPFRs in the size range of PM < 1.1 μm accounted for more than 76% of the EPFRs in PM < 3.3 μm. This might explain the fact that PM < 1.1 emitted from biomass combustion is more dangerous to human health compared to PM_1–2 and PM_2–3.3 and are therefore deserving of more detailed study (Xiao et al. 2021). Ultrafine particulate matter (PM < 0.1 μm) possesses a higher level of EPFRs than coarse PM (2.5 μm < PM < 10 μm) and are highly stable in the environment, with a half-life of up to 4.5 years in road dust (Li et al. 2023).

Exposure to EPFRs

The primary route of human exposure to EPFRs is likely through inhalation of pollutants from both outdoor or indoor sources. Sources of inhaled EPFRs include atmospheric aerosol particles, coal fires, traffic and industrial emissions, fossil fuels, road dust, surface swabs, vehicle exhaust, and biomass combustion (Arangio et al. 2016; Chen et al. 2019; Guo et al. 2020; Y. Wang et al. 2019; J. Zhao et al. 2022). Where smoking is permitted, indoor air quality is markedly influenced by cigarette and tobacco smoke, which represent significant indoor sources of EPFRs (Dellinger et al. 2011; Nazaroff and Singer 2004; Repace and Lowrey 1980). These EPFRs are often associated with respirable particles – those with aerodynamic diameters typically ≤4 μm – that may penetrate deep into the lungs, where these particles may exert toxic effects. EPFR concentrations in these sources vary widely, ranging from 10¹⁷ spins/g to 10²³ spins/g, depending on the type, location of the pollutant source, and season (Arangio et al. 2016; Chen et al. 2018; Feng et al. 2022; Hwang et al. 2021; Li et al. 2023). It is worth noting that Chen et al. (2019) reported that the outdoor sources of EPFRs to which both human and plant species are exposed in Xian, China, vary by season. Outdoor coal combustion and traffic sources were the largest contributors to EPFR levels in the spring, while the major sources of EPFRs in the summer and autumn were traffic and biomass burning. Chen et al. (2019) further suggested that fossil fuel combustion is a primary source by which humans are exposed to EPFRs outdoors. Seasonal variations also affect EPFR concentrations; PM collected during the heating or winter season, when more biomass and coal is burned in urban areas of China, showed a higher concentration of EPFRs, suggesting a greater inhalation risk compared to the non-heating season (Jia et al. 2023). Because EPFRs in both indoor and outdoor air vary significantly by season, it is possible that the risk of human exposure to EPFRs also differs by season.

While published reports specifically determining EPFR exposure assessments were severely lacking, a few investigations examined levels of EPFRs in samples derived from areas in which human exposures are likely. Sly et al. (2019) noted findings from a community-based longitudinal study established to investigate the patterns of acquisition of common respiratory viruses and bacteria in the upper airways of children. These investigators described a detection of EPFRs in 99% of dust samples collected from household vacuum cleaners. Penetration of traffic-related air pollution (TRAP) into homes, along with indoor gas heating and cigarette smoking, was noted as significant sources of indoor EPFR exposure. Similarly, J. Zhao et al. (2021) reported that the use of crop residue and bitumite as residential cookstove fuels resulted in high levels of EPFRs associated with indoor PM_2.5, ranging from 10¹⁷ to 10²⁰ spins/g. Similarly, J. Zhao et al. (2022) found that EPFRs were associated with the indoor burning of biomass and coal using cook stoves in rural China. Both studies suggested the potential for human exposures to EPFRs generated from the incomplete combustion of solid fuels (i.e., using cook stoves) within indoor air. Finally, in a recent report from W. R. Lee et al. (2025), findings demonstrated the inextricable link between EPFR source and human exposure. Further, W. R. Lee et al. (2025) identified several household practices that were correlated with EPFR concentrations. These included construction type, exposure to TRAP, indoor combustion activities, ambient PM_2.5 levels, seasonal variation, cleaning practices, and ventilation. This study also reported that using an extractor fan while cooking consistently lowered the levels of EPFRs, which demonstrates the dependence of air quality upon household cooking practices (W. R. Lee et al. 2025). While studies that detail how humans are exposed to EPFRs are limited, additional research in this area appears warranted.

Animal toxicology

Both acute and chronic exposure to PM air pollution may lead to a wide range of diseases, including stroke, cardiovascular diseases, chronic obstructive pulmonary diseases, aggravated asthma, and lower respiratory infections (Figure 3) (Table 2). It is important to recognize that the interpretation of toxicological and epidemiological endpoints is inherently influenced by factors such as (1) duration and intensity of exposure, (2) route and dose of exposure, and (3) species or population studied. An important aspect of EPFR studies is their ability to generate ROS. The dominant types of ROS formed include hydroxyl and peroxyl radicals (Morrison et al. 2022; L. Wang et al. 2022). After inhalation of PM-containing EPFRs, ROS are generated through redox cycling and play a key role in the pathogenesis of the toxicological responses (Thevenot et al. 2013). Several investigators demonstrated that ROS generated from EPFRs contribute to endpoints related to cardiovascular and respiratory diseases and DNA damage (Burn and Varner 2015; Chuang et al. 2017; X. Wang et al. 2024). Using lab generated EPFRs, Kelley et al. (2013) using the observed the formation of hydroxyl radicals when EPFRs were suspended in either bronchoalveolar lavage fluid (BALF) or plasma, an indication that EPFRs possess the potential to generate ROS even in biological fluids and tissues. Arangio et al. (2016) reported a strong correlation between the EPFR concentration and ROS produced by PM, with fine particles containing greater quantities of EPFRs and producing higher levels of ROS than coarse particles. In other studies (Lakey et al. 2016), it was demonstrated that deposition of atmospheric particles containing redox active metals and organic aerosols in lung lining fluid produced ROS at levels considered characteristic of respiratory diseases. Further, EPFRs were found to oxidize biomolecules in-vitro, thus inducing oxidative stress and cellular toxicity (Balakrishna et al. 2009).

Figure 3.

Proposed health effect mechanisms and exposure routes. EPFR are generated during thermal processes including combustion processes from vehicle exhaust, fossil fuel burning, cigarette smoke and forest fire. They are bound to particles and once inhaled, they can deposit in the various locations within the lungs, depending on their size. Once they are in the lungs, they generate free radicals, increasing the oxidative stress. If small enough they can translocate into the blood. Pulmonary oxidative stress and systemic inflammation as a result of inhalation promote the development of various diseases including respiratory diseases, cardiovascular diseases, neurological disorders and metabolic syndrome.

Table 2.

Animal toxicology studies suggest significant negative effects of EPFR exposure on cardiovascular and respiratory disease progression in differing animal models and organisms.

Pollutant	Animal model	Experimental findings	Reference
Cardiovascular diseases
Lab generated EPFR	Healthy male Sprague Dawley rats	↓ Baseline cardiac function and ↑ systemic inflammation and oxidative stress. markers in the left ventricle.	Mahne et al. (2012).
Lab generated EPFR	Male C57BL/6 mice	↓ Vascular responsiveness associated with an altered pulmonary function.	Harmon et al. (2021).
Lab generated EPFR	Male Brown - Norway rats	↓ Cardiac function before and after ischemia/reperfusion injury.	Lord et al. (2011).
Lab generated EPFR	Male and female C57BL/6 mice	EFPR exposure induces endothelial dysfunction, ↓ vasorelaxation.	Aryal et al. (2024).
Respiratory diseases
Lab generated EPFR	BEAS-2B cells, mouse neonatal airway epithelial cells and C57BL/6 pups.	↑ Lysosomal membrane permeabilization, oxidative stress, and lipid peroxidation in BEAS-2B cells. ↑ Epithelial-to-mesenchymal transition in infant mice	Thevenot et al. (2013).
Lab generated EPFR	C57BL/6NHsd and IL10 knock out mice (B6.129P2-I/10tm1Cgn/J)	↑ Tregs and IL-10 ↓ Suppressed adaptive T cell responses and ↑ influenza disease severity.	Jaligama et al. (2017).
Lab generated EPFR	Mouse Model of Chronic Alcohol Ingestion (C57BL/6 mice)	↑ Severity in the alcoholic lung, oxidative stress, airway collagen content and compromised alveolar integrity	Thevenot et al. (2013).
Lab generated EPFR	Mouse model of asthma (BALB/c mice)	EPFR-containing PM induced a Th17-biased phenotype in lung, accompanied by significant pulmonary neutrophilia.	Wang et al., 2011.
Lab generated EPFR	Human bronchial epithelial (HBE) cells and airways of mouse with impaired autophagy.	PM was endocytosed by cells and triggered endocytosis and autophagy. Autophagy is essential for PM-induced activation of NFKB1 and AP-1, and thus, airway inflammation and mucus hyperproduction.	Chen et al. (2016).
Lab generated EPFR	Neonatal mouse model of flu infection (C57/BL6 mice)	↑ Postnatal asthma development, eosinophilia and pulmonary Th2 responses.	Wang et al. 2013.
coal fly ashes containing quartz and iron	rat lung epithelial cells	↑ Hydroxyl radical, oxidative DNA damage.	Maanen et al. (1999).
Infectious diseases
Lab generated EPFR (DCB230)	Neonatal mice (< seven days of age)	↑ Severity of influenza virus infection and oxidative stress.	G. I. Lee et al. (2014).
Lab generated EPFR (DCB230)	Neonatal mice (< seven days of age)	↑ Severity of influenza virus infection, ↓ IL22.	Kumar et al. (2021).
Lab generated EPFR (DCB230)	Nasal epithelial cells	↑ SARS-CoV-2 replication, pro-inflammatory cytokines.	Yamamoto et al. (2023).
Other Toxic effects
Clean ambient air plus traffic related ammonium sulfate particles	Male and female Sprague Dawley rats	↑ Stillbirths, ↓ birth weight and gestation period, ↑ plasma glucose and fatty acids, lipid in liver and ↓ endothelial function.	Wu et al. (2019).
Biochar	Caenorhabditis elegans	Free radicals in biochar acts as a weak neurotoxin.	Lieke et al. (2018).
Biochar	Corn, wheat and rice seedlings	Inhibit roots and shoots of corn, wheat and rice seedlings.	Liao et al. (2014).

Earlier studies with neonatal Brown-Norway rats demonstrated that inhalation of PM containing EPFRs for 1 week reduced lung function including increased airway hyperreactivity, reduced compliance and elevated hysteresis while increasing systemic and pulmonary oxidative stress, but importantly, that PM devoid of EPFRs failed to induce pulmonary oxidative stress or change in lung compliance (Balakrishna et al. 2011). Follow-on studies showed that neonatal Brown-Norway rats exposed to PM containing EPFRs exhibited airway hyperreactivity associated with an altered expression of key proteins linked to oxidative stress and immune regulation, suggesting a potential mechanism by which EPFRs may contribute to asthma development or exacerbation in early life (Balakrishna et al. 2011). In addition, in an asthma model, EPFR exposure exacerbated pulmonary inflammation and altered the immune response, inducing pulmonary neutrophilia and T cell production of IL17 (Balakrishna et al. 2011).

PM-containing EPFRs was also found to exacerbate the severity of respiratory tract viral infections including influenza virus and SARs-CoV-2 in neonatal mouse and human ALI culture models (Kumar et al. 2021; G. I. Lee et al. 2014; Yamamoto et al. 2023). EPFR exposure might further enhance the severity of flu infection in C57BL/6 neonatal mice by inhibiting the production of interleukin-22 (IL22), a key protein for lung repair. PM containing EPFRs also disrupted the lung microbiome and reduced levels of indole, a metabolite that normally stimulates IL22 production (Kumar et al. 2021). In another study (Yamamoto et al. 2023) examining SARs-CoV-2 infections demonstrated that EPFR exposure prior to infection increased viral replication, inflammation, and cellular damage, while decreasing antioxidant defenses in well-differentiated human nasal epithelium. These findings suggest that air pollution-induced oxidative stress may contribute to the link between poor air quality and worsened COVID-19 outcomes. Several investigators also demonstrated that EPFRs inhibit the activity of cytochrome P450 enzymes in the livers of rabbits and rats (Reed et al. 2014, 2015a, 2015b). Cytochrome P450 enzymes are crucial for metabolizing foreign compounds. This disruption might interfere with the enzymic ability to metabolize foreign substances.

A growing body of evidence also suggests that EPFR inhalation induces cardiovascular dysfunction in lab animals. Recently, Aryal et al. (2024) using adult male C57BL6 mice demonstrated that EPFR exposure induced a dose-dependent reduction in lung function and promoted vascular dysfunction distal to the lungs. Using male and female mice selectively deficient in the aryl hydrocarbon receptor (AhR) in alveolar Type II cells, Aryal et al. (2024) further noted that vascular dysfunction distal to the lungs is mediated by activation of the AhR expressed at the air-blood interface and stimulation of its downstream signaling pathway. AhR activation regulates the transcription of genes associated with vascular physiological signaling, inflammation, and detoxification pathways (Bahman et al. 2024). In adult rats exposed to EPFRs, cardiac impairments such as reduced cardiac function and elevated pulmonary artery pressure were observed. These effects were evident both at baseline and following ischemia/reperfusion, suggesting that EPFR exposure may contribute to cardiovascular risk, even in otherwise healthy animals (Lord et al. 2011; Mahne et al. 2012).

While it is unclear whether EPFRs cross the blood-brain barrier, two studies suggested that chronic PM_2.5 exposure triggers oxidative stress and inflammation in the brain, and exacerbates cognitive decline in adults (Calderón-Garcidueñas et al. 2008; Younan et al. 2020). EPFRs in biochar were also found to induce neurotoxicity in Caenorhabditis elegans, implying that EPFRs in biochar may act as neurotoxicants to invertebrate soil organisms, with potential implications for soil health and ecological balance (Lieke et al. 2018).

In summary, studies in lab animals support that EPFRs present in PM contribute to oxidative stress and inflammation, leading to respiratory, cardiovascular, and potentially neurological dysfunction. Given their ability to persist in the environment and their potential to alter numerous biological pathways, EPFRs may represent a critical and emerging factor in air pollution-related health risks including asthma exacerbation, impaired lung function, cardiovascular complications, and worsened outcomes of viral infections such as influenza and COVID-19.

Epidemiology

An initial screen for epidemiological evidence of EPFR-associated health risks produced no reports directly examining the link between EPFR exposure and health outcomes. However, a substantial body of literature documents the detrimental effects of combustion-derived PM on human health and indicates oxidative stress as a major mechanism. Of these, 6 articles specifically highlighted the presence and impact of free radicals within air pollution. In a study conducted in Beijing, where air pollution, particularly fine and ultrafine PM, is a major concern, Langrish et al. (2012) investigated the health benefits of reducing personal exposure to air pollution in patients with coronary heart disease. The study involved 98 patients who walked a predefined route while wearing a highly efficient face mask on one occasion and not wearing it on another. Data demonstrated that wearing the face mask led to several improvements in cardiovascular health indices, such as reduced myocardial ischemia, lower blood pressure, and increased heart rate variability. While Langrish et al. (2012) highlighted the presence of organic carbon, polycyclic aromatic hydrocarbons, and the high oxidative potential of the PM, these investigators did not measure EPFRs in the PM or definitively establish a causal relationship between these radicals and the observed health improvements.

X. Zhang et al. (2016) found a significant increase in airway oxidative stress and inflammation in individuals exposed to air pollutants, particularly ultrafine particles and transition metals. However, while X. Zhang et al. (2016) measured the oxidative potential of the PM collected, these investigators did not find a significant association between this parameter and airway inflammation, indicating that the specific characteristics of ultrafine particles, rather than their overall oxidative potential per se may be more important in predicting pulmonary inflammation. Similarly, Steenhof et al. (2013) examined 31 volunteers exposed to ambient air pollution at sites in the Netherlands, and air pollution levels and pro-inflammatory biomarkers were collected. Steenhof et al. (2013) postulated that the particle’s oxidative potential might be most strongly correlated with biomarkers for acute nasal airway inflammation. Steenhof et al. (2013) concluded that in nasal lavage samples, there were no consistent correlations between PM mass concentration nor oxidative potential and inflammatory biomarkers such as IL-6, IL-8 and lactoferrin. Rather, associations were detected between nasal inflammatory markers and other PM characteristics, including organic carbon, endotoxin, and NO₂ (Steenhof et al. 2013).

Several investigators employed the only method electron paramagnetic resonance spectroscopy to enable direct, specific, and sensitive detection of radicals and their intermediates (Runberg et al. 2020; Y. Wang et al. 2022). Sly et al. (2019) measured EPFRs in dust samples collected from the homes of children with persistent wheeze. These investigators found that EPFRs were present in almost all dust samples, and higher levels of EPFRs were associated with an elevated likelihood of children reporting persistent wheeze. Data further suggested that EPFRs, potentially originating from traffic-related air pollution and/or indoor combustion sources, may contribute to respiratory problems in children. Sly et al. (2019) proposed that boosting antioxidant defenses may help protect children from the adverse effects of air pollution.

Finally, in an early study by N. Dalal et al. (1989), EPR analysis demonstrated that freshly ground anthracite coal generates more reactive free radicals and enhanced toxicity compared to bituminous coal. These findings, in conjunction with their detection of coal-based free radicals in the lung tissue of autopsied coal miners, suggested that these radicals may contribute to development and/or severity of coal workers’ pneumoconiosis (CWP). In a follow-on study, N. S. Dalal et al. (1991) autopsied lung tissue from coal miners and noted that stable coal radicals were significantly higher in those with longer mining tenure, and these individuals exhibited more severe CWP.

Discussion

EPFRs are emerging as a unique component of combustion-derived PM that display the ability to redox cycling, leading to the formation of ROS and induction of oxidative stress in biological systems (Balakrishna et al. 2009; Kelley et al. 2013). Combustion-generated PM from both natural and human activities, including wildfires, contribute to EPFR formation (Table 3). An interesting finding by (Y. Xu et al. 2020) was that EPFRs in PM_2.5 and cigarette tar exhibited similar pulmonary toxicities as in humans. Further, Gehling and Dellinger (2013) estimated that the risk of EPFR exposure from non-extreme air quality in the U.S. was equivalent to smoking 0.4–0.9 cigarettes per day. During the combustion process, the interaction between organic precursors and transition metals present on the surface of the particle form EPFRs that are persistent with long half-lives (Khachatryan et al. 2014; Vejerano, Lomnicki, and Dellinger 2012). A significant correlation was observed between EPFR concentrations and chemical constituents of the pollutant-particle systems including metal concentrations, organic components and concentrations, and PM size (Table 1). EPFR concentrations in environmental air samples range from 10¹⁵ to 10²² spin/g (Table 3).

Table 3.

EPFR concentration and g-value depend on their source and pollutant size.

Source of Pollutant	Pollutant	Exposure type	Levels pollutants	g-value	Principal findings	Reference(s)
Atmosphere
	PM(0–56 nm)	Outdoor	1.60 × 1013 −3.11 × 1022	2.0016–2.0052	EPFRs positively correlated to metal concentration. Radical concentration and g-values depend on particle size and season.	Arangio et al. (2016), Runberg et al. (2020), X. Wang et al. (2024), Y. Xu et al. (2020), Jia et al. (2023), Chen, Sun et al. 2020b, Chen et al. (2019), Guo et al. (2020), J. Zhao et al. (2022).
Biomass Combustion
	PM(1.1–3.3)	Outdoor/Indoor	0.67 × 1015 spins/m3. 6.11 xl 019 spins/g	2.0029–2.0039	Secondary radicals formed from EPFRs are 20–30% of original concentration.	Guo, Wang et al. (2023), Xiao et al. (2021), J. Zhao et al. (2022), Yang et al. (2017), Z. Zhao et al. (2024), J. Zhao et al. (2022).
Coal combustion
	PM(10–1um)	Outdoor	3.06 × 1019– 1.52 × 10 22 spins/g),	2.00316–2.00371	Coal combustion particulates are more hazardous to humans than those from biomass combustion.	Yang et al. (2017), Yang et al. (2017), Y. Wang et al. (2019), J. Zhao et al. (2022), J. Zhao et al. (2022).
Automobile exhaust
	PM(10–1um) PM2.5	Outdoor	76–154 μg/m3 −3.0 × 1022 spins/g),	2.0028–2.0040.	EPFRs correlated with vehicle exhaust markers (CO, NO, and EC) and non-exhaust markers (Fe and Cu)	Yang et al. (2017) Chen et al. (2018) Hwang et al. (2021), Y. Wang et al. (2019).
Crop residue, firewood, and coal	PM2.5	Outdoor	1.79 – 5.63 ± 2.53 1020 spins/kg	2.003–2.004	EPFR concentration derived from combustion of crop residues is significantly higher than that from firewood.	Zhao et al., 2020, J. Zhao et al. (2021)
Dust
	PM10–PM2.5	Indoor/outdoor	4 × 1014 −1.41 × 1020 spins/g.	2.0024–2.0036	Wheeze status in children associated with household dust containing EPFRs. PM10 fraction is primarily deposited in respiratory tract and pulmonary regions.	Sly et al. (2019). Filipi et al. (2022). Feng et al. (2022). Li et al. (2023).
Surface swabs
	PM	Indoor/outdoor	0.2–7.1 × 1011spins/μg		Asian sandstorms can increase the risk of exposure to EPFRs.	Fillipi et al., 2022.
Asian Storm dust
	PM2.5	Outdoor	3.7×17–1.1×1018 spins/g	~2.0031	The risk of exposure to EPFRs is enhanced by the Asian sandstorms.	Chen et al. (2018).
Soil and sediment
	Penta-chlorophenol	outdoor	10^17 spins/g	2.00275–2.0036	EPFRs detected in contaminated soil and sediment decades after their deposition.	Dela Cruz et al. (2011).
Tree leaves
	PM2.5	Outdoor	7.5× 1016– 4.5 × 1019 spins/g.	2.0028–2.0037	Vehicle emissions, especially from heavy-duty vehicles, are the main sources of EPFRs accumulated on tree leaves.	Wang et al. (2019).
Wildfire
	Charcoal PM(1–10)	Outdoor	~1019 spins/g EPFR	2.0021–2.0036	Forest fires produce significantly higher concentration of EPFRs compared to airborne PM and soils not derived from forest fires.	Sigmund et al. (2021), Fang et al. (2023).

The concentration of EPFRs significantly influences ROS formation in biological systems, potentially impacting their toxicological impact on human health. In addition, the combined effects of EPFRs with other pollutants in PM such as pathogens, other organic pollutants, and metals might lead to enhanced toxicological effects (Kumar et al. 2021; G. I. Lee et al. 2014; Yamamoto et al. 2023). It is also important to note that inhaled particles interact with pulmonary defense mechanisms such as alveolar macrophages. Alveolar macrophages attempt to clear them via phagocytosis, but particles that persist may become more harmful over time. For example, particles initially low in iron (Fe) may accumulate Fe in the lung environment and thereby gain the potential to catalyze Fentontype reactions, producing ROS and stimulating oxidative stress (Beck-Speier et al. 2009; Recalcati and Cairo 2021). Particle size further modulates fate (1) as smaller particles penetrate deeper into the respiratory tract, (2) are more likely to reach the alveolar region, and (3) experience slower clearance than larger particles, altering retention time and influencing toxicity. Impairment of clearance mechanisms, such as occurring in dust overloading or macrophage dysfunction, might exacerbate these effects and contribute to sustained oxidative and inflammatory responses (Borm, Cassee, and Oberdörster 2015; Lazaridis 2023; Morrow 1992).

Epidemiological studies clearly demonstrate higher morbidity and mortality rates linked to PM exposure in areas near industrial sites, heavy traffic, and indoor combustion sources (Table 4). Further, animal studies suggested that EPFR-associated PM induce oxidative stress, reducing lung, cardiovascular and cardiac, metabolic and neurologic functions (Table 2). In addition, EPFRs were found to affect the (1) morphology and biochemical composition of agricultural crops (Baltrėnaitė-Gedienė, Lomnicki, and Guo 2022); (2) are toxic to soil microbial communities (Y. Zhang et al. 2019); and (3) adversely harm terrestrial invertebrates (Zhu et al. 2022), emphazing their broader ecotoxicological impact. Given the ubiquity of EPFR precursors in common combustion and combustion-like processes, prolonged lifetimes, and ability to generate ROS, EPFRs may pose a significant adverse risk to human health. Future studies focused on elucidating the toxicological mechanisms following EPFR exposure, characterizing and quantifying EPFR types and levels in PM, as well as comprehensive epidemiological investigations, may help to enhance our understanding of their health risks.

Table 4.

Studies show significant negative effects of free-radical producing PM_2.5 on human health.

Exposure type	Air pollutants	Study type/Subjects	Results	Reference(s)
Urban airborne PM	Personal pollution exposure, PM components, metals, nitrate and sulfate. oxidative potential of PM assessed using EPR.	Patients with coronary heart disease with and without highly efficient face mask.	↓ PM exposure was associated with consistent improvements in myocardial ischemia, exercise-related increases in blood pressure, and heart rate variability in patients with coronary heart disease.	Langrish et al. (2012).
Traffic-related air pollutants, ultrafine particles and transition metals	NOx, black carbon, PM, O3, PAHs, transition metals (V, Cr, Mn, Ni, Cu, and Fe), oxidative potential.	97 Elderly, nonsmoking adults (age ≥65)	↑ Airway oxidative stress and inflammation correlated with air pollutants, ultrafine particles, and transition metals. ↑ But nonsignificant associations observed for the oxidative potential of PM. Strongest links observed with ultrafine PM.	Zhang et al. (2017)
Underground train, farms, urban sites	PM, elemental carbon, organic carbon, oxidative potential, endotoxins, O3, and NO2	31 volunteers	↑ Nasal inflammatory markers for other PM characteristics, specifically OC, endotoxin and NO2.	Steenhof et al. (2013).
No specific exposure	EPFRs in household dust	Children under two years of age	↑ EPFRs in house dust samples; EPFR characteristics associated with wheeze status.	Sly et al. (2019).
Coal mine, coal dust	Stable coal radicals	Coal miners with and without pneumoconiosis and history of cigarette smoking	Possible correlation between stable coal radicals and coal workers’ pneumoconiosis.	N. S. Dalal et al. (1991).
Coal mine, coal dust	Freshly ground and aged anthracite and bituminous coal samples	Autopsy specimens from coal miners	Coal-based free radicals detected in lung tissue of autopsied coal miners, suggestive of persistent reactivity by the embedded coal dust leading to the progressive disease.	N. Dalal et al. (1989).

Conclusion

Environmentally persistent free radicals (EPFRs) represent a distinct and understudied class of pollutants within combustion-derived PM. Their unique stability, prolonged environmental persistence, and capacity for continuous redox cycling make them potent sources of ROS, with the potential to drive oxidative stress and related pathological processes. Evidence from atmospheric chemistry, exposure assessments, and animal toxicology studies demonstrated that EPFRs are (1) widespread in both indoor and outdoor environments, (2) originate from diverse combustion sources, and (3) linked to adverse respiratory, cardiovascular, and metabolic outcomes in experimental models. However, epidemiological evidence directly connecting EPFR exposure to specific adverse human health outcomes remains scarce, underscoring a critical knowledge gap. To address these gaps and enable accurate risk assessment, future research needs to prioritize well-designed population-based epidemiological studies in communities near combustion or waste treatment sources, incorporating longitudinal exposure assessments and health outcome tracking. Development and deployment of real-time EPFR monitoring technologies, coupled with detailed exposure characterization and dose–response modeling, are essential for establishing causal links between EPFR levels and disease endpoints. Integrating these human studies with mechanistic and translational toxicology research might provide comprehensive data necessary to guide evidence-based regulatory policies and effective mitigation strategies to protect human and environmental health.

Abbreviations

AhR –

Aryl Hydrocarbon Receptor

BALF

Bronchoalveolar Lavage Fluid

COPD

Chronic Obstructive Pulmonary Disease

COVID-19

Coronavirus Disease 2019

CWP

Coal Workers’ Pneumoconiosis

EPFR

Environmentally Persistent Free Radical

EPR

Electron Paramagnetic Resonance

IL22

Interleukin-22

PAHs

Polycyclic Aromatic Hydrocarbons

PFAS

Per- and Polyfluoroalkyl Substances

PM

Particulate Matter

PRISMA

Preferred Reporting Items for Systematic Reviews and Meta-Analyses

ROS

Reactive Oxygen Species

SARS-CoV-2

Severe Acute Respiratory Syndrome Coronavirus 2

TMO

Transition Metal Oxide

TRAP

Traffic-Related Air Pollution

UV

Ultraviolet

Data availability statement

Data will be made available on request.